X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=datapath%2Fdatapath.c;h=f990b738a3b72f103f21b304c56352f0fdeed91f;hb=72e8bf28bb38e8816435c64859fb350215b6a9e6;hp=1ea5d1e9a7c2d850cf989b34c6a5f3e7b53d25ac;hpb=6161d3fd928edf7016abae60f549a135a2f83f09;p=sliver-openvswitch.git diff --git a/datapath/datapath.c b/datapath/datapath.c index 1ea5d1e9a..f990b738a 100644 --- a/datapath/datapath.c +++ b/datapath/datapath.c @@ -1,5 +1,5 @@ /* - * Copyright (c) 2007-2011 Nicira Networks. + * Copyright (c) 2007-2012 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 @@ -39,7 +39,6 @@ #include #include #include -#include #include #include #include @@ -50,21 +49,30 @@ #include #include #include +#include +#include #include "checksum.h" #include "datapath.h" #include "flow.h" +#include "genl_exec.h" #include "vlan.h" #include "tunnel.h" #include "vport-internal_dev.h" #if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,18) || \ - LINUX_VERSION_CODE > KERNEL_VERSION(3,2,0) -#error Kernels before 2.6.18 or after 3.2 are not supported by this version of Open vSwitch. + LINUX_VERSION_CODE >= KERNEL_VERSION(3,7,0) +#error Kernels before 2.6.18 or after 3.6 are not supported by this version of Open vSwitch. #endif -int (*dp_ioctl_hook)(struct net_device *dev, struct ifreq *rq, int cmd); -EXPORT_SYMBOL(dp_ioctl_hook); +#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; + +int (*ovs_dp_ioctl_hook)(struct net_device *dev, struct ifreq *rq, int cmd); +EXPORT_SYMBOL(ovs_dp_ioctl_hook); /** * DOC: Locking: @@ -83,27 +91,23 @@ EXPORT_SYMBOL(dp_ioctl_hook); * each other. */ -/* Global list of datapaths to enable dumping them all out. - * Protected by genl_mutex. - */ -static LIST_HEAD(dps); - static struct vport *new_vport(const struct vport_parms *); -static int queue_gso_packets(int dp_ifindex, struct sk_buff *, +static int queue_gso_packets(struct net *, int dp_ifindex, struct sk_buff *, const struct dp_upcall_info *); -static int queue_userspace_packet(int dp_ifindex, struct sk_buff *, +static int queue_userspace_packet(struct net *, int dp_ifindex, + struct sk_buff *, const struct dp_upcall_info *); /* Must be called with rcu_read_lock, genl_mutex, or RTNL lock. */ -static struct datapath *get_dp(int dp_ifindex) +static struct datapath *get_dp(struct net *net, int dp_ifindex) { struct datapath *dp = NULL; struct net_device *dev; rcu_read_lock(); - dev = dev_get_by_index_rcu(&init_net, dp_ifindex); + dev = dev_get_by_index_rcu(net, dp_ifindex); if (dev) { - struct vport *vport = internal_dev_get_vport(dev); + struct vport *vport = ovs_internal_dev_get_vport(dev); if (vport) dp = vport->dp; } @@ -112,22 +116,10 @@ static struct datapath *get_dp(int dp_ifindex) return dp; } -/* Must be called with genl_mutex. */ -static struct flow_table *get_table_protected(struct datapath *dp) -{ - return rcu_dereference_protected(dp->table, lockdep_genl_is_held()); -} - /* Must be called with rcu_read_lock or RTNL lock. */ -static struct vport *get_vport_protected(struct datapath *dp, u16 port_no) +const char *ovs_dp_name(const struct datapath *dp) { - return rcu_dereference_rtnl(dp->ports[port_no]); -} - -/* Must be called with rcu_read_lock or RTNL lock. */ -const char *dp_name(const struct datapath *dp) -{ - struct vport *vport = rcu_dereference_rtnl(dp->ports[OVSP_LOCAL]); + struct vport *vport = ovs_vport_rtnl_rcu(dp, OVSP_LOCAL); return vport->ops->get_name(vport); } @@ -138,7 +130,7 @@ static int get_dpifindex(struct datapath *dp) rcu_read_lock(); - local = get_vport_protected(dp, OVSP_LOCAL); + local = ovs_vport_rcu(dp, OVSP_LOCAL); if (local) ifindex = local->ops->get_ifindex(local); else @@ -183,17 +175,17 @@ static int dp_fill_ifinfo(struct sk_buff *skb, hdr->ifi_flags = port->ops->get_dev_flags(port); hdr->ifi_change = 0; - NLA_PUT_STRING(skb, IFLA_IFNAME, port->ops->get_name(port)); - NLA_PUT_U32(skb, IFLA_MASTER, get_dpifindex(dp)); - NLA_PUT_U32(skb, IFLA_MTU, port->ops->get_mtu(port)); + if (nla_put_string(skb, IFLA_IFNAME, port->ops->get_name(port)) || + nla_put_u32(skb, IFLA_MASTER, get_dpifindex(dp)) || + nla_put_u32(skb, IFLA_MTU, port->ops->get_mtu(port)) || #ifdef IFLA_OPERSTATE - NLA_PUT_U8(skb, IFLA_OPERSTATE, - port->ops->is_running(port) - ? port->ops->get_operstate(port) - : IF_OPER_DOWN); + nla_put_u8(skb, IFLA_OPERSTATE, + port->ops->is_running(port) ? + port->ops->get_operstate(port) : + IF_OPER_DOWN) || #endif - - NLA_PUT(skb, IFLA_ADDRESS, ETH_ALEN, port->ops->get_addr(port)); + nla_put(skb, IFLA_ADDRESS, ETH_ALEN, port->ops->get_addr(port))) + goto nla_put_failure; return nlmsg_end(skb, nlh); @@ -225,11 +217,11 @@ static void dp_ifinfo_notify(int event, struct vport *port) } } - rtnl_notify(skb, &init_net, 0, RTNLGRP_LINK, NULL, GFP_KERNEL); + rtnl_notify(skb, ovs_dp_get_net(port->dp), 0, RTNLGRP_LINK, NULL, GFP_KERNEL); return; err: - rtnl_set_sk_err(&init_net, RTNLGRP_LINK, err); + rtnl_set_sk_err(ovs_dp_get_net(port->dp), RTNLGRP_LINK, err); out: kfree_skb(skb); } @@ -248,48 +240,68 @@ static void destroy_dp_rcu(struct rcu_head *rcu) { struct datapath *dp = container_of(rcu, struct datapath, rcu); - flow_tbl_destroy((__force struct flow_table *)dp->table); + ovs_flow_tbl_destroy((__force struct flow_table *)dp->table); free_percpu(dp->stats_percpu); + release_net(ovs_dp_get_net(dp)); + kfree(dp->ports); kobject_put(&dp->ifobj); } +static struct hlist_head *vport_hash_bucket(const struct datapath *dp, + u16 port_no) +{ + return &dp->ports[port_no & (DP_VPORT_HASH_BUCKETS - 1)]; +} + +struct vport *ovs_lookup_vport(const struct datapath *dp, u16 port_no) +{ + struct vport *vport; + struct hlist_node *n; + struct hlist_head *head; + + head = vport_hash_bucket(dp, port_no); + hlist_for_each_entry_rcu(vport, n, head, dp_hash_node) { + if (vport->port_no == port_no) + return vport; + } + return NULL; +} + /* Called with RTNL lock and genl_lock. */ static struct vport *new_vport(const struct vport_parms *parms) { struct vport *vport; - vport = vport_add(parms); + vport = ovs_vport_add(parms); if (!IS_ERR(vport)) { struct datapath *dp = parms->dp; + struct hlist_head *head = vport_hash_bucket(dp, vport->port_no); - rcu_assign_pointer(dp->ports[parms->port_no], vport); - list_add(&vport->node, &dp->port_list); - + hlist_add_head_rcu(&vport->dp_hash_node, head); dp_ifinfo_notify(RTM_NEWLINK, vport); } - return vport; } /* Called with RTNL lock. */ -void dp_detach_port(struct vport *p) +void ovs_dp_detach_port(struct vport *p) { ASSERT_RTNL(); if (p->port_no != OVSP_LOCAL) - dp_sysfs_del_if(p); + ovs_dp_sysfs_del_if(p); + dp_ifinfo_notify(RTM_DELLINK, p); /* First drop references to device. */ - list_del(&p->node); - rcu_assign_pointer(p->dp->ports[p->port_no], NULL); + hlist_del_rcu(&p->dp_hash_node); /* Then destroy it. */ - vport_del(p); + ovs_vport_del(p); } /* Must be called with rcu_read_lock. */ -void dp_process_received_packet(struct vport *p, struct sk_buff *skb) +void ovs_dp_process_received_packet(struct vport *p, struct sk_buff *skb) { struct datapath *dp = p->dp; struct sw_flow *flow; @@ -297,22 +309,22 @@ void dp_process_received_packet(struct vport *p, struct sk_buff *skb) u64 *stats_counter; int error; - stats = per_cpu_ptr(dp->stats_percpu, smp_processor_id()); + stats = this_cpu_ptr(dp->stats_percpu); if (!OVS_CB(skb)->flow) { struct sw_flow_key key; int key_len; /* Extract flow from 'skb' into 'key'. */ - error = flow_extract(skb, p->port_no, &key, &key_len); + error = ovs_flow_extract(skb, p->port_no, &key, &key_len); if (unlikely(error)) { kfree_skb(skb); return; } /* Look up flow. */ - flow = flow_tbl_lookup(rcu_dereference(dp->table), - &key, key_len); + flow = ovs_flow_tbl_lookup(rcu_dereference(dp->table), + &key, key_len); if (unlikely(!flow)) { struct dp_upcall_info upcall; @@ -320,7 +332,7 @@ void dp_process_received_packet(struct vport *p, struct sk_buff *skb) upcall.key = &key; upcall.userdata = NULL; upcall.pid = p->upcall_pid; - dp_upcall(dp, skb, &upcall); + ovs_dp_upcall(dp, skb, &upcall); consume_skb(skb); stats_counter = &stats->n_missed; goto out; @@ -330,15 +342,14 @@ void dp_process_received_packet(struct vport *p, struct sk_buff *skb) } stats_counter = &stats->n_hit; - flow_used(OVS_CB(skb)->flow, skb); - execute_actions(dp, skb); + ovs_flow_used(OVS_CB(skb)->flow, skb); + ovs_execute_actions(dp, skb); out: /* Update datapath statistics. */ - - write_seqcount_begin(&stats->seqlock); + u64_stats_update_begin(&stats->sync); (*stats_counter)++; - write_seqcount_end(&stats->seqlock); + u64_stats_update_end(&stats->sync); } static struct genl_family dp_packet_genl_family = { @@ -346,11 +357,12 @@ static struct genl_family dp_packet_genl_family = { .hdrsize = sizeof(struct ovs_header), .name = OVS_PACKET_FAMILY, .version = OVS_PACKET_VERSION, - .maxattr = OVS_PACKET_ATTR_MAX + .maxattr = OVS_PACKET_ATTR_MAX, + SET_NETNSOK }; -int dp_upcall(struct datapath *dp, struct sk_buff *skb, - const struct dp_upcall_info *upcall_info) +int ovs_dp_upcall(struct datapath *dp, struct sk_buff *skb, + const struct dp_upcall_info *upcall_info) { struct dp_stats_percpu *stats; int dp_ifindex; @@ -370,46 +382,48 @@ int dp_upcall(struct datapath *dp, struct sk_buff *skb, forward_ip_summed(skb, true); if (!skb_is_gso(skb)) - err = queue_userspace_packet(dp_ifindex, skb, upcall_info); + err = queue_userspace_packet(ovs_dp_get_net(dp), dp_ifindex, skb, upcall_info); else - err = queue_gso_packets(dp_ifindex, skb, upcall_info); + err = queue_gso_packets(ovs_dp_get_net(dp), dp_ifindex, skb, upcall_info); if (err) goto err; return 0; err: - stats = per_cpu_ptr(dp->stats_percpu, smp_processor_id()); + stats = this_cpu_ptr(dp->stats_percpu); - write_seqcount_begin(&stats->seqlock); + u64_stats_update_begin(&stats->sync); stats->n_lost++; - write_seqcount_end(&stats->seqlock); + u64_stats_update_end(&stats->sync); return err; } -static int queue_gso_packets(int dp_ifindex, struct sk_buff *skb, +static int queue_gso_packets(struct net *net, int dp_ifindex, + struct sk_buff *skb, 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); - if (IS_ERR(skb)) - return PTR_ERR(skb); + if (IS_ERR(segs)) + return PTR_ERR(segs); /* Queue all of the segments. */ skb = segs; do { - err = queue_userspace_packet(dp_ifindex, skb, upcall_info); + err = queue_userspace_packet(net, dp_ifindex, skb, upcall_info); if (err) break; - if (skb == segs && skb_shinfo(skb)->gso_type & SKB_GSO_UDP) { - /* The initial flow key extracted by flow_extract() in - * this case is for a first fragment, so we need to + 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; @@ -433,7 +447,8 @@ static int queue_gso_packets(int dp_ifindex, struct sk_buff *skb, return err; } -static int queue_userspace_packet(int dp_ifindex, struct sk_buff *skb, +static int queue_userspace_packet(struct net *net, int dp_ifindex, + struct sk_buff *skb, const struct dp_upcall_info *upcall_info) { struct ovs_header *upcall; @@ -477,7 +492,7 @@ static int queue_userspace_packet(int dp_ifindex, struct sk_buff *skb, upcall->dp_ifindex = dp_ifindex; nla = nla_nest_start(user_skb, OVS_PACKET_ATTR_KEY); - flow_to_nlattrs(upcall_info->key, user_skb); + ovs_flow_to_nlattrs(upcall_info->key, user_skb); nla_nest_end(user_skb, nla); if (upcall_info->userdata) @@ -488,7 +503,7 @@ static int queue_userspace_packet(int dp_ifindex, struct sk_buff *skb, skb_copy_and_csum_dev(skb, nla_data(nla)); - err = genlmsg_unicast(&init_net, user_skb, upcall_info->pid); + err = genlmsg_unicast(net, user_skb, upcall_info->pid); out: kfree_skb(nskb); @@ -496,24 +511,19 @@ out: } /* Called with genl_mutex. */ -static int flush_flows(int dp_ifindex) +static int flush_flows(struct datapath *dp) { struct flow_table *old_table; struct flow_table *new_table; - struct datapath *dp; - - dp = get_dp(dp_ifindex); - if (!dp) - return -ENODEV; - old_table = get_table_protected(dp); - new_table = flow_tbl_alloc(TBL_MIN_BUCKETS); + old_table = genl_dereference(dp->table); + new_table = ovs_flow_tbl_alloc(TBL_MIN_BUCKETS); if (!new_table) return -ENOMEM; rcu_assign_pointer(dp->table, new_table); - flow_tbl_deferred_destroy(old_table); + ovs_flow_tbl_deferred_destroy(old_table); return 0; } @@ -548,6 +558,19 @@ static int validate_sample(const struct nlattr *attr, 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; + } + + return -EINVAL; +} + static int validate_set(const struct nlattr *a, const struct sw_flow_key *flow_key) { @@ -564,17 +587,32 @@ static int validate_set(const struct nlattr *a, switch (key_type) { const struct ovs_key_ipv4 *ipv4_key; + const struct ovs_key_ipv4_tunnel *tun_key; + const struct ovs_key_ipv6 *ipv6_key; case OVS_KEY_ATTR_PRIORITY: case OVS_KEY_ATTR_TUN_ID: case OVS_KEY_ATTR_ETHERNET: break; + case OVS_KEY_ATTR_SKB_MARK: +#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,20) && !defined(CONFIG_NETFILTER) + if (nla_get_u32(ovs_key) != 0) + return -EINVAL; +#endif + break; + + case OVS_KEY_ATTR_IPV4_TUNNEL: + tun_key = nla_data(ovs_key); + if (!tun_key->ipv4_dst) + return -EINVAL; + break; + case OVS_KEY_ATTR_IPV4: if (flow_key->eth.type != htons(ETH_P_IP)) return -EINVAL; - if (!flow_key->ipv4.addr.src || !flow_key->ipv4.addr.dst) + if (!flow_key->ip.proto) return -EINVAL; ipv4_key = nla_data(ovs_key); @@ -586,22 +624,36 @@ static int validate_set(const struct nlattr *a, break; - case OVS_KEY_ATTR_TCP: - if (flow_key->ip.proto != IPPROTO_TCP) + case OVS_KEY_ATTR_IPV6: + if (flow_key->eth.type != htons(ETH_P_IPV6)) + return -EINVAL; + + if (!flow_key->ip.proto) return -EINVAL; - if (!flow_key->ipv4.tp.src || !flow_key->ipv4.tp.dst) + 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; - if (!flow_key->ipv4.tp.src || !flow_key->ipv4.tp.dst) - return -EINVAL; - break; + return validate_tp_port(flow_key); default: return -EINVAL; @@ -756,55 +808,51 @@ static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info) packet->protocol = htons(ETH_P_802_2); /* Build an sw_flow for sending this packet. */ - flow = flow_alloc(); + flow = ovs_flow_alloc(); err = PTR_ERR(flow); if (IS_ERR(flow)) goto err_kfree_skb; - err = flow_extract(packet, -1, &flow->key, &key_len); + err = ovs_flow_extract(packet, -1, &flow->key, &key_len); if (err) - goto err_flow_put; + goto err_flow_free; - err = flow_metadata_from_nlattrs(&flow->key.phy.priority, - &flow->key.phy.in_port, - &flow->key.phy.tun_id, - a[OVS_PACKET_ATTR_KEY]); + err = ovs_flow_metadata_from_nlattrs(flow, key_len, a[OVS_PACKET_ATTR_KEY]); if (err) - goto err_flow_put; + goto err_flow_free; err = validate_actions(a[OVS_PACKET_ATTR_ACTIONS], &flow->key, 0); if (err) - goto err_flow_put; - - flow->hash = flow_hash(&flow->key, key_len); + goto err_flow_free; - acts = flow_actions_alloc(a[OVS_PACKET_ATTR_ACTIONS]); + acts = ovs_flow_actions_alloc(a[OVS_PACKET_ATTR_ACTIONS]); err = PTR_ERR(acts); if (IS_ERR(acts)) - goto err_flow_put; + goto err_flow_free; rcu_assign_pointer(flow->sf_acts, acts); OVS_CB(packet)->flow = flow; packet->priority = flow->key.phy.priority; + skb_set_mark(packet, flow->key.phy.skb_mark); rcu_read_lock(); - dp = get_dp(ovs_header->dp_ifindex); + dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); err = -ENODEV; if (!dp) goto err_unlock; local_bh_disable(); - err = execute_actions(dp, packet); + err = ovs_execute_actions(dp, packet); local_bh_enable(); rcu_read_unlock(); - flow_put(flow); + ovs_flow_free(flow); return err; err_unlock: rcu_read_unlock(); -err_flow_put: - flow_put(flow); +err_flow_free: + ovs_flow_free(flow); err_kfree_skb: kfree_skb(packet); err: @@ -828,22 +876,22 @@ static struct genl_ops dp_packet_genl_ops[] = { static void get_dp_stats(struct datapath *dp, struct ovs_dp_stats *stats) { int i; - struct flow_table *table = get_table_protected(dp); + struct flow_table *table = genl_dereference(dp->table); - stats->n_flows = flow_tbl_count(table); + stats->n_flows = ovs_flow_tbl_count(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; - unsigned seqcount; + unsigned int start; percpu_stats = per_cpu_ptr(dp->stats_percpu, i); do { - seqcount = read_seqcount_begin(&percpu_stats->seqlock); + start = u64_stats_fetch_begin_bh(&percpu_stats->sync); local_stats = *percpu_stats; - } while (read_seqcount_retry(&percpu_stats->seqlock, seqcount)); + } while (u64_stats_fetch_retry_bh(&percpu_stats->sync, start)); stats->n_hit += local_stats.n_hit; stats->n_missed += local_stats.n_missed; @@ -862,10 +910,11 @@ static struct genl_family dp_flow_genl_family = { .hdrsize = sizeof(struct ovs_header), .name = OVS_FLOW_FAMILY, .version = OVS_FLOW_VERSION, - .maxattr = OVS_FLOW_ATTR_MAX + .maxattr = OVS_FLOW_ATTR_MAX, + SET_NETNSOK }; -static struct genl_multicast_group dp_flow_multicast_group = { +static struct genl_multicast_group ovs_dp_flow_multicast_group = { .name = OVS_FLOW_MCGROUP }; @@ -895,7 +944,7 @@ static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp, nla = nla_nest_start(skb, OVS_FLOW_ATTR_KEY); if (!nla) goto nla_put_failure; - err = flow_to_nlattrs(&flow->key, skb); + err = ovs_flow_to_nlattrs(&flow->key, skb); if (err) goto error; nla_nest_end(skb, nla); @@ -907,15 +956,18 @@ static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp, tcp_flags = flow->tcp_flags; spin_unlock_bh(&flow->lock); - if (used) - NLA_PUT_U64(skb, OVS_FLOW_ATTR_USED, flow_used_time(used)); + if (used && + nla_put_u64(skb, OVS_FLOW_ATTR_USED, ovs_flow_used_time(used))) + goto nla_put_failure; - if (stats.n_packets) - NLA_PUT(skb, OVS_FLOW_ATTR_STATS, - sizeof(struct ovs_flow_stats), &stats); + if (stats.n_packets && + nla_put(skb, OVS_FLOW_ATTR_STATS, + sizeof(struct ovs_flow_stats), &stats)) + goto nla_put_failure; - if (tcp_flags) - NLA_PUT_U8(skb, OVS_FLOW_ATTR_TCP_FLAGS, tcp_flags); + if (tcp_flags && + nla_put_u8(skb, OVS_FLOW_ATTR_TCP_FLAGS, tcp_flags)) + goto nla_put_failure; /* 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 @@ -997,7 +1049,7 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) error = -EINVAL; if (!a[OVS_FLOW_ATTR_KEY]) goto error; - error = flow_from_nlattrs(&key, &key_len, a[OVS_FLOW_ATTR_KEY]); + error = ovs_flow_from_nlattrs(&key, &key_len, a[OVS_FLOW_ATTR_KEY]); if (error) goto error; @@ -1011,13 +1063,13 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) goto error; } - dp = get_dp(ovs_header->dp_ifindex); + dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); error = -ENODEV; if (!dp) goto error; - table = get_table_protected(dp); - flow = flow_tbl_lookup(table, &key, key_len); + table = genl_dereference(dp->table); + flow = ovs_flow_tbl_lookup(table, &key, key_len); if (!flow) { struct sw_flow_actions *acts; @@ -1027,36 +1079,34 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) goto error; /* Expand table, if necessary, to make room. */ - if (flow_tbl_need_to_expand(table)) { + if (ovs_flow_tbl_need_to_expand(table)) { struct flow_table *new_table; - new_table = flow_tbl_expand(table); + new_table = ovs_flow_tbl_expand(table); if (!IS_ERR(new_table)) { rcu_assign_pointer(dp->table, new_table); - flow_tbl_deferred_destroy(table); - table = get_table_protected(dp); + ovs_flow_tbl_deferred_destroy(table); + table = genl_dereference(dp->table); } } /* Allocate flow. */ - flow = flow_alloc(); + flow = ovs_flow_alloc(); if (IS_ERR(flow)) { error = PTR_ERR(flow); goto error; } - flow->key = key; clear_stats(flow); /* Obtain actions. */ - acts = flow_actions_alloc(a[OVS_FLOW_ATTR_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); /* Put flow in bucket. */ - flow->hash = flow_hash(&key, key_len); - flow_tbl_insert(table, flow); + ovs_flow_tbl_insert(table, flow, &key, key_len); reply = ovs_flow_cmd_build_info(flow, dp, info->snd_pid, info->snd_seq, @@ -1087,13 +1137,13 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) old_acts->actions_len))) { struct sw_flow_actions *new_acts; - new_acts = flow_actions_alloc(acts_attrs); + new_acts = ovs_flow_actions_alloc(acts_attrs); error = PTR_ERR(new_acts); if (IS_ERR(new_acts)) goto error; rcu_assign_pointer(flow->sf_acts, new_acts); - flow_deferred_free_acts(old_acts); + ovs_flow_deferred_free_acts(old_acts); } reply = ovs_flow_cmd_build_info(flow, dp, info->snd_pid, @@ -1109,14 +1159,15 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) if (!IS_ERR(reply)) genl_notify(reply, genl_info_net(info), info->snd_pid, - dp_flow_multicast_group.id, info->nlhdr, GFP_KERNEL); + ovs_dp_flow_multicast_group.id, info->nlhdr, + GFP_KERNEL); else - netlink_set_err(INIT_NET_GENL_SOCK, 0, - dp_flow_multicast_group.id, PTR_ERR(reply)); + netlink_set_err(GENL_SOCK(sock_net(skb->sk)), 0, + ovs_dp_flow_multicast_group.id, PTR_ERR(reply)); return 0; error_free_flow: - flow_put(flow); + ovs_flow_free(flow); error: return error; } @@ -1135,16 +1186,16 @@ static int ovs_flow_cmd_get(struct sk_buff *skb, struct genl_info *info) if (!a[OVS_FLOW_ATTR_KEY]) return -EINVAL; - err = flow_from_nlattrs(&key, &key_len, a[OVS_FLOW_ATTR_KEY]); + err = ovs_flow_from_nlattrs(&key, &key_len, a[OVS_FLOW_ATTR_KEY]); if (err) return err; - dp = get_dp(ovs_header->dp_ifindex); + dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); if (!dp) return -ENODEV; - table = get_table_protected(dp); - flow = flow_tbl_lookup(table, &key, key_len); + table = genl_dereference(dp->table); + flow = ovs_flow_tbl_lookup(table, &key, key_len); if (!flow) return -ENOENT; @@ -1168,18 +1219,19 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info) int err; int key_len; + dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); + if (!dp) + return -ENODEV; + if (!a[OVS_FLOW_ATTR_KEY]) - return flush_flows(ovs_header->dp_ifindex); - err = flow_from_nlattrs(&key, &key_len, a[OVS_FLOW_ATTR_KEY]); + return flush_flows(dp); + + err = ovs_flow_from_nlattrs(&key, &key_len, a[OVS_FLOW_ATTR_KEY]); if (err) return err; - dp = get_dp(ovs_header->dp_ifindex); - if (!dp) - return -ENODEV; - - table = get_table_protected(dp); - flow = flow_tbl_lookup(table, &key, key_len); + table = genl_dereference(dp->table); + flow = ovs_flow_tbl_lookup(table, &key, key_len); if (!flow) return -ENOENT; @@ -1187,16 +1239,16 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info) if (!reply) return -ENOMEM; - flow_tbl_remove(table, flow); + ovs_flow_tbl_remove(table, flow); err = ovs_flow_cmd_fill_info(flow, dp, reply, info->snd_pid, info->snd_seq, 0, OVS_FLOW_CMD_DEL); BUG_ON(err < 0); - flow_deferred_free(flow); + ovs_flow_deferred_free(flow); genl_notify(reply, genl_info_net(info), info->snd_pid, - dp_flow_multicast_group.id, info->nlhdr, GFP_KERNEL); + ovs_dp_flow_multicast_group.id, info->nlhdr, GFP_KERNEL); return 0; } @@ -1204,18 +1256,21 @@ 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 datapath *dp; + struct flow_table *table; - dp = get_dp(ovs_header->dp_ifindex); + dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); if (!dp) return -ENODEV; + table = genl_dereference(dp->table); + for (;;) { struct sw_flow *flow; u32 bucket, obj; bucket = cb->args[0]; obj = cb->args[1]; - flow = flow_tbl_next(get_table_protected(dp), &bucket, &obj); + flow = ovs_flow_tbl_next(table, &bucket, &obj); if (!flow) break; @@ -1267,10 +1322,11 @@ static struct genl_family dp_datapath_genl_family = { .hdrsize = sizeof(struct ovs_header), .name = OVS_DATAPATH_FAMILY, .version = OVS_DATAPATH_VERSION, - .maxattr = OVS_DP_ATTR_MAX + .maxattr = OVS_DP_ATTR_MAX, + SET_NETNSOK }; -static struct genl_multicast_group dp_datapath_multicast_group = { +static struct genl_multicast_group ovs_dp_datapath_multicast_group = { .name = OVS_DATAPATH_MCGROUP }; @@ -1289,13 +1345,14 @@ static int ovs_dp_cmd_fill_info(struct datapath *dp, struct sk_buff *skb, ovs_header->dp_ifindex = get_dpifindex(dp); rcu_read_lock(); - err = nla_put_string(skb, OVS_DP_ATTR_NAME, dp_name(dp)); + 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); - NLA_PUT(skb, OVS_DP_ATTR_STATS, sizeof(struct ovs_dp_stats), &dp_stats); + if (nla_put(skb, OVS_DP_ATTR_STATS, sizeof(struct ovs_dp_stats), &dp_stats)) + goto nla_put_failure; return genlmsg_end(skb, ovs_header); @@ -1329,18 +1386,19 @@ static int ovs_dp_cmd_validate(struct nlattr *a[OVS_DP_ATTR_MAX + 1]) } /* Called with genl_mutex and optionally with RTNL lock also. */ -static struct datapath *lookup_datapath(struct ovs_header *ovs_header, +static struct datapath *lookup_datapath(struct net *net, + struct ovs_header *ovs_header, struct nlattr *a[OVS_DP_ATTR_MAX + 1]) { struct datapath *dp; if (!a[OVS_DP_ATTR_NAME]) - dp = get_dp(ovs_header->dp_ifindex); + dp = get_dp(net, ovs_header->dp_ifindex); else { struct vport *vport; rcu_read_lock(); - vport = vport_locate(nla_data(a[OVS_DP_ATTR_NAME])); + 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(); } @@ -1354,7 +1412,8 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) struct sk_buff *reply; struct datapath *dp; struct vport *vport; - int err; + struct ovs_net *ovs_net; + int err, i; err = -EINVAL; if (!a[OVS_DP_ATTR_NAME] || !a[OVS_DP_ATTR_UPCALL_PID]) @@ -1365,24 +1424,22 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) goto err; rtnl_lock(); - err = -ENODEV; - if (!try_module_get(THIS_MODULE)) - goto err_unlock_rtnl; err = -ENOMEM; dp = kzalloc(sizeof(*dp), GFP_KERNEL); if (dp == NULL) - goto err_put_module; - INIT_LIST_HEAD(&dp->port_list); + goto err_unlock_rtnl; /* Initialize kobject for bridge. This will be added as * /sys/class/net//brif later, if sysfs is enabled. */ dp->ifobj.kset = NULL; kobject_init(&dp->ifobj, &dp_ktype); + ovs_dp_set_net(dp, hold_net(sock_net(skb->sk))); + /* Allocate table. */ err = -ENOMEM; - rcu_assign_pointer(dp->table, flow_tbl_alloc(TBL_MIN_BUCKETS)); + rcu_assign_pointer(dp->table, ovs_flow_tbl_alloc(TBL_MIN_BUCKETS)); if (!dp->table) goto err_free_dp; @@ -1392,6 +1449,16 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) goto err_destroy_table; } + dp->ports = kmalloc(DP_VPORT_HASH_BUCKETS * sizeof(struct hlist_head), + GFP_KERNEL); + if (!dp->ports) { + err = -ENOMEM; + goto err_destroy_percpu; + } + + for (i = 0; i < DP_VPORT_HASH_BUCKETS; i++) + INIT_HLIST_HEAD(&dp->ports[i]); + /* Set up our datapath device. */ parms.name = nla_data(a[OVS_DP_ATTR_NAME]); parms.type = OVS_VPORT_TYPE_INTERNAL; @@ -1406,7 +1473,7 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) if (err == -EBUSY) err = -EEXIST; - goto err_destroy_percpu; + goto err_destroy_ports_array; } reply = ovs_dp_cmd_build_info(dp, info->snd_pid, @@ -1415,81 +1482,92 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) if (IS_ERR(reply)) goto err_destroy_local_port; - list_add_tail(&dp->list_node, &dps); - dp_sysfs_add_dp(dp); + ovs_net = net_generic(ovs_dp_get_net(dp), ovs_net_id); + list_add_tail(&dp->list_node, &ovs_net->dps); + ovs_dp_sysfs_add_dp(dp); rtnl_unlock(); genl_notify(reply, genl_info_net(info), info->snd_pid, - dp_datapath_multicast_group.id, info->nlhdr, GFP_KERNEL); + ovs_dp_datapath_multicast_group.id, info->nlhdr, + GFP_KERNEL); return 0; err_destroy_local_port: - dp_detach_port(get_vport_protected(dp, OVSP_LOCAL)); + ovs_dp_detach_port(ovs_vport_rtnl(dp, OVSP_LOCAL)); +err_destroy_ports_array: + kfree(dp->ports); err_destroy_percpu: free_percpu(dp->stats_percpu); err_destroy_table: - flow_tbl_destroy(get_table_protected(dp)); + ovs_flow_tbl_destroy(genl_dereference(dp->table)); err_free_dp: + release_net(ovs_dp_get_net(dp)); kfree(dp); -err_put_module: - module_put(THIS_MODULE); err_unlock_rtnl: rtnl_unlock(); err: return err; } +/* Called with genl_mutex. */ +static void __dp_destroy(struct datapath *dp) +{ + int i; + + rtnl_lock(); + + for (i = 0; i < DP_VPORT_HASH_BUCKETS; i++) { + struct vport *vport; + struct hlist_node *node, *n; + + hlist_for_each_entry_safe(vport, node, n, &dp->ports[i], dp_hash_node) + if (vport->port_no != OVSP_LOCAL) + ovs_dp_detach_port(vport); + } + + ovs_dp_sysfs_del_dp(dp); + list_del(&dp->list_node); + ovs_dp_detach_port(ovs_vport_rtnl(dp, OVSP_LOCAL)); + + /* rtnl_unlock() will wait until all the references to devices that + * are pending unregistration have been dropped. We do it here to + * ensure that any internal devices (which contain DP pointers) are + * fully destroyed before freeing the datapath. + */ + rtnl_unlock(); + + call_rcu(&dp->rcu, destroy_dp_rcu); +} + static int ovs_dp_cmd_del(struct sk_buff *skb, struct genl_info *info) { - struct vport *vport, *next_vport; struct sk_buff *reply; struct datapath *dp; int err; err = ovs_dp_cmd_validate(info->attrs); if (err) - goto exit; + return err; - rtnl_lock(); - dp = lookup_datapath(info->userhdr, info->attrs); + dp = lookup_datapath(sock_net(skb->sk), info->userhdr, info->attrs); err = PTR_ERR(dp); if (IS_ERR(dp)) - goto exit_unlock; + return err; reply = ovs_dp_cmd_build_info(dp, info->snd_pid, info->snd_seq, OVS_DP_CMD_DEL); err = PTR_ERR(reply); if (IS_ERR(reply)) - goto exit_unlock; - - list_for_each_entry_safe(vport, next_vport, &dp->port_list, node) - if (vport->port_no != OVSP_LOCAL) - dp_detach_port(vport); - - dp_sysfs_del_dp(dp); - list_del(&dp->list_node); - dp_detach_port(get_vport_protected(dp, OVSP_LOCAL)); - - /* rtnl_unlock() will wait until all the references to devices that - * are pending unregistration have been dropped. We do it here to - * ensure that any internal devices (which contain DP pointers) are - * fully destroyed before freeing the datapath. - */ - rtnl_unlock(); + return err; - call_rcu(&dp->rcu, destroy_dp_rcu); - module_put(THIS_MODULE); + __dp_destroy(dp); genl_notify(reply, genl_info_net(info), info->snd_pid, - dp_datapath_multicast_group.id, info->nlhdr, GFP_KERNEL); + ovs_dp_datapath_multicast_group.id, info->nlhdr, + GFP_KERNEL); return 0; - -exit_unlock: - rtnl_unlock(); -exit: - return err; } static int ovs_dp_cmd_set(struct sk_buff *skb, struct genl_info *info) @@ -1502,7 +1580,7 @@ static int ovs_dp_cmd_set(struct sk_buff *skb, struct genl_info *info) if (err) return err; - dp = lookup_datapath(info->userhdr, info->attrs); + dp = lookup_datapath(sock_net(skb->sk), info->userhdr, info->attrs); if (IS_ERR(dp)) return PTR_ERR(dp); @@ -1510,13 +1588,15 @@ static int ovs_dp_cmd_set(struct sk_buff *skb, struct genl_info *info) info->snd_seq, OVS_DP_CMD_NEW); if (IS_ERR(reply)) { err = PTR_ERR(reply); - netlink_set_err(INIT_NET_GENL_SOCK, 0, - dp_datapath_multicast_group.id, err); + netlink_set_err(GENL_SOCK(sock_net(skb->sk)), 0, + ovs_dp_datapath_multicast_group.id, err); return 0; } genl_notify(reply, genl_info_net(info), info->snd_pid, - dp_datapath_multicast_group.id, info->nlhdr, GFP_KERNEL); + ovs_dp_datapath_multicast_group.id, info->nlhdr, + GFP_KERNEL); + return 0; } @@ -1530,7 +1610,7 @@ static int ovs_dp_cmd_get(struct sk_buff *skb, struct genl_info *info) if (err) return err; - dp = lookup_datapath(info->userhdr, info->attrs); + dp = lookup_datapath(sock_net(skb->sk), info->userhdr, info->attrs); if (IS_ERR(dp)) return PTR_ERR(dp); @@ -1544,14 +1624,14 @@ static int ovs_dp_cmd_get(struct sk_buff *skb, struct genl_info *info) static int ovs_dp_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb) { + struct ovs_net *ovs_net = net_generic(sock_net(skb->sk), ovs_net_id); struct datapath *dp; int skip = cb->args[0]; int i = 0; - list_for_each_entry(dp, &dps, list_node) { - if (i < skip) - continue; - if (ovs_dp_cmd_fill_info(dp, skb, NETLINK_CB(cb->skb).pid, + list_for_each_entry(dp, &ovs_net->dps, list_node) { + if (i >= skip && + ovs_dp_cmd_fill_info(dp, skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, NLM_F_MULTI, OVS_DP_CMD_NEW) < 0) break; @@ -1607,10 +1687,11 @@ static struct genl_family dp_vport_genl_family = { .hdrsize = sizeof(struct ovs_header), .name = OVS_VPORT_FAMILY, .version = OVS_VPORT_VERSION, - .maxattr = OVS_VPORT_ATTR_MAX + .maxattr = OVS_VPORT_ATTR_MAX, + SET_NETNSOK }; -struct genl_multicast_group dp_vport_multicast_group = { +struct genl_multicast_group ovs_dp_vport_multicast_group = { .name = OVS_VPORT_MCGROUP }; @@ -1629,19 +1710,22 @@ static int ovs_vport_cmd_fill_info(struct vport *vport, struct sk_buff *skb, ovs_header->dp_ifindex = get_dpifindex(vport->dp); - 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_pid); + 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_pid)) + goto nla_put_failure; - vport_get_stats(vport, &vport_stats); - NLA_PUT(skb, OVS_VPORT_ATTR_STATS, sizeof(struct ovs_vport_stats), - &vport_stats); + ovs_vport_get_stats(vport, &vport_stats); + if (nla_put(skb, OVS_VPORT_ATTR_STATS, sizeof(struct ovs_vport_stats), + &vport_stats)) + goto nla_put_failure; - NLA_PUT(skb, OVS_VPORT_ATTR_ADDRESS, ETH_ALEN, - vport->ops->get_addr(vport)); + if (nla_put(skb, OVS_VPORT_ATTR_ADDRESS, ETH_ALEN, + vport->ops->get_addr(vport))) + goto nla_put_failure; - err = vport_get_options(vport, skb); + err = ovs_vport_get_options(vport, skb); if (err == -EMSGSIZE) goto error; @@ -1679,16 +1763,20 @@ static int ovs_vport_cmd_validate(struct nlattr *a[OVS_VPORT_ATTR_MAX + 1]) } /* Called with RTNL lock or RCU read lock. */ -static struct vport *lookup_vport(struct ovs_header *ovs_header, +static struct vport *lookup_vport(struct net *net, + struct ovs_header *ovs_header, struct nlattr *a[OVS_VPORT_ATTR_MAX + 1]) { struct datapath *dp; struct vport *vport; if (a[OVS_VPORT_ATTR_NAME]) { - vport = vport_locate(nla_data(a[OVS_VPORT_ATTR_NAME])); + vport = ovs_vport_locate(net, nla_data(a[OVS_VPORT_ATTR_NAME])); if (!vport) return ERR_PTR(-ENODEV); + if (ovs_header->dp_ifindex && + ovs_header->dp_ifindex != get_dpifindex(vport->dp)) + return ERR_PTR(-ENODEV); return vport; } else if (a[OVS_VPORT_ATTR_PORT_NO]) { u32 port_no = nla_get_u32(a[OVS_VPORT_ATTR_PORT_NO]); @@ -1696,11 +1784,11 @@ static struct vport *lookup_vport(struct ovs_header *ovs_header, if (port_no >= DP_MAX_PORTS) return ERR_PTR(-EFBIG); - dp = get_dp(ovs_header->dp_ifindex); + dp = get_dp(net, ovs_header->dp_ifindex); if (!dp) return ERR_PTR(-ENODEV); - vport = get_vport_protected(dp, port_no); + vport = ovs_vport_rtnl_rcu(dp, port_no); if (!vport) return ERR_PTR(-ENOENT); return vport; @@ -1715,10 +1803,10 @@ static int change_vport(struct vport *vport, int err = 0; if (a[OVS_VPORT_ATTR_STATS]) - vport_set_stats(vport, nla_data(a[OVS_VPORT_ATTR_STATS])); + ovs_vport_set_stats(vport, nla_data(a[OVS_VPORT_ATTR_STATS])); if (a[OVS_VPORT_ATTR_ADDRESS]) - err = vport_set_addr(vport, nla_data(a[OVS_VPORT_ATTR_ADDRESS])); + err = ovs_vport_set_addr(vport, nla_data(a[OVS_VPORT_ATTR_ADDRESS])); return err; } @@ -1744,7 +1832,7 @@ static int ovs_vport_cmd_new(struct sk_buff *skb, struct genl_info *info) goto exit; rtnl_lock(); - dp = get_dp(ovs_header->dp_ifindex); + dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); err = -ENODEV; if (!dp) goto exit_unlock; @@ -1756,7 +1844,7 @@ static int ovs_vport_cmd_new(struct sk_buff *skb, struct genl_info *info) if (port_no >= DP_MAX_PORTS) goto exit_unlock; - vport = get_vport_protected(dp, port_no); + vport = ovs_vport_rtnl(dp, port_no); err = -EBUSY; if (vport) goto exit_unlock; @@ -1766,7 +1854,7 @@ static int ovs_vport_cmd_new(struct sk_buff *skb, struct genl_info *info) err = -EFBIG; goto exit_unlock; } - vport = get_vport_protected(dp, port_no); + vport = ovs_vport_rtnl(dp, port_no); if (!vport) break; } @@ -1784,7 +1872,7 @@ static int ovs_vport_cmd_new(struct sk_buff *skb, struct genl_info *info) if (IS_ERR(vport)) goto exit_unlock; - dp_sysfs_add_if(vport); + ovs_dp_sysfs_add_if(vport); err = change_vport(vport, a); if (!err) { @@ -1795,12 +1883,11 @@ static int ovs_vport_cmd_new(struct sk_buff *skb, struct genl_info *info) err = PTR_ERR(reply); } if (err) { - dp_detach_port(vport); + ovs_dp_detach_port(vport); goto exit_unlock; } genl_notify(reply, genl_info_net(info), info->snd_pid, - dp_vport_multicast_group.id, info->nlhdr, GFP_KERNEL); - + ovs_dp_vport_multicast_group.id, info->nlhdr, GFP_KERNEL); exit_unlock: rtnl_unlock(); @@ -1820,7 +1907,7 @@ static int ovs_vport_cmd_set(struct sk_buff *skb, struct genl_info *info) goto exit; rtnl_lock(); - vport = lookup_vport(info->userhdr, a); + vport = lookup_vport(sock_net(skb->sk), info->userhdr, a); err = PTR_ERR(vport); if (IS_ERR(vport)) goto exit_unlock; @@ -1831,23 +1918,24 @@ static int ovs_vport_cmd_set(struct sk_buff *skb, struct genl_info *info) err = -EINVAL; if (!err && a[OVS_VPORT_ATTR_OPTIONS]) - err = vport_set_options(vport, a[OVS_VPORT_ATTR_OPTIONS]); + err = ovs_vport_set_options(vport, a[OVS_VPORT_ATTR_OPTIONS]); if (!err) err = change_vport(vport, a); + else + goto exit_unlock; if (!err && a[OVS_VPORT_ATTR_UPCALL_PID]) vport->upcall_pid = nla_get_u32(a[OVS_VPORT_ATTR_UPCALL_PID]); reply = ovs_vport_cmd_build_info(vport, info->snd_pid, info->snd_seq, OVS_VPORT_CMD_NEW); if (IS_ERR(reply)) { - err = PTR_ERR(reply); - netlink_set_err(INIT_NET_GENL_SOCK, 0, - dp_vport_multicast_group.id, err); - return 0; + netlink_set_err(GENL_SOCK(sock_net(skb->sk)), 0, + ovs_dp_vport_multicast_group.id, PTR_ERR(reply)); + goto exit_unlock; } genl_notify(reply, genl_info_net(info), info->snd_pid, - dp_vport_multicast_group.id, info->nlhdr, GFP_KERNEL); + ovs_dp_vport_multicast_group.id, info->nlhdr, GFP_KERNEL); exit_unlock: rtnl_unlock(); @@ -1867,7 +1955,7 @@ static int ovs_vport_cmd_del(struct sk_buff *skb, struct genl_info *info) goto exit; rtnl_lock(); - vport = lookup_vport(info->userhdr, a); + vport = lookup_vport(sock_net(skb->sk), info->userhdr, a); err = PTR_ERR(vport); if (IS_ERR(vport)) goto exit_unlock; @@ -1883,10 +1971,10 @@ static int ovs_vport_cmd_del(struct sk_buff *skb, struct genl_info *info) if (IS_ERR(reply)) goto exit_unlock; - dp_detach_port(vport); + ovs_dp_detach_port(vport); genl_notify(reply, genl_info_net(info), info->snd_pid, - dp_vport_multicast_group.id, info->nlhdr, GFP_KERNEL); + ovs_dp_vport_multicast_group.id, info->nlhdr, GFP_KERNEL); exit_unlock: rtnl_unlock(); @@ -1907,7 +1995,7 @@ static int ovs_vport_cmd_get(struct sk_buff *skb, struct genl_info *info) goto exit; rcu_read_lock(); - vport = lookup_vport(ovs_header, a); + vport = lookup_vport(sock_net(skb->sk), ovs_header, a); err = PTR_ERR(vport); if (IS_ERR(vport)) goto exit_unlock; @@ -1932,32 +2020,39 @@ static int ovs_vport_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb) { struct ovs_header *ovs_header = genlmsg_data(nlmsg_data(cb->nlh)); struct datapath *dp; - u32 port_no; - int retval; + int bucket = cb->args[0], skip = cb->args[1]; + int i, j = 0; - dp = get_dp(ovs_header->dp_ifindex); + dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); if (!dp) return -ENODEV; rcu_read_lock(); - for (port_no = cb->args[0]; port_no < DP_MAX_PORTS; port_no++) { + for (i = bucket; i < DP_VPORT_HASH_BUCKETS; i++) { struct vport *vport; - - vport = get_vport_protected(dp, port_no); - if (!vport) - continue; - - if (ovs_vport_cmd_fill_info(vport, skb, NETLINK_CB(cb->skb).pid, - cb->nlh->nlmsg_seq, NLM_F_MULTI, - OVS_VPORT_CMD_NEW) < 0) - break; + struct hlist_node *n; + + j = 0; + hlist_for_each_entry_rcu(vport, n, &dp->ports[i], dp_hash_node) { + if (j >= skip && + ovs_vport_cmd_fill_info(vport, skb, + NETLINK_CB(cb->skb).pid, + cb->nlh->nlmsg_seq, + NLM_F_MULTI, + OVS_VPORT_CMD_NEW) < 0) + goto out; + + j++; + } + skip = 0; } +out: rcu_read_unlock(); - cb->args[0] = port_no; - retval = skb->len; + cb->args[0] = i; + cb->args[1] = j; - return retval; + return skb->len; } static struct genl_ops dp_vport_genl_ops[] = { @@ -1994,13 +2089,13 @@ struct genl_family_and_ops { static const struct genl_family_and_ops dp_genl_families[] = { { &dp_datapath_genl_family, dp_datapath_genl_ops, ARRAY_SIZE(dp_datapath_genl_ops), - &dp_datapath_multicast_group }, + &ovs_dp_datapath_multicast_group }, { &dp_vport_genl_family, dp_vport_genl_ops, ARRAY_SIZE(dp_vport_genl_ops), - &dp_vport_multicast_group }, + &ovs_dp_vport_multicast_group }, { &dp_flow_genl_family, dp_flow_genl_ops, ARRAY_SIZE(dp_flow_genl_ops), - &dp_flow_multicast_group }, + &ovs_dp_flow_multicast_group }, { &dp_packet_genl_family, dp_packet_genl_ops, ARRAY_SIZE(dp_packet_genl_ops), NULL }, @@ -2044,6 +2139,69 @@ error: return err; } +static int __rehash_flow_table(void *dummy) +{ + struct datapath *dp; + struct net *net; + + 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 = genl_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(); + return 0; +} + +static void rehash_flow_table(struct work_struct *work) +{ + genl_exec(__rehash_flow_table, NULL); + schedule_delayed_work(&rehash_flow_wq, REHASH_FLOW_INTERVAL); +} + +static int dp_destroy_all(void *data) +{ + struct datapath *dp, *dp_next; + struct ovs_net *ovs_net = data; + + list_for_each_entry_safe(dp, dp_next, &ovs_net->dps, list_node) + __dp_destroy(dp); + + return 0; +} + +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); + return 0; +} + +static void __net_exit ovs_exit_net(struct net *net) +{ + struct ovs_net *ovs_net = net_generic(net, ovs_net_id); + + genl_exec(dp_destroy_all, ovs_net); +} + +static struct pernet_operations ovs_net_ops = { + .init = ovs_init_net, + .exit = ovs_exit_net, + .id = &ovs_net_id, + .size = sizeof(struct ovs_net), +}; + static int __init dp_init(void) { struct sk_buff *dummy_skb; @@ -2052,50 +2210,74 @@ static int __init dp_init(void) BUILD_BUG_ON(sizeof(struct ovs_skb_cb) > sizeof(dummy_skb->cb)); pr_info("Open vSwitch switching datapath %s, built "__DATE__" "__TIME__"\n", - VERSION BUILDNR); + VERSION); - err = tnl_init(); + err = genl_exec_init(); if (err) goto error; - err = flow_init(); + err = ovs_workqueues_init(); + if (err) + goto error_genl_exec; + + err = ovs_tnl_init(); + if (err) + goto error_wq; + + err = ovs_flow_init(); if (err) goto error_tnl_exit; - err = vport_init(); + err = ovs_vport_init(); if (err) goto error_flow_exit; - err = register_netdevice_notifier(&dp_device_notifier); + err = register_pernet_device(&ovs_net_ops); if (err) goto error_vport_exit; + err = register_netdevice_notifier(&ovs_dp_device_notifier); + if (err) + goto error_netns_exit; + err = dp_register_genl(); if (err < 0) goto error_unreg_notifier; + schedule_delayed_work(&rehash_flow_wq, REHASH_FLOW_INTERVAL); + return 0; error_unreg_notifier: - unregister_netdevice_notifier(&dp_device_notifier); + unregister_netdevice_notifier(&ovs_dp_device_notifier); +error_netns_exit: + unregister_pernet_device(&ovs_net_ops); error_vport_exit: - vport_exit(); + ovs_vport_exit(); error_flow_exit: - flow_exit(); + ovs_flow_exit(); error_tnl_exit: - tnl_exit(); + ovs_tnl_exit(); +error_wq: + ovs_workqueues_exit(); +error_genl_exec: + genl_exec_exit(); error: return err; } static void dp_cleanup(void) { - rcu_barrier(); + cancel_delayed_work_sync(&rehash_flow_wq); dp_unregister_genl(ARRAY_SIZE(dp_genl_families)); - unregister_netdevice_notifier(&dp_device_notifier); - vport_exit(); - flow_exit(); - tnl_exit(); + unregister_netdevice_notifier(&ovs_dp_device_notifier); + unregister_pernet_device(&ovs_net_ops); + rcu_barrier(); + ovs_vport_exit(); + ovs_flow_exit(); + ovs_tnl_exit(); + ovs_workqueues_exit(); + genl_exec_exit(); } module_init(dp_init); @@ -2103,3 +2285,4 @@ module_exit(dp_cleanup); MODULE_DESCRIPTION("Open vSwitch switching datapath"); MODULE_LICENSE("GPL"); +MODULE_VERSION(VERSION);