diff options
Diffstat (limited to 'net')
294 files changed, 9494 insertions, 5096 deletions
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c index 175273f38cb..44ebd5c2cd4 100644 --- a/net/8021q/vlan.c +++ b/net/8021q/vlan.c @@ -169,6 +169,7 @@ int register_vlan_dev(struct net_device *dev) if (err < 0) goto out_uninit_mvrp; + vlan->nest_level = dev_get_nest_level(real_dev, is_vlan_dev) + 1; err = register_netdevice(dev); if (err < 0) goto out_uninit_mvrp; diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c index 3c32bd257b7..9012b1c922b 100644 --- a/net/8021q/vlan_core.c +++ b/net/8021q/vlan_core.c @@ -63,7 +63,7 @@ bool vlan_do_receive(struct sk_buff **skbp) } /* Must be invoked with rcu_read_lock. */ -struct net_device *__vlan_find_dev_deep(struct net_device *dev, +struct net_device *__vlan_find_dev_deep_rcu(struct net_device *dev, __be16 vlan_proto, u16 vlan_id) { struct vlan_info *vlan_info = rcu_dereference(dev->vlan_info); @@ -81,13 +81,13 @@ struct net_device *__vlan_find_dev_deep(struct net_device *dev, upper_dev = netdev_master_upper_dev_get_rcu(dev); if (upper_dev) - return __vlan_find_dev_deep(upper_dev, + return __vlan_find_dev_deep_rcu(upper_dev, vlan_proto, vlan_id); } return NULL; } -EXPORT_SYMBOL(__vlan_find_dev_deep); +EXPORT_SYMBOL(__vlan_find_dev_deep_rcu); struct net_device *vlan_dev_real_dev(const struct net_device *dev) { diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 6f142f03716..ad2ac3c0039 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -524,6 +524,11 @@ static void vlan_dev_set_lockdep_class(struct net_device *dev, int subclass) netdev_for_each_tx_queue(dev, vlan_dev_set_lockdep_one, &subclass); } +static int vlan_dev_get_lock_subclass(struct net_device *dev) +{ + return vlan_dev_priv(dev)->nest_level; +} + static const struct header_ops vlan_header_ops = { .create = vlan_dev_hard_header, .rebuild = vlan_dev_rebuild_header, @@ -559,7 +564,6 @@ static const struct net_device_ops vlan_netdev_ops; static int vlan_dev_init(struct net_device *dev) { struct net_device *real_dev = vlan_dev_priv(dev)->real_dev; - int subclass = 0; netif_carrier_off(dev); @@ -608,10 +612,7 @@ static int vlan_dev_init(struct net_device *dev) SET_NETDEV_DEVTYPE(dev, &vlan_type); - if (is_vlan_dev(real_dev)) - subclass = 1; - - vlan_dev_set_lockdep_class(dev, subclass); + vlan_dev_set_lockdep_class(dev, vlan_dev_get_lock_subclass(dev)); vlan_dev_priv(dev)->vlan_pcpu_stats = netdev_alloc_pcpu_stats(struct vlan_pcpu_stats); if (!vlan_dev_priv(dev)->vlan_pcpu_stats) @@ -642,9 +643,9 @@ static netdev_features_t vlan_dev_fix_features(struct net_device *dev, struct net_device *real_dev = vlan_dev_priv(dev)->real_dev; netdev_features_t old_features = features; - features &= real_dev->vlan_features; + features = netdev_intersect_features(features, real_dev->vlan_features); features |= NETIF_F_RXCSUM; - features &= real_dev->features; + features = netdev_intersect_features(features, real_dev->features); features |= old_features & NETIF_F_SOFT_FEATURES; features |= NETIF_F_LLTX; @@ -670,38 +671,36 @@ static void vlan_ethtool_get_drvinfo(struct net_device *dev, static struct rtnl_link_stats64 *vlan_dev_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats) { + struct vlan_pcpu_stats *p; + u32 rx_errors = 0, tx_dropped = 0; + int i; - if (vlan_dev_priv(dev)->vlan_pcpu_stats) { - struct vlan_pcpu_stats *p; - u32 rx_errors = 0, tx_dropped = 0; - int i; - - for_each_possible_cpu(i) { - u64 rxpackets, rxbytes, rxmulticast, txpackets, txbytes; - unsigned int start; - - p = per_cpu_ptr(vlan_dev_priv(dev)->vlan_pcpu_stats, i); - do { - start = u64_stats_fetch_begin_irq(&p->syncp); - rxpackets = p->rx_packets; - rxbytes = p->rx_bytes; - rxmulticast = p->rx_multicast; - txpackets = p->tx_packets; - txbytes = p->tx_bytes; - } while (u64_stats_fetch_retry_irq(&p->syncp, start)); - - stats->rx_packets += rxpackets; - stats->rx_bytes += rxbytes; - stats->multicast += rxmulticast; - stats->tx_packets += txpackets; - stats->tx_bytes += txbytes; - /* rx_errors & tx_dropped are u32 */ - rx_errors += p->rx_errors; - tx_dropped += p->tx_dropped; - } - stats->rx_errors = rx_errors; - stats->tx_dropped = tx_dropped; + for_each_possible_cpu(i) { + u64 rxpackets, rxbytes, rxmulticast, txpackets, txbytes; + unsigned int start; + + p = per_cpu_ptr(vlan_dev_priv(dev)->vlan_pcpu_stats, i); + do { + start = u64_stats_fetch_begin_irq(&p->syncp); + rxpackets = p->rx_packets; + rxbytes = p->rx_bytes; + rxmulticast = p->rx_multicast; + txpackets = p->tx_packets; + txbytes = p->tx_bytes; + } while (u64_stats_fetch_retry_irq(&p->syncp, start)); + + stats->rx_packets += rxpackets; + stats->rx_bytes += rxbytes; + stats->multicast += rxmulticast; + stats->tx_packets += txpackets; + stats->tx_bytes += txbytes; + /* rx_errors & tx_dropped are u32 */ + rx_errors += p->rx_errors; + tx_dropped += p->tx_dropped; } + stats->rx_errors = rx_errors; + stats->tx_dropped = tx_dropped; + return stats; } @@ -783,6 +782,7 @@ static const struct net_device_ops vlan_netdev_ops = { .ndo_netpoll_cleanup = vlan_dev_netpoll_cleanup, #endif .ndo_fix_features = vlan_dev_fix_features, + .ndo_get_lock_subclass = vlan_dev_get_lock_subclass, }; void vlan_setup(struct net_device *dev) diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c index 786ee2f83d5..01a1082e02b 100644 --- a/net/appletalk/ddp.c +++ b/net/appletalk/ddp.c @@ -1669,7 +1669,7 @@ static int atalk_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr goto out; } - if (sk->sk_no_check == 1) + if (sk->sk_no_check_tx) ddp->deh_sum = 0; else ddp->deh_sum = atalk_checksum(skb, len + sizeof(*ddp)); diff --git a/net/atm/svc.c b/net/atm/svc.c index 1281049c135..d8e5d0c2ebb 100644 --- a/net/atm/svc.c +++ b/net/atm/svc.c @@ -263,17 +263,11 @@ static int svc_connect(struct socket *sock, struct sockaddr *sockaddr, goto out; } } -/* - * Not supported yet - * - * #ifndef CONFIG_SINGLE_SIGITF - */ + vcc->qos.txtp.max_pcr = SELECT_TOP_PCR(vcc->qos.txtp); vcc->qos.txtp.pcr = 0; vcc->qos.txtp.min_pcr = 0; -/* - * #endif - */ + error = vcc_connect(sock, vcc->itf, vcc->vpi, vcc->vci); if (!error) sock->state = SS_CONNECTED; diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c index b3bd4ec3fd9..f04224c3200 100644 --- a/net/batman-adv/bat_iv_ogm.c +++ b/net/batman-adv/bat_iv_ogm.c @@ -1545,6 +1545,8 @@ out_neigh: if ((orig_neigh_node) && (!is_single_hop_neigh)) batadv_orig_node_free_ref(orig_neigh_node); out: + if (router_ifinfo) + batadv_neigh_ifinfo_free_ref(router_ifinfo); if (router) batadv_neigh_node_free_ref(router); if (router_router) diff --git a/net/batman-adv/debugfs.c b/net/batman-adv/debugfs.c index b758881be10..a12e25efaf6 100644 --- a/net/batman-adv/debugfs.c +++ b/net/batman-adv/debugfs.c @@ -245,6 +245,7 @@ static int batadv_algorithms_open(struct inode *inode, struct file *file) static int batadv_originators_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; + return single_open(file, batadv_orig_seq_print_text, net_dev); } @@ -258,18 +259,21 @@ static int batadv_originators_hardif_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; + return single_open(file, batadv_orig_hardif_seq_print_text, net_dev); } static int batadv_gateways_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; + return single_open(file, batadv_gw_client_seq_print_text, net_dev); } static int batadv_transtable_global_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; + return single_open(file, batadv_tt_global_seq_print_text, net_dev); } @@ -277,6 +281,7 @@ static int batadv_transtable_global_open(struct inode *inode, struct file *file) static int batadv_bla_claim_table_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; + return single_open(file, batadv_bla_claim_table_seq_print_text, net_dev); } @@ -285,6 +290,7 @@ static int batadv_bla_backbone_table_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; + return single_open(file, batadv_bla_backbone_table_seq_print_text, net_dev); } @@ -300,6 +306,7 @@ static int batadv_bla_backbone_table_open(struct inode *inode, static int batadv_dat_cache_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; + return single_open(file, batadv_dat_cache_seq_print_text, net_dev); } #endif @@ -307,6 +314,7 @@ static int batadv_dat_cache_open(struct inode *inode, struct file *file) static int batadv_transtable_local_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; + return single_open(file, batadv_tt_local_seq_print_text, net_dev); } @@ -319,6 +327,7 @@ struct batadv_debuginfo { static int batadv_nc_nodes_open(struct inode *inode, struct file *file) { struct net_device *net_dev = (struct net_device *)inode->i_private; + return single_open(file, batadv_nc_nodes_seq_print_text, net_dev); } #endif @@ -333,7 +342,7 @@ struct batadv_debuginfo batadv_debuginfo_##_name = { \ .llseek = seq_lseek, \ .release = single_release, \ } \ -}; +} /* the following attributes are general and therefore they will be directly * placed in the BATADV_DEBUGFS_SUBDIR subdirectory of debugfs @@ -395,7 +404,7 @@ struct batadv_debuginfo batadv_hardif_debuginfo_##_name = { \ .llseek = seq_lseek, \ .release = single_release, \ }, \ -}; +} static BATADV_HARDIF_DEBUGINFO(originators, S_IRUGO, batadv_originators_hardif_open); diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c index b25fd64d727..dcd99b2bea3 100644 --- a/net/batman-adv/distributed-arp-table.c +++ b/net/batman-adv/distributed-arp-table.c @@ -662,6 +662,7 @@ static void batadv_dat_tvlv_container_update(struct batadv_priv *bat_priv) void batadv_dat_status_update(struct net_device *net_dev) { struct batadv_priv *bat_priv = netdev_priv(net_dev); + batadv_dat_tvlv_container_update(bat_priv); } @@ -940,8 +941,7 @@ bool batadv_dat_snoop_outgoing_arp_request(struct batadv_priv *bat_priv, * additional DAT answer may trigger kernel warnings about * a packet coming from the wrong port. */ - if (batadv_is_my_client(bat_priv, dat_entry->mac_addr, - BATADV_NO_FLAGS)) { + if (batadv_is_my_client(bat_priv, dat_entry->mac_addr, vid)) { ret = true; goto out; } diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c index bcc4bea632f..f14e54a0569 100644 --- a/net/batman-adv/fragmentation.c +++ b/net/batman-adv/fragmentation.c @@ -418,12 +418,13 @@ bool batadv_frag_send_packet(struct sk_buff *skb, struct batadv_neigh_node *neigh_node) { struct batadv_priv *bat_priv; - struct batadv_hard_iface *primary_if; + struct batadv_hard_iface *primary_if = NULL; struct batadv_frag_packet frag_header; struct sk_buff *skb_fragment; unsigned mtu = neigh_node->if_incoming->net_dev->mtu; unsigned header_size = sizeof(frag_header); unsigned max_fragment_size, max_packet_size; + bool ret = false; /* To avoid merge and refragmentation at next-hops we never send * fragments larger than BATADV_FRAG_MAX_FRAG_SIZE @@ -483,7 +484,11 @@ bool batadv_frag_send_packet(struct sk_buff *skb, skb->len + ETH_HLEN); batadv_send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr); - return true; + ret = true; + out_err: - return false; + if (primary_if) + batadv_hardif_free_ref(primary_if); + + return ret; } diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c index c835e137423..90cff585b37 100644 --- a/net/batman-adv/gateway_client.c +++ b/net/batman-adv/gateway_client.c @@ -42,8 +42,10 @@ static void batadv_gw_node_free_ref(struct batadv_gw_node *gw_node) { - if (atomic_dec_and_test(&gw_node->refcount)) + if (atomic_dec_and_test(&gw_node->refcount)) { + batadv_orig_node_free_ref(gw_node->orig_node); kfree_rcu(gw_node, rcu); + } } static struct batadv_gw_node * @@ -406,9 +408,14 @@ static void batadv_gw_node_add(struct batadv_priv *bat_priv, if (gateway->bandwidth_down == 0) return; + if (!atomic_inc_not_zero(&orig_node->refcount)) + return; + gw_node = kzalloc(sizeof(*gw_node), GFP_ATOMIC); - if (!gw_node) + if (!gw_node) { + batadv_orig_node_free_ref(orig_node); return; + } INIT_HLIST_NODE(&gw_node->list); gw_node->orig_node = orig_node; diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c index b851cc58085..fbda6b54baf 100644 --- a/net/batman-adv/hard-interface.c +++ b/net/batman-adv/hard-interface.c @@ -83,7 +83,7 @@ static bool batadv_is_on_batman_iface(const struct net_device *net_dev) return true; /* no more parents..stop recursion */ - if (net_dev->iflink == net_dev->ifindex) + if (net_dev->iflink == 0 || net_dev->iflink == net_dev->ifindex) return false; /* recurse over the parent device */ diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h index 770dc890cee..118b990bae2 100644 --- a/net/batman-adv/main.h +++ b/net/batman-adv/main.h @@ -24,7 +24,7 @@ #define BATADV_DRIVER_DEVICE "batman-adv" #ifndef BATADV_SOURCE_VERSION -#define BATADV_SOURCE_VERSION "2014.2.0" +#define BATADV_SOURCE_VERSION "2014.3.0" #endif /* B.A.T.M.A.N. parameters */ diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c index 8c7ca811de6..96b66fd30f9 100644 --- a/net/batman-adv/multicast.c +++ b/net/batman-adv/multicast.c @@ -415,7 +415,7 @@ batadv_mcast_forw_ipv4_node_get(struct batadv_priv *bat_priv) hlist_for_each_entry_rcu(tmp_orig_node, &bat_priv->mcast.want_all_ipv4_list, mcast_want_all_ipv4_node) { - if (!atomic_inc_not_zero(&orig_node->refcount)) + if (!atomic_inc_not_zero(&tmp_orig_node->refcount)) continue; orig_node = tmp_orig_node; @@ -442,7 +442,7 @@ batadv_mcast_forw_ipv6_node_get(struct batadv_priv *bat_priv) hlist_for_each_entry_rcu(tmp_orig_node, &bat_priv->mcast.want_all_ipv6_list, mcast_want_all_ipv6_node) { - if (!atomic_inc_not_zero(&orig_node->refcount)) + if (!atomic_inc_not_zero(&tmp_orig_node->refcount)) continue; orig_node = tmp_orig_node; @@ -493,7 +493,7 @@ batadv_mcast_forw_unsnoop_node_get(struct batadv_priv *bat_priv) hlist_for_each_entry_rcu(tmp_orig_node, &bat_priv->mcast.want_all_unsnoopables_list, mcast_want_all_unsnoopables_node) { - if (!atomic_inc_not_zero(&orig_node->refcount)) + if (!atomic_inc_not_zero(&tmp_orig_node->refcount)) continue; orig_node = tmp_orig_node; diff --git a/net/batman-adv/network-coding.c b/net/batman-adv/network-coding.c index a9546fe541e..40a2fc4bcf4 100644 --- a/net/batman-adv/network-coding.c +++ b/net/batman-adv/network-coding.c @@ -86,6 +86,7 @@ static void batadv_nc_tvlv_container_update(struct batadv_priv *bat_priv) void batadv_nc_status_update(struct net_device *net_dev) { struct batadv_priv *bat_priv = netdev_priv(net_dev); + batadv_nc_tvlv_container_update(bat_priv); } diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c index ffd9dfbd9b0..6a484514cd3 100644 --- a/net/batman-adv/originator.c +++ b/net/batman-adv/originator.c @@ -501,12 +501,17 @@ batadv_neigh_node_get(const struct batadv_orig_node *orig_node, static void batadv_orig_ifinfo_free_rcu(struct rcu_head *rcu) { struct batadv_orig_ifinfo *orig_ifinfo; + struct batadv_neigh_node *router; orig_ifinfo = container_of(rcu, struct batadv_orig_ifinfo, rcu); if (orig_ifinfo->if_outgoing != BATADV_IF_DEFAULT) batadv_hardif_free_ref_now(orig_ifinfo->if_outgoing); + /* this is the last reference to this object */ + router = rcu_dereference_protected(orig_ifinfo->router, true); + if (router) + batadv_neigh_node_free_ref_now(router); kfree(orig_ifinfo); } @@ -702,6 +707,47 @@ free_orig_node: } /** + * batadv_purge_neigh_ifinfo - purge obsolete ifinfo entries from neighbor + * @bat_priv: the bat priv with all the soft interface information + * @neigh: orig node which is to be checked + */ +static void +batadv_purge_neigh_ifinfo(struct batadv_priv *bat_priv, + struct batadv_neigh_node *neigh) +{ + struct batadv_neigh_ifinfo *neigh_ifinfo; + struct batadv_hard_iface *if_outgoing; + struct hlist_node *node_tmp; + + spin_lock_bh(&neigh->ifinfo_lock); + + /* for all ifinfo objects for this neighinator */ + hlist_for_each_entry_safe(neigh_ifinfo, node_tmp, + &neigh->ifinfo_list, list) { + if_outgoing = neigh_ifinfo->if_outgoing; + + /* always keep the default interface */ + if (if_outgoing == BATADV_IF_DEFAULT) + continue; + + /* don't purge if the interface is not (going) down */ + if ((if_outgoing->if_status != BATADV_IF_INACTIVE) && + (if_outgoing->if_status != BATADV_IF_NOT_IN_USE) && + (if_outgoing->if_status != BATADV_IF_TO_BE_REMOVED)) + continue; + + batadv_dbg(BATADV_DBG_BATMAN, bat_priv, + "neighbor/ifinfo purge: neighbor %pM, iface: %s\n", + neigh->addr, if_outgoing->net_dev->name); + + hlist_del_rcu(&neigh_ifinfo->list); + batadv_neigh_ifinfo_free_ref(neigh_ifinfo); + } + + spin_unlock_bh(&neigh->ifinfo_lock); +} + +/** * batadv_purge_orig_ifinfo - purge obsolete ifinfo entries from originator * @bat_priv: the bat priv with all the soft interface information * @orig_node: orig node which is to be checked @@ -800,6 +846,11 @@ batadv_purge_orig_neighbors(struct batadv_priv *bat_priv, hlist_del_rcu(&neigh_node->list); batadv_neigh_node_free_ref(neigh_node); + } else { + /* only necessary if not the whole neighbor is to be + * deleted, but some interface has been removed. + */ + batadv_purge_neigh_ifinfo(bat_priv, neigh_node); } } @@ -857,7 +908,7 @@ static bool batadv_purge_orig_node(struct batadv_priv *bat_priv, { struct batadv_neigh_node *best_neigh_node; struct batadv_hard_iface *hard_iface; - bool changed; + bool changed_ifinfo, changed_neigh; if (batadv_has_timed_out(orig_node->last_seen, 2 * BATADV_PURGE_TIMEOUT)) { @@ -867,10 +918,10 @@ static bool batadv_purge_orig_node(struct batadv_priv *bat_priv, jiffies_to_msecs(orig_node->last_seen)); return true; } - changed = batadv_purge_orig_ifinfo(bat_priv, orig_node); - changed = changed || batadv_purge_orig_neighbors(bat_priv, orig_node); + changed_ifinfo = batadv_purge_orig_ifinfo(bat_priv, orig_node); + changed_neigh = batadv_purge_orig_neighbors(bat_priv, orig_node); - if (!changed) + if (!changed_ifinfo && !changed_neigh) return false; /* first for NULL ... */ @@ -1028,7 +1079,8 @@ int batadv_orig_hardif_seq_print_text(struct seq_file *seq, void *offset) bat_priv->bat_algo_ops->bat_orig_print(bat_priv, seq, hard_iface); out: - batadv_hardif_free_ref(hard_iface); + if (hard_iface) + batadv_hardif_free_ref(hard_iface); return 0; } diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c index 744a59b85e1..e7ee65dc20b 100644 --- a/net/batman-adv/soft-interface.c +++ b/net/batman-adv/soft-interface.c @@ -884,7 +884,7 @@ static void batadv_softif_init_early(struct net_device *dev) /* generate random address */ eth_hw_addr_random(dev); - SET_ETHTOOL_OPS(dev, &batadv_ethtool_ops); + dev->ethtool_ops = &batadv_ethtool_ops; memset(priv, 0, sizeof(*priv)); } diff --git a/net/batman-adv/sysfs.c b/net/batman-adv/sysfs.c index 1ebb0d9e2ea..fc47baa888c 100644 --- a/net/batman-adv/sysfs.c +++ b/net/batman-adv/sysfs.c @@ -29,12 +29,14 @@ static struct net_device *batadv_kobj_to_netdev(struct kobject *obj) { struct device *dev = container_of(obj->parent, struct device, kobj); + return to_net_dev(dev); } static struct batadv_priv *batadv_kobj_to_batpriv(struct kobject *obj) { struct net_device *net_dev = batadv_kobj_to_netdev(obj); + return netdev_priv(net_dev); } @@ -106,7 +108,7 @@ struct batadv_attribute batadv_attr_vlan_##_name = { \ .mode = _mode }, \ .show = _show, \ .store = _store, \ -}; +} /* Use this, if you have customized show and store functions */ #define BATADV_ATTR(_name, _mode, _show, _store) \ @@ -115,7 +117,7 @@ struct batadv_attribute batadv_attr_##_name = { \ .mode = _mode }, \ .show = _show, \ .store = _store, \ -}; +} #define BATADV_ATTR_SIF_STORE_BOOL(_name, _post_func) \ ssize_t batadv_store_##_name(struct kobject *kobj, \ @@ -124,6 +126,7 @@ ssize_t batadv_store_##_name(struct kobject *kobj, \ { \ struct net_device *net_dev = batadv_kobj_to_netdev(kobj); \ struct batadv_priv *bat_priv = netdev_priv(net_dev); \ + \ return __batadv_store_bool_attr(buff, count, _post_func, attr, \ &bat_priv->_name, net_dev); \ } @@ -133,6 +136,7 @@ ssize_t batadv_show_##_name(struct kobject *kobj, \ struct attribute *attr, char *buff) \ { \ struct batadv_priv *bat_priv = batadv_kobj_to_batpriv(kobj); \ + \ return sprintf(buff, "%s\n", \ atomic_read(&bat_priv->_name) == 0 ? \ "disabled" : "enabled"); \ @@ -155,6 +159,7 @@ ssize_t batadv_store_##_name(struct kobject *kobj, \ { \ struct net_device *net_dev = batadv_kobj_to_netdev(kobj); \ struct batadv_priv *bat_priv = netdev_priv(net_dev); \ + \ return __batadv_store_uint_attr(buff, count, _min, _max, \ _post_func, attr, \ &bat_priv->_name, net_dev); \ @@ -165,6 +170,7 @@ ssize_t batadv_show_##_name(struct kobject *kobj, \ struct attribute *attr, char *buff) \ { \ struct batadv_priv *bat_priv = batadv_kobj_to_batpriv(kobj); \ + \ return sprintf(buff, "%i\n", atomic_read(&bat_priv->_name)); \ } \ @@ -188,6 +194,7 @@ ssize_t batadv_store_vlan_##_name(struct kobject *kobj, \ size_t res = __batadv_store_bool_attr(buff, count, _post_func, \ attr, &vlan->_name, \ bat_priv->soft_iface); \ + \ batadv_softif_vlan_free_ref(vlan); \ return res; \ } @@ -202,6 +209,7 @@ ssize_t batadv_show_vlan_##_name(struct kobject *kobj, \ size_t res = sprintf(buff, "%s\n", \ atomic_read(&vlan->_name) == 0 ? \ "disabled" : "enabled"); \ + \ batadv_softif_vlan_free_ref(vlan); \ return res; \ } @@ -324,12 +332,14 @@ static ssize_t batadv_show_bat_algo(struct kobject *kobj, struct attribute *attr, char *buff) { struct batadv_priv *bat_priv = batadv_kobj_to_batpriv(kobj); + return sprintf(buff, "%s\n", bat_priv->bat_algo_ops->name); } static void batadv_post_gw_reselect(struct net_device *net_dev) { struct batadv_priv *bat_priv = netdev_priv(net_dev); + batadv_gw_reselect(bat_priv); } diff --git a/net/bridge/Makefile b/net/bridge/Makefile index e85498b2f16..8590b942bff 100644 --- a/net/bridge/Makefile +++ b/net/bridge/Makefile @@ -5,7 +5,7 @@ obj-$(CONFIG_BRIDGE) += bridge.o bridge-y := br.o br_device.o br_fdb.o br_forward.o br_if.o br_input.o \ - br_ioctl.o br_notify.o br_stp.o br_stp_bpdu.o \ + br_ioctl.o br_stp.o br_stp_bpdu.o \ br_stp_if.o br_stp_timer.o br_netlink.o bridge-$(CONFIG_SYSFS) += br_sysfs_if.o br_sysfs_br.o @@ -16,4 +16,4 @@ bridge-$(CONFIG_BRIDGE_IGMP_SNOOPING) += br_multicast.o br_mdb.o bridge-$(CONFIG_BRIDGE_VLAN_FILTERING) += br_vlan.o -obj-$(CONFIG_BRIDGE_NF_EBTABLES) += netfilter/ +obj-$(CONFIG_NETFILTER) += netfilter/ diff --git a/net/bridge/br.c b/net/bridge/br.c index 19311aafcf5..1a755a1e541 100644 --- a/net/bridge/br.c +++ b/net/bridge/br.c @@ -22,6 +22,104 @@ #include "br_private.h" +/* + * Handle changes in state of network devices enslaved to a bridge. + * + * Note: don't care about up/down if bridge itself is down, because + * port state is checked when bridge is brought up. + */ +static int br_device_event(struct notifier_block *unused, unsigned long event, void *ptr) +{ + struct net_device *dev = netdev_notifier_info_to_dev(ptr); + struct net_bridge_port *p; + struct net_bridge *br; + bool changed_addr; + int err; + + /* register of bridge completed, add sysfs entries */ + if ((dev->priv_flags & IFF_EBRIDGE) && event == NETDEV_REGISTER) { + br_sysfs_addbr(dev); + return NOTIFY_DONE; + } + + /* not a port of a bridge */ + p = br_port_get_rtnl(dev); + if (!p) + return NOTIFY_DONE; + + br = p->br; + + switch (event) { + case NETDEV_CHANGEMTU: + dev_set_mtu(br->dev, br_min_mtu(br)); + break; + + case NETDEV_CHANGEADDR: + spin_lock_bh(&br->lock); + br_fdb_changeaddr(p, dev->dev_addr); + changed_addr = br_stp_recalculate_bridge_id(br); + spin_unlock_bh(&br->lock); + + if (changed_addr) + call_netdevice_notifiers(NETDEV_CHANGEADDR, br->dev); + + break; + + case NETDEV_CHANGE: + br_port_carrier_check(p); + break; + + case NETDEV_FEAT_CHANGE: + netdev_update_features(br->dev); + break; + + case NETDEV_DOWN: + spin_lock_bh(&br->lock); + if (br->dev->flags & IFF_UP) + br_stp_disable_port(p); + spin_unlock_bh(&br->lock); + break; + + case NETDEV_UP: + if (netif_running(br->dev) && netif_oper_up(dev)) { + spin_lock_bh(&br->lock); + br_stp_enable_port(p); + spin_unlock_bh(&br->lock); + } + break; + + case NETDEV_UNREGISTER: + br_del_if(br, dev); + break; + + case NETDEV_CHANGENAME: + err = br_sysfs_renameif(p); + if (err) + return notifier_from_errno(err); + break; + + case NETDEV_PRE_TYPE_CHANGE: + /* Forbid underlaying device to change its type. */ + return NOTIFY_BAD; + + case NETDEV_RESEND_IGMP: + /* Propagate to master device */ + call_netdevice_notifiers(event, br->dev); + break; + } + + /* Events that may cause spanning tree to refresh */ + if (event == NETDEV_CHANGEADDR || event == NETDEV_UP || + event == NETDEV_CHANGE || event == NETDEV_DOWN) + br_ifinfo_notify(RTM_NEWLINK, p); + + return NOTIFY_DONE; +} + +static struct notifier_block br_device_notifier = { + .notifier_call = br_device_event +}; + static void __net_exit br_net_exit(struct net *net) { struct net_device *dev; diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 3e2da2cb72d..d77e2f0ff0e 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -112,6 +112,12 @@ static void br_dev_set_multicast_list(struct net_device *dev) { } +static void br_dev_change_rx_flags(struct net_device *dev, int change) +{ + if (change & IFF_PROMISC) + br_manage_promisc(netdev_priv(dev)); +} + static int br_dev_stop(struct net_device *dev) { struct net_bridge *br = netdev_priv(dev); @@ -309,6 +315,7 @@ static const struct net_device_ops br_netdev_ops = { .ndo_get_stats64 = br_get_stats64, .ndo_set_mac_address = br_set_mac_address, .ndo_set_rx_mode = br_dev_set_multicast_list, + .ndo_change_rx_flags = br_dev_change_rx_flags, .ndo_change_mtu = br_change_mtu, .ndo_do_ioctl = br_dev_ioctl, #ifdef CONFIG_NET_POLL_CONTROLLER @@ -348,7 +355,7 @@ void br_dev_setup(struct net_device *dev) dev->netdev_ops = &br_netdev_ops; dev->destructor = br_dev_free; - SET_ETHTOOL_OPS(dev, &br_ethtool_ops); + dev->ethtool_ops = &br_ethtool_ops; SET_NETDEV_DEVTYPE(dev, &br_type); dev->tx_queue_len = 0; dev->priv_flags = IFF_EBRIDGE; diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index 9203d5a1943..b524c36c127 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -85,8 +85,58 @@ static void fdb_rcu_free(struct rcu_head *head) kmem_cache_free(br_fdb_cache, ent); } +/* When a static FDB entry is added, the mac address from the entry is + * added to the bridge private HW address list and all required ports + * are then updated with the new information. + * Called under RTNL. + */ +static void fdb_add_hw(struct net_bridge *br, const unsigned char *addr) +{ + int err; + struct net_bridge_port *p, *tmp; + + ASSERT_RTNL(); + + list_for_each_entry(p, &br->port_list, list) { + if (!br_promisc_port(p)) { + err = dev_uc_add(p->dev, addr); + if (err) + goto undo; + } + } + + return; +undo: + list_for_each_entry(tmp, &br->port_list, list) { + if (tmp == p) + break; + if (!br_promisc_port(tmp)) + dev_uc_del(tmp->dev, addr); + } +} + +/* When a static FDB entry is deleted, the HW address from that entry is + * also removed from the bridge private HW address list and updates all + * the ports with needed information. + * Called under RTNL. + */ +static void fdb_del_hw(struct net_bridge *br, const unsigned char *addr) +{ + struct net_bridge_port *p; + + ASSERT_RTNL(); + + list_for_each_entry(p, &br->port_list, list) { + if (!br_promisc_port(p)) + dev_uc_del(p->dev, addr); + } +} + static void fdb_delete(struct net_bridge *br, struct net_bridge_fdb_entry *f) { + if (f->is_static) + fdb_del_hw(br, f->addr.addr); + hlist_del_rcu(&f->hlist); fdb_notify(br, f, RTM_DELNEIGH); call_rcu(&f->rcu, fdb_rcu_free); @@ -466,6 +516,7 @@ static int fdb_insert(struct net_bridge *br, struct net_bridge_port *source, return -ENOMEM; fdb->is_local = fdb->is_static = 1; + fdb_add_hw(br, addr); fdb_notify(br, fdb, RTM_NEWNEIGH); return 0; } @@ -487,6 +538,7 @@ void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source, { struct hlist_head *head = &br->hash[br_mac_hash(addr, vid)]; struct net_bridge_fdb_entry *fdb; + bool fdb_modified = false; /* some users want to always flood. */ if (hold_time(br) == 0) @@ -507,10 +559,15 @@ void br_fdb_update(struct net_bridge *br, struct net_bridge_port *source, source->dev->name); } else { /* fastpath: update of existing entry */ - fdb->dst = source; + if (unlikely(source != fdb->dst)) { + fdb->dst = source; + fdb_modified = true; + } fdb->updated = jiffies; if (unlikely(added_by_user)) fdb->added_by_user = 1; + if (unlikely(fdb_modified)) + fdb_notify(br, fdb, RTM_NEWNEIGH); } } else { spin_lock(&br->hash_lock); @@ -565,6 +622,8 @@ static int fdb_fill_info(struct sk_buff *skb, const struct net_bridge *br, if (nla_put(skb, NDA_LLADDR, ETH_ALEN, &fdb->addr)) goto nla_put_failure; + if (nla_put_u32(skb, NDA_MASTER, br->dev->ifindex)) + goto nla_put_failure; ci.ndm_used = jiffies_to_clock_t(now - fdb->used); ci.ndm_confirmed = 0; ci.ndm_updated = jiffies_to_clock_t(now - fdb->updated); @@ -586,6 +645,7 @@ static inline size_t fdb_nlmsg_size(void) { return NLMSG_ALIGN(sizeof(struct ndmsg)) + nla_total_size(ETH_ALEN) /* NDA_LLADDR */ + + nla_total_size(sizeof(u32)) /* NDA_MASTER */ + nla_total_size(sizeof(u16)) /* NDA_VLAN */ + nla_total_size(sizeof(struct nda_cacheinfo)); } @@ -678,13 +738,25 @@ static int fdb_add_entry(struct net_bridge_port *source, const __u8 *addr, } if (fdb_to_nud(fdb) != state) { - if (state & NUD_PERMANENT) - fdb->is_local = fdb->is_static = 1; - else if (state & NUD_NOARP) { + if (state & NUD_PERMANENT) { + fdb->is_local = 1; + if (!fdb->is_static) { + fdb->is_static = 1; + fdb_add_hw(br, addr); + } + } else if (state & NUD_NOARP) { + fdb->is_local = 0; + if (!fdb->is_static) { + fdb->is_static = 1; + fdb_add_hw(br, addr); + } + } else { fdb->is_local = 0; - fdb->is_static = 1; - } else - fdb->is_local = fdb->is_static = 0; + if (fdb->is_static) { + fdb->is_static = 0; + fdb_del_hw(br, addr); + } + } modified = true; } @@ -874,3 +946,59 @@ int br_fdb_delete(struct ndmsg *ndm, struct nlattr *tb[], out: return err; } + +int br_fdb_sync_static(struct net_bridge *br, struct net_bridge_port *p) +{ + struct net_bridge_fdb_entry *fdb, *tmp; + int i; + int err; + + ASSERT_RTNL(); + + for (i = 0; i < BR_HASH_SIZE; i++) { + hlist_for_each_entry(fdb, &br->hash[i], hlist) { + /* We only care for static entries */ + if (!fdb->is_static) + continue; + + err = dev_uc_add(p->dev, fdb->addr.addr); + if (err) + goto rollback; + } + } + return 0; + +rollback: + for (i = 0; i < BR_HASH_SIZE; i++) { + hlist_for_each_entry(tmp, &br->hash[i], hlist) { + /* If we reached the fdb that failed, we can stop */ + if (tmp == fdb) + break; + + /* We only care for static entries */ + if (!tmp->is_static) + continue; + + dev_uc_del(p->dev, tmp->addr.addr); + } + } + return err; +} + +void br_fdb_unsync_static(struct net_bridge *br, struct net_bridge_port *p) +{ + struct net_bridge_fdb_entry *fdb; + int i; + + ASSERT_RTNL(); + + for (i = 0; i < BR_HASH_SIZE; i++) { + hlist_for_each_entry_rcu(fdb, &br->hash[i], hlist) { + /* We only care for static entries */ + if (!fdb->is_static) + continue; + + dev_uc_del(p->dev, fdb->addr.addr); + } + } +} diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index 5262b8617eb..3eca3fdf8fe 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -85,6 +85,111 @@ void br_port_carrier_check(struct net_bridge_port *p) spin_unlock_bh(&br->lock); } +static void br_port_set_promisc(struct net_bridge_port *p) +{ + int err = 0; + + if (br_promisc_port(p)) + return; + + err = dev_set_promiscuity(p->dev, 1); + if (err) + return; + + br_fdb_unsync_static(p->br, p); + p->flags |= BR_PROMISC; +} + +static void br_port_clear_promisc(struct net_bridge_port *p) +{ + int err; + + /* Check if the port is already non-promisc or if it doesn't + * support UNICAST filtering. Without unicast filtering support + * we'll end up re-enabling promisc mode anyway, so just check for + * it here. + */ + if (!br_promisc_port(p) || !(p->dev->priv_flags & IFF_UNICAST_FLT)) + return; + + /* Since we'll be clearing the promisc mode, program the port + * first so that we don't have interruption in traffic. + */ + err = br_fdb_sync_static(p->br, p); + if (err) + return; + + dev_set_promiscuity(p->dev, -1); + p->flags &= ~BR_PROMISC; +} + +/* When a port is added or removed or when certain port flags + * change, this function is called to automatically manage + * promiscuity setting of all the bridge ports. We are always called + * under RTNL so can skip using rcu primitives. + */ +void br_manage_promisc(struct net_bridge *br) +{ + struct net_bridge_port *p; + bool set_all = false; + + /* If vlan filtering is disabled or bridge interface is placed + * into promiscuous mode, place all ports in promiscuous mode. + */ + if ((br->dev->flags & IFF_PROMISC) || !br_vlan_enabled(br)) + set_all = true; + + list_for_each_entry(p, &br->port_list, list) { + if (set_all) { + br_port_set_promisc(p); + } else { + /* If the number of auto-ports is <= 1, then all other + * ports will have their output configuration + * statically specified through fdbs. Since ingress + * on the auto-port becomes forwarding/egress to other + * ports and egress configuration is statically known, + * we can say that ingress configuration of the + * auto-port is also statically known. + * This lets us disable promiscuous mode and write + * this config to hw. + */ + if (br->auto_cnt == 0 || + (br->auto_cnt == 1 && br_auto_port(p))) + br_port_clear_promisc(p); + else + br_port_set_promisc(p); + } + } +} + +static void nbp_update_port_count(struct net_bridge *br) +{ + struct net_bridge_port *p; + u32 cnt = 0; + + list_for_each_entry(p, &br->port_list, list) { + if (br_auto_port(p)) + cnt++; + } + if (br->auto_cnt != cnt) { + br->auto_cnt = cnt; + br_manage_promisc(br); + } +} + +static void nbp_delete_promisc(struct net_bridge_port *p) +{ + /* If port is currently promiscuous, unset promiscuity. + * Otherwise, it is a static port so remove all addresses + * from it. + */ + dev_set_allmulti(p->dev, -1); + if (br_promisc_port(p)) + dev_set_promiscuity(p->dev, -1); + else + br_fdb_unsync_static(p->br, p); +} + static void release_nbp(struct kobject *kobj) { struct net_bridge_port *p @@ -133,7 +238,7 @@ static void del_nbp(struct net_bridge_port *p) sysfs_remove_link(br->ifobj, p->dev->name); - dev_set_promiscuity(dev, -1); + nbp_delete_promisc(p); spin_lock_bh(&br->lock); br_stp_disable_port(p); @@ -141,10 +246,11 @@ static void del_nbp(struct net_bridge_port *p) br_ifinfo_notify(RTM_DELLINK, p); + list_del_rcu(&p->list); + nbp_vlan_flush(p); br_fdb_delete_by_port(br, p, 1); - - list_del_rcu(&p->list); + nbp_update_port_count(br); dev->priv_flags &= ~IFF_BRIDGE_PORT; @@ -353,7 +459,7 @@ int br_add_if(struct net_bridge *br, struct net_device *dev) call_netdevice_notifiers(NETDEV_JOIN, dev); - err = dev_set_promiscuity(dev, 1); + err = dev_set_allmulti(dev, 1); if (err) goto put_back; @@ -384,6 +490,8 @@ int br_add_if(struct net_bridge *br, struct net_device *dev) list_add_rcu(&p->list, &br->port_list); + nbp_update_port_count(br); + netdev_update_features(br->dev); if (br->dev->needed_headroom < dev->needed_headroom) @@ -421,7 +529,7 @@ err2: kobject_put(&p->kobj); p = NULL; /* kobject_put frees */ err1: - dev_set_promiscuity(dev, -1); + dev_set_allmulti(dev, -1); put_back: dev_put(dev); kfree(p); @@ -455,3 +563,11 @@ int br_del_if(struct net_bridge *br, struct net_device *dev) return 0; } + +void br_port_flags_change(struct net_bridge_port *p, unsigned long mask) +{ + struct net_bridge *br = p->br; + + if (mask & BR_AUTO_MASK) + nbp_update_port_count(br); +} diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c index 7985deaff52..04d6348fd53 100644 --- a/net/bridge/br_input.c +++ b/net/bridge/br_input.c @@ -147,8 +147,8 @@ static int br_handle_local_finish(struct sk_buff *skb) struct net_bridge_port *p = br_port_get_rcu(skb->dev); u16 vid = 0; - br_vlan_get_tag(skb, &vid); - if (p->flags & BR_LEARNING) + /* check if vlan is allowed, to avoid spoofing */ + if (p->flags & BR_LEARNING && br_should_learn(p, skb, &vid)) br_fdb_update(p->br, p, eth_hdr(skb)->h_source, vid, false); return 0; /* process further */ } diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index 80e1b0f60a3..a615264cf01 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -535,7 +535,7 @@ static struct net_device *brnf_get_logical_dev(struct sk_buff *skb, const struct if (brnf_pass_vlan_indev == 0 || !vlan_tx_tag_present(skb)) return br; - vlan = __vlan_find_dev_deep(br, skb->vlan_proto, + vlan = __vlan_find_dev_deep_rcu(br, skb->vlan_proto, vlan_tx_tag_get(skb) & VLAN_VID_MASK); return vlan ? vlan : br; @@ -859,12 +859,12 @@ static unsigned int br_nf_forward_arp(const struct nf_hook_ops *ops, return NF_STOLEN; } -#if IS_ENABLED(CONFIG_NF_CONNTRACK_IPV4) +#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV4) static int br_nf_dev_queue_xmit(struct sk_buff *skb) { int ret; - if (skb->nfct != NULL && skb->protocol == htons(ETH_P_IP) && + if (skb->protocol == htons(ETH_P_IP) && skb->len + nf_bridge_mtu_reduction(skb) > skb->dev->mtu && !skb_is_gso(skb)) { if (br_parse_ip_options(skb)) diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index e74b6d530cb..26edb518b83 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -328,6 +328,7 @@ static void br_set_port_flag(struct net_bridge_port *p, struct nlattr *tb[], static int br_setport(struct net_bridge_port *p, struct nlattr *tb[]) { int err; + unsigned long old_flags = p->flags; br_set_port_flag(p, tb, IFLA_BRPORT_MODE, BR_HAIRPIN_MODE); br_set_port_flag(p, tb, IFLA_BRPORT_GUARD, BR_BPDU_GUARD); @@ -353,6 +354,8 @@ static int br_setport(struct net_bridge_port *p, struct nlattr *tb[]) if (err) return err; } + + br_port_flags_change(p, old_flags ^ p->flags); return 0; } @@ -445,6 +448,20 @@ static int br_validate(struct nlattr *tb[], struct nlattr *data[]) return 0; } +static int br_dev_newlink(struct net *src_net, struct net_device *dev, + struct nlattr *tb[], struct nlattr *data[]) +{ + struct net_bridge *br = netdev_priv(dev); + + if (tb[IFLA_ADDRESS]) { + spin_lock_bh(&br->lock); + br_stp_change_bridge_id(br, nla_data(tb[IFLA_ADDRESS])); + spin_unlock_bh(&br->lock); + } + + return register_netdevice(dev); +} + static size_t br_get_link_af_size(const struct net_device *dev) { struct net_port_vlans *pv; @@ -473,6 +490,7 @@ struct rtnl_link_ops br_link_ops __read_mostly = { .priv_size = sizeof(struct net_bridge), .setup = br_dev_setup, .validate = br_validate, + .newlink = br_dev_newlink, .dellink = br_dev_delete, }; diff --git a/net/bridge/br_notify.c b/net/bridge/br_notify.c deleted file mode 100644 index 2998dd1769a..00000000000 --- a/net/bridge/br_notify.c +++ /dev/null @@ -1,118 +0,0 @@ -/* - * Device event handling - * Linux ethernet bridge - * - * Authors: - * Lennert Buytenhek <buytenh@gnu.org> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#include <linux/kernel.h> -#include <linux/rtnetlink.h> -#include <net/net_namespace.h> - -#include "br_private.h" - -static int br_device_event(struct notifier_block *unused, unsigned long event, void *ptr); - -struct notifier_block br_device_notifier = { - .notifier_call = br_device_event -}; - -/* - * Handle changes in state of network devices enslaved to a bridge. - * - * Note: don't care about up/down if bridge itself is down, because - * port state is checked when bridge is brought up. - */ -static int br_device_event(struct notifier_block *unused, unsigned long event, void *ptr) -{ - struct net_device *dev = netdev_notifier_info_to_dev(ptr); - struct net_bridge_port *p; - struct net_bridge *br; - bool changed_addr; - int err; - - /* register of bridge completed, add sysfs entries */ - if ((dev->priv_flags & IFF_EBRIDGE) && event == NETDEV_REGISTER) { - br_sysfs_addbr(dev); - return NOTIFY_DONE; - } - - /* not a port of a bridge */ - p = br_port_get_rtnl(dev); - if (!p) - return NOTIFY_DONE; - - br = p->br; - - switch (event) { - case NETDEV_CHANGEMTU: - dev_set_mtu(br->dev, br_min_mtu(br)); - break; - - case NETDEV_CHANGEADDR: - spin_lock_bh(&br->lock); - br_fdb_changeaddr(p, dev->dev_addr); - changed_addr = br_stp_recalculate_bridge_id(br); - spin_unlock_bh(&br->lock); - - if (changed_addr) - call_netdevice_notifiers(NETDEV_CHANGEADDR, br->dev); - - break; - - case NETDEV_CHANGE: - br_port_carrier_check(p); - break; - - case NETDEV_FEAT_CHANGE: - netdev_update_features(br->dev); - break; - - case NETDEV_DOWN: - spin_lock_bh(&br->lock); - if (br->dev->flags & IFF_UP) - br_stp_disable_port(p); - spin_unlock_bh(&br->lock); - break; - - case NETDEV_UP: - if (netif_running(br->dev) && netif_oper_up(dev)) { - spin_lock_bh(&br->lock); - br_stp_enable_port(p); - spin_unlock_bh(&br->lock); - } - break; - - case NETDEV_UNREGISTER: - br_del_if(br, dev); - break; - - case NETDEV_CHANGENAME: - err = br_sysfs_renameif(p); - if (err) - return notifier_from_errno(err); - break; - - case NETDEV_PRE_TYPE_CHANGE: - /* Forbid underlaying device to change its type. */ - return NOTIFY_BAD; - - case NETDEV_RESEND_IGMP: - /* Propagate to master device */ - call_netdevice_notifiers(event, br->dev); - break; - } - - /* Events that may cause spanning tree to refresh */ - if (event == NETDEV_CHANGEADDR || event == NETDEV_UP || - event == NETDEV_CHANGE || event == NETDEV_DOWN) - br_ifinfo_notify(RTM_NEWLINK, p); - - return NOTIFY_DONE; -} diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 06811d79f89..bc17210d4c5 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -174,6 +174,8 @@ struct net_bridge_port #define BR_ADMIN_COST 0x00000010 #define BR_LEARNING 0x00000020 #define BR_FLOOD 0x00000040 +#define BR_AUTO_MASK (BR_FLOOD | BR_LEARNING) +#define BR_PROMISC 0x00000080 #ifdef CONFIG_BRIDGE_IGMP_SNOOPING struct bridge_mcast_query ip4_query; @@ -198,6 +200,9 @@ struct net_bridge_port #endif }; +#define br_auto_port(p) ((p)->flags & BR_AUTO_MASK) +#define br_promisc_port(p) ((p)->flags & BR_PROMISC) + #define br_port_exists(dev) (dev->priv_flags & IFF_BRIDGE_PORT) static inline struct net_bridge_port *br_port_get_rcu(const struct net_device *dev) @@ -290,6 +295,7 @@ struct net_bridge struct timer_list topology_change_timer; struct timer_list gc_timer; struct kobject *ifobj; + u32 auto_cnt; #ifdef CONFIG_BRIDGE_VLAN_FILTERING u8 vlan_enabled; struct net_port_vlans __rcu *vlan_info; @@ -327,8 +333,6 @@ struct br_input_skb_cb { #define br_debug(br, format, args...) \ pr_debug("%s: " format, (br)->dev->name, ##args) -extern struct notifier_block br_device_notifier; - /* called under bridge lock */ static inline int br_is_root_bridge(const struct net_bridge *br) { @@ -395,6 +399,8 @@ int br_fdb_add(struct ndmsg *nlh, struct nlattr *tb[], struct net_device *dev, const unsigned char *addr, u16 nlh_flags); int br_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb, struct net_device *dev, int idx); +int br_fdb_sync_static(struct net_bridge *br, struct net_bridge_port *p); +void br_fdb_unsync_static(struct net_bridge *br, struct net_bridge_port *p); /* br_forward.c */ void br_deliver(const struct net_bridge_port *to, struct sk_buff *skb); @@ -415,6 +421,8 @@ int br_del_if(struct net_bridge *br, struct net_device *dev); int br_min_mtu(const struct net_bridge *br); netdev_features_t br_features_recompute(struct net_bridge *br, netdev_features_t features); +void br_port_flags_change(struct net_bridge_port *port, unsigned long mask); +void br_manage_promisc(struct net_bridge *br); /* br_input.c */ int br_handle_frame_finish(struct sk_buff *skb); @@ -581,6 +589,7 @@ bool br_allowed_ingress(struct net_bridge *br, struct net_port_vlans *v, struct sk_buff *skb, u16 *vid); bool br_allowed_egress(struct net_bridge *br, const struct net_port_vlans *v, const struct sk_buff *skb); +bool br_should_learn(struct net_bridge_port *p, struct sk_buff *skb, u16 *vid); struct sk_buff *br_handle_vlan(struct net_bridge *br, const struct net_port_vlans *v, struct sk_buff *skb); @@ -632,6 +641,10 @@ static inline u16 br_get_pvid(const struct net_port_vlans *v) return v->pvid ?: VLAN_N_VID; } +static inline int br_vlan_enabled(struct net_bridge *br) +{ + return br->vlan_enabled; +} #else static inline bool br_allowed_ingress(struct net_bridge *br, struct net_port_vlans *v, @@ -648,6 +661,12 @@ static inline bool br_allowed_egress(struct net_bridge *br, return true; } +static inline bool br_should_learn(struct net_bridge_port *p, + struct sk_buff *skb, u16 *vid) +{ + return true; +} + static inline struct sk_buff *br_handle_vlan(struct net_bridge *br, const struct net_port_vlans *v, struct sk_buff *skb) @@ -712,6 +731,11 @@ static inline u16 br_get_pvid(const struct net_port_vlans *v) { return VLAN_N_VID; /* Returns invalid vid */ } + +static inline int br_vlan_enabled(struct net_bridge *br) +{ + return 0; +} #endif /* br_netfilter.c */ diff --git a/net/bridge/br_sysfs_if.c b/net/bridge/br_sysfs_if.c index dd595bd7fa8..e561cd59b8a 100644 --- a/net/bridge/br_sysfs_if.c +++ b/net/bridge/br_sysfs_if.c @@ -41,20 +41,30 @@ static ssize_t show_##_name(struct net_bridge_port *p, char *buf) \ } \ static int store_##_name(struct net_bridge_port *p, unsigned long v) \ { \ - unsigned long flags = p->flags; \ - if (v) \ - flags |= _mask; \ - else \ - flags &= ~_mask; \ - if (flags != p->flags) { \ - p->flags = flags; \ - br_ifinfo_notify(RTM_NEWLINK, p); \ - } \ - return 0; \ + return store_flag(p, v, _mask); \ } \ static BRPORT_ATTR(_name, S_IRUGO | S_IWUSR, \ show_##_name, store_##_name) +static int store_flag(struct net_bridge_port *p, unsigned long v, + unsigned long mask) +{ + unsigned long flags; + + flags = p->flags; + + if (v) + flags |= mask; + else + flags &= ~mask; + + if (flags != p->flags) { + p->flags = flags; + br_port_flags_change(p, mask); + br_ifinfo_notify(RTM_NEWLINK, p); + } + return 0; +} static ssize_t show_path_cost(struct net_bridge_port *p, char *buf) { diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c index 4a371610278..fcc95390f86 100644 --- a/net/bridge/br_vlan.c +++ b/net/bridge/br_vlan.c @@ -241,6 +241,34 @@ bool br_allowed_egress(struct net_bridge *br, return false; } +/* Called under RCU */ +bool br_should_learn(struct net_bridge_port *p, struct sk_buff *skb, u16 *vid) +{ + struct net_bridge *br = p->br; + struct net_port_vlans *v; + + if (!br->vlan_enabled) + return true; + + v = rcu_dereference(p->vlan_info); + if (!v) + return false; + + br_vlan_get_tag(skb, vid); + if (!*vid) { + *vid = br_get_pvid(v); + if (*vid == VLAN_N_VID) + return false; + + return true; + } + + if (test_bit(*vid, v->vlan_bitmap)) + return true; + + return false; +} + /* Must be protected by RTNL. * Must be called with vid in range from 1 to 4094 inclusive. */ @@ -332,6 +360,7 @@ int br_vlan_filter_toggle(struct net_bridge *br, unsigned long val) goto unlock; br->vlan_enabled = val; + br_manage_promisc(br); unlock: rtnl_unlock(); diff --git a/net/bridge/netfilter/Kconfig b/net/bridge/netfilter/Kconfig index 5ca74a0e595..629dc77874a 100644 --- a/net/bridge/netfilter/Kconfig +++ b/net/bridge/netfilter/Kconfig @@ -2,14 +2,23 @@ # Bridge netfilter configuration # # -config NF_TABLES_BRIDGE - depends on NF_TABLES +menuconfig NF_TABLES_BRIDGE + depends on BRIDGE && NETFILTER && NF_TABLES tristate "Ethernet Bridge nf_tables support" +if NF_TABLES_BRIDGE + +config NFT_BRIDGE_META + tristate "Netfilter nf_table bridge meta support" + depends on NFT_META + help + Add support for bridge dedicated meta key. + +endif # NF_TABLES_BRIDGE + menuconfig BRIDGE_NF_EBTABLES tristate "Ethernet Bridge tables (ebtables) support" - depends on BRIDGE && NETFILTER - select NETFILTER_XTABLES + depends on BRIDGE && NETFILTER && NETFILTER_XTABLES help ebtables is a general, extensible frame/packet identification framework. Say 'Y' or 'M' here if you want to do Ethernet diff --git a/net/bridge/netfilter/Makefile b/net/bridge/netfilter/Makefile index ea7629f58b3..6f2f3943d66 100644 --- a/net/bridge/netfilter/Makefile +++ b/net/bridge/netfilter/Makefile @@ -3,6 +3,7 @@ # obj-$(CONFIG_NF_TABLES_BRIDGE) += nf_tables_bridge.o +obj-$(CONFIG_NFT_BRIDGE_META) += nft_meta_bridge.o obj-$(CONFIG_BRIDGE_NF_EBTABLES) += ebtables.o diff --git a/net/bridge/netfilter/nft_meta_bridge.c b/net/bridge/netfilter/nft_meta_bridge.c new file mode 100644 index 00000000000..4f02109d708 --- /dev/null +++ b/net/bridge/netfilter/nft_meta_bridge.c @@ -0,0 +1,139 @@ +/* + * Copyright (c) 2014 Intel Corporation + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ + +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/netlink.h> +#include <linux/netfilter.h> +#include <linux/netfilter/nf_tables.h> +#include <net/netfilter/nf_tables.h> +#include <net/netfilter/nft_meta.h> + +#include "../br_private.h" + +static void nft_meta_bridge_get_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt) +{ + const struct nft_meta *priv = nft_expr_priv(expr); + const struct net_device *in = pkt->in, *out = pkt->out; + struct nft_data *dest = &data[priv->dreg]; + const struct net_bridge_port *p; + + switch (priv->key) { + case NFT_META_BRI_IIFNAME: + if (in == NULL || (p = br_port_get_rcu(in)) == NULL) + goto err; + break; + case NFT_META_BRI_OIFNAME: + if (out == NULL || (p = br_port_get_rcu(out)) == NULL) + goto err; + break; + default: + goto out; + } + + strncpy((char *)dest->data, p->br->dev->name, sizeof(dest->data)); + return; +out: + return nft_meta_get_eval(expr, data, pkt); +err: + data[NFT_REG_VERDICT].verdict = NFT_BREAK; +} + +static int nft_meta_bridge_get_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) +{ + struct nft_meta *priv = nft_expr_priv(expr); + int err; + + priv->key = ntohl(nla_get_be32(tb[NFTA_META_KEY])); + switch (priv->key) { + case NFT_META_BRI_IIFNAME: + case NFT_META_BRI_OIFNAME: + break; + default: + return nft_meta_get_init(ctx, expr, tb); + } + + priv->dreg = ntohl(nla_get_be32(tb[NFTA_META_DREG])); + err = nft_validate_output_register(priv->dreg); + if (err < 0) + return err; + + err = nft_validate_data_load(ctx, priv->dreg, NULL, NFT_DATA_VALUE); + if (err < 0) + return err; + + return 0; +} + +static struct nft_expr_type nft_meta_bridge_type; +static const struct nft_expr_ops nft_meta_bridge_get_ops = { + .type = &nft_meta_bridge_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_meta)), + .eval = nft_meta_bridge_get_eval, + .init = nft_meta_bridge_get_init, + .dump = nft_meta_get_dump, +}; + +static const struct nft_expr_ops nft_meta_bridge_set_ops = { + .type = &nft_meta_bridge_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_meta)), + .eval = nft_meta_set_eval, + .init = nft_meta_set_init, + .dump = nft_meta_set_dump, +}; + +static const struct nft_expr_ops * +nft_meta_bridge_select_ops(const struct nft_ctx *ctx, + const struct nlattr * const tb[]) +{ + if (tb[NFTA_META_KEY] == NULL) + return ERR_PTR(-EINVAL); + + if (tb[NFTA_META_DREG] && tb[NFTA_META_SREG]) + return ERR_PTR(-EINVAL); + + if (tb[NFTA_META_DREG]) + return &nft_meta_bridge_get_ops; + + if (tb[NFTA_META_SREG]) + return &nft_meta_bridge_set_ops; + + return ERR_PTR(-EINVAL); +} + +static struct nft_expr_type nft_meta_bridge_type __read_mostly = { + .family = NFPROTO_BRIDGE, + .name = "meta", + .select_ops = &nft_meta_bridge_select_ops, + .policy = nft_meta_policy, + .maxattr = NFTA_META_MAX, + .owner = THIS_MODULE, +}; + +static int __init nft_meta_bridge_module_init(void) +{ + return nft_register_expr(&nft_meta_bridge_type); +} + +static void __exit nft_meta_bridge_module_exit(void) +{ + nft_unregister_expr(&nft_meta_bridge_type); +} + +module_init(nft_meta_bridge_module_init); +module_exit(nft_meta_bridge_module_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>"); +MODULE_ALIAS_NFT_AF_EXPR(AF_BRIDGE, "meta"); diff --git a/net/can/af_can.c b/net/can/af_can.c index a27f8aad9e9..ce82337521f 100644 --- a/net/can/af_can.c +++ b/net/can/af_can.c @@ -338,6 +338,29 @@ static struct dev_rcv_lists *find_dev_rcv_lists(struct net_device *dev) } /** + * effhash - hash function for 29 bit CAN identifier reduction + * @can_id: 29 bit CAN identifier + * + * Description: + * To reduce the linear traversal in one linked list of _single_ EFF CAN + * frame subscriptions the 29 bit identifier is mapped to 10 bits. + * (see CAN_EFF_RCV_HASH_BITS definition) + * + * Return: + * Hash value from 0x000 - 0x3FF ( enforced by CAN_EFF_RCV_HASH_BITS mask ) + */ +static unsigned int effhash(canid_t can_id) +{ + unsigned int hash; + + hash = can_id; + hash ^= can_id >> CAN_EFF_RCV_HASH_BITS; + hash ^= can_id >> (2 * CAN_EFF_RCV_HASH_BITS); + + return hash & ((1 << CAN_EFF_RCV_HASH_BITS) - 1); +} + +/** * find_rcv_list - determine optimal filterlist inside device filter struct * @can_id: pointer to CAN identifier of a given can_filter * @mask: pointer to CAN mask of a given can_filter @@ -400,10 +423,8 @@ static struct hlist_head *find_rcv_list(canid_t *can_id, canid_t *mask, !(*can_id & CAN_RTR_FLAG)) { if (*can_id & CAN_EFF_FLAG) { - if (*mask == (CAN_EFF_MASK | CAN_EFF_RTR_FLAGS)) { - /* RFC: a future use-case for hash-tables? */ - return &d->rx[RX_EFF]; - } + if (*mask == (CAN_EFF_MASK | CAN_EFF_RTR_FLAGS)) + return &d->rx_eff[effhash(*can_id)]; } else { if (*mask == (CAN_SFF_MASK | CAN_EFF_RTR_FLAGS)) return &d->rx_sff[*can_id]; @@ -632,7 +653,7 @@ static int can_rcv_filter(struct dev_rcv_lists *d, struct sk_buff *skb) return matches; if (can_id & CAN_EFF_FLAG) { - hlist_for_each_entry_rcu(r, &d->rx[RX_EFF], list) { + hlist_for_each_entry_rcu(r, &d->rx_eff[effhash(can_id)], list) { if (r->can_id == can_id) { deliver(skb, r); matches++; diff --git a/net/can/af_can.h b/net/can/af_can.h index 6de58b40535..fca0fe9fc45 100644 --- a/net/can/af_can.h +++ b/net/can/af_can.h @@ -59,12 +59,17 @@ struct receiver { char *ident; }; -enum { RX_ERR, RX_ALL, RX_FIL, RX_INV, RX_EFF, RX_MAX }; +#define CAN_SFF_RCV_ARRAY_SZ (1 << CAN_SFF_ID_BITS) +#define CAN_EFF_RCV_HASH_BITS 10 +#define CAN_EFF_RCV_ARRAY_SZ (1 << CAN_EFF_RCV_HASH_BITS) + +enum { RX_ERR, RX_ALL, RX_FIL, RX_INV, RX_MAX }; /* per device receive filters linked at dev->ml_priv */ struct dev_rcv_lists { struct hlist_head rx[RX_MAX]; - struct hlist_head rx_sff[0x800]; + struct hlist_head rx_sff[CAN_SFF_RCV_ARRAY_SZ]; + struct hlist_head rx_eff[CAN_EFF_RCV_ARRAY_SZ]; int remove_on_zero_entries; int entries; }; diff --git a/net/can/gw.c b/net/can/gw.c index ac31891967d..050a2110d43 100644 --- a/net/can/gw.c +++ b/net/can/gw.c @@ -804,7 +804,7 @@ static int cgw_create_job(struct sk_buff *skb, struct nlmsghdr *nlh) u8 limhops = 0; int err = 0; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if (nlmsg_len(nlh) < sizeof(*r)) @@ -893,7 +893,7 @@ static int cgw_remove_job(struct sk_buff *skb, struct nlmsghdr *nlh) u8 limhops = 0; int err = 0; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if (nlmsg_len(nlh) < sizeof(*r)) diff --git a/net/can/proc.c b/net/can/proc.c index b543470c8f8..1a19b985a86 100644 --- a/net/can/proc.c +++ b/net/can/proc.c @@ -80,7 +80,6 @@ static const char rx_list_name[][8] = { [RX_ALL] = "rx_all", [RX_FIL] = "rx_fil", [RX_INV] = "rx_inv", - [RX_EFF] = "rx_eff", }; /* @@ -389,25 +388,26 @@ static const struct file_operations can_rcvlist_proc_fops = { .release = single_release, }; -static inline void can_rcvlist_sff_proc_show_one(struct seq_file *m, - struct net_device *dev, - struct dev_rcv_lists *d) +static inline void can_rcvlist_proc_show_array(struct seq_file *m, + struct net_device *dev, + struct hlist_head *rcv_array, + unsigned int rcv_array_sz) { - int i; + unsigned int i; int all_empty = 1; /* check whether at least one list is non-empty */ - for (i = 0; i < 0x800; i++) - if (!hlist_empty(&d->rx_sff[i])) { + for (i = 0; i < rcv_array_sz; i++) + if (!hlist_empty(&rcv_array[i])) { all_empty = 0; break; } if (!all_empty) { can_print_recv_banner(m); - for (i = 0; i < 0x800; i++) { - if (!hlist_empty(&d->rx_sff[i])) - can_print_rcvlist(m, &d->rx_sff[i], dev); + for (i = 0; i < rcv_array_sz; i++) { + if (!hlist_empty(&rcv_array[i])) + can_print_rcvlist(m, &rcv_array[i], dev); } } else seq_printf(m, " (%s: no entry)\n", DNAME(dev)); @@ -425,12 +425,15 @@ static int can_rcvlist_sff_proc_show(struct seq_file *m, void *v) /* sff receive list for 'all' CAN devices (dev == NULL) */ d = &can_rx_alldev_list; - can_rcvlist_sff_proc_show_one(m, NULL, d); + can_rcvlist_proc_show_array(m, NULL, d->rx_sff, ARRAY_SIZE(d->rx_sff)); /* sff receive list for registered CAN devices */ for_each_netdev_rcu(&init_net, dev) { - if (dev->type == ARPHRD_CAN && dev->ml_priv) - can_rcvlist_sff_proc_show_one(m, dev, dev->ml_priv); + if (dev->type == ARPHRD_CAN && dev->ml_priv) { + d = dev->ml_priv; + can_rcvlist_proc_show_array(m, dev, d->rx_sff, + ARRAY_SIZE(d->rx_sff)); + } } rcu_read_unlock(); @@ -452,6 +455,49 @@ static const struct file_operations can_rcvlist_sff_proc_fops = { .release = single_release, }; + +static int can_rcvlist_eff_proc_show(struct seq_file *m, void *v) +{ + struct net_device *dev; + struct dev_rcv_lists *d; + + /* RX_EFF */ + seq_puts(m, "\nreceive list 'rx_eff':\n"); + + rcu_read_lock(); + + /* eff receive list for 'all' CAN devices (dev == NULL) */ + d = &can_rx_alldev_list; + can_rcvlist_proc_show_array(m, NULL, d->rx_eff, ARRAY_SIZE(d->rx_eff)); + + /* eff receive list for registered CAN devices */ + for_each_netdev_rcu(&init_net, dev) { + if (dev->type == ARPHRD_CAN && dev->ml_priv) { + d = dev->ml_priv; + can_rcvlist_proc_show_array(m, dev, d->rx_eff, + ARRAY_SIZE(d->rx_eff)); + } + } + + rcu_read_unlock(); + + seq_putc(m, '\n'); + return 0; +} + +static int can_rcvlist_eff_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, can_rcvlist_eff_proc_show, NULL); +} + +static const struct file_operations can_rcvlist_eff_proc_fops = { + .owner = THIS_MODULE, + .open = can_rcvlist_eff_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + /* * proc utility functions */ @@ -491,8 +537,8 @@ void can_init_proc(void) &can_rcvlist_proc_fops, (void *)RX_FIL); pde_rcvlist_inv = proc_create_data(CAN_PROC_RCVLIST_INV, 0644, can_dir, &can_rcvlist_proc_fops, (void *)RX_INV); - pde_rcvlist_eff = proc_create_data(CAN_PROC_RCVLIST_EFF, 0644, can_dir, - &can_rcvlist_proc_fops, (void *)RX_EFF); + pde_rcvlist_eff = proc_create(CAN_PROC_RCVLIST_EFF, 0644, can_dir, + &can_rcvlist_eff_proc_fops); pde_rcvlist_sff = proc_create(CAN_PROC_RCVLIST_SFF, 0644, can_dir, &can_rcvlist_sff_proc_fops); } diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c index dac7f9b9868..1948d592aa5 100644 --- a/net/ceph/messenger.c +++ b/net/ceph/messenger.c @@ -557,7 +557,7 @@ static int ceph_tcp_sendmsg(struct socket *sock, struct kvec *iov, return r; } -static int ceph_tcp_sendpage(struct socket *sock, struct page *page, +static int __ceph_tcp_sendpage(struct socket *sock, struct page *page, int offset, size_t size, bool more) { int flags = MSG_DONTWAIT | MSG_NOSIGNAL | (more ? MSG_MORE : MSG_EOR); @@ -570,6 +570,24 @@ static int ceph_tcp_sendpage(struct socket *sock, struct page *page, return ret; } +static int ceph_tcp_sendpage(struct socket *sock, struct page *page, + int offset, size_t size, bool more) +{ + int ret; + struct kvec iov; + + /* sendpage cannot properly handle pages with page_count == 0, + * we need to fallback to sendmsg if that's the case */ + if (page_count(page) >= 1) + return __ceph_tcp_sendpage(sock, page, offset, size, more); + + iov.iov_base = kmap(page) + offset; + iov.iov_len = size; + ret = ceph_tcp_sendmsg(sock, &iov, 1, size, more); + kunmap(page); + + return ret; +} /* * Shutdown/close the socket for the given connection. diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c index e632b5a52f5..c547e46084d 100644 --- a/net/ceph/osdmap.c +++ b/net/ceph/osdmap.c @@ -329,6 +329,11 @@ static struct crush_map *crush_decode(void *pbyval, void *end) dout("crush decode tunable chooseleaf_descend_once = %d", c->chooseleaf_descend_once); + ceph_decode_need(p, end, sizeof(u8), done); + c->chooseleaf_vary_r = ceph_decode_8(p); + dout("crush decode tunable chooseleaf_vary_r = %d", + c->chooseleaf_vary_r); + done: dout("crush_decode success\n"); return c; @@ -1548,8 +1553,10 @@ static void apply_primary_affinity(struct ceph_osdmap *osdmap, u32 pps, return; for (i = 0; i < len; i++) { - if (osds[i] != CRUSH_ITEM_NONE && - osdmap->osd_primary_affinity[i] != + int osd = osds[i]; + + if (osd != CRUSH_ITEM_NONE && + osdmap->osd_primary_affinity[osd] != CEPH_OSD_DEFAULT_PRIMARY_AFFINITY) { break; } @@ -1563,10 +1570,9 @@ static void apply_primary_affinity(struct ceph_osdmap *osdmap, u32 pps, * osd's pgs get rejected as primary. */ for (i = 0; i < len; i++) { - int osd; + int osd = osds[i]; u32 aff; - osd = osds[i]; if (osd == CRUSH_ITEM_NONE) continue; diff --git a/net/core/Makefile b/net/core/Makefile index 826b925aa45..71093d94ad2 100644 --- a/net/core/Makefile +++ b/net/core/Makefile @@ -9,7 +9,7 @@ obj-$(CONFIG_SYSCTL) += sysctl_net_core.o obj-y += dev.o ethtool.o dev_addr_lists.o dst.o netevent.o \ neighbour.o rtnetlink.o utils.o link_watch.o filter.o \ - sock_diag.o dev_ioctl.o + sock_diag.o dev_ioctl.o tso.o obj-$(CONFIG_XFRM) += flow.o obj-y += net-sysfs.o diff --git a/net/core/dev.c b/net/core/dev.c index 5b3042e69f8..ed8fe62d41a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1661,6 +1661,29 @@ bool is_skb_forwardable(struct net_device *dev, struct sk_buff *skb) } EXPORT_SYMBOL_GPL(is_skb_forwardable); +int __dev_forward_skb(struct net_device *dev, struct sk_buff *skb) +{ + if (skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) { + if (skb_copy_ubufs(skb, GFP_ATOMIC)) { + atomic_long_inc(&dev->rx_dropped); + kfree_skb(skb); + return NET_RX_DROP; + } + } + + if (unlikely(!is_skb_forwardable(dev, skb))) { + atomic_long_inc(&dev->rx_dropped); + kfree_skb(skb); + return NET_RX_DROP; + } + + skb_scrub_packet(skb, true); + skb->protocol = eth_type_trans(skb, dev); + + return 0; +} +EXPORT_SYMBOL_GPL(__dev_forward_skb); + /** * dev_forward_skb - loopback an skb to another netif * @@ -1681,24 +1704,7 @@ EXPORT_SYMBOL_GPL(is_skb_forwardable); */ int dev_forward_skb(struct net_device *dev, struct sk_buff *skb) { - if (skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) { - if (skb_copy_ubufs(skb, GFP_ATOMIC)) { - atomic_long_inc(&dev->rx_dropped); - kfree_skb(skb); - return NET_RX_DROP; - } - } - - if (unlikely(!is_skb_forwardable(dev, skb))) { - atomic_long_inc(&dev->rx_dropped); - kfree_skb(skb); - return NET_RX_DROP; - } - - skb_scrub_packet(skb, true); - skb->protocol = eth_type_trans(skb, dev); - - return netif_rx_internal(skb); + return __dev_forward_skb(dev, skb) ?: netif_rx_internal(skb); } EXPORT_SYMBOL_GPL(dev_forward_skb); @@ -2283,8 +2289,8 @@ EXPORT_SYMBOL(skb_checksum_help); __be16 skb_network_protocol(struct sk_buff *skb, int *depth) { + unsigned int vlan_depth = skb->mac_len; __be16 type = skb->protocol; - int vlan_depth = skb->mac_len; /* Tunnel gso handlers can set protocol to ethernet. */ if (type == htons(ETH_P_TEB)) { @@ -2297,15 +2303,30 @@ __be16 skb_network_protocol(struct sk_buff *skb, int *depth) type = eth->h_proto; } - while (type == htons(ETH_P_8021Q) || type == htons(ETH_P_8021AD)) { - struct vlan_hdr *vh; + /* if skb->protocol is 802.1Q/AD then the header should already be + * present at mac_len - VLAN_HLEN (if mac_len > 0), or at + * ETH_HLEN otherwise + */ + if (type == htons(ETH_P_8021Q) || type == htons(ETH_P_8021AD)) { + if (vlan_depth) { + if (unlikely(WARN_ON(vlan_depth < VLAN_HLEN))) + return 0; + vlan_depth -= VLAN_HLEN; + } else { + vlan_depth = ETH_HLEN; + } + do { + struct vlan_hdr *vh; - if (unlikely(!pskb_may_pull(skb, vlan_depth + VLAN_HLEN))) - return 0; + if (unlikely(!pskb_may_pull(skb, + vlan_depth + VLAN_HLEN))) + return 0; - vh = (struct vlan_hdr *)(skb->data + vlan_depth); - type = vh->h_vlan_encapsulated_proto; - vlan_depth += VLAN_HLEN; + vh = (struct vlan_hdr *)(skb->data + vlan_depth); + type = vh->h_vlan_encapsulated_proto; + vlan_depth += VLAN_HLEN; + } while (type == htons(ETH_P_8021Q) || + type == htons(ETH_P_8021AD)); } *depth = vlan_depth; @@ -2418,7 +2439,7 @@ EXPORT_SYMBOL(netdev_rx_csum_fault); * 2. No high memory really exists on this machine. */ -static int illegal_highdma(const struct net_device *dev, struct sk_buff *skb) +static int illegal_highdma(struct net_device *dev, struct sk_buff *skb) { #ifdef CONFIG_HIGHMEM int i; @@ -2492,39 +2513,63 @@ static int dev_gso_segment(struct sk_buff *skb, netdev_features_t features) return 0; } +/* If MPLS offload request, verify we are testing hardware MPLS features + * instead of standard features for the netdev. + */ +#ifdef CONFIG_NET_MPLS_GSO +static netdev_features_t net_mpls_features(struct sk_buff *skb, + netdev_features_t features, + __be16 type) +{ + if (type == htons(ETH_P_MPLS_UC) || type == htons(ETH_P_MPLS_MC)) + features &= skb->dev->mpls_features; + + return features; +} +#else +static netdev_features_t net_mpls_features(struct sk_buff *skb, + netdev_features_t features, + __be16 type) +{ + return features; +} +#endif + static netdev_features_t harmonize_features(struct sk_buff *skb, - const struct net_device *dev, - netdev_features_t features) + netdev_features_t features) { int tmp; + __be16 type; + + type = skb_network_protocol(skb, &tmp); + features = net_mpls_features(skb, features, type); if (skb->ip_summed != CHECKSUM_NONE && - !can_checksum_protocol(features, skb_network_protocol(skb, &tmp))) { + !can_checksum_protocol(features, type)) { features &= ~NETIF_F_ALL_CSUM; - } else if (illegal_highdma(dev, skb)) { + } else if (illegal_highdma(skb->dev, skb)) { features &= ~NETIF_F_SG; } return features; } -netdev_features_t netif_skb_dev_features(struct sk_buff *skb, - const struct net_device *dev) +netdev_features_t netif_skb_features(struct sk_buff *skb) { __be16 protocol = skb->protocol; - netdev_features_t features = dev->features; + netdev_features_t features = skb->dev->features; - if (skb_shinfo(skb)->gso_segs > dev->gso_max_segs) + if (skb_shinfo(skb)->gso_segs > skb->dev->gso_max_segs) features &= ~NETIF_F_GSO_MASK; if (protocol == htons(ETH_P_8021Q) || protocol == htons(ETH_P_8021AD)) { struct vlan_ethhdr *veh = (struct vlan_ethhdr *)skb->data; protocol = veh->h_vlan_encapsulated_proto; } else if (!vlan_tx_tag_present(skb)) { - return harmonize_features(skb, dev, features); + return harmonize_features(skb, features); } - features &= (dev->vlan_features | NETIF_F_HW_VLAN_CTAG_TX | + features &= (skb->dev->vlan_features | NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX); if (protocol == htons(ETH_P_8021Q) || protocol == htons(ETH_P_8021AD)) @@ -2532,9 +2577,9 @@ netdev_features_t netif_skb_dev_features(struct sk_buff *skb, NETIF_F_GEN_CSUM | NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX; - return harmonize_features(skb, dev, features); + return harmonize_features(skb, features); } -EXPORT_SYMBOL(netif_skb_dev_features); +EXPORT_SYMBOL(netif_skb_features); int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, struct netdev_queue *txq) @@ -3953,6 +3998,7 @@ static enum gro_result dev_gro_receive(struct napi_struct *napi, struct sk_buff } NAPI_GRO_CB(skb)->count = 1; NAPI_GRO_CB(skb)->age = jiffies; + NAPI_GRO_CB(skb)->last = skb; skb_shinfo(skb)->gso_size = skb_gro_len(skb); skb->next = napi->gro_list; napi->gro_list = skb; @@ -4543,6 +4589,32 @@ void *netdev_adjacent_get_private(struct list_head *adj_list) EXPORT_SYMBOL(netdev_adjacent_get_private); /** + * netdev_upper_get_next_dev_rcu - Get the next dev from upper list + * @dev: device + * @iter: list_head ** of the current position + * + * Gets the next device from the dev's upper list, starting from iter + * position. The caller must hold RCU read lock. + */ +struct net_device *netdev_upper_get_next_dev_rcu(struct net_device *dev, + struct list_head **iter) +{ + struct netdev_adjacent *upper; + + WARN_ON_ONCE(!rcu_read_lock_held() && !lockdep_rtnl_is_held()); + + upper = list_entry_rcu((*iter)->next, struct netdev_adjacent, list); + + if (&upper->list == &dev->adj_list.upper) + return NULL; + + *iter = &upper->list; + + return upper->dev; +} +EXPORT_SYMBOL(netdev_upper_get_next_dev_rcu); + +/** * netdev_all_upper_get_next_dev_rcu - Get the next dev from upper list * @dev: device * @iter: list_head ** of the current position @@ -4624,6 +4696,32 @@ void *netdev_lower_get_next_private_rcu(struct net_device *dev, EXPORT_SYMBOL(netdev_lower_get_next_private_rcu); /** + * netdev_lower_get_next - Get the next device from the lower neighbour + * list + * @dev: device + * @iter: list_head ** of the current position + * + * Gets the next netdev_adjacent from the dev's lower neighbour + * list, starting from iter position. The caller must hold RTNL lock or + * its own locking that guarantees that the neighbour lower + * list will remain unchainged. + */ +void *netdev_lower_get_next(struct net_device *dev, struct list_head **iter) +{ + struct netdev_adjacent *lower; + + lower = list_entry((*iter)->next, struct netdev_adjacent, list); + + if (&lower->list == &dev->adj_list.lower) + return NULL; + + *iter = &lower->list; + + return lower->dev; +} +EXPORT_SYMBOL(netdev_lower_get_next); + +/** * netdev_lower_get_first_private_rcu - Get the first ->private from the * lower neighbour list, RCU * variant @@ -5073,6 +5171,30 @@ void *netdev_lower_dev_get_private(struct net_device *dev, } EXPORT_SYMBOL(netdev_lower_dev_get_private); + +int dev_get_nest_level(struct net_device *dev, + bool (*type_check)(struct net_device *dev)) +{ + struct net_device *lower = NULL; + struct list_head *iter; + int max_nest = -1; + int nest; + + ASSERT_RTNL(); + + netdev_for_each_lower_dev(dev, lower, iter) { + nest = dev_get_nest_level(lower, type_check); + if (max_nest < nest) + max_nest = nest; + } + + if (type_check(dev)) + max_nest++; + + return max_nest; +} +EXPORT_SYMBOL(dev_get_nest_level); + static void dev_change_rx_flags(struct net_device *dev, int flags) { const struct net_device_ops *ops = dev->netdev_ops; @@ -5542,7 +5664,7 @@ static int dev_new_index(struct net *net) /* Delayed registration/unregisteration */ static LIST_HEAD(net_todo_list); -static DECLARE_WAIT_QUEUE_HEAD(netdev_unregistering_wq); +DECLARE_WAIT_QUEUE_HEAD(netdev_unregistering_wq); static void net_set_todo(struct net_device *dev) { @@ -5599,10 +5721,6 @@ static void rollback_registered_many(struct list_head *head) */ call_netdevice_notifiers(NETDEV_UNREGISTER, dev); - if (!dev->rtnl_link_ops || - dev->rtnl_link_state == RTNL_LINK_INITIALIZED) - rtmsg_ifinfo(RTM_DELLINK, dev, ~0U, GFP_KERNEL); - /* * Flush the unicast and multicast chains */ @@ -5612,6 +5730,10 @@ static void rollback_registered_many(struct list_head *head) if (dev->netdev_ops->ndo_uninit) dev->netdev_ops->ndo_uninit(dev); + if (!dev->rtnl_link_ops || + dev->rtnl_link_state == RTNL_LINK_INITIALIZED) + rtmsg_ifinfo(RTM_DELLINK, dev, ~0U, GFP_KERNEL); + /* Notifier chain MUST detach us all upper devices. */ WARN_ON(netdev_has_any_upper_dev(dev)); @@ -5837,10 +5959,7 @@ static void netdev_init_one_queue(struct net_device *dev, static void netif_free_tx_queues(struct net_device *dev) { - if (is_vmalloc_addr(dev->_tx)) - vfree(dev->_tx); - else - kfree(dev->_tx); + kvfree(dev->_tx); } static int netif_alloc_netdev_queues(struct net_device *dev) @@ -6314,10 +6433,7 @@ void netdev_freemem(struct net_device *dev) { char *addr = (char *)dev - dev->padded; - if (is_vmalloc_addr(addr)) - vfree(addr); - else - kfree(addr); + kvfree(addr); } /** @@ -6422,11 +6538,6 @@ free_all: free_pcpu: free_percpu(dev->pcpu_refcnt); - netif_free_tx_queues(dev); -#ifdef CONFIG_SYSFS - kfree(dev->_rx); -#endif - free_dev: netdev_freemem(dev); return NULL; diff --git a/net/core/dev_addr_lists.c b/net/core/dev_addr_lists.c index 329d5794e7d..b6b230600b9 100644 --- a/net/core/dev_addr_lists.c +++ b/net/core/dev_addr_lists.c @@ -225,6 +225,91 @@ void __hw_addr_unsync(struct netdev_hw_addr_list *to_list, } EXPORT_SYMBOL(__hw_addr_unsync); +/** + * __hw_addr_sync_dev - Synchonize device's multicast list + * @list: address list to syncronize + * @dev: device to sync + * @sync: function to call if address should be added + * @unsync: function to call if address should be removed + * + * This funciton is intended to be called from the ndo_set_rx_mode + * function of devices that require explicit address add/remove + * notifications. The unsync function may be NULL in which case + * the addresses requiring removal will simply be removed without + * any notification to the device. + **/ +int __hw_addr_sync_dev(struct netdev_hw_addr_list *list, + struct net_device *dev, + int (*sync)(struct net_device *, const unsigned char *), + int (*unsync)(struct net_device *, + const unsigned char *)) +{ + struct netdev_hw_addr *ha, *tmp; + int err; + + /* first go through and flush out any stale entries */ + list_for_each_entry_safe(ha, tmp, &list->list, list) { + if (!ha->sync_cnt || ha->refcount != 1) + continue; + + /* if unsync is defined and fails defer unsyncing address */ + if (unsync && unsync(dev, ha->addr)) + continue; + + ha->sync_cnt--; + __hw_addr_del_entry(list, ha, false, false); + } + + /* go through and sync new entries to the list */ + list_for_each_entry_safe(ha, tmp, &list->list, list) { + if (ha->sync_cnt) + continue; + + err = sync(dev, ha->addr); + if (err) + return err; + + ha->sync_cnt++; + ha->refcount++; + } + + return 0; +} +EXPORT_SYMBOL(__hw_addr_sync_dev); + +/** + * __hw_addr_unsync_dev - Remove synchonized addresses from device + * @list: address list to remove syncronized addresses from + * @dev: device to sync + * @unsync: function to call if address should be removed + * + * Remove all addresses that were added to the device by __hw_addr_sync_dev(). + * This function is intended to be called from the ndo_stop or ndo_open + * functions on devices that require explicit address add/remove + * notifications. If the unsync function pointer is NULL then this function + * can be used to just reset the sync_cnt for the addresses in the list. + **/ +void __hw_addr_unsync_dev(struct netdev_hw_addr_list *list, + struct net_device *dev, + int (*unsync)(struct net_device *, + const unsigned char *)) +{ + struct netdev_hw_addr *ha, *tmp; + + list_for_each_entry_safe(ha, tmp, &list->list, list) { + if (!ha->sync_cnt) + continue; + + /* if unsync is defined and fails defer unsyncing address */ + if (unsync && unsync(dev, ha->addr)) + continue; + + ha->sync_cnt--; + __hw_addr_del_entry(list, ha, false, false); + } +} +EXPORT_SYMBOL(__hw_addr_unsync_dev); + static void __hw_addr_flush(struct netdev_hw_addr_list *list) { struct netdev_hw_addr *ha, *tmp; diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 640ba0e5831..17cb912793f 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -557,6 +557,23 @@ err_out: return ret; } +static int ethtool_copy_validate_indir(u32 *indir, void __user *useraddr, + struct ethtool_rxnfc *rx_rings, + u32 size) +{ + int i; + + if (copy_from_user(indir, useraddr, size * sizeof(indir[0]))) + return -EFAULT; + + /* Validate ring indices */ + for (i = 0; i < size; i++) + if (indir[i] >= rx_rings->data) + return -EINVAL; + + return 0; +} + static noinline_for_stack int ethtool_get_rxfh_indir(struct net_device *dev, void __user *useraddr) { @@ -565,7 +582,7 @@ static noinline_for_stack int ethtool_get_rxfh_indir(struct net_device *dev, int ret; if (!dev->ethtool_ops->get_rxfh_indir_size || - !dev->ethtool_ops->get_rxfh_indir) + !dev->ethtool_ops->get_rxfh) return -EOPNOTSUPP; dev_size = dev->ethtool_ops->get_rxfh_indir_size(dev); if (dev_size == 0) @@ -591,7 +608,7 @@ static noinline_for_stack int ethtool_get_rxfh_indir(struct net_device *dev, if (!indir) return -ENOMEM; - ret = dev->ethtool_ops->get_rxfh_indir(dev, indir); + ret = dev->ethtool_ops->get_rxfh(dev, indir, NULL); if (ret) goto out; @@ -613,8 +630,9 @@ static noinline_for_stack int ethtool_set_rxfh_indir(struct net_device *dev, u32 *indir; const struct ethtool_ops *ops = dev->ethtool_ops; int ret; + u32 ringidx_offset = offsetof(struct ethtool_rxfh_indir, ring_index[0]); - if (!ops->get_rxfh_indir_size || !ops->set_rxfh_indir || + if (!ops->get_rxfh_indir_size || !ops->set_rxfh || !ops->get_rxnfc) return -EOPNOTSUPP; @@ -643,28 +661,184 @@ static noinline_for_stack int ethtool_set_rxfh_indir(struct net_device *dev, for (i = 0; i < dev_size; i++) indir[i] = ethtool_rxfh_indir_default(i, rx_rings.data); } else { - if (copy_from_user(indir, - useraddr + - offsetof(struct ethtool_rxfh_indir, - ring_index[0]), - dev_size * sizeof(indir[0]))) { + ret = ethtool_copy_validate_indir(indir, + useraddr + ringidx_offset, + &rx_rings, + dev_size); + if (ret) + goto out; + } + + ret = ops->set_rxfh(dev, indir, NULL); + +out: + kfree(indir); + return ret; +} + +static noinline_for_stack int ethtool_get_rxfh(struct net_device *dev, + void __user *useraddr) +{ + int ret; + const struct ethtool_ops *ops = dev->ethtool_ops; + u32 user_indir_size, user_key_size; + u32 dev_indir_size = 0, dev_key_size = 0; + struct ethtool_rxfh rxfh; + u32 total_size; + u32 indir_bytes; + u32 *indir = NULL; + u8 *hkey = NULL; + u8 *rss_config; + + if (!(dev->ethtool_ops->get_rxfh_indir_size || + dev->ethtool_ops->get_rxfh_key_size) || + !dev->ethtool_ops->get_rxfh) + return -EOPNOTSUPP; + + if (ops->get_rxfh_indir_size) + dev_indir_size = ops->get_rxfh_indir_size(dev); + if (ops->get_rxfh_key_size) + dev_key_size = ops->get_rxfh_key_size(dev); + + if ((dev_key_size + dev_indir_size) == 0) + return -EOPNOTSUPP; + + if (copy_from_user(&rxfh, useraddr, sizeof(rxfh))) + return -EFAULT; + user_indir_size = rxfh.indir_size; + user_key_size = rxfh.key_size; + + /* Check that reserved fields are 0 for now */ + if (rxfh.rss_context || rxfh.rsvd[0] || rxfh.rsvd[1]) + return -EINVAL; + + rxfh.indir_size = dev_indir_size; + rxfh.key_size = dev_key_size; + if (copy_to_user(useraddr, &rxfh, sizeof(rxfh))) + return -EFAULT; + + /* If the user buffer size is 0, this is just a query for the + * device table size and key size. Otherwise, if the User size is + * not equal to device table size or key size it's an error. + */ + if (!user_indir_size && !user_key_size) + return 0; + + if ((user_indir_size && (user_indir_size != dev_indir_size)) || + (user_key_size && (user_key_size != dev_key_size))) + return -EINVAL; + + indir_bytes = user_indir_size * sizeof(indir[0]); + total_size = indir_bytes + user_key_size; + rss_config = kzalloc(total_size, GFP_USER); + if (!rss_config) + return -ENOMEM; + + if (user_indir_size) + indir = (u32 *)rss_config; + + if (user_key_size) + hkey = rss_config + indir_bytes; + + ret = dev->ethtool_ops->get_rxfh(dev, indir, hkey); + if (!ret) { + if (copy_to_user(useraddr + + offsetof(struct ethtool_rxfh, rss_config[0]), + rss_config, total_size)) ret = -EFAULT; + } + + kfree(rss_config); + + return ret; +} + +static noinline_for_stack int ethtool_set_rxfh(struct net_device *dev, + void __user *useraddr) +{ + int ret; + const struct ethtool_ops *ops = dev->ethtool_ops; + struct ethtool_rxnfc rx_rings; + struct ethtool_rxfh rxfh; + u32 dev_indir_size = 0, dev_key_size = 0, i; + u32 *indir = NULL, indir_bytes = 0; + u8 *hkey = NULL; + u8 *rss_config; + u32 rss_cfg_offset = offsetof(struct ethtool_rxfh, rss_config[0]); + + if (!(ops->get_rxfh_indir_size || ops->get_rxfh_key_size) || + !ops->get_rxnfc || !ops->set_rxfh) + return -EOPNOTSUPP; + + if (ops->get_rxfh_indir_size) + dev_indir_size = ops->get_rxfh_indir_size(dev); + if (ops->get_rxfh_key_size) + dev_key_size = dev->ethtool_ops->get_rxfh_key_size(dev); + if ((dev_key_size + dev_indir_size) == 0) + return -EOPNOTSUPP; + + if (copy_from_user(&rxfh, useraddr, sizeof(rxfh))) + return -EFAULT; + + /* Check that reserved fields are 0 for now */ + if (rxfh.rss_context || rxfh.rsvd[0] || rxfh.rsvd[1]) + return -EINVAL; + + /* If either indir or hash key is valid, proceed further. + * It is not valid to request that both be unchanged. + */ + if ((rxfh.indir_size && + rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE && + rxfh.indir_size != dev_indir_size) || + (rxfh.key_size && (rxfh.key_size != dev_key_size)) || + (rxfh.indir_size == ETH_RXFH_INDIR_NO_CHANGE && + rxfh.key_size == 0)) + return -EINVAL; + + if (rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE) + indir_bytes = dev_indir_size * sizeof(indir[0]); + + rss_config = kzalloc(indir_bytes + rxfh.key_size, GFP_USER); + if (!rss_config) + return -ENOMEM; + + rx_rings.cmd = ETHTOOL_GRXRINGS; + ret = ops->get_rxnfc(dev, &rx_rings, NULL); + if (ret) + goto out; + + /* rxfh.indir_size == 0 means reset the indir table to default. + * rxfh.indir_size == ETH_RXFH_INDIR_NO_CHANGE means leave it unchanged. + */ + if (rxfh.indir_size && + rxfh.indir_size != ETH_RXFH_INDIR_NO_CHANGE) { + indir = (u32 *)rss_config; + ret = ethtool_copy_validate_indir(indir, + useraddr + rss_cfg_offset, + &rx_rings, + rxfh.indir_size); + if (ret) goto out; - } + } else if (rxfh.indir_size == 0) { + indir = (u32 *)rss_config; + for (i = 0; i < dev_indir_size; i++) + indir[i] = ethtool_rxfh_indir_default(i, rx_rings.data); + } - /* Validate ring indices */ - for (i = 0; i < dev_size; i++) { - if (indir[i] >= rx_rings.data) { - ret = -EINVAL; - goto out; - } + if (rxfh.key_size) { + hkey = rss_config + indir_bytes; + if (copy_from_user(hkey, + useraddr + rss_cfg_offset + indir_bytes, + rxfh.key_size)) { + ret = -EFAULT; + goto out; } } - ret = ops->set_rxfh_indir(dev, indir); + ret = ops->set_rxfh(dev, indir, hkey); out: - kfree(indir); + kfree(rss_config); return ret; } @@ -1491,6 +1665,7 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) case ETHTOOL_GRXCLSRULE: case ETHTOOL_GRXCLSRLALL: case ETHTOOL_GRXFHINDIR: + case ETHTOOL_GRSSH: case ETHTOOL_GFEATURES: case ETHTOOL_GCHANNELS: case ETHTOOL_GET_TS_INFO: @@ -1628,6 +1803,12 @@ int dev_ethtool(struct net *net, struct ifreq *ifr) case ETHTOOL_SRXFHINDIR: rc = ethtool_set_rxfh_indir(dev, useraddr); break; + case ETHTOOL_GRSSH: + rc = ethtool_get_rxfh(dev, useraddr); + break; + case ETHTOOL_SRSSH: + rc = ethtool_set_rxfh(dev, useraddr); + break; case ETHTOOL_GFEATURES: rc = ethtool_get_features(dev, useraddr); break; diff --git a/net/core/filter.c b/net/core/filter.c index cd58614660c..6bd2e350e75 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -45,6 +45,27 @@ #include <linux/seccomp.h> #include <linux/if_vlan.h> +/* Registers */ +#define BPF_R0 regs[BPF_REG_0] +#define BPF_R1 regs[BPF_REG_1] +#define BPF_R2 regs[BPF_REG_2] +#define BPF_R3 regs[BPF_REG_3] +#define BPF_R4 regs[BPF_REG_4] +#define BPF_R5 regs[BPF_REG_5] +#define BPF_R6 regs[BPF_REG_6] +#define BPF_R7 regs[BPF_REG_7] +#define BPF_R8 regs[BPF_REG_8] +#define BPF_R9 regs[BPF_REG_9] +#define BPF_R10 regs[BPF_REG_10] + +/* Named registers */ +#define A regs[insn->a_reg] +#define X regs[insn->x_reg] +#define FP regs[BPF_REG_FP] +#define ARG1 regs[BPF_REG_ARG1] +#define CTX regs[BPF_REG_CTX] +#define K insn->imm + /* No hurry in this branch * * Exported for the bpf jit load helper. @@ -57,9 +78,9 @@ void *bpf_internal_load_pointer_neg_helper(const struct sk_buff *skb, int k, uns ptr = skb_network_header(skb) + k - SKF_NET_OFF; else if (k >= SKF_LL_OFF) ptr = skb_mac_header(skb) + k - SKF_LL_OFF; - if (ptr >= skb->head && ptr + size <= skb_tail_pointer(skb)) return ptr; + return NULL; } @@ -68,6 +89,7 @@ static inline void *load_pointer(const struct sk_buff *skb, int k, { if (k >= 0) return skb_header_pointer(skb, k, size, buffer); + return bpf_internal_load_pointer_neg_helper(skb, k, size); } @@ -131,210 +153,213 @@ noinline u64 __bpf_call_base(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) * keep, 0 for none. @ctx is the data we are operating on, @insn is the * array of filter instructions. */ -unsigned int __sk_run_filter(void *ctx, const struct sock_filter_int *insn) +static unsigned int __sk_run_filter(void *ctx, const struct sock_filter_int *insn) { u64 stack[MAX_BPF_STACK / sizeof(u64)]; u64 regs[MAX_BPF_REG], tmp; - void *ptr; - int off; - -#define K insn->imm -#define A regs[insn->a_reg] -#define X regs[insn->x_reg] -#define R0 regs[0] - -#define CONT ({insn++; goto select_insn; }) -#define CONT_JMP ({insn++; goto select_insn; }) - static const void *jumptable[256] = { [0 ... 255] = &&default_label, /* Now overwrite non-defaults ... */ -#define DL(A, B, C) [A|B|C] = &&A##_##B##_##C - DL(BPF_ALU, BPF_ADD, BPF_X), - DL(BPF_ALU, BPF_ADD, BPF_K), - DL(BPF_ALU, BPF_SUB, BPF_X), - DL(BPF_ALU, BPF_SUB, BPF_K), - DL(BPF_ALU, BPF_AND, BPF_X), - DL(BPF_ALU, BPF_AND, BPF_K), - DL(BPF_ALU, BPF_OR, BPF_X), - DL(BPF_ALU, BPF_OR, BPF_K), - DL(BPF_ALU, BPF_LSH, BPF_X), - DL(BPF_ALU, BPF_LSH, BPF_K), - DL(BPF_ALU, BPF_RSH, BPF_X), - DL(BPF_ALU, BPF_RSH, BPF_K), - DL(BPF_ALU, BPF_XOR, BPF_X), - DL(BPF_ALU, BPF_XOR, BPF_K), - DL(BPF_ALU, BPF_MUL, BPF_X), - DL(BPF_ALU, BPF_MUL, BPF_K), - DL(BPF_ALU, BPF_MOV, BPF_X), - DL(BPF_ALU, BPF_MOV, BPF_K), - DL(BPF_ALU, BPF_DIV, BPF_X), - DL(BPF_ALU, BPF_DIV, BPF_K), - DL(BPF_ALU, BPF_MOD, BPF_X), - DL(BPF_ALU, BPF_MOD, BPF_K), - DL(BPF_ALU, BPF_NEG, 0), - DL(BPF_ALU, BPF_END, BPF_TO_BE), - DL(BPF_ALU, BPF_END, BPF_TO_LE), - DL(BPF_ALU64, BPF_ADD, BPF_X), - DL(BPF_ALU64, BPF_ADD, BPF_K), - DL(BPF_ALU64, BPF_SUB, BPF_X), - DL(BPF_ALU64, BPF_SUB, BPF_K), - DL(BPF_ALU64, BPF_AND, BPF_X), - DL(BPF_ALU64, BPF_AND, BPF_K), - DL(BPF_ALU64, BPF_OR, BPF_X), - DL(BPF_ALU64, BPF_OR, BPF_K), - DL(BPF_ALU64, BPF_LSH, BPF_X), - DL(BPF_ALU64, BPF_LSH, BPF_K), - DL(BPF_ALU64, BPF_RSH, BPF_X), - DL(BPF_ALU64, BPF_RSH, BPF_K), - DL(BPF_ALU64, BPF_XOR, BPF_X), - DL(BPF_ALU64, BPF_XOR, BPF_K), - DL(BPF_ALU64, BPF_MUL, BPF_X), - DL(BPF_ALU64, BPF_MUL, BPF_K), - DL(BPF_ALU64, BPF_MOV, BPF_X), - DL(BPF_ALU64, BPF_MOV, BPF_K), - DL(BPF_ALU64, BPF_ARSH, BPF_X), - DL(BPF_ALU64, BPF_ARSH, BPF_K), - DL(BPF_ALU64, BPF_DIV, BPF_X), - DL(BPF_ALU64, BPF_DIV, BPF_K), - DL(BPF_ALU64, BPF_MOD, BPF_X), - DL(BPF_ALU64, BPF_MOD, BPF_K), - DL(BPF_ALU64, BPF_NEG, 0), - DL(BPF_JMP, BPF_CALL, 0), - DL(BPF_JMP, BPF_JA, 0), - DL(BPF_JMP, BPF_JEQ, BPF_X), - DL(BPF_JMP, BPF_JEQ, BPF_K), - DL(BPF_JMP, BPF_JNE, BPF_X), - DL(BPF_JMP, BPF_JNE, BPF_K), - DL(BPF_JMP, BPF_JGT, BPF_X), - DL(BPF_JMP, BPF_JGT, BPF_K), - DL(BPF_JMP, BPF_JGE, BPF_X), - DL(BPF_JMP, BPF_JGE, BPF_K), - DL(BPF_JMP, BPF_JSGT, BPF_X), - DL(BPF_JMP, BPF_JSGT, BPF_K), - DL(BPF_JMP, BPF_JSGE, BPF_X), - DL(BPF_JMP, BPF_JSGE, BPF_K), - DL(BPF_JMP, BPF_JSET, BPF_X), - DL(BPF_JMP, BPF_JSET, BPF_K), - DL(BPF_JMP, BPF_EXIT, 0), - DL(BPF_STX, BPF_MEM, BPF_B), - DL(BPF_STX, BPF_MEM, BPF_H), - DL(BPF_STX, BPF_MEM, BPF_W), - DL(BPF_STX, BPF_MEM, BPF_DW), - DL(BPF_STX, BPF_XADD, BPF_W), - DL(BPF_STX, BPF_XADD, BPF_DW), - DL(BPF_ST, BPF_MEM, BPF_B), - DL(BPF_ST, BPF_MEM, BPF_H), - DL(BPF_ST, BPF_MEM, BPF_W), - DL(BPF_ST, BPF_MEM, BPF_DW), - DL(BPF_LDX, BPF_MEM, BPF_B), - DL(BPF_LDX, BPF_MEM, BPF_H), - DL(BPF_LDX, BPF_MEM, BPF_W), - DL(BPF_LDX, BPF_MEM, BPF_DW), - DL(BPF_LD, BPF_ABS, BPF_W), - DL(BPF_LD, BPF_ABS, BPF_H), - DL(BPF_LD, BPF_ABS, BPF_B), - DL(BPF_LD, BPF_IND, BPF_W), - DL(BPF_LD, BPF_IND, BPF_H), - DL(BPF_LD, BPF_IND, BPF_B), -#undef DL + /* 32 bit ALU operations */ + [BPF_ALU | BPF_ADD | BPF_X] = &&ALU_ADD_X, + [BPF_ALU | BPF_ADD | BPF_K] = &&ALU_ADD_K, + [BPF_ALU | BPF_SUB | BPF_X] = &&ALU_SUB_X, + [BPF_ALU | BPF_SUB | BPF_K] = &&ALU_SUB_K, + [BPF_ALU | BPF_AND | BPF_X] = &&ALU_AND_X, + [BPF_ALU | BPF_AND | BPF_K] = &&ALU_AND_K, + [BPF_ALU | BPF_OR | BPF_X] = &&ALU_OR_X, + [BPF_ALU | BPF_OR | BPF_K] = &&ALU_OR_K, + [BPF_ALU | BPF_LSH | BPF_X] = &&ALU_LSH_X, + [BPF_ALU | BPF_LSH | BPF_K] = &&ALU_LSH_K, + [BPF_ALU | BPF_RSH | BPF_X] = &&ALU_RSH_X, + [BPF_ALU | BPF_RSH | BPF_K] = &&ALU_RSH_K, + [BPF_ALU | BPF_XOR | BPF_X] = &&ALU_XOR_X, + [BPF_ALU | BPF_XOR | BPF_K] = &&ALU_XOR_K, + [BPF_ALU | BPF_MUL | BPF_X] = &&ALU_MUL_X, + [BPF_ALU | BPF_MUL | BPF_K] = &&ALU_MUL_K, + [BPF_ALU | BPF_MOV | BPF_X] = &&ALU_MOV_X, + [BPF_ALU | BPF_MOV | BPF_K] = &&ALU_MOV_K, + [BPF_ALU | BPF_DIV | BPF_X] = &&ALU_DIV_X, + [BPF_ALU | BPF_DIV | BPF_K] = &&ALU_DIV_K, + [BPF_ALU | BPF_MOD | BPF_X] = &&ALU_MOD_X, + [BPF_ALU | BPF_MOD | BPF_K] = &&ALU_MOD_K, + [BPF_ALU | BPF_NEG] = &&ALU_NEG, + [BPF_ALU | BPF_END | BPF_TO_BE] = &&ALU_END_TO_BE, + [BPF_ALU | BPF_END | BPF_TO_LE] = &&ALU_END_TO_LE, + /* 64 bit ALU operations */ + [BPF_ALU64 | BPF_ADD | BPF_X] = &&ALU64_ADD_X, + [BPF_ALU64 | BPF_ADD | BPF_K] = &&ALU64_ADD_K, + [BPF_ALU64 | BPF_SUB | BPF_X] = &&ALU64_SUB_X, + [BPF_ALU64 | BPF_SUB | BPF_K] = &&ALU64_SUB_K, + [BPF_ALU64 | BPF_AND | BPF_X] = &&ALU64_AND_X, + [BPF_ALU64 | BPF_AND | BPF_K] = &&ALU64_AND_K, + [BPF_ALU64 | BPF_OR | BPF_X] = &&ALU64_OR_X, + [BPF_ALU64 | BPF_OR | BPF_K] = &&ALU64_OR_K, + [BPF_ALU64 | BPF_LSH | BPF_X] = &&ALU64_LSH_X, + [BPF_ALU64 | BPF_LSH | BPF_K] = &&ALU64_LSH_K, + [BPF_ALU64 | BPF_RSH | BPF_X] = &&ALU64_RSH_X, + [BPF_ALU64 | BPF_RSH | BPF_K] = &&ALU64_RSH_K, + [BPF_ALU64 | BPF_XOR | BPF_X] = &&ALU64_XOR_X, + [BPF_ALU64 | BPF_XOR | BPF_K] = &&ALU64_XOR_K, + [BPF_ALU64 | BPF_MUL | BPF_X] = &&ALU64_MUL_X, + [BPF_ALU64 | BPF_MUL | BPF_K] = &&ALU64_MUL_K, + [BPF_ALU64 | BPF_MOV | BPF_X] = &&ALU64_MOV_X, + [BPF_ALU64 | BPF_MOV | BPF_K] = &&ALU64_MOV_K, + [BPF_ALU64 | BPF_ARSH | BPF_X] = &&ALU64_ARSH_X, + [BPF_ALU64 | BPF_ARSH | BPF_K] = &&ALU64_ARSH_K, + [BPF_ALU64 | BPF_DIV | BPF_X] = &&ALU64_DIV_X, + [BPF_ALU64 | BPF_DIV | BPF_K] = &&ALU64_DIV_K, + [BPF_ALU64 | BPF_MOD | BPF_X] = &&ALU64_MOD_X, + [BPF_ALU64 | BPF_MOD | BPF_K] = &&ALU64_MOD_K, + [BPF_ALU64 | BPF_NEG] = &&ALU64_NEG, + /* Call instruction */ + [BPF_JMP | BPF_CALL] = &&JMP_CALL, + /* Jumps */ + [BPF_JMP | BPF_JA] = &&JMP_JA, + [BPF_JMP | BPF_JEQ | BPF_X] = &&JMP_JEQ_X, + [BPF_JMP | BPF_JEQ | BPF_K] = &&JMP_JEQ_K, + [BPF_JMP | BPF_JNE | BPF_X] = &&JMP_JNE_X, + [BPF_JMP | BPF_JNE | BPF_K] = &&JMP_JNE_K, + [BPF_JMP | BPF_JGT | BPF_X] = &&JMP_JGT_X, + [BPF_JMP | BPF_JGT | BPF_K] = &&JMP_JGT_K, + [BPF_JMP | BPF_JGE | BPF_X] = &&JMP_JGE_X, + [BPF_JMP | BPF_JGE | BPF_K] = &&JMP_JGE_K, + [BPF_JMP | BPF_JSGT | BPF_X] = &&JMP_JSGT_X, + [BPF_JMP | BPF_JSGT | BPF_K] = &&JMP_JSGT_K, + [BPF_JMP | BPF_JSGE | BPF_X] = &&JMP_JSGE_X, + [BPF_JMP | BPF_JSGE | BPF_K] = &&JMP_JSGE_K, + [BPF_JMP | BPF_JSET | BPF_X] = &&JMP_JSET_X, + [BPF_JMP | BPF_JSET | BPF_K] = &&JMP_JSET_K, + /* Program return */ + [BPF_JMP | BPF_EXIT] = &&JMP_EXIT, + /* Store instructions */ + [BPF_STX | BPF_MEM | BPF_B] = &&STX_MEM_B, + [BPF_STX | BPF_MEM | BPF_H] = &&STX_MEM_H, + [BPF_STX | BPF_MEM | BPF_W] = &&STX_MEM_W, + [BPF_STX | BPF_MEM | BPF_DW] = &&STX_MEM_DW, + [BPF_STX | BPF_XADD | BPF_W] = &&STX_XADD_W, + [BPF_STX | BPF_XADD | BPF_DW] = &&STX_XADD_DW, + [BPF_ST | BPF_MEM | BPF_B] = &&ST_MEM_B, + [BPF_ST | BPF_MEM | BPF_H] = &&ST_MEM_H, + [BPF_ST | BPF_MEM | BPF_W] = &&ST_MEM_W, + [BPF_ST | BPF_MEM | BPF_DW] = &&ST_MEM_DW, + /* Load instructions */ + [BPF_LDX | BPF_MEM | BPF_B] = &&LDX_MEM_B, + [BPF_LDX | BPF_MEM | BPF_H] = &&LDX_MEM_H, + [BPF_LDX | BPF_MEM | BPF_W] = &&LDX_MEM_W, + [BPF_LDX | BPF_MEM | BPF_DW] = &&LDX_MEM_DW, + [BPF_LD | BPF_ABS | BPF_W] = &&LD_ABS_W, + [BPF_LD | BPF_ABS | BPF_H] = &&LD_ABS_H, + [BPF_LD | BPF_ABS | BPF_B] = &&LD_ABS_B, + [BPF_LD | BPF_IND | BPF_W] = &&LD_IND_W, + [BPF_LD | BPF_IND | BPF_H] = &&LD_IND_H, + [BPF_LD | BPF_IND | BPF_B] = &&LD_IND_B, }; + void *ptr; + int off; + +#define CONT ({ insn++; goto select_insn; }) +#define CONT_JMP ({ insn++; goto select_insn; }) + + FP = (u64) (unsigned long) &stack[ARRAY_SIZE(stack)]; + ARG1 = (u64) (unsigned long) ctx; - regs[FP_REG] = (u64) (unsigned long) &stack[ARRAY_SIZE(stack)]; - regs[ARG1_REG] = (u64) (unsigned long) ctx; + /* Register for user BPF programs need to be reset first. */ + regs[BPF_REG_A] = 0; + regs[BPF_REG_X] = 0; select_insn: goto *jumptable[insn->code]; /* ALU */ #define ALU(OPCODE, OP) \ - BPF_ALU64_##OPCODE##_BPF_X: \ + ALU64_##OPCODE##_X: \ A = A OP X; \ CONT; \ - BPF_ALU_##OPCODE##_BPF_X: \ + ALU_##OPCODE##_X: \ A = (u32) A OP (u32) X; \ CONT; \ - BPF_ALU64_##OPCODE##_BPF_K: \ + ALU64_##OPCODE##_K: \ A = A OP K; \ CONT; \ - BPF_ALU_##OPCODE##_BPF_K: \ + ALU_##OPCODE##_K: \ A = (u32) A OP (u32) K; \ CONT; - ALU(BPF_ADD, +) - ALU(BPF_SUB, -) - ALU(BPF_AND, &) - ALU(BPF_OR, |) - ALU(BPF_LSH, <<) - ALU(BPF_RSH, >>) - ALU(BPF_XOR, ^) - ALU(BPF_MUL, *) + ALU(ADD, +) + ALU(SUB, -) + ALU(AND, &) + ALU(OR, |) + ALU(LSH, <<) + ALU(RSH, >>) + ALU(XOR, ^) + ALU(MUL, *) #undef ALU - BPF_ALU_BPF_NEG_0: + ALU_NEG: A = (u32) -A; CONT; - BPF_ALU64_BPF_NEG_0: + ALU64_NEG: A = -A; CONT; - BPF_ALU_BPF_MOV_BPF_X: + ALU_MOV_X: A = (u32) X; CONT; - BPF_ALU_BPF_MOV_BPF_K: + ALU_MOV_K: A = (u32) K; CONT; - BPF_ALU64_BPF_MOV_BPF_X: + ALU64_MOV_X: A = X; CONT; - BPF_ALU64_BPF_MOV_BPF_K: + ALU64_MOV_K: A = K; CONT; - BPF_ALU64_BPF_ARSH_BPF_X: + ALU64_ARSH_X: (*(s64 *) &A) >>= X; CONT; - BPF_ALU64_BPF_ARSH_BPF_K: + ALU64_ARSH_K: (*(s64 *) &A) >>= K; CONT; - BPF_ALU64_BPF_MOD_BPF_X: + ALU64_MOD_X: if (unlikely(X == 0)) return 0; tmp = A; A = do_div(tmp, X); CONT; - BPF_ALU_BPF_MOD_BPF_X: + ALU_MOD_X: if (unlikely(X == 0)) return 0; tmp = (u32) A; A = do_div(tmp, (u32) X); CONT; - BPF_ALU64_BPF_MOD_BPF_K: + ALU64_MOD_K: tmp = A; A = do_div(tmp, K); CONT; - BPF_ALU_BPF_MOD_BPF_K: + ALU_MOD_K: tmp = (u32) A; A = do_div(tmp, (u32) K); CONT; - BPF_ALU64_BPF_DIV_BPF_X: + ALU64_DIV_X: if (unlikely(X == 0)) return 0; do_div(A, X); CONT; - BPF_ALU_BPF_DIV_BPF_X: + ALU_DIV_X: if (unlikely(X == 0)) return 0; tmp = (u32) A; do_div(tmp, (u32) X); A = (u32) tmp; CONT; - BPF_ALU64_BPF_DIV_BPF_K: + ALU64_DIV_K: do_div(A, K); CONT; - BPF_ALU_BPF_DIV_BPF_K: + ALU_DIV_K: tmp = (u32) A; do_div(tmp, (u32) K); A = (u32) tmp; CONT; - BPF_ALU_BPF_END_BPF_TO_BE: + ALU_END_TO_BE: switch (K) { case 16: A = (__force u16) cpu_to_be16(A); @@ -347,7 +372,7 @@ select_insn: break; } CONT; - BPF_ALU_BPF_END_BPF_TO_LE: + ALU_END_TO_LE: switch (K) { case 16: A = (__force u16) cpu_to_le16(A); @@ -362,142 +387,144 @@ select_insn: CONT; /* CALL */ - BPF_JMP_BPF_CALL_0: - /* Function call scratches R1-R5 registers, preserves R6-R9, - * and stores return value into R0. + JMP_CALL: + /* Function call scratches BPF_R1-BPF_R5 registers, + * preserves BPF_R6-BPF_R9, and stores return value + * into BPF_R0. */ - R0 = (__bpf_call_base + insn->imm)(regs[1], regs[2], regs[3], - regs[4], regs[5]); + BPF_R0 = (__bpf_call_base + insn->imm)(BPF_R1, BPF_R2, BPF_R3, + BPF_R4, BPF_R5); CONT; /* JMP */ - BPF_JMP_BPF_JA_0: + JMP_JA: insn += insn->off; CONT; - BPF_JMP_BPF_JEQ_BPF_X: + JMP_JEQ_X: if (A == X) { insn += insn->off; CONT_JMP; } CONT; - BPF_JMP_BPF_JEQ_BPF_K: + JMP_JEQ_K: if (A == K) { insn += insn->off; CONT_JMP; } CONT; - BPF_JMP_BPF_JNE_BPF_X: + JMP_JNE_X: if (A != X) { insn += insn->off; CONT_JMP; } CONT; - BPF_JMP_BPF_JNE_BPF_K: + JMP_JNE_K: if (A != K) { insn += insn->off; CONT_JMP; } CONT; - BPF_JMP_BPF_JGT_BPF_X: + JMP_JGT_X: if (A > X) { insn += insn->off; CONT_JMP; } CONT; - BPF_JMP_BPF_JGT_BPF_K: + JMP_JGT_K: if (A > K) { insn += insn->off; CONT_JMP; } CONT; - BPF_JMP_BPF_JGE_BPF_X: + JMP_JGE_X: if (A >= X) { insn += insn->off; CONT_JMP; } CONT; - BPF_JMP_BPF_JGE_BPF_K: + JMP_JGE_K: if (A >= K) { insn += insn->off; CONT_JMP; } CONT; - BPF_JMP_BPF_JSGT_BPF_X: - if (((s64)A) > ((s64)X)) { + JMP_JSGT_X: + if (((s64) A) > ((s64) X)) { insn += insn->off; CONT_JMP; } CONT; - BPF_JMP_BPF_JSGT_BPF_K: - if (((s64)A) > ((s64)K)) { + JMP_JSGT_K: + if (((s64) A) > ((s64) K)) { insn += insn->off; CONT_JMP; } CONT; - BPF_JMP_BPF_JSGE_BPF_X: - if (((s64)A) >= ((s64)X)) { + JMP_JSGE_X: + if (((s64) A) >= ((s64) X)) { insn += insn->off; CONT_JMP; } CONT; - BPF_JMP_BPF_JSGE_BPF_K: - if (((s64)A) >= ((s64)K)) { + JMP_JSGE_K: + if (((s64) A) >= ((s64) K)) { insn += insn->off; CONT_JMP; } CONT; - BPF_JMP_BPF_JSET_BPF_X: + JMP_JSET_X: if (A & X) { insn += insn->off; CONT_JMP; } CONT; - BPF_JMP_BPF_JSET_BPF_K: + JMP_JSET_K: if (A & K) { insn += insn->off; CONT_JMP; } CONT; - BPF_JMP_BPF_EXIT_0: - return R0; + JMP_EXIT: + return BPF_R0; /* STX and ST and LDX*/ #define LDST(SIZEOP, SIZE) \ - BPF_STX_BPF_MEM_##SIZEOP: \ + STX_MEM_##SIZEOP: \ *(SIZE *)(unsigned long) (A + insn->off) = X; \ CONT; \ - BPF_ST_BPF_MEM_##SIZEOP: \ + ST_MEM_##SIZEOP: \ *(SIZE *)(unsigned long) (A + insn->off) = K; \ CONT; \ - BPF_LDX_BPF_MEM_##SIZEOP: \ + LDX_MEM_##SIZEOP: \ A = *(SIZE *)(unsigned long) (X + insn->off); \ CONT; - LDST(BPF_B, u8) - LDST(BPF_H, u16) - LDST(BPF_W, u32) - LDST(BPF_DW, u64) + LDST(B, u8) + LDST(H, u16) + LDST(W, u32) + LDST(DW, u64) #undef LDST - BPF_STX_BPF_XADD_BPF_W: /* lock xadd *(u32 *)(A + insn->off) += X */ + STX_XADD_W: /* lock xadd *(u32 *)(A + insn->off) += X */ atomic_add((u32) X, (atomic_t *)(unsigned long) (A + insn->off)); CONT; - BPF_STX_BPF_XADD_BPF_DW: /* lock xadd *(u64 *)(A + insn->off) += X */ + STX_XADD_DW: /* lock xadd *(u64 *)(A + insn->off) += X */ atomic64_add((u64) X, (atomic64_t *)(unsigned long) (A + insn->off)); CONT; - BPF_LD_BPF_ABS_BPF_W: /* R0 = ntohl(*(u32 *) (skb->data + K)) */ + LD_ABS_W: /* BPF_R0 = ntohl(*(u32 *) (skb->data + K)) */ off = K; load_word: - /* BPF_LD + BPD_ABS and BPF_LD + BPF_IND insns are only - * appearing in the programs where ctx == skb. All programs - * keep 'ctx' in regs[CTX_REG] == R6, sk_convert_filter() - * saves it in R6, internal BPF verifier will check that - * R6 == ctx. + /* BPF_LD + BPD_ABS and BPF_LD + BPF_IND insns are + * only appearing in the programs where ctx == + * skb. All programs keep 'ctx' in regs[BPF_REG_CTX] + * == BPF_R6, sk_convert_filter() saves it in BPF_R6, + * internal BPF verifier will check that BPF_R6 == + * ctx. * - * BPF_ABS and BPF_IND are wrappers of function calls, so - * they scratch R1-R5 registers, preserve R6-R9, and store - * return value into R0. + * BPF_ABS and BPF_IND are wrappers of function calls, + * so they scratch BPF_R1-BPF_R5 registers, preserve + * BPF_R6-BPF_R9, and store return value into BPF_R0. * * Implicit input: * ctx @@ -507,39 +534,43 @@ load_word: * K == 32-bit immediate * * Output: - * R0 - 8/16/32-bit skb data converted to cpu endianness + * BPF_R0 - 8/16/32-bit skb data converted to cpu endianness */ + ptr = load_pointer((struct sk_buff *) ctx, off, 4, &tmp); if (likely(ptr != NULL)) { - R0 = get_unaligned_be32(ptr); + BPF_R0 = get_unaligned_be32(ptr); CONT; } + return 0; - BPF_LD_BPF_ABS_BPF_H: /* R0 = ntohs(*(u16 *) (skb->data + K)) */ + LD_ABS_H: /* BPF_R0 = ntohs(*(u16 *) (skb->data + K)) */ off = K; load_half: ptr = load_pointer((struct sk_buff *) ctx, off, 2, &tmp); if (likely(ptr != NULL)) { - R0 = get_unaligned_be16(ptr); + BPF_R0 = get_unaligned_be16(ptr); CONT; } + return 0; - BPF_LD_BPF_ABS_BPF_B: /* R0 = *(u8 *) (ctx + K) */ + LD_ABS_B: /* BPF_R0 = *(u8 *) (ctx + K) */ off = K; load_byte: ptr = load_pointer((struct sk_buff *) ctx, off, 1, &tmp); if (likely(ptr != NULL)) { - R0 = *(u8 *)ptr; + BPF_R0 = *(u8 *)ptr; CONT; } + return 0; - BPF_LD_BPF_IND_BPF_W: /* R0 = ntohl(*(u32 *) (skb->data + X + K)) */ + LD_IND_W: /* BPF_R0 = ntohl(*(u32 *) (skb->data + X + K)) */ off = K + X; goto load_word; - BPF_LD_BPF_IND_BPF_H: /* R0 = ntohs(*(u16 *) (skb->data + X + K)) */ + LD_IND_H: /* BPF_R0 = ntohs(*(u16 *) (skb->data + X + K)) */ off = K + X; goto load_half; - BPF_LD_BPF_IND_BPF_B: /* R0 = *(u8 *) (skb->data + X + K) */ + LD_IND_B: /* BPF_R0 = *(u8 *) (skb->data + X + K) */ off = K + X; goto load_byte; @@ -547,29 +578,17 @@ load_byte: /* If we ever reach this, we have a bug somewhere. */ WARN_RATELIMIT(1, "unknown opcode %02x\n", insn->code); return 0; -#undef CONT_JMP -#undef CONT - -#undef R0 -#undef X -#undef A -#undef K } -u32 sk_run_filter_int_seccomp(const struct seccomp_data *ctx, - const struct sock_filter_int *insni) - __attribute__ ((alias ("__sk_run_filter"))); - -u32 sk_run_filter_int_skb(const struct sk_buff *ctx, - const struct sock_filter_int *insni) - __attribute__ ((alias ("__sk_run_filter"))); -EXPORT_SYMBOL_GPL(sk_run_filter_int_skb); - /* Helper to find the offset of pkt_type in sk_buff structure. We want * to make sure its still a 3bit field starting at a byte boundary; * taken from arch/x86/net/bpf_jit_comp.c. */ +#ifdef __BIG_ENDIAN_BITFIELD +#define PKT_TYPE_MAX (7 << 5) +#else #define PKT_TYPE_MAX 7 +#endif static unsigned int pkt_type_offset(void) { struct sk_buff skb_probe = { .pkt_type = ~0, }; @@ -585,16 +604,14 @@ static unsigned int pkt_type_offset(void) return -1; } -static u64 __skb_get_pay_offset(u64 ctx, u64 A, u64 X, u64 r4, u64 r5) +static u64 __skb_get_pay_offset(u64 ctx, u64 a, u64 x, u64 r4, u64 r5) { - struct sk_buff *skb = (struct sk_buff *)(long) ctx; - - return __skb_get_poff(skb); + return __skb_get_poff((struct sk_buff *)(unsigned long) ctx); } -static u64 __skb_get_nlattr(u64 ctx, u64 A, u64 X, u64 r4, u64 r5) +static u64 __skb_get_nlattr(u64 ctx, u64 a, u64 x, u64 r4, u64 r5) { - struct sk_buff *skb = (struct sk_buff *)(long) ctx; + struct sk_buff *skb = (struct sk_buff *)(unsigned long) ctx; struct nlattr *nla; if (skb_is_nonlinear(skb)) @@ -603,19 +620,19 @@ static u64 __skb_get_nlattr(u64 ctx, u64 A, u64 X, u64 r4, u64 r5) if (skb->len < sizeof(struct nlattr)) return 0; - if (A > skb->len - sizeof(struct nlattr)) + if (a > skb->len - sizeof(struct nlattr)) return 0; - nla = nla_find((struct nlattr *) &skb->data[A], skb->len - A, X); + nla = nla_find((struct nlattr *) &skb->data[a], skb->len - a, x); if (nla) return (void *) nla - (void *) skb->data; return 0; } -static u64 __skb_get_nlattr_nest(u64 ctx, u64 A, u64 X, u64 r4, u64 r5) +static u64 __skb_get_nlattr_nest(u64 ctx, u64 a, u64 x, u64 r4, u64 r5) { - struct sk_buff *skb = (struct sk_buff *)(long) ctx; + struct sk_buff *skb = (struct sk_buff *)(unsigned long) ctx; struct nlattr *nla; if (skb_is_nonlinear(skb)) @@ -624,31 +641,30 @@ static u64 __skb_get_nlattr_nest(u64 ctx, u64 A, u64 X, u64 r4, u64 r5) if (skb->len < sizeof(struct nlattr)) return 0; - if (A > skb->len - sizeof(struct nlattr)) + if (a > skb->len - sizeof(struct nlattr)) return 0; - nla = (struct nlattr *) &skb->data[A]; - if (nla->nla_len > skb->len - A) + nla = (struct nlattr *) &skb->data[a]; + if (nla->nla_len > skb->len - a) return 0; - nla = nla_find_nested(nla, X); + nla = nla_find_nested(nla, x); if (nla) return (void *) nla - (void *) skb->data; return 0; } -static u64 __get_raw_cpu_id(u64 ctx, u64 A, u64 X, u64 r4, u64 r5) +static u64 __get_raw_cpu_id(u64 ctx, u64 a, u64 x, u64 r4, u64 r5) { return raw_smp_processor_id(); } -/* Register mappings for user programs. */ -#define A_REG 0 -#define X_REG 7 -#define TMP_REG 8 -#define ARG2_REG 2 -#define ARG3_REG 3 +/* note that this only generates 32-bit random numbers */ +static u64 __get_random_u32(u64 ctx, u64 a, u64 x, u64 r4, u64 r5) +{ + return prandom_u32(); +} static bool convert_bpf_extensions(struct sock_filter *fp, struct sock_filter_int **insnp) @@ -659,119 +675,83 @@ static bool convert_bpf_extensions(struct sock_filter *fp, case SKF_AD_OFF + SKF_AD_PROTOCOL: BUILD_BUG_ON(FIELD_SIZEOF(struct sk_buff, protocol) != 2); - insn->code = BPF_LDX | BPF_MEM | BPF_H; - insn->a_reg = A_REG; - insn->x_reg = CTX_REG; - insn->off = offsetof(struct sk_buff, protocol); - insn++; - + /* A = *(u16 *) (ctx + offsetof(protocol)) */ + *insn++ = BPF_LDX_MEM(BPF_H, BPF_REG_A, BPF_REG_CTX, + offsetof(struct sk_buff, protocol)); /* A = ntohs(A) [emitting a nop or swap16] */ - insn->code = BPF_ALU | BPF_END | BPF_FROM_BE; - insn->a_reg = A_REG; - insn->imm = 16; + *insn = BPF_ENDIAN(BPF_FROM_BE, BPF_REG_A, 16); break; case SKF_AD_OFF + SKF_AD_PKTTYPE: - insn->code = BPF_LDX | BPF_MEM | BPF_B; - insn->a_reg = A_REG; - insn->x_reg = CTX_REG; - insn->off = pkt_type_offset(); + *insn = BPF_LDX_MEM(BPF_B, BPF_REG_A, BPF_REG_CTX, + pkt_type_offset()); if (insn->off < 0) return false; insn++; - - insn->code = BPF_ALU | BPF_AND | BPF_K; - insn->a_reg = A_REG; - insn->imm = PKT_TYPE_MAX; + *insn = BPF_ALU32_IMM(BPF_AND, BPF_REG_A, PKT_TYPE_MAX); +#ifdef __BIG_ENDIAN_BITFIELD + insn++; + *insn = BPF_ALU32_IMM(BPF_RSH, BPF_REG_A, 5); +#endif break; case SKF_AD_OFF + SKF_AD_IFINDEX: case SKF_AD_OFF + SKF_AD_HATYPE: - if (FIELD_SIZEOF(struct sk_buff, dev) == 8) - insn->code = BPF_LDX | BPF_MEM | BPF_DW; - else - insn->code = BPF_LDX | BPF_MEM | BPF_W; - insn->a_reg = TMP_REG; - insn->x_reg = CTX_REG; - insn->off = offsetof(struct sk_buff, dev); - insn++; - - insn->code = BPF_JMP | BPF_JNE | BPF_K; - insn->a_reg = TMP_REG; - insn->imm = 0; - insn->off = 1; - insn++; - - insn->code = BPF_JMP | BPF_EXIT; - insn++; - BUILD_BUG_ON(FIELD_SIZEOF(struct net_device, ifindex) != 4); BUILD_BUG_ON(FIELD_SIZEOF(struct net_device, type) != 2); - - insn->a_reg = A_REG; - insn->x_reg = TMP_REG; - - if (fp->k == SKF_AD_OFF + SKF_AD_IFINDEX) { - insn->code = BPF_LDX | BPF_MEM | BPF_W; - insn->off = offsetof(struct net_device, ifindex); - } else { - insn->code = BPF_LDX | BPF_MEM | BPF_H; - insn->off = offsetof(struct net_device, type); - } + BUILD_BUG_ON(bytes_to_bpf_size(FIELD_SIZEOF(struct sk_buff, dev)) < 0); + + *insn++ = BPF_LDX_MEM(bytes_to_bpf_size(FIELD_SIZEOF(struct sk_buff, dev)), + BPF_REG_TMP, BPF_REG_CTX, + offsetof(struct sk_buff, dev)); + /* if (tmp != 0) goto pc + 1 */ + *insn++ = BPF_JMP_IMM(BPF_JNE, BPF_REG_TMP, 0, 1); + *insn++ = BPF_EXIT_INSN(); + if (fp->k == SKF_AD_OFF + SKF_AD_IFINDEX) + *insn = BPF_LDX_MEM(BPF_W, BPF_REG_A, BPF_REG_TMP, + offsetof(struct net_device, ifindex)); + else + *insn = BPF_LDX_MEM(BPF_H, BPF_REG_A, BPF_REG_TMP, + offsetof(struct net_device, type)); break; case SKF_AD_OFF + SKF_AD_MARK: BUILD_BUG_ON(FIELD_SIZEOF(struct sk_buff, mark) != 4); - insn->code = BPF_LDX | BPF_MEM | BPF_W; - insn->a_reg = A_REG; - insn->x_reg = CTX_REG; - insn->off = offsetof(struct sk_buff, mark); + *insn = BPF_LDX_MEM(BPF_W, BPF_REG_A, BPF_REG_CTX, + offsetof(struct sk_buff, mark)); break; case SKF_AD_OFF + SKF_AD_RXHASH: BUILD_BUG_ON(FIELD_SIZEOF(struct sk_buff, hash) != 4); - insn->code = BPF_LDX | BPF_MEM | BPF_W; - insn->a_reg = A_REG; - insn->x_reg = CTX_REG; - insn->off = offsetof(struct sk_buff, hash); + *insn = BPF_LDX_MEM(BPF_W, BPF_REG_A, BPF_REG_CTX, + offsetof(struct sk_buff, hash)); break; case SKF_AD_OFF + SKF_AD_QUEUE: BUILD_BUG_ON(FIELD_SIZEOF(struct sk_buff, queue_mapping) != 2); - insn->code = BPF_LDX | BPF_MEM | BPF_H; - insn->a_reg = A_REG; - insn->x_reg = CTX_REG; - insn->off = offsetof(struct sk_buff, queue_mapping); + *insn = BPF_LDX_MEM(BPF_H, BPF_REG_A, BPF_REG_CTX, + offsetof(struct sk_buff, queue_mapping)); break; case SKF_AD_OFF + SKF_AD_VLAN_TAG: case SKF_AD_OFF + SKF_AD_VLAN_TAG_PRESENT: BUILD_BUG_ON(FIELD_SIZEOF(struct sk_buff, vlan_tci) != 2); - - insn->code = BPF_LDX | BPF_MEM | BPF_H; - insn->a_reg = A_REG; - insn->x_reg = CTX_REG; - insn->off = offsetof(struct sk_buff, vlan_tci); - insn++; - BUILD_BUG_ON(VLAN_TAG_PRESENT != 0x1000); + /* A = *(u16 *) (ctx + offsetof(vlan_tci)) */ + *insn++ = BPF_LDX_MEM(BPF_H, BPF_REG_A, BPF_REG_CTX, + offsetof(struct sk_buff, vlan_tci)); if (fp->k == SKF_AD_OFF + SKF_AD_VLAN_TAG) { - insn->code = BPF_ALU | BPF_AND | BPF_K; - insn->a_reg = A_REG; - insn->imm = ~VLAN_TAG_PRESENT; + *insn = BPF_ALU32_IMM(BPF_AND, BPF_REG_A, + ~VLAN_TAG_PRESENT); } else { - insn->code = BPF_ALU | BPF_RSH | BPF_K; - insn->a_reg = A_REG; - insn->imm = 12; - insn++; - - insn->code = BPF_ALU | BPF_AND | BPF_K; - insn->a_reg = A_REG; - insn->imm = 1; + /* A >>= 12 */ + *insn++ = BPF_ALU32_IMM(BPF_RSH, BPF_REG_A, 12); + /* A &= 1 */ + *insn = BPF_ALU32_IMM(BPF_AND, BPF_REG_A, 1); } break; @@ -779,46 +759,36 @@ static bool convert_bpf_extensions(struct sock_filter *fp, case SKF_AD_OFF + SKF_AD_NLATTR: case SKF_AD_OFF + SKF_AD_NLATTR_NEST: case SKF_AD_OFF + SKF_AD_CPU: + case SKF_AD_OFF + SKF_AD_RANDOM: /* arg1 = ctx */ - insn->code = BPF_ALU64 | BPF_MOV | BPF_X; - insn->a_reg = ARG1_REG; - insn->x_reg = CTX_REG; - insn++; - + *insn++ = BPF_MOV64_REG(BPF_REG_ARG1, BPF_REG_CTX); /* arg2 = A */ - insn->code = BPF_ALU64 | BPF_MOV | BPF_X; - insn->a_reg = ARG2_REG; - insn->x_reg = A_REG; - insn++; - + *insn++ = BPF_MOV64_REG(BPF_REG_ARG2, BPF_REG_A); /* arg3 = X */ - insn->code = BPF_ALU64 | BPF_MOV | BPF_X; - insn->a_reg = ARG3_REG; - insn->x_reg = X_REG; - insn++; - + *insn++ = BPF_MOV64_REG(BPF_REG_ARG3, BPF_REG_X); /* Emit call(ctx, arg2=A, arg3=X) */ - insn->code = BPF_JMP | BPF_CALL; switch (fp->k) { case SKF_AD_OFF + SKF_AD_PAY_OFFSET: - insn->imm = __skb_get_pay_offset - __bpf_call_base; + *insn = BPF_EMIT_CALL(__skb_get_pay_offset); break; case SKF_AD_OFF + SKF_AD_NLATTR: - insn->imm = __skb_get_nlattr - __bpf_call_base; + *insn = BPF_EMIT_CALL(__skb_get_nlattr); break; case SKF_AD_OFF + SKF_AD_NLATTR_NEST: - insn->imm = __skb_get_nlattr_nest - __bpf_call_base; + *insn = BPF_EMIT_CALL(__skb_get_nlattr_nest); break; case SKF_AD_OFF + SKF_AD_CPU: - insn->imm = __get_raw_cpu_id - __bpf_call_base; + *insn = BPF_EMIT_CALL(__get_raw_cpu_id); + break; + case SKF_AD_OFF + SKF_AD_RANDOM: + *insn = BPF_EMIT_CALL(__get_random_u32); break; } break; case SKF_AD_OFF + SKF_AD_ALU_XOR_X: - insn->code = BPF_ALU | BPF_XOR | BPF_X; - insn->a_reg = A_REG; - insn->x_reg = X_REG; + /* A ^= X */ + *insn = BPF_ALU32_REG(BPF_XOR, BPF_REG_A, BPF_REG_X); break; default: @@ -868,7 +838,7 @@ int sk_convert_filter(struct sock_filter *prog, int len, u8 bpf_src; BUILD_BUG_ON(BPF_MEMWORDS * sizeof(u32) > MAX_BPF_STACK); - BUILD_BUG_ON(FP_REG + 1 != MAX_BPF_REG); + BUILD_BUG_ON(BPF_REG_FP + 1 != MAX_BPF_REG); if (len <= 0 || len >= BPF_MAXINSNS) return -EINVAL; @@ -883,11 +853,8 @@ do_pass: new_insn = new_prog; fp = prog; - if (new_insn) { - new_insn->code = BPF_ALU64 | BPF_MOV | BPF_X; - new_insn->a_reg = CTX_REG; - new_insn->x_reg = ARG1_REG; - } + if (new_insn) + *new_insn = BPF_MOV64_REG(BPF_REG_CTX, BPF_REG_ARG1); new_insn++; for (i = 0; i < len; fp++, i++) { @@ -935,17 +902,16 @@ do_pass: convert_bpf_extensions(fp, &insn)) break; - insn->code = fp->code; - insn->a_reg = A_REG; - insn->x_reg = X_REG; - insn->imm = fp->k; + *insn = BPF_RAW_INSN(fp->code, BPF_REG_A, BPF_REG_X, 0, fp->k); break; - /* Jump opcodes map as-is, but offsets need adjustment. */ - case BPF_JMP | BPF_JA: - target = i + fp->k + 1; - insn->code = fp->code; -#define EMIT_JMP \ + /* Jump transformation cannot use BPF block macros + * everywhere as offset calculation and target updates + * require a bit more work than the rest, i.e. jump + * opcodes map as-is, but offsets need adjustment. + */ + +#define BPF_EMIT_JMP \ do { \ if (target >= len || target < 0) \ goto err; \ @@ -954,7 +920,10 @@ do_pass: insn->off -= insn - tmp_insns; \ } while (0) - EMIT_JMP; + case BPF_JMP | BPF_JA: + target = i + fp->k + 1; + insn->code = fp->code; + BPF_EMIT_JMP; break; case BPF_JMP | BPF_JEQ | BPF_K: @@ -970,17 +939,14 @@ do_pass: * immediate into tmp register and use it * in compare insn. */ - insn->code = BPF_ALU | BPF_MOV | BPF_K; - insn->a_reg = TMP_REG; - insn->imm = fp->k; - insn++; + *insn++ = BPF_MOV32_IMM(BPF_REG_TMP, fp->k); - insn->a_reg = A_REG; - insn->x_reg = TMP_REG; + insn->a_reg = BPF_REG_A; + insn->x_reg = BPF_REG_TMP; bpf_src = BPF_X; } else { - insn->a_reg = A_REG; - insn->x_reg = X_REG; + insn->a_reg = BPF_REG_A; + insn->x_reg = BPF_REG_X; insn->imm = fp->k; bpf_src = BPF_SRC(fp->code); } @@ -989,7 +955,7 @@ do_pass: if (fp->jf == 0) { insn->code = BPF_JMP | BPF_OP(fp->code) | bpf_src; target = i + fp->jt + 1; - EMIT_JMP; + BPF_EMIT_JMP; break; } @@ -997,127 +963,94 @@ do_pass: if (fp->jt == 0 && BPF_OP(fp->code) == BPF_JEQ) { insn->code = BPF_JMP | BPF_JNE | bpf_src; target = i + fp->jf + 1; - EMIT_JMP; + BPF_EMIT_JMP; break; } /* Other jumps are mapped into two insns: Jxx and JA. */ target = i + fp->jt + 1; insn->code = BPF_JMP | BPF_OP(fp->code) | bpf_src; - EMIT_JMP; + BPF_EMIT_JMP; insn++; insn->code = BPF_JMP | BPF_JA; target = i + fp->jf + 1; - EMIT_JMP; + BPF_EMIT_JMP; break; /* ldxb 4 * ([14] & 0xf) is remaped into 6 insns. */ case BPF_LDX | BPF_MSH | BPF_B: - insn->code = BPF_ALU64 | BPF_MOV | BPF_X; - insn->a_reg = TMP_REG; - insn->x_reg = A_REG; - insn++; - - insn->code = BPF_LD | BPF_ABS | BPF_B; - insn->a_reg = A_REG; - insn->imm = fp->k; - insn++; - - insn->code = BPF_ALU | BPF_AND | BPF_K; - insn->a_reg = A_REG; - insn->imm = 0xf; - insn++; - - insn->code = BPF_ALU | BPF_LSH | BPF_K; - insn->a_reg = A_REG; - insn->imm = 2; - insn++; - - insn->code = BPF_ALU64 | BPF_MOV | BPF_X; - insn->a_reg = X_REG; - insn->x_reg = A_REG; - insn++; - - insn->code = BPF_ALU64 | BPF_MOV | BPF_X; - insn->a_reg = A_REG; - insn->x_reg = TMP_REG; + /* tmp = A */ + *insn++ = BPF_MOV64_REG(BPF_REG_TMP, BPF_REG_A); + /* A = BPF_R0 = *(u8 *) (skb->data + K) */ + *insn++ = BPF_LD_ABS(BPF_B, fp->k); + /* A &= 0xf */ + *insn++ = BPF_ALU32_IMM(BPF_AND, BPF_REG_A, 0xf); + /* A <<= 2 */ + *insn++ = BPF_ALU32_IMM(BPF_LSH, BPF_REG_A, 2); + /* X = A */ + *insn++ = BPF_MOV64_REG(BPF_REG_X, BPF_REG_A); + /* A = tmp */ + *insn = BPF_MOV64_REG(BPF_REG_A, BPF_REG_TMP); break; /* RET_K, RET_A are remaped into 2 insns. */ case BPF_RET | BPF_A: case BPF_RET | BPF_K: - insn->code = BPF_ALU | BPF_MOV | - (BPF_RVAL(fp->code) == BPF_K ? - BPF_K : BPF_X); - insn->a_reg = 0; - insn->x_reg = A_REG; - insn->imm = fp->k; - insn++; - - insn->code = BPF_JMP | BPF_EXIT; + *insn++ = BPF_MOV32_RAW(BPF_RVAL(fp->code) == BPF_K ? + BPF_K : BPF_X, BPF_REG_0, + BPF_REG_A, fp->k); + *insn = BPF_EXIT_INSN(); break; /* Store to stack. */ case BPF_ST: case BPF_STX: - insn->code = BPF_STX | BPF_MEM | BPF_W; - insn->a_reg = FP_REG; - insn->x_reg = fp->code == BPF_ST ? A_REG : X_REG; - insn->off = -(BPF_MEMWORDS - fp->k) * 4; + *insn = BPF_STX_MEM(BPF_W, BPF_REG_FP, BPF_CLASS(fp->code) == + BPF_ST ? BPF_REG_A : BPF_REG_X, + -(BPF_MEMWORDS - fp->k) * 4); break; /* Load from stack. */ case BPF_LD | BPF_MEM: case BPF_LDX | BPF_MEM: - insn->code = BPF_LDX | BPF_MEM | BPF_W; - insn->a_reg = BPF_CLASS(fp->code) == BPF_LD ? - A_REG : X_REG; - insn->x_reg = FP_REG; - insn->off = -(BPF_MEMWORDS - fp->k) * 4; + *insn = BPF_LDX_MEM(BPF_W, BPF_CLASS(fp->code) == BPF_LD ? + BPF_REG_A : BPF_REG_X, BPF_REG_FP, + -(BPF_MEMWORDS - fp->k) * 4); break; /* A = K or X = K */ case BPF_LD | BPF_IMM: case BPF_LDX | BPF_IMM: - insn->code = BPF_ALU | BPF_MOV | BPF_K; - insn->a_reg = BPF_CLASS(fp->code) == BPF_LD ? - A_REG : X_REG; - insn->imm = fp->k; + *insn = BPF_MOV32_IMM(BPF_CLASS(fp->code) == BPF_LD ? + BPF_REG_A : BPF_REG_X, fp->k); break; /* X = A */ case BPF_MISC | BPF_TAX: - insn->code = BPF_ALU64 | BPF_MOV | BPF_X; - insn->a_reg = X_REG; - insn->x_reg = A_REG; + *insn = BPF_MOV64_REG(BPF_REG_X, BPF_REG_A); break; /* A = X */ case BPF_MISC | BPF_TXA: - insn->code = BPF_ALU64 | BPF_MOV | BPF_X; - insn->a_reg = A_REG; - insn->x_reg = X_REG; + *insn = BPF_MOV64_REG(BPF_REG_A, BPF_REG_X); break; /* A = skb->len or X = skb->len */ case BPF_LD | BPF_W | BPF_LEN: case BPF_LDX | BPF_W | BPF_LEN: - insn->code = BPF_LDX | BPF_MEM | BPF_W; - insn->a_reg = BPF_CLASS(fp->code) == BPF_LD ? - A_REG : X_REG; - insn->x_reg = CTX_REG; - insn->off = offsetof(struct sk_buff, len); + *insn = BPF_LDX_MEM(BPF_W, BPF_CLASS(fp->code) == BPF_LD ? + BPF_REG_A : BPF_REG_X, BPF_REG_CTX, + offsetof(struct sk_buff, len)); break; - /* access seccomp_data fields */ + /* Access seccomp_data fields. */ case BPF_LDX | BPF_ABS | BPF_W: - insn->code = BPF_LDX | BPF_MEM | BPF_W; - insn->a_reg = A_REG; - insn->x_reg = CTX_REG; - insn->off = fp->k; + /* A = *(u32 *) (ctx + K) */ + *insn = BPF_LDX_MEM(BPF_W, BPF_REG_A, BPF_REG_CTX, fp->k); break; + /* Unkown instruction. */ default: goto err; } @@ -1126,7 +1059,6 @@ do_pass: if (new_prog) memcpy(new_insn, tmp_insns, sizeof(*insn) * (insn - tmp_insns)); - new_insn += insn - tmp_insns; } @@ -1141,7 +1073,6 @@ do_pass: new_flen = new_insn - new_prog; if (pass > 2) goto err; - goto do_pass; } @@ -1165,44 +1096,46 @@ err: */ static int check_load_and_stores(struct sock_filter *filter, int flen) { - u16 *masks, memvalid = 0; /* one bit per cell, 16 cells */ + u16 *masks, memvalid = 0; /* One bit per cell, 16 cells */ int pc, ret = 0; BUILD_BUG_ON(BPF_MEMWORDS > 16); + masks = kmalloc(flen * sizeof(*masks), GFP_KERNEL); if (!masks) return -ENOMEM; + memset(masks, 0xff, flen * sizeof(*masks)); for (pc = 0; pc < flen; pc++) { memvalid &= masks[pc]; switch (filter[pc].code) { - case BPF_S_ST: - case BPF_S_STX: + case BPF_ST: + case BPF_STX: memvalid |= (1 << filter[pc].k); break; - case BPF_S_LD_MEM: - case BPF_S_LDX_MEM: + case BPF_LD | BPF_MEM: + case BPF_LDX | BPF_MEM: if (!(memvalid & (1 << filter[pc].k))) { ret = -EINVAL; goto error; } break; - case BPF_S_JMP_JA: - /* a jump must set masks on target */ + case BPF_JMP | BPF_JA: + /* A jump must set masks on target */ masks[pc + 1 + filter[pc].k] &= memvalid; memvalid = ~0; break; - case BPF_S_JMP_JEQ_K: - case BPF_S_JMP_JEQ_X: - case BPF_S_JMP_JGE_K: - case BPF_S_JMP_JGE_X: - case BPF_S_JMP_JGT_K: - case BPF_S_JMP_JGT_X: - case BPF_S_JMP_JSET_X: - case BPF_S_JMP_JSET_K: - /* a jump must set masks on targets */ + case BPF_JMP | BPF_JEQ | BPF_K: + case BPF_JMP | BPF_JEQ | BPF_X: + case BPF_JMP | BPF_JGE | BPF_K: + case BPF_JMP | BPF_JGE | BPF_X: + case BPF_JMP | BPF_JGT | BPF_K: + case BPF_JMP | BPF_JGT | BPF_X: + case BPF_JMP | BPF_JSET | BPF_K: + case BPF_JMP | BPF_JSET | BPF_X: + /* A jump must set masks on targets */ masks[pc + 1 + filter[pc].jt] &= memvalid; masks[pc + 1 + filter[pc].jf] &= memvalid; memvalid = ~0; @@ -1214,6 +1147,72 @@ error: return ret; } +static bool chk_code_allowed(u16 code_to_probe) +{ + static const bool codes[] = { + /* 32 bit ALU operations */ + [BPF_ALU | BPF_ADD | BPF_K] = true, + [BPF_ALU | BPF_ADD | BPF_X] = true, + [BPF_ALU | BPF_SUB | BPF_K] = true, + [BPF_ALU | BPF_SUB | BPF_X] = true, + [BPF_ALU | BPF_MUL | BPF_K] = true, + [BPF_ALU | BPF_MUL | BPF_X] = true, + [BPF_ALU | BPF_DIV | BPF_K] = true, + [BPF_ALU | BPF_DIV | BPF_X] = true, + [BPF_ALU | BPF_MOD | BPF_K] = true, + [BPF_ALU | BPF_MOD | BPF_X] = true, + [BPF_ALU | BPF_AND | BPF_K] = true, + [BPF_ALU | BPF_AND | BPF_X] = true, + [BPF_ALU | BPF_OR | BPF_K] = true, + [BPF_ALU | BPF_OR | BPF_X] = true, + [BPF_ALU | BPF_XOR | BPF_K] = true, + [BPF_ALU | BPF_XOR | BPF_X] = true, + [BPF_ALU | BPF_LSH | BPF_K] = true, + [BPF_ALU | BPF_LSH | BPF_X] = true, + [BPF_ALU | BPF_RSH | BPF_K] = true, + [BPF_ALU | BPF_RSH | BPF_X] = true, + [BPF_ALU | BPF_NEG] = true, + /* Load instructions */ + [BPF_LD | BPF_W | BPF_ABS] = true, + [BPF_LD | BPF_H | BPF_ABS] = true, + [BPF_LD | BPF_B | BPF_ABS] = true, + [BPF_LD | BPF_W | BPF_LEN] = true, + [BPF_LD | BPF_W | BPF_IND] = true, + [BPF_LD | BPF_H | BPF_IND] = true, + [BPF_LD | BPF_B | BPF_IND] = true, + [BPF_LD | BPF_IMM] = true, + [BPF_LD | BPF_MEM] = true, + [BPF_LDX | BPF_W | BPF_LEN] = true, + [BPF_LDX | BPF_B | BPF_MSH] = true, + [BPF_LDX | BPF_IMM] = true, + [BPF_LDX | BPF_MEM] = true, + /* Store instructions */ + [BPF_ST] = true, + [BPF_STX] = true, + /* Misc instructions */ + [BPF_MISC | BPF_TAX] = true, + [BPF_MISC | BPF_TXA] = true, + /* Return instructions */ + [BPF_RET | BPF_K] = true, + [BPF_RET | BPF_A] = true, + /* Jump instructions */ + [BPF_JMP | BPF_JA] = true, + [BPF_JMP | BPF_JEQ | BPF_K] = true, + [BPF_JMP | BPF_JEQ | BPF_X] = true, + [BPF_JMP | BPF_JGE | BPF_K] = true, + [BPF_JMP | BPF_JGE | BPF_X] = true, + [BPF_JMP | BPF_JGT | BPF_K] = true, + [BPF_JMP | BPF_JGT | BPF_X] = true, + [BPF_JMP | BPF_JSET | BPF_K] = true, + [BPF_JMP | BPF_JSET | BPF_X] = true, + }; + + if (code_to_probe >= ARRAY_SIZE(codes)) + return false; + + return codes[code_to_probe]; +} + /** * sk_chk_filter - verify socket filter code * @filter: filter to verify @@ -1230,153 +1229,76 @@ error: */ int sk_chk_filter(struct sock_filter *filter, unsigned int flen) { - /* - * Valid instructions are initialized to non-0. - * Invalid instructions are initialized to 0. - */ - static const u8 codes[] = { - [BPF_ALU|BPF_ADD|BPF_K] = BPF_S_ALU_ADD_K, - [BPF_ALU|BPF_ADD|BPF_X] = BPF_S_ALU_ADD_X, - [BPF_ALU|BPF_SUB|BPF_K] = BPF_S_ALU_SUB_K, - [BPF_ALU|BPF_SUB|BPF_X] = BPF_S_ALU_SUB_X, - [BPF_ALU|BPF_MUL|BPF_K] = BPF_S_ALU_MUL_K, - [BPF_ALU|BPF_MUL|BPF_X] = BPF_S_ALU_MUL_X, - [BPF_ALU|BPF_DIV|BPF_X] = BPF_S_ALU_DIV_X, - [BPF_ALU|BPF_MOD|BPF_K] = BPF_S_ALU_MOD_K, - [BPF_ALU|BPF_MOD|BPF_X] = BPF_S_ALU_MOD_X, - [BPF_ALU|BPF_AND|BPF_K] = BPF_S_ALU_AND_K, - [BPF_ALU|BPF_AND|BPF_X] = BPF_S_ALU_AND_X, - [BPF_ALU|BPF_OR|BPF_K] = BPF_S_ALU_OR_K, - [BPF_ALU|BPF_OR|BPF_X] = BPF_S_ALU_OR_X, - [BPF_ALU|BPF_XOR|BPF_K] = BPF_S_ALU_XOR_K, - [BPF_ALU|BPF_XOR|BPF_X] = BPF_S_ALU_XOR_X, - [BPF_ALU|BPF_LSH|BPF_K] = BPF_S_ALU_LSH_K, - [BPF_ALU|BPF_LSH|BPF_X] = BPF_S_ALU_LSH_X, - [BPF_ALU|BPF_RSH|BPF_K] = BPF_S_ALU_RSH_K, - [BPF_ALU|BPF_RSH|BPF_X] = BPF_S_ALU_RSH_X, - [BPF_ALU|BPF_NEG] = BPF_S_ALU_NEG, - [BPF_LD|BPF_W|BPF_ABS] = BPF_S_LD_W_ABS, - [BPF_LD|BPF_H|BPF_ABS] = BPF_S_LD_H_ABS, - [BPF_LD|BPF_B|BPF_ABS] = BPF_S_LD_B_ABS, - [BPF_LD|BPF_W|BPF_LEN] = BPF_S_LD_W_LEN, - [BPF_LD|BPF_W|BPF_IND] = BPF_S_LD_W_IND, - [BPF_LD|BPF_H|BPF_IND] = BPF_S_LD_H_IND, - [BPF_LD|BPF_B|BPF_IND] = BPF_S_LD_B_IND, - [BPF_LD|BPF_IMM] = BPF_S_LD_IMM, - [BPF_LDX|BPF_W|BPF_LEN] = BPF_S_LDX_W_LEN, - [BPF_LDX|BPF_B|BPF_MSH] = BPF_S_LDX_B_MSH, - [BPF_LDX|BPF_IMM] = BPF_S_LDX_IMM, - [BPF_MISC|BPF_TAX] = BPF_S_MISC_TAX, - [BPF_MISC|BPF_TXA] = BPF_S_MISC_TXA, - [BPF_RET|BPF_K] = BPF_S_RET_K, - [BPF_RET|BPF_A] = BPF_S_RET_A, - [BPF_ALU|BPF_DIV|BPF_K] = BPF_S_ALU_DIV_K, - [BPF_LD|BPF_MEM] = BPF_S_LD_MEM, - [BPF_LDX|BPF_MEM] = BPF_S_LDX_MEM, - [BPF_ST] = BPF_S_ST, - [BPF_STX] = BPF_S_STX, - [BPF_JMP|BPF_JA] = BPF_S_JMP_JA, - [BPF_JMP|BPF_JEQ|BPF_K] = BPF_S_JMP_JEQ_K, - [BPF_JMP|BPF_JEQ|BPF_X] = BPF_S_JMP_JEQ_X, - [BPF_JMP|BPF_JGE|BPF_K] = BPF_S_JMP_JGE_K, - [BPF_JMP|BPF_JGE|BPF_X] = BPF_S_JMP_JGE_X, - [BPF_JMP|BPF_JGT|BPF_K] = BPF_S_JMP_JGT_K, - [BPF_JMP|BPF_JGT|BPF_X] = BPF_S_JMP_JGT_X, - [BPF_JMP|BPF_JSET|BPF_K] = BPF_S_JMP_JSET_K, - [BPF_JMP|BPF_JSET|BPF_X] = BPF_S_JMP_JSET_X, - }; - int pc; bool anc_found; + int pc; if (flen == 0 || flen > BPF_MAXINSNS) return -EINVAL; - /* check the filter code now */ + /* Check the filter code now */ for (pc = 0; pc < flen; pc++) { struct sock_filter *ftest = &filter[pc]; - u16 code = ftest->code; - if (code >= ARRAY_SIZE(codes)) - return -EINVAL; - code = codes[code]; - if (!code) + /* May we actually operate on this code? */ + if (!chk_code_allowed(ftest->code)) return -EINVAL; + /* Some instructions need special checks */ - switch (code) { - case BPF_S_ALU_DIV_K: - case BPF_S_ALU_MOD_K: - /* check for division by zero */ + switch (ftest->code) { + case BPF_ALU | BPF_DIV | BPF_K: + case BPF_ALU | BPF_MOD | BPF_K: + /* Check for division by zero */ if (ftest->k == 0) return -EINVAL; break; - case BPF_S_LD_MEM: - case BPF_S_LDX_MEM: - case BPF_S_ST: - case BPF_S_STX: - /* check for invalid memory addresses */ + case BPF_LD | BPF_MEM: + case BPF_LDX | BPF_MEM: + case BPF_ST: + case BPF_STX: + /* Check for invalid memory addresses */ if (ftest->k >= BPF_MEMWORDS) return -EINVAL; break; - case BPF_S_JMP_JA: - /* - * Note, the large ftest->k might cause loops. + case BPF_JMP | BPF_JA: + /* Note, the large ftest->k might cause loops. * Compare this with conditional jumps below, * where offsets are limited. --ANK (981016) */ - if (ftest->k >= (unsigned int)(flen-pc-1)) + if (ftest->k >= (unsigned int)(flen - pc - 1)) return -EINVAL; break; - case BPF_S_JMP_JEQ_K: - case BPF_S_JMP_JEQ_X: - case BPF_S_JMP_JGE_K: - case BPF_S_JMP_JGE_X: - case BPF_S_JMP_JGT_K: - case BPF_S_JMP_JGT_X: - case BPF_S_JMP_JSET_X: - case BPF_S_JMP_JSET_K: - /* for conditionals both must be safe */ + case BPF_JMP | BPF_JEQ | BPF_K: + case BPF_JMP | BPF_JEQ | BPF_X: + case BPF_JMP | BPF_JGE | BPF_K: + case BPF_JMP | BPF_JGE | BPF_X: + case BPF_JMP | BPF_JGT | BPF_K: + case BPF_JMP | BPF_JGT | BPF_X: + case BPF_JMP | BPF_JSET | BPF_K: + case BPF_JMP | BPF_JSET | BPF_X: + /* Both conditionals must be safe */ if (pc + ftest->jt + 1 >= flen || pc + ftest->jf + 1 >= flen) return -EINVAL; break; - case BPF_S_LD_W_ABS: - case BPF_S_LD_H_ABS: - case BPF_S_LD_B_ABS: + case BPF_LD | BPF_W | BPF_ABS: + case BPF_LD | BPF_H | BPF_ABS: + case BPF_LD | BPF_B | BPF_ABS: anc_found = false; -#define ANCILLARY(CODE) case SKF_AD_OFF + SKF_AD_##CODE: \ - code = BPF_S_ANC_##CODE; \ - anc_found = true; \ - break - switch (ftest->k) { - ANCILLARY(PROTOCOL); - ANCILLARY(PKTTYPE); - ANCILLARY(IFINDEX); - ANCILLARY(NLATTR); - ANCILLARY(NLATTR_NEST); - ANCILLARY(MARK); - ANCILLARY(QUEUE); - ANCILLARY(HATYPE); - ANCILLARY(RXHASH); - ANCILLARY(CPU); - ANCILLARY(ALU_XOR_X); - ANCILLARY(VLAN_TAG); - ANCILLARY(VLAN_TAG_PRESENT); - ANCILLARY(PAY_OFFSET); - } - - /* ancillary operation unknown or unsupported */ + if (bpf_anc_helper(ftest) & BPF_ANC) + anc_found = true; + /* Ancillary operation unknown or unsupported */ if (anc_found == false && ftest->k >= SKF_AD_OFF) return -EINVAL; } - ftest->code = code; } - /* last instruction must be a RET code */ + /* Last instruction must be a RET code */ switch (filter[flen - 1].code) { - case BPF_S_RET_K: - case BPF_S_RET_A: + case BPF_RET | BPF_K: + case BPF_RET | BPF_A: return check_load_and_stores(filter, flen); } + return -EINVAL; } EXPORT_SYMBOL(sk_chk_filter); @@ -1421,7 +1343,7 @@ static void sk_filter_release_rcu(struct rcu_head *rcu) struct sk_filter *fp = container_of(rcu, struct sk_filter, rcu); sk_release_orig_filter(fp); - bpf_jit_free(fp); + sk_filter_free(fp); } /** @@ -1459,7 +1381,7 @@ static struct sk_filter *__sk_migrate_realloc(struct sk_filter *fp, fp_new = sock_kmalloc(sk, len, GFP_KERNEL); if (fp_new) { - memcpy(fp_new, fp, sizeof(struct sk_filter)); + *fp_new = *fp; /* As we're kepping orig_prog in fp_new along, * we need to make sure we're not evicting it * from the old fp. @@ -1476,7 +1398,7 @@ static struct sk_filter *__sk_migrate_filter(struct sk_filter *fp, { struct sock_filter *old_prog; struct sk_filter *old_fp; - int i, err, new_len, old_len = fp->len; + int err, new_len, old_len = fp->len; /* We are free to overwrite insns et al right here as it * won't be used at this point in time anymore internally @@ -1486,13 +1408,6 @@ static struct sk_filter *__sk_migrate_filter(struct sk_filter *fp, BUILD_BUG_ON(sizeof(struct sock_filter) != sizeof(struct sock_filter_int)); - /* For now, we need to unfiddle BPF_S_* identifiers in place. - * This can sooner or later on be subject to removal, e.g. when - * JITs have been converted. - */ - for (i = 0; i < fp->len; i++) - sk_decode_filter(&fp->insns[i], &fp->insns[i]); - /* Conversion cannot happen on overlapping memory areas, * so we need to keep the user BPF around until the 2nd * pass. At this time, the user BPF is stored in fp->insns. @@ -1521,7 +1436,6 @@ static struct sk_filter *__sk_migrate_filter(struct sk_filter *fp, goto out_err_free; } - fp->bpf_func = sk_run_filter_int_skb; fp->len = new_len; /* 2nd pass: remap sock_filter insns into sock_filter_int insns. */ @@ -1534,6 +1448,8 @@ static struct sk_filter *__sk_migrate_filter(struct sk_filter *fp, */ goto out_err_free; + sk_filter_select_runtime(fp); + kfree(old_prog); return fp; @@ -1548,6 +1464,33 @@ out_err: return ERR_PTR(err); } +void __weak bpf_int_jit_compile(struct sk_filter *prog) +{ +} + +/** + * sk_filter_select_runtime - select execution runtime for BPF program + * @fp: sk_filter populated with internal BPF program + * + * try to JIT internal BPF program, if JIT is not available select interpreter + * BPF program will be executed via SK_RUN_FILTER() macro + */ +void sk_filter_select_runtime(struct sk_filter *fp) +{ + fp->bpf_func = (void *) __sk_run_filter; + + /* Probe if internal BPF can be JITed */ + bpf_int_jit_compile(fp); +} +EXPORT_SYMBOL_GPL(sk_filter_select_runtime); + +/* free internal BPF program */ +void sk_filter_free(struct sk_filter *fp) +{ + bpf_jit_free(fp); +} +EXPORT_SYMBOL_GPL(sk_filter_free); + static struct sk_filter *__sk_prepare_filter(struct sk_filter *fp, struct sock *sk) { @@ -1557,8 +1500,13 @@ static struct sk_filter *__sk_prepare_filter(struct sk_filter *fp, fp->jited = 0; err = sk_chk_filter(fp->insns, fp->len); - if (err) + if (err) { + if (sk != NULL) + sk_filter_uncharge(sk, fp); + else + kfree(fp); return ERR_PTR(err); + } /* Probe if we can JIT compile the filter and if so, do * the compilation of the filter. @@ -1585,7 +1533,7 @@ static struct sk_filter *__sk_prepare_filter(struct sk_filter *fp, * a negative errno code is returned. On success the return is zero. */ int sk_unattached_filter_create(struct sk_filter **pfp, - struct sock_fprog *fprog) + struct sock_fprog_kern *fprog) { unsigned int fsize = sk_filter_proglen(fprog); struct sk_filter *fp; @@ -1706,83 +1654,6 @@ int sk_detach_filter(struct sock *sk) } EXPORT_SYMBOL_GPL(sk_detach_filter); -void sk_decode_filter(struct sock_filter *filt, struct sock_filter *to) -{ - static const u16 decodes[] = { - [BPF_S_ALU_ADD_K] = BPF_ALU|BPF_ADD|BPF_K, - [BPF_S_ALU_ADD_X] = BPF_ALU|BPF_ADD|BPF_X, - [BPF_S_ALU_SUB_K] = BPF_ALU|BPF_SUB|BPF_K, - [BPF_S_ALU_SUB_X] = BPF_ALU|BPF_SUB|BPF_X, - [BPF_S_ALU_MUL_K] = BPF_ALU|BPF_MUL|BPF_K, - [BPF_S_ALU_MUL_X] = BPF_ALU|BPF_MUL|BPF_X, - [BPF_S_ALU_DIV_X] = BPF_ALU|BPF_DIV|BPF_X, - [BPF_S_ALU_MOD_K] = BPF_ALU|BPF_MOD|BPF_K, - [BPF_S_ALU_MOD_X] = BPF_ALU|BPF_MOD|BPF_X, - [BPF_S_ALU_AND_K] = BPF_ALU|BPF_AND|BPF_K, - [BPF_S_ALU_AND_X] = BPF_ALU|BPF_AND|BPF_X, - [BPF_S_ALU_OR_K] = BPF_ALU|BPF_OR|BPF_K, - [BPF_S_ALU_OR_X] = BPF_ALU|BPF_OR|BPF_X, - [BPF_S_ALU_XOR_K] = BPF_ALU|BPF_XOR|BPF_K, - [BPF_S_ALU_XOR_X] = BPF_ALU|BPF_XOR|BPF_X, - [BPF_S_ALU_LSH_K] = BPF_ALU|BPF_LSH|BPF_K, - [BPF_S_ALU_LSH_X] = BPF_ALU|BPF_LSH|BPF_X, - [BPF_S_ALU_RSH_K] = BPF_ALU|BPF_RSH|BPF_K, - [BPF_S_ALU_RSH_X] = BPF_ALU|BPF_RSH|BPF_X, - [BPF_S_ALU_NEG] = BPF_ALU|BPF_NEG, - [BPF_S_LD_W_ABS] = BPF_LD|BPF_W|BPF_ABS, - [BPF_S_LD_H_ABS] = BPF_LD|BPF_H|BPF_ABS, - [BPF_S_LD_B_ABS] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_ANC_PROTOCOL] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_ANC_PKTTYPE] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_ANC_IFINDEX] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_ANC_NLATTR] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_ANC_NLATTR_NEST] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_ANC_MARK] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_ANC_QUEUE] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_ANC_HATYPE] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_ANC_RXHASH] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_ANC_CPU] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_ANC_ALU_XOR_X] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_ANC_VLAN_TAG] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_ANC_VLAN_TAG_PRESENT] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_ANC_PAY_OFFSET] = BPF_LD|BPF_B|BPF_ABS, - [BPF_S_LD_W_LEN] = BPF_LD|BPF_W|BPF_LEN, - [BPF_S_LD_W_IND] = BPF_LD|BPF_W|BPF_IND, - [BPF_S_LD_H_IND] = BPF_LD|BPF_H|BPF_IND, - [BPF_S_LD_B_IND] = BPF_LD|BPF_B|BPF_IND, - [BPF_S_LD_IMM] = BPF_LD|BPF_IMM, - [BPF_S_LDX_W_LEN] = BPF_LDX|BPF_W|BPF_LEN, - [BPF_S_LDX_B_MSH] = BPF_LDX|BPF_B|BPF_MSH, - [BPF_S_LDX_IMM] = BPF_LDX|BPF_IMM, - [BPF_S_MISC_TAX] = BPF_MISC|BPF_TAX, - [BPF_S_MISC_TXA] = BPF_MISC|BPF_TXA, - [BPF_S_RET_K] = BPF_RET|BPF_K, - [BPF_S_RET_A] = BPF_RET|BPF_A, - [BPF_S_ALU_DIV_K] = BPF_ALU|BPF_DIV|BPF_K, - [BPF_S_LD_MEM] = BPF_LD|BPF_MEM, - [BPF_S_LDX_MEM] = BPF_LDX|BPF_MEM, - [BPF_S_ST] = BPF_ST, - [BPF_S_STX] = BPF_STX, - [BPF_S_JMP_JA] = BPF_JMP|BPF_JA, - [BPF_S_JMP_JEQ_K] = BPF_JMP|BPF_JEQ|BPF_K, - [BPF_S_JMP_JEQ_X] = BPF_JMP|BPF_JEQ|BPF_X, - [BPF_S_JMP_JGE_K] = BPF_JMP|BPF_JGE|BPF_K, - [BPF_S_JMP_JGE_X] = BPF_JMP|BPF_JGE|BPF_X, - [BPF_S_JMP_JGT_K] = BPF_JMP|BPF_JGT|BPF_K, - [BPF_S_JMP_JGT_X] = BPF_JMP|BPF_JGT|BPF_X, - [BPF_S_JMP_JSET_K] = BPF_JMP|BPF_JSET|BPF_K, - [BPF_S_JMP_JSET_X] = BPF_JMP|BPF_JSET|BPF_X, - }; - u16 code; - - code = filt->code; - - to->code = decodes[code]; - to->jt = filt->jt; - to->jf = filt->jf; - to->k = filt->k; -} - int sk_get_filter(struct sock *sk, struct sock_filter __user *ubuf, unsigned int len) { diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 8f8a96ef9f3..32d872eec7f 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1248,8 +1248,8 @@ void __neigh_set_probe_once(struct neighbour *neigh) neigh->updated = jiffies; if (!(neigh->nud_state & NUD_FAILED)) return; - neigh->nud_state = NUD_PROBE; - atomic_set(&neigh->probes, NEIGH_VAR(neigh->parms, UCAST_PROBES)); + neigh->nud_state = NUD_INCOMPLETE; + atomic_set(&neigh->probes, neigh_max_probes(neigh)); neigh_add_timer(neigh, jiffies + NEIGH_VAR(neigh->parms, RETRANS_TIME)); } diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index 81d3a9a0845..85b62691f4f 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -24,7 +24,7 @@ static LIST_HEAD(pernet_list); static struct list_head *first_device = &pernet_list; -static DEFINE_MUTEX(net_mutex); +DEFINE_MUTEX(net_mutex); LIST_HEAD(net_namespace_list); EXPORT_SYMBOL_GPL(net_namespace_list); @@ -273,7 +273,7 @@ static void cleanup_net(struct work_struct *work) { const struct pernet_operations *ops; struct net *net, *tmp; - LIST_HEAD(net_kill_list); + struct list_head net_kill_list; LIST_HEAD(net_exit_list); /* Atomically snapshot the list of namespaces to cleanup */ diff --git a/net/core/pktgen.c b/net/core/pktgen.c index 0304f981f7f..fc17a9d309a 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -573,7 +573,7 @@ static int pktgen_if_show(struct seq_file *seq, void *v) is_zero_ether_addr(pkt_dev->src_mac) ? pkt_dev->odev->dev_addr : pkt_dev->src_mac); - seq_printf(seq, "dst_mac: "); + seq_puts(seq, "dst_mac: "); seq_printf(seq, "%pM\n", pkt_dev->dst_mac); seq_printf(seq, @@ -588,7 +588,7 @@ static int pktgen_if_show(struct seq_file *seq, void *v) if (pkt_dev->nr_labels) { unsigned int i; - seq_printf(seq, " mpls: "); + seq_puts(seq, " mpls: "); for (i = 0; i < pkt_dev->nr_labels; i++) seq_printf(seq, "%08x%s", ntohl(pkt_dev->labels[i]), i == pkt_dev->nr_labels-1 ? "\n" : ", "); @@ -613,67 +613,67 @@ static int pktgen_if_show(struct seq_file *seq, void *v) if (pkt_dev->node >= 0) seq_printf(seq, " node: %d\n", pkt_dev->node); - seq_printf(seq, " Flags: "); + seq_puts(seq, " Flags: "); if (pkt_dev->flags & F_IPV6) - seq_printf(seq, "IPV6 "); + seq_puts(seq, "IPV6 "); if (pkt_dev->flags & F_IPSRC_RND) - seq_printf(seq, "IPSRC_RND "); + seq_puts(seq, "IPSRC_RND "); if (pkt_dev->flags & F_IPDST_RND) - seq_printf(seq, "IPDST_RND "); + seq_puts(seq, "IPDST_RND "); if (pkt_dev->flags & F_TXSIZE_RND) - seq_printf(seq, "TXSIZE_RND "); + seq_puts(seq, "TXSIZE_RND "); if (pkt_dev->flags & F_UDPSRC_RND) - seq_printf(seq, "UDPSRC_RND "); + seq_puts(seq, "UDPSRC_RND "); if (pkt_dev->flags & F_UDPDST_RND) - seq_printf(seq, "UDPDST_RND "); + seq_puts(seq, "UDPDST_RND "); if (pkt_dev->flags & F_UDPCSUM) - seq_printf(seq, "UDPCSUM "); + seq_puts(seq, "UDPCSUM "); if (pkt_dev->flags & F_MPLS_RND) - seq_printf(seq, "MPLS_RND "); + seq_puts(seq, "MPLS_RND "); if (pkt_dev->flags & F_QUEUE_MAP_RND) - seq_printf(seq, "QUEUE_MAP_RND "); + seq_puts(seq, "QUEUE_MAP_RND "); if (pkt_dev->flags & F_QUEUE_MAP_CPU) - seq_printf(seq, "QUEUE_MAP_CPU "); + seq_puts(seq, "QUEUE_MAP_CPU "); if (pkt_dev->cflows) { if (pkt_dev->flags & F_FLOW_SEQ) - seq_printf(seq, "FLOW_SEQ "); /*in sequence flows*/ + seq_puts(seq, "FLOW_SEQ "); /*in sequence flows*/ else - seq_printf(seq, "FLOW_RND "); + seq_puts(seq, "FLOW_RND "); } #ifdef CONFIG_XFRM if (pkt_dev->flags & F_IPSEC_ON) { - seq_printf(seq, "IPSEC "); + seq_puts(seq, "IPSEC "); if (pkt_dev->spi) seq_printf(seq, "spi:%u", pkt_dev->spi); } #endif if (pkt_dev->flags & F_MACSRC_RND) - seq_printf(seq, "MACSRC_RND "); + seq_puts(seq, "MACSRC_RND "); if (pkt_dev->flags & F_MACDST_RND) - seq_printf(seq, "MACDST_RND "); + seq_puts(seq, "MACDST_RND "); if (pkt_dev->flags & F_VID_RND) - seq_printf(seq, "VID_RND "); + seq_puts(seq, "VID_RND "); if (pkt_dev->flags & F_SVID_RND) - seq_printf(seq, "SVID_RND "); + seq_puts(seq, "SVID_RND "); if (pkt_dev->flags & F_NODE) - seq_printf(seq, "NODE_ALLOC "); + seq_puts(seq, "NODE_ALLOC "); seq_puts(seq, "\n"); @@ -716,7 +716,7 @@ static int pktgen_if_show(struct seq_file *seq, void *v) if (pkt_dev->result[0]) seq_printf(seq, "Result: %s\n", pkt_dev->result); else - seq_printf(seq, "Result: Idle\n"); + seq_puts(seq, "Result: Idle\n"); return 0; } @@ -1735,14 +1735,14 @@ static int pktgen_thread_show(struct seq_file *seq, void *v) BUG_ON(!t); - seq_printf(seq, "Running: "); + seq_puts(seq, "Running: "); if_lock(t); list_for_each_entry(pkt_dev, &t->if_list, list) if (pkt_dev->running) seq_printf(seq, "%s ", pkt_dev->odevname); - seq_printf(seq, "\nStopped: "); + seq_puts(seq, "\nStopped: "); list_for_each_entry(pkt_dev, &t->if_list, list) if (!pkt_dev->running) @@ -1751,7 +1751,7 @@ static int pktgen_thread_show(struct seq_file *seq, void *v) if (t->result[0]) seq_printf(seq, "\nResult: %s\n", t->result); else - seq_printf(seq, "\nResult: NA\n"); + seq_puts(seq, "\nResult: NA\n"); if_unlock(t); diff --git a/net/core/ptp_classifier.c b/net/core/ptp_classifier.c index eaba0f68f86..d3027a73fd4 100644 --- a/net/core/ptp_classifier.c +++ b/net/core/ptp_classifier.c @@ -88,7 +88,7 @@ EXPORT_SYMBOL_GPL(ptp_classify_raw); void __init ptp_classifier_init(void) { - static struct sock_filter ptp_filter[] = { + static struct sock_filter ptp_filter[] __initdata = { { 0x28, 0, 0, 0x0000000c }, { 0x15, 0, 12, 0x00000800 }, { 0x30, 0, 0, 0x00000017 }, @@ -133,7 +133,7 @@ void __init ptp_classifier_init(void) { 0x16, 0, 0, 0x00000000 }, { 0x06, 0, 0, 0x00000000 }, }; - struct sock_fprog ptp_prog = { + struct sock_fprog_kern ptp_prog = { .len = ARRAY_SIZE(ptp_filter), .filter = ptp_filter, }; diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index d4ff41739b0..741b22c62ac 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -353,15 +353,46 @@ void __rtnl_link_unregister(struct rtnl_link_ops *ops) } EXPORT_SYMBOL_GPL(__rtnl_link_unregister); +/* Return with the rtnl_lock held when there are no network + * devices unregistering in any network namespace. + */ +static void rtnl_lock_unregistering_all(void) +{ + struct net *net; + bool unregistering; + DEFINE_WAIT(wait); + + for (;;) { + prepare_to_wait(&netdev_unregistering_wq, &wait, + TASK_UNINTERRUPTIBLE); + unregistering = false; + rtnl_lock(); + for_each_net(net) { + if (net->dev_unreg_count > 0) { + unregistering = true; + break; + } + } + if (!unregistering) + break; + __rtnl_unlock(); + schedule(); + } + finish_wait(&netdev_unregistering_wq, &wait); +} + /** * rtnl_link_unregister - Unregister rtnl_link_ops from rtnetlink. * @ops: struct rtnl_link_ops * to unregister */ void rtnl_link_unregister(struct rtnl_link_ops *ops) { - rtnl_lock(); + /* Close the race with cleanup_net() */ + mutex_lock(&net_mutex); + rtnl_lock_unregistering_all(); __rtnl_link_unregister(ops); rtnl_unlock(); + mutex_unlock(&net_mutex); } EXPORT_SYMBOL_GPL(rtnl_link_unregister); @@ -767,14 +798,15 @@ static inline int rtnl_vfinfo_size(const struct net_device *dev, size += num_vfs * (nla_total_size(sizeof(struct ifla_vf_mac)) + nla_total_size(sizeof(struct ifla_vf_vlan)) + - nla_total_size(sizeof(struct ifla_vf_tx_rate)) + - nla_total_size(sizeof(struct ifla_vf_spoofchk))); + nla_total_size(sizeof(struct ifla_vf_spoofchk)) + + nla_total_size(sizeof(struct ifla_vf_rate))); return size; } else return 0; } -static size_t rtnl_port_size(const struct net_device *dev) +static size_t rtnl_port_size(const struct net_device *dev, + u32 ext_filter_mask) { size_t port_size = nla_total_size(4) /* PORT_VF */ + nla_total_size(PORT_PROFILE_MAX) /* PORT_PROFILE */ @@ -790,7 +822,8 @@ static size_t rtnl_port_size(const struct net_device *dev) size_t port_self_size = nla_total_size(sizeof(struct nlattr)) + port_size; - if (!dev->netdev_ops->ndo_get_vf_port || !dev->dev.parent) + if (!dev->netdev_ops->ndo_get_vf_port || !dev->dev.parent || + !(ext_filter_mask & RTEXT_FILTER_VF)) return 0; if (dev_num_vf(dev->dev.parent)) return port_self_size + vf_ports_size + @@ -826,7 +859,7 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev, + nla_total_size(ext_filter_mask & RTEXT_FILTER_VF ? 4 : 0) /* IFLA_NUM_VF */ + rtnl_vfinfo_size(dev, ext_filter_mask) /* IFLA_VFINFO_LIST */ - + rtnl_port_size(dev) /* IFLA_VF_PORTS + IFLA_PORT_SELF */ + + rtnl_port_size(dev, ext_filter_mask) /* IFLA_VF_PORTS + IFLA_PORT_SELF */ + rtnl_link_get_size(dev) /* IFLA_LINKINFO */ + rtnl_link_get_af_size(dev) /* IFLA_AF_SPEC */ + nla_total_size(MAX_PHYS_PORT_ID_LEN); /* IFLA_PHYS_PORT_ID */ @@ -888,11 +921,13 @@ static int rtnl_port_self_fill(struct sk_buff *skb, struct net_device *dev) return 0; } -static int rtnl_port_fill(struct sk_buff *skb, struct net_device *dev) +static int rtnl_port_fill(struct sk_buff *skb, struct net_device *dev, + u32 ext_filter_mask) { int err; - if (!dev->netdev_ops->ndo_get_vf_port || !dev->dev.parent) + if (!dev->netdev_ops->ndo_get_vf_port || !dev->dev.parent || + !(ext_filter_mask & RTEXT_FILTER_VF)) return 0; err = rtnl_port_self_fill(skb, dev); @@ -1030,6 +1065,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, struct ifla_vf_info ivi; struct ifla_vf_mac vf_mac; struct ifla_vf_vlan vf_vlan; + struct ifla_vf_rate vf_rate; struct ifla_vf_tx_rate vf_tx_rate; struct ifla_vf_spoofchk vf_spoofchk; struct ifla_vf_link_state vf_linkstate; @@ -1050,6 +1086,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, break; vf_mac.vf = vf_vlan.vf = + vf_rate.vf = vf_tx_rate.vf = vf_spoofchk.vf = vf_linkstate.vf = ivi.vf; @@ -1057,7 +1094,9 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, memcpy(vf_mac.mac, ivi.mac, sizeof(ivi.mac)); vf_vlan.vlan = ivi.vlan; vf_vlan.qos = ivi.qos; - vf_tx_rate.rate = ivi.tx_rate; + vf_tx_rate.rate = ivi.max_tx_rate; + vf_rate.min_tx_rate = ivi.min_tx_rate; + vf_rate.max_tx_rate = ivi.max_tx_rate; vf_spoofchk.setting = ivi.spoofchk; vf_linkstate.link_state = ivi.linkstate; vf = nla_nest_start(skb, IFLA_VF_INFO); @@ -1067,6 +1106,8 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, } if (nla_put(skb, IFLA_VF_MAC, sizeof(vf_mac), &vf_mac) || nla_put(skb, IFLA_VF_VLAN, sizeof(vf_vlan), &vf_vlan) || + nla_put(skb, IFLA_VF_RATE, sizeof(vf_rate), + &vf_rate) || nla_put(skb, IFLA_VF_TX_RATE, sizeof(vf_tx_rate), &vf_tx_rate) || nla_put(skb, IFLA_VF_SPOOFCHK, sizeof(vf_spoofchk), @@ -1079,7 +1120,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, nla_nest_end(skb, vfinfo); } - if (rtnl_port_fill(skb, dev)) + if (rtnl_port_fill(skb, dev, ext_filter_mask)) goto nla_put_failure; if (dev->rtnl_link_ops || rtnl_have_link_slave_info(dev)) { @@ -1173,6 +1214,8 @@ static const struct nla_policy ifla_vf_policy[IFLA_VF_MAX+1] = { .len = sizeof(struct ifla_vf_tx_rate) }, [IFLA_VF_SPOOFCHK] = { .type = NLA_BINARY, .len = sizeof(struct ifla_vf_spoofchk) }, + [IFLA_VF_RATE] = { .type = NLA_BINARY, + .len = sizeof(struct ifla_vf_rate) }, }; static const struct nla_policy ifla_port_policy[IFLA_PORT_MAX+1] = { @@ -1198,6 +1241,7 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) struct hlist_head *head; struct nlattr *tb[IFLA_MAX+1]; u32 ext_filter_mask = 0; + int err; s_h = cb->args[0]; s_idx = cb->args[1]; @@ -1218,11 +1262,17 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) hlist_for_each_entry_rcu(dev, head, index_hlist) { if (idx < s_idx) goto cont; - if (rtnl_fill_ifinfo(skb, dev, RTM_NEWLINK, - NETLINK_CB(cb->skb).portid, - cb->nlh->nlmsg_seq, 0, - NLM_F_MULTI, - ext_filter_mask) <= 0) + err = rtnl_fill_ifinfo(skb, dev, RTM_NEWLINK, + NETLINK_CB(cb->skb).portid, + cb->nlh->nlmsg_seq, 0, + NLM_F_MULTI, + ext_filter_mask); + /* If we ran out of room on the first message, + * we're in trouble + */ + WARN_ON((err == -EMSGSIZE) && (skb->len == 0)); + + if (err <= 0) goto out; nl_dump_check_consistent(cb, nlmsg_hdr(skb)); @@ -1325,11 +1375,29 @@ static int do_setvfinfo(struct net_device *dev, struct nlattr *attr) } case IFLA_VF_TX_RATE: { struct ifla_vf_tx_rate *ivt; + struct ifla_vf_info ivf; ivt = nla_data(vf); err = -EOPNOTSUPP; - if (ops->ndo_set_vf_tx_rate) - err = ops->ndo_set_vf_tx_rate(dev, ivt->vf, - ivt->rate); + if (ops->ndo_get_vf_config) + err = ops->ndo_get_vf_config(dev, ivt->vf, + &ivf); + if (err) + break; + err = -EOPNOTSUPP; + if (ops->ndo_set_vf_rate) + err = ops->ndo_set_vf_rate(dev, ivt->vf, + ivf.min_tx_rate, + ivt->rate); + break; + } + case IFLA_VF_RATE: { + struct ifla_vf_rate *ivt; + ivt = nla_data(vf); + err = -EOPNOTSUPP; + if (ops->ndo_set_vf_rate) + err = ops->ndo_set_vf_rate(dev, ivt->vf, + ivt->min_tx_rate, + ivt->max_tx_rate); break; } case IFLA_VF_SPOOFCHK: { @@ -1395,7 +1463,8 @@ static int do_set_master(struct net_device *dev, int ifindex) return 0; } -static int do_setlink(struct net_device *dev, struct ifinfomsg *ifm, +static int do_setlink(const struct sk_buff *skb, + struct net_device *dev, struct ifinfomsg *ifm, struct nlattr **tb, char *ifname, int modified) { const struct net_device_ops *ops = dev->netdev_ops; @@ -1407,7 +1476,7 @@ static int do_setlink(struct net_device *dev, struct ifinfomsg *ifm, err = PTR_ERR(net); goto errout; } - if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) { + if (!netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) { err = -EPERM; goto errout; } @@ -1661,7 +1730,7 @@ static int rtnl_setlink(struct sk_buff *skb, struct nlmsghdr *nlh) if (err < 0) goto errout; - err = do_setlink(dev, ifm, tb, ifname, 0); + err = do_setlink(skb, dev, ifm, tb, ifname, 0); errout: return err; } @@ -1778,7 +1847,8 @@ err: } EXPORT_SYMBOL(rtnl_create_link); -static int rtnl_group_changelink(struct net *net, int group, +static int rtnl_group_changelink(const struct sk_buff *skb, + struct net *net, int group, struct ifinfomsg *ifm, struct nlattr **tb) { @@ -1787,7 +1857,7 @@ static int rtnl_group_changelink(struct net *net, int group, for_each_netdev(net, dev) { if (dev->group == group) { - err = do_setlink(dev, ifm, tb, NULL, 0); + err = do_setlink(skb, dev, ifm, tb, NULL, 0); if (err < 0) return err; } @@ -1929,12 +1999,12 @@ replay: modified = 1; } - return do_setlink(dev, ifm, tb, ifname, modified); + return do_setlink(skb, dev, ifm, tb, ifname, modified); } if (!(nlh->nlmsg_flags & NLM_F_CREATE)) { if (ifm->ifi_index == 0 && tb[IFLA_GROUP]) - return rtnl_group_changelink(net, + return rtnl_group_changelink(skb, net, nla_get_u32(tb[IFLA_GROUP]), ifm, tb); return -ENODEV; @@ -1975,11 +2045,15 @@ replay: if (ops->newlink) { err = ops->newlink(net, dev, tb, data); /* Drivers should call free_netdev() in ->destructor - * and unregister it on failure so that device could be - * finally freed in rtnl_unlock. + * and unregister it on failure after registration + * so that device could be finally freed in rtnl_unlock. */ - if (err < 0) + if (err < 0) { + /* If device is not registered at all, free it now */ + if (dev->reg_state == NETREG_UNINITIALIZED) + free_netdev(dev); goto out; + } } else { err = register_netdevice(dev); if (err < 0) { @@ -2321,7 +2395,7 @@ static int rtnl_fdb_del(struct sk_buff *skb, struct nlmsghdr *nlh) int err = -EINVAL; __u8 *addr; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; err = nlmsg_parse(nlh, sizeof(*ndm), tb, NDA_MAX, NULL); @@ -2773,7 +2847,7 @@ static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) sz_idx = type>>2; kind = type&3; - if (kind != 2 && !ns_capable(net->user_ns, CAP_NET_ADMIN)) + if (kind != 2 && !netlink_net_capable(skb, CAP_NET_ADMIN)) return -EPERM; if (kind == 2 && nlh->nlmsg_flags&NLM_F_DUMP) { diff --git a/net/core/secure_seq.c b/net/core/secure_seq.c index 897da56f3af..ba71212f025 100644 --- a/net/core/secure_seq.c +++ b/net/core/secure_seq.c @@ -85,31 +85,6 @@ EXPORT_SYMBOL(secure_ipv6_port_ephemeral); #endif #ifdef CONFIG_INET -__u32 secure_ip_id(__be32 daddr) -{ - u32 hash[MD5_DIGEST_WORDS]; - - net_secret_init(); - hash[0] = (__force __u32) daddr; - hash[1] = net_secret[13]; - hash[2] = net_secret[14]; - hash[3] = net_secret[15]; - - md5_transform(hash, net_secret); - - return hash[0]; -} - -__u32 secure_ipv6_id(const __be32 daddr[4]) -{ - __u32 hash[4]; - - net_secret_init(); - memcpy(hash, daddr, 16); - md5_transform(hash, net_secret); - - return hash[0]; -} __u32 secure_tcp_sequence_number(__be32 saddr, __be32 daddr, __be16 sport, __be16 dport) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 1b62343f583..05f4bef2ce1 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -694,7 +694,7 @@ static void __copy_skb_header(struct sk_buff *new, const struct sk_buff *old) #endif memcpy(new->cb, old->cb, sizeof(old->cb)); new->csum = old->csum; - new->local_df = old->local_df; + new->ignore_df = old->ignore_df; new->pkt_type = old->pkt_type; new->ip_summed = old->ip_summed; skb_copy_queue_mapping(new, old); @@ -2885,7 +2885,9 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb, if (unlikely(!proto)) return ERR_PTR(-EINVAL); - csum = !!can_checksum_protocol(features, proto); + csum = !head_skb->encap_hdr_csum && + !!can_checksum_protocol(features, proto); + __skb_push(head_skb, doffset); headroom = skb_headroom(head_skb); pos = skb_headlen(head_skb); @@ -2983,6 +2985,8 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb, nskb->csum = skb_copy_and_csum_bits(head_skb, offset, skb_put(nskb, len), len, 0); + SKB_GSO_CB(nskb)->csum_start = + skb_headroom(nskb) + offset; continue; } @@ -3052,6 +3056,8 @@ perform_csum_check: nskb->csum = skb_checksum(nskb, doffset, nskb->len - doffset, 0); nskb->ip_summed = CHECKSUM_NONE; + SKB_GSO_CB(nskb)->csum_start = + skb_headroom(nskb) + doffset; } } while ((offset += len) < head_skb->len); @@ -3076,7 +3082,7 @@ int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb) if (unlikely(p->len + len >= 65536)) return -E2BIG; - lp = NAPI_GRO_CB(p)->last ?: p; + lp = NAPI_GRO_CB(p)->last; pinfo = skb_shinfo(lp); if (headlen <= offset) { @@ -3192,7 +3198,7 @@ merge: __skb_pull(skb, offset); - if (!NAPI_GRO_CB(p)->last) + if (NAPI_GRO_CB(p)->last == p) skb_shinfo(p)->frag_list = skb; else NAPI_GRO_CB(p)->last->next = skb; @@ -3913,7 +3919,7 @@ void skb_scrub_packet(struct sk_buff *skb, bool xnet) skb->tstamp.tv64 = 0; skb->pkt_type = PACKET_HOST; skb->skb_iif = 0; - skb->local_df = 0; + skb->ignore_df = 0; skb_dst_drop(skb); skb->mark = 0; secpath_reset(skb); diff --git a/net/core/sock.c b/net/core/sock.c index b4fff008136..026e01f7027 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -145,6 +145,55 @@ static DEFINE_MUTEX(proto_list_mutex); static LIST_HEAD(proto_list); +/** + * sk_ns_capable - General socket capability test + * @sk: Socket to use a capability on or through + * @user_ns: The user namespace of the capability to use + * @cap: The capability to use + * + * Test to see if the opener of the socket had when the socket was + * created and the current process has the capability @cap in the user + * namespace @user_ns. + */ +bool sk_ns_capable(const struct sock *sk, + struct user_namespace *user_ns, int cap) +{ + return file_ns_capable(sk->sk_socket->file, user_ns, cap) && + ns_capable(user_ns, cap); +} +EXPORT_SYMBOL(sk_ns_capable); + +/** + * sk_capable - Socket global capability test + * @sk: Socket to use a capability on or through + * @cap: The global capbility to use + * + * Test to see if the opener of the socket had when the socket was + * created and the current process has the capability @cap in all user + * namespaces. + */ +bool sk_capable(const struct sock *sk, int cap) +{ + return sk_ns_capable(sk, &init_user_ns, cap); +} +EXPORT_SYMBOL(sk_capable); + +/** + * sk_net_capable - Network namespace socket capability test + * @sk: Socket to use a capability on or through + * @cap: The capability to use + * + * Test to see if the opener of the socket had when the socke was created + * and the current process has the capability @cap over the network namespace + * the socket is a member of. + */ +bool sk_net_capable(const struct sock *sk, int cap) +{ + return sk_ns_capable(sk, sock_net(sk)->user_ns, cap); +} +EXPORT_SYMBOL(sk_net_capable); + + #ifdef CONFIG_MEMCG_KMEM int mem_cgroup_sockets_init(struct mem_cgroup *memcg, struct cgroup_subsys *ss) { @@ -735,7 +784,7 @@ set_rcvbuf: break; case SO_NO_CHECK: - sk->sk_no_check = valbool; + sk->sk_no_check_tx = valbool; break; case SO_PRIORITY: @@ -1015,7 +1064,7 @@ int sock_getsockopt(struct socket *sock, int level, int optname, break; case SO_NO_CHECK: - v.val = sk->sk_no_check; + v.val = sk->sk_no_check_tx; break; case SO_PRIORITY: diff --git a/net/core/sock_diag.c b/net/core/sock_diag.c index d7af1885932..a4216a4c957 100644 --- a/net/core/sock_diag.c +++ b/net/core/sock_diag.c @@ -49,7 +49,7 @@ int sock_diag_put_meminfo(struct sock *sk, struct sk_buff *skb, int attrtype) } EXPORT_SYMBOL_GPL(sock_diag_put_meminfo); -int sock_diag_put_filterinfo(struct user_namespace *user_ns, struct sock *sk, +int sock_diag_put_filterinfo(bool may_report_filterinfo, struct sock *sk, struct sk_buff *skb, int attrtype) { struct sock_fprog_kern *fprog; @@ -58,7 +58,7 @@ int sock_diag_put_filterinfo(struct user_namespace *user_ns, struct sock *sk, unsigned int flen; int err = 0; - if (!ns_capable(user_ns, CAP_NET_ADMIN)) { + if (!may_report_filterinfo) { nla_reserve(skb, attrtype, 0); return 0; } diff --git a/net/core/tso.c b/net/core/tso.c new file mode 100644 index 00000000000..8c3203c585b --- /dev/null +++ b/net/core/tso.c @@ -0,0 +1,77 @@ +#include <linux/export.h> +#include <net/ip.h> +#include <net/tso.h> + +/* Calculate expected number of TX descriptors */ +int tso_count_descs(struct sk_buff *skb) +{ + /* The Marvell Way */ + return skb_shinfo(skb)->gso_segs * 2 + skb_shinfo(skb)->nr_frags; +} +EXPORT_SYMBOL(tso_count_descs); + +void tso_build_hdr(struct sk_buff *skb, char *hdr, struct tso_t *tso, + int size, bool is_last) +{ + struct iphdr *iph; + struct tcphdr *tcph; + int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb); + int mac_hdr_len = skb_network_offset(skb); + + memcpy(hdr, skb->data, hdr_len); + iph = (struct iphdr *)(hdr + mac_hdr_len); + iph->id = htons(tso->ip_id); + iph->tot_len = htons(size + hdr_len - mac_hdr_len); + tcph = (struct tcphdr *)(hdr + skb_transport_offset(skb)); + tcph->seq = htonl(tso->tcp_seq); + tso->ip_id++; + + if (!is_last) { + /* Clear all special flags for not last packet */ + tcph->psh = 0; + tcph->fin = 0; + tcph->rst = 0; + } +} +EXPORT_SYMBOL(tso_build_hdr); + +void tso_build_data(struct sk_buff *skb, struct tso_t *tso, int size) +{ + tso->tcp_seq += size; + tso->size -= size; + tso->data += size; + + if ((tso->size == 0) && + (tso->next_frag_idx < skb_shinfo(skb)->nr_frags)) { + skb_frag_t *frag = &skb_shinfo(skb)->frags[tso->next_frag_idx]; + + /* Move to next segment */ + tso->size = frag->size; + tso->data = page_address(frag->page.p) + frag->page_offset; + tso->next_frag_idx++; + } +} +EXPORT_SYMBOL(tso_build_data); + +void tso_start(struct sk_buff *skb, struct tso_t *tso) +{ + int hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb); + + tso->ip_id = ntohs(ip_hdr(skb)->id); + tso->tcp_seq = ntohl(tcp_hdr(skb)->seq); + tso->next_frag_idx = 0; + + /* Build first data */ + tso->size = skb_headlen(skb) - hdr_len; + tso->data = skb->data + hdr_len; + if ((tso->size == 0) && + (tso->next_frag_idx < skb_shinfo(skb)->nr_frags)) { + skb_frag_t *frag = &skb_shinfo(skb)->frags[tso->next_frag_idx]; + + /* Move to next segment */ + tso->size = frag->size; + tso->data = page_address(frag->page.p) + frag->page_offset; + tso->next_frag_idx++; + } +} +EXPORT_SYMBOL(tso_start); diff --git a/net/core/utils.c b/net/core/utils.c index 2f737bf90b3..eed34338736 100644 --- a/net/core/utils.c +++ b/net/core/utils.c @@ -348,8 +348,8 @@ static void __net_random_once_deferred(struct work_struct *w) { struct __net_random_once_work *work = container_of(w, struct __net_random_once_work, work); - if (!static_key_enabled(work->key)) - static_key_slow_inc(work->key); + BUG_ON(!static_key_enabled(work->key)); + static_key_slow_dec(work->key); kfree(work); } @@ -367,7 +367,7 @@ static void __net_random_once_disable_jump(struct static_key *key) } bool __net_get_random_once(void *buf, int nbytes, bool *done, - struct static_key *done_key) + struct static_key *once_key) { static DEFINE_SPINLOCK(lock); unsigned long flags; @@ -382,7 +382,7 @@ bool __net_get_random_once(void *buf, int nbytes, bool *done, *done = true; spin_unlock_irqrestore(&lock, flags); - __net_random_once_disable_jump(done_key); + __net_random_once_disable_jump(once_key); return true; } diff --git a/net/dcb/dcbnl.c b/net/dcb/dcbnl.c index 55364440267..f8b98d89c28 100644 --- a/net/dcb/dcbnl.c +++ b/net/dcb/dcbnl.c @@ -1669,7 +1669,7 @@ static int dcb_doit(struct sk_buff *skb, struct nlmsghdr *nlh) struct nlmsghdr *reply_nlh = NULL; const struct reply_func *fn; - if ((nlh->nlmsg_type == RTM_SETDCB) && !capable(CAP_NET_ADMIN)) + if ((nlh->nlmsg_type == RTM_SETDCB) && !netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; ret = nlmsg_parse(nlh, sizeof(*dcb), tb, DCB_ATTR_MAX, diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 22b5d818b20..6ca645c4b48 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -1024,7 +1024,6 @@ static struct inet_protosw dccp_v4_protosw = { .protocol = IPPROTO_DCCP, .prot = &dccp_v4_prot, .ops = &inet_dccp_ops, - .no_check = 0, .flags = INET_PROTOSW_ICSK, }; diff --git a/net/dccp/proto.c b/net/dccp/proto.c index eb892b4f481..de2c1e71930 100644 --- a/net/dccp/proto.c +++ b/net/dccp/proto.c @@ -1084,14 +1084,15 @@ EXPORT_SYMBOL_GPL(dccp_shutdown); static inline int dccp_mib_init(void) { - return snmp_mib_init((void __percpu **)dccp_statistics, - sizeof(struct dccp_mib), - __alignof__(struct dccp_mib)); + dccp_statistics = alloc_percpu(struct dccp_mib); + if (!dccp_statistics) + return -ENOMEM; + return 0; } static inline void dccp_mib_exit(void) { - snmp_mib_free((void __percpu **)dccp_statistics); + free_percpu(dccp_statistics); } static int thash_entries; diff --git a/net/dccp/sysctl.c b/net/dccp/sysctl.c index 607ab71b5a0..53731e45403 100644 --- a/net/dccp/sysctl.c +++ b/net/dccp/sysctl.c @@ -20,6 +20,7 @@ /* Boundary values */ static int zero = 0, + one = 1, u8_max = 0xFF; static unsigned long seqw_min = DCCPF_SEQ_WMIN, seqw_max = 0xFFFFFFFF; /* maximum on 32 bit */ @@ -58,7 +59,7 @@ static struct ctl_table dccp_default_table[] = { .maxlen = sizeof(sysctl_dccp_request_retries), .mode = 0644, .proc_handler = proc_dointvec_minmax, - .extra1 = &zero, + .extra1 = &one, .extra2 = &u8_max, }, { diff --git a/net/dccp/timer.c b/net/dccp/timer.c index 16f0b223102..1cd46a345cb 100644 --- a/net/dccp/timer.c +++ b/net/dccp/timer.c @@ -280,7 +280,7 @@ static ktime_t dccp_timestamp_seed; */ u32 dccp_timestamp(void) { - s64 delta = ktime_us_delta(ktime_get_real(), dccp_timestamp_seed); + u64 delta = (u64)ktime_us_delta(ktime_get_real(), dccp_timestamp_seed); do_div(delta, 10); return delta; diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c index 4c04848953b..ae011b46c07 100644 --- a/net/decnet/af_decnet.c +++ b/net/decnet/af_decnet.c @@ -481,7 +481,7 @@ static struct sock *dn_alloc_sock(struct net *net, struct socket *sock, gfp_t gf sk->sk_backlog_rcv = dn_nsp_backlog_rcv; sk->sk_destruct = dn_destruct; - sk->sk_no_check = 1; + sk->sk_no_check_tx = 1; sk->sk_family = PF_DECnet; sk->sk_protocol = 0; sk->sk_allocation = gfp; diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index a603823a3e2..3b726f31c64 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -574,7 +574,7 @@ static int dn_nl_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh) struct dn_ifaddr __rcu **ifap; int err = -EINVAL; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if (!net_eq(net, &init_net)) @@ -618,7 +618,7 @@ static int dn_nl_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh) struct dn_ifaddr *ifa; int err; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if (!net_eq(net, &init_net)) diff --git a/net/decnet/dn_fib.c b/net/decnet/dn_fib.c index 57dc159245e..d332aefb084 100644 --- a/net/decnet/dn_fib.c +++ b/net/decnet/dn_fib.c @@ -505,7 +505,7 @@ static int dn_fib_rtm_delroute(struct sk_buff *skb, struct nlmsghdr *nlh) struct nlattr *attrs[RTA_MAX+1]; int err; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if (!net_eq(net, &init_net)) @@ -530,7 +530,7 @@ static int dn_fib_rtm_newroute(struct sk_buff *skb, struct nlmsghdr *nlh) struct nlattr *attrs[RTA_MAX+1]; int err; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if (!net_eq(net, &init_net)) diff --git a/net/decnet/netfilter/dn_rtmsg.c b/net/decnet/netfilter/dn_rtmsg.c index e83015cecfa..e4d9560a910 100644 --- a/net/decnet/netfilter/dn_rtmsg.c +++ b/net/decnet/netfilter/dn_rtmsg.c @@ -107,7 +107,7 @@ static inline void dnrmg_receive_user_skb(struct sk_buff *skb) if (nlh->nlmsg_len < sizeof(*nlh) || skb->len < nlh->nlmsg_len) return; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) RCV_SKB_FAIL(-EPERM); /* Eventually we might send routing messages too */ diff --git a/net/dns_resolver/dns_query.c b/net/dns_resolver/dns_query.c index e7b6d53eef8..5b499589613 100644 --- a/net/dns_resolver/dns_query.c +++ b/net/dns_resolver/dns_query.c @@ -93,8 +93,8 @@ int dns_query(const char *type, const char *name, size_t namelen, } if (!namelen) - namelen = strlen(name); - if (namelen < 3) + namelen = strnlen(name, 256); + if (namelen < 3 || namelen > 255) return -EINVAL; desclen += namelen + 1; diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 0eb5d5e76df..5db37cef50a 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -406,8 +406,9 @@ static int dsa_of_probe(struct platform_device *pdev) goto out_free; } - chip_index = 0; + chip_index = -1; for_each_available_child_of_node(np, child) { + chip_index++; cd = &pd->chip[chip_index]; cd->mii_bus = &mdio_bus->dev; diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 02c0e1716f6..64c5af0a10d 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -346,7 +346,7 @@ dsa_slave_create(struct dsa_switch *ds, struct device *parent, return slave_dev; slave_dev->features = master->vlan_features; - SET_ETHTOOL_OPS(slave_dev, &dsa_slave_ethtool_ops); + slave_dev->ethtool_ops = &dsa_slave_ethtool_ops; eth_hw_addr_inherit(slave_dev, master); slave_dev->tx_queue_len = 0; diff --git a/net/ieee802154/6lowpan_rtnl.c b/net/ieee802154/6lowpan_rtnl.c index 0f5a69ed746..fe6bd7a7108 100644 --- a/net/ieee802154/6lowpan_rtnl.c +++ b/net/ieee802154/6lowpan_rtnl.c @@ -92,6 +92,7 @@ static int lowpan_header_create(struct sk_buff *skb, const u8 *saddr = _saddr; const u8 *daddr = _daddr; struct ieee802154_addr sa, da; + struct ieee802154_mac_cb *cb = mac_cb_init(skb); /* TODO: * if this package isn't ipv6 one, where should it be routed? @@ -115,8 +116,7 @@ static int lowpan_header_create(struct sk_buff *skb, * from MAC subif of the 'dev' and 'real_dev' network devices, but * this isn't implemented in mainline yet, so currently we assign 0xff */ - mac_cb(skb)->flags = IEEE802154_FC_TYPE_DATA; - mac_cb(skb)->seq = ieee802154_mlme_ops(dev)->get_dsn(dev); + cb->type = IEEE802154_FC_TYPE_DATA; /* prepare wpan address data */ sa.mode = IEEE802154_ADDR_LONG; @@ -135,11 +135,10 @@ static int lowpan_header_create(struct sk_buff *skb, } else { da.mode = IEEE802154_ADDR_LONG; da.extended_addr = ieee802154_devaddr_from_raw(daddr); - - /* request acknowledgment */ - mac_cb(skb)->flags |= MAC_CB_FLAG_ACKREQ; } + cb->ackreq = !lowpan_is_addr_broadcast(daddr); + return dev_hard_header(skb, lowpan_dev_info(dev)->real_dev, type, (void *)&da, (void *)&sa, 0); } @@ -221,139 +220,149 @@ static int lowpan_set_address(struct net_device *dev, void *p) return 0; } -static int -lowpan_fragment_xmit(struct sk_buff *skb, u8 *head, - int mlen, int plen, int offset, int type) +static struct sk_buff* +lowpan_alloc_frag(struct sk_buff *skb, int size, + const struct ieee802154_hdr *master_hdr) { + struct net_device *real_dev = lowpan_dev_info(skb->dev)->real_dev; struct sk_buff *frag; - int hlen; - - hlen = (type == LOWPAN_DISPATCH_FRAG1) ? - LOWPAN_FRAG1_HEAD_SIZE : LOWPAN_FRAGN_HEAD_SIZE; - - raw_dump_inline(__func__, "6lowpan fragment header", head, hlen); + int rc; + + frag = alloc_skb(real_dev->hard_header_len + + real_dev->needed_tailroom + size, + GFP_ATOMIC); + + if (likely(frag)) { + frag->dev = real_dev; + frag->priority = skb->priority; + skb_reserve(frag, real_dev->hard_header_len); + skb_reset_network_header(frag); + *mac_cb(frag) = *mac_cb(skb); + + rc = dev_hard_header(frag, real_dev, 0, &master_hdr->dest, + &master_hdr->source, size); + if (rc < 0) { + kfree_skb(frag); + return ERR_PTR(-rc); + } + } else { + frag = ERR_PTR(ENOMEM); + } - frag = netdev_alloc_skb(skb->dev, - hlen + mlen + plen + IEEE802154_MFR_SIZE); - if (!frag) - return -ENOMEM; + return frag; +} - frag->priority = skb->priority; +static int +lowpan_xmit_fragment(struct sk_buff *skb, const struct ieee802154_hdr *wpan_hdr, + u8 *frag_hdr, int frag_hdrlen, + int offset, int len) +{ + struct sk_buff *frag; - /* copy header, MFR and payload */ - skb_put(frag, mlen); - skb_copy_to_linear_data(frag, skb_mac_header(skb), mlen); + raw_dump_inline(__func__, " fragment header", frag_hdr, frag_hdrlen); - skb_put(frag, hlen); - skb_copy_to_linear_data_offset(frag, mlen, head, hlen); + frag = lowpan_alloc_frag(skb, frag_hdrlen + len, wpan_hdr); + if (IS_ERR(frag)) + return -PTR_ERR(frag); - skb_put(frag, plen); - skb_copy_to_linear_data_offset(frag, mlen + hlen, - skb_network_header(skb) + offset, plen); + memcpy(skb_put(frag, frag_hdrlen), frag_hdr, frag_hdrlen); + memcpy(skb_put(frag, len), skb_network_header(skb) + offset, len); - raw_dump_table(__func__, " raw fragment dump", frag->data, frag->len); + raw_dump_table(__func__, " fragment dump", frag->data, frag->len); return dev_queue_xmit(frag); } static int -lowpan_skb_fragmentation(struct sk_buff *skb, struct net_device *dev) +lowpan_xmit_fragmented(struct sk_buff *skb, struct net_device *dev, + const struct ieee802154_hdr *wpan_hdr) { - int err; - u16 dgram_offset, dgram_size, payload_length, header_length, - lowpan_size, frag_plen, offset; - __be16 tag; - u8 head[5]; - - header_length = skb->mac_len; - payload_length = skb->len - header_length; - tag = lowpan_dev_info(dev)->fragment_tag++; - lowpan_size = skb_network_header_len(skb); + u16 dgram_size, dgram_offset; + __be16 frag_tag; + u8 frag_hdr[5]; + int frag_cap, frag_len, payload_cap, rc; + int skb_unprocessed, skb_offset; + dgram_size = lowpan_uncompress_size(skb, &dgram_offset) - - header_length; + skb->mac_len; + frag_tag = lowpan_dev_info(dev)->fragment_tag++; - /* first fragment header */ - head[0] = LOWPAN_DISPATCH_FRAG1 | ((dgram_size >> 8) & 0x7); - head[1] = dgram_size & 0xff; - memcpy(head + 2, &tag, sizeof(tag)); + frag_hdr[0] = LOWPAN_DISPATCH_FRAG1 | ((dgram_size >> 8) & 0x07); + frag_hdr[1] = dgram_size & 0xff; + memcpy(frag_hdr + 2, &frag_tag, sizeof(frag_tag)); - /* calc the nearest payload length(divided to 8) for first fragment - * which fits into a IEEE802154_MTU - */ - frag_plen = round_down(IEEE802154_MTU - header_length - - LOWPAN_FRAG1_HEAD_SIZE - lowpan_size - - IEEE802154_MFR_SIZE, 8); - - err = lowpan_fragment_xmit(skb, head, header_length, - frag_plen + lowpan_size, 0, - LOWPAN_DISPATCH_FRAG1); - if (err) { - pr_debug("%s unable to send FRAG1 packet (tag: %d)", - __func__, tag); - goto exit; - } + payload_cap = ieee802154_max_payload(wpan_hdr); - offset = lowpan_size + frag_plen; - dgram_offset += frag_plen; + frag_len = round_down(payload_cap - LOWPAN_FRAG1_HEAD_SIZE - + skb_network_header_len(skb), 8); - /* next fragment header */ - head[0] &= ~LOWPAN_DISPATCH_FRAG1; - head[0] |= LOWPAN_DISPATCH_FRAGN; + skb_offset = skb_network_header_len(skb); + skb_unprocessed = skb->len - skb->mac_len - skb_offset; - frag_plen = round_down(IEEE802154_MTU - header_length - - LOWPAN_FRAGN_HEAD_SIZE - IEEE802154_MFR_SIZE, 8); + rc = lowpan_xmit_fragment(skb, wpan_hdr, frag_hdr, + LOWPAN_FRAG1_HEAD_SIZE, 0, + frag_len + skb_network_header_len(skb)); + if (rc) { + pr_debug("%s unable to send FRAG1 packet (tag: %d)", + __func__, frag_tag); + goto err; + } - while (payload_length - offset > 0) { - int len = frag_plen; + frag_hdr[0] &= ~LOWPAN_DISPATCH_FRAG1; + frag_hdr[0] |= LOWPAN_DISPATCH_FRAGN; + frag_cap = round_down(payload_cap - LOWPAN_FRAGN_HEAD_SIZE, 8); - head[4] = dgram_offset >> 3; + do { + dgram_offset += frag_len; + skb_offset += frag_len; + skb_unprocessed -= frag_len; + frag_len = min(frag_cap, skb_unprocessed); - if (payload_length - offset < len) - len = payload_length - offset; + frag_hdr[4] = dgram_offset >> 3; - err = lowpan_fragment_xmit(skb, head, header_length, len, - offset, LOWPAN_DISPATCH_FRAGN); - if (err) { + rc = lowpan_xmit_fragment(skb, wpan_hdr, frag_hdr, + LOWPAN_FRAGN_HEAD_SIZE, skb_offset, + frag_len); + if (rc) { pr_debug("%s unable to send a FRAGN packet. (tag: %d, offset: %d)\n", - __func__, tag, offset); - goto exit; + __func__, frag_tag, skb_offset); + goto err; } + } while (skb_unprocessed > frag_cap); - offset += len; - dgram_offset += len; - } + consume_skb(skb); + return NET_XMIT_SUCCESS; -exit: - return err; +err: + kfree_skb(skb); + return rc; } static netdev_tx_t lowpan_xmit(struct sk_buff *skb, struct net_device *dev) { - int err = -1; + struct ieee802154_hdr wpan_hdr; + int max_single; pr_debug("package xmit\n"); - skb->dev = lowpan_dev_info(dev)->real_dev; - if (skb->dev == NULL) { - pr_debug("ERROR: no real wpan device found\n"); - goto error; + if (ieee802154_hdr_peek(skb, &wpan_hdr) < 0) { + kfree_skb(skb); + return NET_XMIT_DROP; } - /* Send directly if less than the MTU minus the 2 checksum bytes. */ - if (skb->len <= IEEE802154_MTU - IEEE802154_MFR_SIZE) { - err = dev_queue_xmit(skb); - goto out; - } + max_single = ieee802154_max_payload(&wpan_hdr); - pr_debug("frame is too big, fragmentation is needed\n"); - err = lowpan_skb_fragmentation(skb, dev); -error: - dev_kfree_skb(skb); -out: - if (err) - pr_debug("ERROR: xmit failed\n"); + if (skb_tail_pointer(skb) - skb_network_header(skb) <= max_single) { + skb->dev = lowpan_dev_info(dev)->real_dev; + return dev_queue_xmit(skb); + } else { + netdev_tx_t rc; + + pr_debug("frame is too big, fragmentation is needed\n"); + rc = lowpan_xmit_fragmented(skb, dev, &wpan_hdr); - return (err < 0) ? NET_XMIT_DROP : err; + return rc < 0 ? NET_XMIT_DROP : rc; + } } static struct wpan_phy *lowpan_get_phy(const struct net_device *dev) diff --git a/net/ieee802154/dgram.c b/net/ieee802154/dgram.c index 786437bc0c0..4f0ed878019 100644 --- a/net/ieee802154/dgram.c +++ b/net/ieee802154/dgram.c @@ -21,6 +21,7 @@ * Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> */ +#include <linux/capability.h> #include <linux/net.h> #include <linux/module.h> #include <linux/if_arp.h> @@ -45,7 +46,12 @@ struct dgram_sock { struct ieee802154_addr dst_addr; unsigned int bound:1; + unsigned int connected:1; unsigned int want_ack:1; + unsigned int secen:1; + unsigned int secen_override:1; + unsigned int seclevel:3; + unsigned int seclevel_override:1; }; static inline struct dgram_sock *dgram_sk(const struct sock *sk) @@ -73,10 +79,7 @@ static int dgram_init(struct sock *sk) { struct dgram_sock *ro = dgram_sk(sk); - ro->dst_addr.mode = IEEE802154_ADDR_LONG; - ro->dst_addr.pan_id = cpu_to_le16(IEEE802154_PANID_BROADCAST); ro->want_ack = 1; - memset(&ro->dst_addr.extended_addr, 0xff, IEEE802154_ADDR_LEN); return 0; } @@ -183,6 +186,7 @@ static int dgram_connect(struct sock *sk, struct sockaddr *uaddr, } ieee802154_addr_from_sa(&ro->dst_addr, &addr->addr); + ro->connected = 1; out: release_sock(sk); @@ -194,10 +198,7 @@ static int dgram_disconnect(struct sock *sk, int flags) struct dgram_sock *ro = dgram_sk(sk); lock_sock(sk); - - ro->dst_addr.mode = IEEE802154_ADDR_LONG; - memset(&ro->dst_addr.extended_addr, 0xff, IEEE802154_ADDR_LEN); - + ro->connected = 0; release_sock(sk); return 0; @@ -209,7 +210,9 @@ static int dgram_sendmsg(struct kiocb *iocb, struct sock *sk, struct net_device *dev; unsigned int mtu; struct sk_buff *skb; + struct ieee802154_mac_cb *cb; struct dgram_sock *ro = dgram_sk(sk); + struct ieee802154_addr dst_addr; int hlen, tlen; int err; @@ -218,6 +221,11 @@ static int dgram_sendmsg(struct kiocb *iocb, struct sock *sk, return -EOPNOTSUPP; } + if (!ro->connected && !msg->msg_name) + return -EDESTADDRREQ; + else if (ro->connected && msg->msg_name) + return -EISCONN; + if (!ro->bound) dev = dev_getfirstbyhwtype(sock_net(sk), ARPHRD_IEEE802154); else @@ -249,18 +257,28 @@ static int dgram_sendmsg(struct kiocb *iocb, struct sock *sk, skb_reset_network_header(skb); - mac_cb(skb)->flags = IEEE802154_FC_TYPE_DATA; - if (ro->want_ack) - mac_cb(skb)->flags |= MAC_CB_FLAG_ACKREQ; + cb = mac_cb_init(skb); + cb->type = IEEE802154_FC_TYPE_DATA; + cb->ackreq = ro->want_ack; + + if (msg->msg_name) { + DECLARE_SOCKADDR(struct sockaddr_ieee802154*, daddr, msg->msg_name); - mac_cb(skb)->seq = ieee802154_mlme_ops(dev)->get_dsn(dev); - err = dev_hard_header(skb, dev, ETH_P_IEEE802154, &ro->dst_addr, - ro->bound ? &ro->src_addr : NULL, size); + ieee802154_addr_from_sa(&dst_addr, &daddr->addr); + } else { + dst_addr = ro->dst_addr; + } + + cb->secen = ro->secen; + cb->secen_override = ro->secen_override; + cb->seclevel = ro->seclevel; + cb->seclevel_override = ro->seclevel_override; + + err = dev_hard_header(skb, dev, ETH_P_IEEE802154, &dst_addr, + ro->bound ? &ro->src_addr : NULL, size); if (err < 0) goto out_skb; - skb_reset_mac_header(skb); - err = memcpy_fromiovec(skb_put(skb, size), msg->msg_iov, size); if (err < 0) goto out_skb; @@ -419,6 +437,20 @@ static int dgram_getsockopt(struct sock *sk, int level, int optname, case WPAN_WANTACK: val = ro->want_ack; break; + case WPAN_SECURITY: + if (!ro->secen_override) + val = WPAN_SECURITY_DEFAULT; + else if (ro->secen) + val = WPAN_SECURITY_ON; + else + val = WPAN_SECURITY_OFF; + break; + case WPAN_SECURITY_LEVEL: + if (!ro->seclevel_override) + val = WPAN_SECURITY_LEVEL_DEFAULT; + else + val = ro->seclevel; + break; default: return -ENOPROTOOPT; } @@ -434,6 +466,7 @@ static int dgram_setsockopt(struct sock *sk, int level, int optname, char __user *optval, unsigned int optlen) { struct dgram_sock *ro = dgram_sk(sk); + struct net *net = sock_net(sk); int val; int err = 0; @@ -449,6 +482,47 @@ static int dgram_setsockopt(struct sock *sk, int level, int optname, case WPAN_WANTACK: ro->want_ack = !!val; break; + case WPAN_SECURITY: + if (!ns_capable(net->user_ns, CAP_NET_ADMIN) && + !ns_capable(net->user_ns, CAP_NET_RAW)) { + err = -EPERM; + break; + } + + switch (val) { + case WPAN_SECURITY_DEFAULT: + ro->secen_override = 0; + break; + case WPAN_SECURITY_ON: + ro->secen_override = 1; + ro->secen = 1; + break; + case WPAN_SECURITY_OFF: + ro->secen_override = 1; + ro->secen = 0; + break; + default: + err = -EINVAL; + break; + } + break; + case WPAN_SECURITY_LEVEL: + if (!ns_capable(net->user_ns, CAP_NET_ADMIN) && + !ns_capable(net->user_ns, CAP_NET_RAW)) { + err = -EPERM; + break; + } + + if (val < WPAN_SECURITY_LEVEL_DEFAULT || + val > IEEE802154_SCF_SECLEVEL_ENC_MIC128) { + err = -EINVAL; + } else if (val == WPAN_SECURITY_LEVEL_DEFAULT) { + ro->seclevel_override = 0; + } else { + ro->seclevel_override = 1; + ro->seclevel = val; + } + break; default: err = -ENOPROTOOPT; break; diff --git a/net/ieee802154/header_ops.c b/net/ieee802154/header_ops.c index bed42a48408..c09294e39ca 100644 --- a/net/ieee802154/header_ops.c +++ b/net/ieee802154/header_ops.c @@ -195,15 +195,16 @@ ieee802154_hdr_get_sechdr(const u8 *buf, struct ieee802154_sechdr *hdr) return pos; } +static int ieee802154_sechdr_lengths[4] = { + [IEEE802154_SCF_KEY_IMPLICIT] = 5, + [IEEE802154_SCF_KEY_INDEX] = 6, + [IEEE802154_SCF_KEY_SHORT_INDEX] = 10, + [IEEE802154_SCF_KEY_HW_INDEX] = 14, +}; + static int ieee802154_hdr_sechdr_len(u8 sc) { - switch (IEEE802154_SCF_KEY_ID_MODE(sc)) { - case IEEE802154_SCF_KEY_IMPLICIT: return 5; - case IEEE802154_SCF_KEY_INDEX: return 6; - case IEEE802154_SCF_KEY_SHORT_INDEX: return 10; - case IEEE802154_SCF_KEY_HW_INDEX: return 14; - default: return -EINVAL; - } + return ieee802154_sechdr_lengths[IEEE802154_SCF_KEY_ID_MODE(sc)]; } static int ieee802154_hdr_minlen(const struct ieee802154_hdr *hdr) @@ -285,3 +286,40 @@ ieee802154_hdr_peek_addrs(const struct sk_buff *skb, struct ieee802154_hdr *hdr) return pos; } EXPORT_SYMBOL_GPL(ieee802154_hdr_peek_addrs); + +int +ieee802154_hdr_peek(const struct sk_buff *skb, struct ieee802154_hdr *hdr) +{ + const u8 *buf = skb_mac_header(skb); + int pos; + + pos = ieee802154_hdr_peek_addrs(skb, hdr); + if (pos < 0) + return -EINVAL; + + if (hdr->fc.security_enabled) { + u8 key_id_mode = IEEE802154_SCF_KEY_ID_MODE(*(buf + pos)); + int want = pos + ieee802154_sechdr_lengths[key_id_mode]; + + if (buf + want > skb_tail_pointer(skb)) + return -EINVAL; + + pos += ieee802154_hdr_get_sechdr(buf + pos, &hdr->sec); + } + + return pos; +} +EXPORT_SYMBOL_GPL(ieee802154_hdr_peek); + +int ieee802154_max_payload(const struct ieee802154_hdr *hdr) +{ + int hlen = ieee802154_hdr_minlen(hdr); + + if (hdr->fc.security_enabled) { + hlen += ieee802154_sechdr_lengths[hdr->sec.key_id_mode] - 1; + hlen += ieee802154_sechdr_authtag_len(&hdr->sec); + } + + return IEEE802154_MTU - hlen - IEEE802154_MFR_SIZE; +} +EXPORT_SYMBOL_GPL(ieee802154_max_payload); diff --git a/net/ieee802154/ieee802154.h b/net/ieee802154/ieee802154.h index 6693a5cf01c..8b83a231299 100644 --- a/net/ieee802154/ieee802154.h +++ b/net/ieee802154/ieee802154.h @@ -68,4 +68,23 @@ int ieee802154_list_iface(struct sk_buff *skb, struct genl_info *info); int ieee802154_dump_iface(struct sk_buff *skb, struct netlink_callback *cb); int ieee802154_set_macparams(struct sk_buff *skb, struct genl_info *info); +int ieee802154_llsec_getparams(struct sk_buff *skb, struct genl_info *info); +int ieee802154_llsec_setparams(struct sk_buff *skb, struct genl_info *info); +int ieee802154_llsec_add_key(struct sk_buff *skb, struct genl_info *info); +int ieee802154_llsec_del_key(struct sk_buff *skb, struct genl_info *info); +int ieee802154_llsec_dump_keys(struct sk_buff *skb, + struct netlink_callback *cb); +int ieee802154_llsec_add_dev(struct sk_buff *skb, struct genl_info *info); +int ieee802154_llsec_del_dev(struct sk_buff *skb, struct genl_info *info); +int ieee802154_llsec_dump_devs(struct sk_buff *skb, + struct netlink_callback *cb); +int ieee802154_llsec_add_devkey(struct sk_buff *skb, struct genl_info *info); +int ieee802154_llsec_del_devkey(struct sk_buff *skb, struct genl_info *info); +int ieee802154_llsec_dump_devkeys(struct sk_buff *skb, + struct netlink_callback *cb); +int ieee802154_llsec_add_seclevel(struct sk_buff *skb, struct genl_info *info); +int ieee802154_llsec_del_seclevel(struct sk_buff *skb, struct genl_info *info); +int ieee802154_llsec_dump_seclevels(struct sk_buff *skb, + struct netlink_callback *cb); + #endif diff --git a/net/ieee802154/netlink.c b/net/ieee802154/netlink.c index 04b20589d97..26efcf4fd2f 100644 --- a/net/ieee802154/netlink.c +++ b/net/ieee802154/netlink.c @@ -124,6 +124,26 @@ static const struct genl_ops ieee8021154_ops[] = { IEEE802154_DUMP(IEEE802154_LIST_IFACE, ieee802154_list_iface, ieee802154_dump_iface), IEEE802154_OP(IEEE802154_SET_MACPARAMS, ieee802154_set_macparams), + IEEE802154_OP(IEEE802154_LLSEC_GETPARAMS, ieee802154_llsec_getparams), + IEEE802154_OP(IEEE802154_LLSEC_SETPARAMS, ieee802154_llsec_setparams), + IEEE802154_DUMP(IEEE802154_LLSEC_LIST_KEY, NULL, + ieee802154_llsec_dump_keys), + IEEE802154_OP(IEEE802154_LLSEC_ADD_KEY, ieee802154_llsec_add_key), + IEEE802154_OP(IEEE802154_LLSEC_DEL_KEY, ieee802154_llsec_del_key), + IEEE802154_DUMP(IEEE802154_LLSEC_LIST_DEV, NULL, + ieee802154_llsec_dump_devs), + IEEE802154_OP(IEEE802154_LLSEC_ADD_DEV, ieee802154_llsec_add_dev), + IEEE802154_OP(IEEE802154_LLSEC_DEL_DEV, ieee802154_llsec_del_dev), + IEEE802154_DUMP(IEEE802154_LLSEC_LIST_DEVKEY, NULL, + ieee802154_llsec_dump_devkeys), + IEEE802154_OP(IEEE802154_LLSEC_ADD_DEVKEY, ieee802154_llsec_add_devkey), + IEEE802154_OP(IEEE802154_LLSEC_DEL_DEVKEY, ieee802154_llsec_del_devkey), + IEEE802154_DUMP(IEEE802154_LLSEC_LIST_SECLEVEL, NULL, + ieee802154_llsec_dump_seclevels), + IEEE802154_OP(IEEE802154_LLSEC_ADD_SECLEVEL, + ieee802154_llsec_add_seclevel), + IEEE802154_OP(IEEE802154_LLSEC_DEL_SECLEVEL, + ieee802154_llsec_del_seclevel), }; static const struct genl_multicast_group ieee802154_mcgrps[] = { diff --git a/net/ieee802154/nl-mac.c b/net/ieee802154/nl-mac.c index 5d285498c0f..a3281b8bfd5 100644 --- a/net/ieee802154/nl-mac.c +++ b/net/ieee802154/nl-mac.c @@ -715,3 +715,812 @@ out: dev_put(dev); return rc; } + + + +static int +ieee802154_llsec_parse_key_id(struct genl_info *info, + struct ieee802154_llsec_key_id *desc) +{ + memset(desc, 0, sizeof(*desc)); + + if (!info->attrs[IEEE802154_ATTR_LLSEC_KEY_MODE]) + return -EINVAL; + + desc->mode = nla_get_u8(info->attrs[IEEE802154_ATTR_LLSEC_KEY_MODE]); + + if (desc->mode == IEEE802154_SCF_KEY_IMPLICIT) { + if (!info->attrs[IEEE802154_ATTR_PAN_ID] && + !(info->attrs[IEEE802154_ATTR_SHORT_ADDR] || + info->attrs[IEEE802154_ATTR_HW_ADDR])) + return -EINVAL; + + desc->device_addr.pan_id = nla_get_shortaddr(info->attrs[IEEE802154_ATTR_PAN_ID]); + + if (info->attrs[IEEE802154_ATTR_SHORT_ADDR]) { + desc->device_addr.mode = IEEE802154_ADDR_SHORT; + desc->device_addr.short_addr = nla_get_shortaddr(info->attrs[IEEE802154_ATTR_SHORT_ADDR]); + } else { + desc->device_addr.mode = IEEE802154_ADDR_LONG; + desc->device_addr.extended_addr = nla_get_hwaddr(info->attrs[IEEE802154_ATTR_HW_ADDR]); + } + } + + if (desc->mode != IEEE802154_SCF_KEY_IMPLICIT && + !info->attrs[IEEE802154_ATTR_LLSEC_KEY_ID]) + return -EINVAL; + + if (desc->mode == IEEE802154_SCF_KEY_SHORT_INDEX && + !info->attrs[IEEE802154_ATTR_LLSEC_KEY_SOURCE_SHORT]) + return -EINVAL; + + if (desc->mode == IEEE802154_SCF_KEY_HW_INDEX && + !info->attrs[IEEE802154_ATTR_LLSEC_KEY_SOURCE_EXTENDED]) + return -EINVAL; + + if (desc->mode != IEEE802154_SCF_KEY_IMPLICIT) + desc->id = nla_get_u8(info->attrs[IEEE802154_ATTR_LLSEC_KEY_ID]); + + switch (desc->mode) { + case IEEE802154_SCF_KEY_SHORT_INDEX: + { + u32 source = nla_get_u32(info->attrs[IEEE802154_ATTR_LLSEC_KEY_SOURCE_SHORT]); + desc->short_source = cpu_to_le32(source); + break; + } + case IEEE802154_SCF_KEY_HW_INDEX: + desc->extended_source = nla_get_hwaddr(info->attrs[IEEE802154_ATTR_LLSEC_KEY_SOURCE_EXTENDED]); + break; + } + + return 0; +} + +static int +ieee802154_llsec_fill_key_id(struct sk_buff *msg, + const struct ieee802154_llsec_key_id *desc) +{ + if (nla_put_u8(msg, IEEE802154_ATTR_LLSEC_KEY_MODE, desc->mode)) + return -EMSGSIZE; + + if (desc->mode == IEEE802154_SCF_KEY_IMPLICIT) { + if (nla_put_shortaddr(msg, IEEE802154_ATTR_PAN_ID, + desc->device_addr.pan_id)) + return -EMSGSIZE; + + if (desc->device_addr.mode == IEEE802154_ADDR_SHORT && + nla_put_shortaddr(msg, IEEE802154_ATTR_SHORT_ADDR, + desc->device_addr.short_addr)) + return -EMSGSIZE; + + if (desc->device_addr.mode == IEEE802154_ADDR_LONG && + nla_put_hwaddr(msg, IEEE802154_ATTR_HW_ADDR, + desc->device_addr.extended_addr)) + return -EMSGSIZE; + } + + if (desc->mode != IEEE802154_SCF_KEY_IMPLICIT && + nla_put_u8(msg, IEEE802154_ATTR_LLSEC_KEY_ID, desc->id)) + return -EMSGSIZE; + + if (desc->mode == IEEE802154_SCF_KEY_SHORT_INDEX && + nla_put_u32(msg, IEEE802154_ATTR_LLSEC_KEY_SOURCE_SHORT, + le32_to_cpu(desc->short_source))) + return -EMSGSIZE; + + if (desc->mode == IEEE802154_SCF_KEY_HW_INDEX && + nla_put_hwaddr(msg, IEEE802154_ATTR_LLSEC_KEY_SOURCE_EXTENDED, + desc->extended_source)) + return -EMSGSIZE; + + return 0; +} + +int ieee802154_llsec_getparams(struct sk_buff *skb, struct genl_info *info) +{ + struct sk_buff *msg; + struct net_device *dev = NULL; + int rc = -ENOBUFS; + struct ieee802154_mlme_ops *ops; + void *hdr; + struct ieee802154_llsec_params params; + + pr_debug("%s\n", __func__); + + dev = ieee802154_nl_get_dev(info); + if (!dev) + return -ENODEV; + + ops = ieee802154_mlme_ops(dev); + if (!ops->llsec) { + rc = -EOPNOTSUPP; + goto out_dev; + } + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + goto out_dev; + + hdr = genlmsg_put(msg, 0, info->snd_seq, &nl802154_family, 0, + IEEE802154_LLSEC_GETPARAMS); + if (!hdr) + goto out_free; + + rc = ops->llsec->get_params(dev, ¶ms); + if (rc < 0) + goto out_free; + + if (nla_put_string(msg, IEEE802154_ATTR_DEV_NAME, dev->name) || + nla_put_u32(msg, IEEE802154_ATTR_DEV_INDEX, dev->ifindex) || + nla_put_u8(msg, IEEE802154_ATTR_LLSEC_ENABLED, params.enabled) || + nla_put_u8(msg, IEEE802154_ATTR_LLSEC_SECLEVEL, params.out_level) || + nla_put_u32(msg, IEEE802154_ATTR_LLSEC_FRAME_COUNTER, + be32_to_cpu(params.frame_counter)) || + ieee802154_llsec_fill_key_id(msg, ¶ms.out_key)) + goto out_free; + + dev_put(dev); + + return ieee802154_nl_reply(msg, info); +out_free: + nlmsg_free(msg); +out_dev: + dev_put(dev); + return rc; +} + +int ieee802154_llsec_setparams(struct sk_buff *skb, struct genl_info *info) +{ + struct net_device *dev = NULL; + int rc = -EINVAL; + struct ieee802154_mlme_ops *ops; + struct ieee802154_llsec_params params; + int changed = 0; + + pr_debug("%s\n", __func__); + + dev = ieee802154_nl_get_dev(info); + if (!dev) + return -ENODEV; + + if (!info->attrs[IEEE802154_ATTR_LLSEC_ENABLED] && + !info->attrs[IEEE802154_ATTR_LLSEC_KEY_MODE] && + !info->attrs[IEEE802154_ATTR_LLSEC_SECLEVEL]) + goto out; + + ops = ieee802154_mlme_ops(dev); + if (!ops->llsec) { + rc = -EOPNOTSUPP; + goto out; + } + + if (info->attrs[IEEE802154_ATTR_LLSEC_SECLEVEL] && + nla_get_u8(info->attrs[IEEE802154_ATTR_LLSEC_SECLEVEL]) > 7) + goto out; + + if (info->attrs[IEEE802154_ATTR_LLSEC_ENABLED]) { + params.enabled = nla_get_u8(info->attrs[IEEE802154_ATTR_LLSEC_ENABLED]); + changed |= IEEE802154_LLSEC_PARAM_ENABLED; + } + + if (info->attrs[IEEE802154_ATTR_LLSEC_KEY_MODE]) { + if (ieee802154_llsec_parse_key_id(info, ¶ms.out_key)) + goto out; + + changed |= IEEE802154_LLSEC_PARAM_OUT_KEY; + } + + if (info->attrs[IEEE802154_ATTR_LLSEC_SECLEVEL]) { + params.out_level = nla_get_u8(info->attrs[IEEE802154_ATTR_LLSEC_SECLEVEL]); + changed |= IEEE802154_LLSEC_PARAM_OUT_LEVEL; + } + + if (info->attrs[IEEE802154_ATTR_LLSEC_FRAME_COUNTER]) { + u32 fc = nla_get_u32(info->attrs[IEEE802154_ATTR_LLSEC_FRAME_COUNTER]); + + params.frame_counter = cpu_to_be32(fc); + changed |= IEEE802154_LLSEC_PARAM_FRAME_COUNTER; + } + + rc = ops->llsec->set_params(dev, ¶ms, changed); + + dev_put(dev); + + return rc; +out: + dev_put(dev); + return rc; +} + + + +struct llsec_dump_data { + struct sk_buff *skb; + int s_idx, s_idx2; + int portid; + int nlmsg_seq; + struct net_device *dev; + struct ieee802154_mlme_ops *ops; + struct ieee802154_llsec_table *table; +}; + +static int +ieee802154_llsec_dump_table(struct sk_buff *skb, struct netlink_callback *cb, + int (*step)(struct llsec_dump_data*)) +{ + struct net *net = sock_net(skb->sk); + struct net_device *dev; + struct llsec_dump_data data; + int idx = 0; + int first_dev = cb->args[0]; + int rc; + + for_each_netdev(net, dev) { + if (idx < first_dev || dev->type != ARPHRD_IEEE802154) + goto skip; + + data.ops = ieee802154_mlme_ops(dev); + if (!data.ops->llsec) + goto skip; + + data.skb = skb; + data.s_idx = cb->args[1]; + data.s_idx2 = cb->args[2]; + data.dev = dev; + data.portid = NETLINK_CB(cb->skb).portid; + data.nlmsg_seq = cb->nlh->nlmsg_seq; + + data.ops->llsec->lock_table(dev); + data.ops->llsec->get_table(data.dev, &data.table); + rc = step(&data); + data.ops->llsec->unlock_table(dev); + + if (rc < 0) + break; + +skip: + idx++; + } + cb->args[0] = idx; + + return skb->len; +} + +static int +ieee802154_nl_llsec_change(struct sk_buff *skb, struct genl_info *info, + int (*fn)(struct net_device*, struct genl_info*)) +{ + struct net_device *dev = NULL; + int rc = -EINVAL; + + dev = ieee802154_nl_get_dev(info); + if (!dev) + return -ENODEV; + + if (!ieee802154_mlme_ops(dev)->llsec) + rc = -EOPNOTSUPP; + else + rc = fn(dev, info); + + dev_put(dev); + return rc; +} + + + +static int +ieee802154_llsec_parse_key(struct genl_info *info, + struct ieee802154_llsec_key *key) +{ + u8 frames; + u32 commands[256 / 32]; + + memset(key, 0, sizeof(*key)); + + if (!info->attrs[IEEE802154_ATTR_LLSEC_KEY_USAGE_FRAME_TYPES] || + !info->attrs[IEEE802154_ATTR_LLSEC_KEY_BYTES]) + return -EINVAL; + + frames = nla_get_u8(info->attrs[IEEE802154_ATTR_LLSEC_KEY_USAGE_FRAME_TYPES]); + if ((frames & BIT(IEEE802154_FC_TYPE_MAC_CMD)) && + !info->attrs[IEEE802154_ATTR_LLSEC_KEY_USAGE_COMMANDS]) + return -EINVAL; + + if (info->attrs[IEEE802154_ATTR_LLSEC_KEY_USAGE_COMMANDS]) { + nla_memcpy(commands, + info->attrs[IEEE802154_ATTR_LLSEC_KEY_USAGE_COMMANDS], + 256 / 8); + + if (commands[0] || commands[1] || commands[2] || commands[3] || + commands[4] || commands[5] || commands[6] || + commands[7] >= BIT(IEEE802154_CMD_GTS_REQ + 1)) + return -EINVAL; + + key->cmd_frame_ids = commands[7]; + } + + key->frame_types = frames; + + nla_memcpy(key->key, info->attrs[IEEE802154_ATTR_LLSEC_KEY_BYTES], + IEEE802154_LLSEC_KEY_SIZE); + + return 0; +} + +static int llsec_add_key(struct net_device *dev, struct genl_info *info) +{ + struct ieee802154_mlme_ops *ops = ieee802154_mlme_ops(dev); + struct ieee802154_llsec_key key; + struct ieee802154_llsec_key_id id; + + if (ieee802154_llsec_parse_key(info, &key) || + ieee802154_llsec_parse_key_id(info, &id)) + return -EINVAL; + + return ops->llsec->add_key(dev, &id, &key); +} + +int ieee802154_llsec_add_key(struct sk_buff *skb, struct genl_info *info) +{ + if ((info->nlhdr->nlmsg_flags & (NLM_F_CREATE | NLM_F_EXCL)) != + (NLM_F_CREATE | NLM_F_EXCL)) + return -EINVAL; + + return ieee802154_nl_llsec_change(skb, info, llsec_add_key); +} + +static int llsec_remove_key(struct net_device *dev, struct genl_info *info) +{ + struct ieee802154_mlme_ops *ops = ieee802154_mlme_ops(dev); + struct ieee802154_llsec_key_id id; + + if (ieee802154_llsec_parse_key_id(info, &id)) + return -EINVAL; + + return ops->llsec->del_key(dev, &id); +} + +int ieee802154_llsec_del_key(struct sk_buff *skb, struct genl_info *info) +{ + return ieee802154_nl_llsec_change(skb, info, llsec_remove_key); +} + +static int +ieee802154_nl_fill_key(struct sk_buff *msg, u32 portid, u32 seq, + const struct ieee802154_llsec_key_entry *key, + const struct net_device *dev) +{ + void *hdr; + u32 commands[256 / 32]; + + hdr = genlmsg_put(msg, 0, seq, &nl802154_family, NLM_F_MULTI, + IEEE802154_LLSEC_LIST_KEY); + if (!hdr) + goto out; + + if (nla_put_string(msg, IEEE802154_ATTR_DEV_NAME, dev->name) || + nla_put_u32(msg, IEEE802154_ATTR_DEV_INDEX, dev->ifindex) || + ieee802154_llsec_fill_key_id(msg, &key->id) || + nla_put_u8(msg, IEEE802154_ATTR_LLSEC_KEY_USAGE_FRAME_TYPES, + key->key->frame_types)) + goto nla_put_failure; + + if (key->key->frame_types & BIT(IEEE802154_FC_TYPE_MAC_CMD)) { + memset(commands, 0, sizeof(commands)); + commands[7] = key->key->cmd_frame_ids; + if (nla_put(msg, IEEE802154_ATTR_LLSEC_KEY_USAGE_COMMANDS, + sizeof(commands), commands)) + goto nla_put_failure; + } + + if (nla_put(msg, IEEE802154_ATTR_LLSEC_KEY_BYTES, + IEEE802154_LLSEC_KEY_SIZE, key->key->key)) + goto nla_put_failure; + + genlmsg_end(msg, hdr); + return 0; + +nla_put_failure: + genlmsg_cancel(msg, hdr); +out: + return -EMSGSIZE; +} + +static int llsec_iter_keys(struct llsec_dump_data *data) +{ + struct ieee802154_llsec_key_entry *pos; + int rc = 0, idx = 0; + + list_for_each_entry(pos, &data->table->keys, list) { + if (idx++ < data->s_idx) + continue; + + if (ieee802154_nl_fill_key(data->skb, data->portid, + data->nlmsg_seq, pos, data->dev)) { + rc = -EMSGSIZE; + break; + } + + data->s_idx++; + } + + return rc; +} + +int ieee802154_llsec_dump_keys(struct sk_buff *skb, struct netlink_callback *cb) +{ + return ieee802154_llsec_dump_table(skb, cb, llsec_iter_keys); +} + + + +static int +llsec_parse_dev(struct genl_info *info, + struct ieee802154_llsec_device *dev) +{ + memset(dev, 0, sizeof(*dev)); + + if (!info->attrs[IEEE802154_ATTR_LLSEC_FRAME_COUNTER] || + !info->attrs[IEEE802154_ATTR_HW_ADDR] || + !info->attrs[IEEE802154_ATTR_LLSEC_DEV_OVERRIDE] || + !info->attrs[IEEE802154_ATTR_LLSEC_DEV_KEY_MODE] || + (!!info->attrs[IEEE802154_ATTR_PAN_ID] != + !!info->attrs[IEEE802154_ATTR_SHORT_ADDR])) + return -EINVAL; + + if (info->attrs[IEEE802154_ATTR_PAN_ID]) { + dev->pan_id = nla_get_shortaddr(info->attrs[IEEE802154_ATTR_PAN_ID]); + dev->short_addr = nla_get_shortaddr(info->attrs[IEEE802154_ATTR_SHORT_ADDR]); + } else { + dev->short_addr = cpu_to_le16(IEEE802154_ADDR_UNDEF); + } + + dev->hwaddr = nla_get_hwaddr(info->attrs[IEEE802154_ATTR_HW_ADDR]); + dev->frame_counter = nla_get_u32(info->attrs[IEEE802154_ATTR_LLSEC_FRAME_COUNTER]); + dev->seclevel_exempt = !!nla_get_u8(info->attrs[IEEE802154_ATTR_LLSEC_DEV_OVERRIDE]); + dev->key_mode = nla_get_u8(info->attrs[IEEE802154_ATTR_LLSEC_DEV_KEY_MODE]); + + if (dev->key_mode >= __IEEE802154_LLSEC_DEVKEY_MAX) + return -EINVAL; + + return 0; +} + +static int llsec_add_dev(struct net_device *dev, struct genl_info *info) +{ + struct ieee802154_mlme_ops *ops = ieee802154_mlme_ops(dev); + struct ieee802154_llsec_device desc; + + if (llsec_parse_dev(info, &desc)) + return -EINVAL; + + return ops->llsec->add_dev(dev, &desc); +} + +int ieee802154_llsec_add_dev(struct sk_buff *skb, struct genl_info *info) +{ + if ((info->nlhdr->nlmsg_flags & (NLM_F_CREATE | NLM_F_EXCL)) != + (NLM_F_CREATE | NLM_F_EXCL)) + return -EINVAL; + + return ieee802154_nl_llsec_change(skb, info, llsec_add_dev); +} + +static int llsec_del_dev(struct net_device *dev, struct genl_info *info) +{ + struct ieee802154_mlme_ops *ops = ieee802154_mlme_ops(dev); + __le64 devaddr; + + if (!info->attrs[IEEE802154_ATTR_HW_ADDR]) + return -EINVAL; + + devaddr = nla_get_hwaddr(info->attrs[IEEE802154_ATTR_HW_ADDR]); + + return ops->llsec->del_dev(dev, devaddr); +} + +int ieee802154_llsec_del_dev(struct sk_buff *skb, struct genl_info *info) +{ + return ieee802154_nl_llsec_change(skb, info, llsec_del_dev); +} + +static int +ieee802154_nl_fill_dev(struct sk_buff *msg, u32 portid, u32 seq, + const struct ieee802154_llsec_device *desc, + const struct net_device *dev) +{ + void *hdr; + + hdr = genlmsg_put(msg, 0, seq, &nl802154_family, NLM_F_MULTI, + IEEE802154_LLSEC_LIST_DEV); + if (!hdr) + goto out; + + if (nla_put_string(msg, IEEE802154_ATTR_DEV_NAME, dev->name) || + nla_put_u32(msg, IEEE802154_ATTR_DEV_INDEX, dev->ifindex) || + nla_put_shortaddr(msg, IEEE802154_ATTR_PAN_ID, desc->pan_id) || + nla_put_shortaddr(msg, IEEE802154_ATTR_SHORT_ADDR, + desc->short_addr) || + nla_put_hwaddr(msg, IEEE802154_ATTR_HW_ADDR, desc->hwaddr) || + nla_put_u32(msg, IEEE802154_ATTR_LLSEC_FRAME_COUNTER, + desc->frame_counter) || + nla_put_u8(msg, IEEE802154_ATTR_LLSEC_DEV_OVERRIDE, + desc->seclevel_exempt) || + nla_put_u8(msg, IEEE802154_ATTR_LLSEC_DEV_KEY_MODE, desc->key_mode)) + goto nla_put_failure; + + genlmsg_end(msg, hdr); + return 0; + +nla_put_failure: + genlmsg_cancel(msg, hdr); +out: + return -EMSGSIZE; +} + +static int llsec_iter_devs(struct llsec_dump_data *data) +{ + struct ieee802154_llsec_device *pos; + int rc = 0, idx = 0; + + list_for_each_entry(pos, &data->table->devices, list) { + if (idx++ < data->s_idx) + continue; + + if (ieee802154_nl_fill_dev(data->skb, data->portid, + data->nlmsg_seq, pos, data->dev)) { + rc = -EMSGSIZE; + break; + } + + data->s_idx++; + } + + return rc; +} + +int ieee802154_llsec_dump_devs(struct sk_buff *skb, struct netlink_callback *cb) +{ + return ieee802154_llsec_dump_table(skb, cb, llsec_iter_devs); +} + + + +static int llsec_add_devkey(struct net_device *dev, struct genl_info *info) +{ + struct ieee802154_mlme_ops *ops = ieee802154_mlme_ops(dev); + struct ieee802154_llsec_device_key key; + __le64 devaddr; + + if (!info->attrs[IEEE802154_ATTR_LLSEC_FRAME_COUNTER] || + !info->attrs[IEEE802154_ATTR_HW_ADDR] || + ieee802154_llsec_parse_key_id(info, &key.key_id)) + return -EINVAL; + + devaddr = nla_get_hwaddr(info->attrs[IEEE802154_ATTR_HW_ADDR]); + key.frame_counter = nla_get_u32(info->attrs[IEEE802154_ATTR_LLSEC_FRAME_COUNTER]); + + return ops->llsec->add_devkey(dev, devaddr, &key); +} + +int ieee802154_llsec_add_devkey(struct sk_buff *skb, struct genl_info *info) +{ + if ((info->nlhdr->nlmsg_flags & (NLM_F_CREATE | NLM_F_EXCL)) != + (NLM_F_CREATE | NLM_F_EXCL)) + return -EINVAL; + + return ieee802154_nl_llsec_change(skb, info, llsec_add_devkey); +} + +static int llsec_del_devkey(struct net_device *dev, struct genl_info *info) +{ + struct ieee802154_mlme_ops *ops = ieee802154_mlme_ops(dev); + struct ieee802154_llsec_device_key key; + __le64 devaddr; + + if (!info->attrs[IEEE802154_ATTR_HW_ADDR] || + ieee802154_llsec_parse_key_id(info, &key.key_id)) + return -EINVAL; + + devaddr = nla_get_hwaddr(info->attrs[IEEE802154_ATTR_HW_ADDR]); + + return ops->llsec->del_devkey(dev, devaddr, &key); +} + +int ieee802154_llsec_del_devkey(struct sk_buff *skb, struct genl_info *info) +{ + return ieee802154_nl_llsec_change(skb, info, llsec_del_devkey); +} + +static int +ieee802154_nl_fill_devkey(struct sk_buff *msg, u32 portid, u32 seq, + __le64 devaddr, + const struct ieee802154_llsec_device_key *devkey, + const struct net_device *dev) +{ + void *hdr; + + hdr = genlmsg_put(msg, 0, seq, &nl802154_family, NLM_F_MULTI, + IEEE802154_LLSEC_LIST_DEVKEY); + if (!hdr) + goto out; + + if (nla_put_string(msg, IEEE802154_ATTR_DEV_NAME, dev->name) || + nla_put_u32(msg, IEEE802154_ATTR_DEV_INDEX, dev->ifindex) || + nla_put_hwaddr(msg, IEEE802154_ATTR_HW_ADDR, devaddr) || + nla_put_u32(msg, IEEE802154_ATTR_LLSEC_FRAME_COUNTER, + devkey->frame_counter) || + ieee802154_llsec_fill_key_id(msg, &devkey->key_id)) + goto nla_put_failure; + + genlmsg_end(msg, hdr); + return 0; + +nla_put_failure: + genlmsg_cancel(msg, hdr); +out: + return -EMSGSIZE; +} + +static int llsec_iter_devkeys(struct llsec_dump_data *data) +{ + struct ieee802154_llsec_device *dpos; + struct ieee802154_llsec_device_key *kpos; + int rc = 0, idx = 0, idx2; + + list_for_each_entry(dpos, &data->table->devices, list) { + if (idx++ < data->s_idx) + continue; + + idx2 = 0; + + list_for_each_entry(kpos, &dpos->keys, list) { + if (idx2++ < data->s_idx2) + continue; + + if (ieee802154_nl_fill_devkey(data->skb, data->portid, + data->nlmsg_seq, + dpos->hwaddr, kpos, + data->dev)) { + return rc = -EMSGSIZE; + } + + data->s_idx2++; + } + + data->s_idx++; + } + + return rc; +} + +int ieee802154_llsec_dump_devkeys(struct sk_buff *skb, + struct netlink_callback *cb) +{ + return ieee802154_llsec_dump_table(skb, cb, llsec_iter_devkeys); +} + + + +static int +llsec_parse_seclevel(struct genl_info *info, + struct ieee802154_llsec_seclevel *sl) +{ + memset(sl, 0, sizeof(*sl)); + + if (!info->attrs[IEEE802154_ATTR_LLSEC_FRAME_TYPE] || + !info->attrs[IEEE802154_ATTR_LLSEC_SECLEVELS] || + !info->attrs[IEEE802154_ATTR_LLSEC_DEV_OVERRIDE]) + return -EINVAL; + + sl->frame_type = nla_get_u8(info->attrs[IEEE802154_ATTR_LLSEC_FRAME_TYPE]); + if (sl->frame_type == IEEE802154_FC_TYPE_MAC_CMD) { + if (!info->attrs[IEEE802154_ATTR_LLSEC_CMD_FRAME_ID]) + return -EINVAL; + + sl->cmd_frame_id = nla_get_u8(info->attrs[IEEE802154_ATTR_LLSEC_CMD_FRAME_ID]); + } + + sl->sec_levels = nla_get_u8(info->attrs[IEEE802154_ATTR_LLSEC_SECLEVELS]); + sl->device_override = nla_get_u8(info->attrs[IEEE802154_ATTR_LLSEC_DEV_OVERRIDE]); + + return 0; +} + +static int llsec_add_seclevel(struct net_device *dev, struct genl_info *info) +{ + struct ieee802154_mlme_ops *ops = ieee802154_mlme_ops(dev); + struct ieee802154_llsec_seclevel sl; + + if (llsec_parse_seclevel(info, &sl)) + return -EINVAL; + + return ops->llsec->add_seclevel(dev, &sl); +} + +int ieee802154_llsec_add_seclevel(struct sk_buff *skb, struct genl_info *info) +{ + if ((info->nlhdr->nlmsg_flags & (NLM_F_CREATE | NLM_F_EXCL)) != + (NLM_F_CREATE | NLM_F_EXCL)) + return -EINVAL; + + return ieee802154_nl_llsec_change(skb, info, llsec_add_seclevel); +} + +static int llsec_del_seclevel(struct net_device *dev, struct genl_info *info) +{ + struct ieee802154_mlme_ops *ops = ieee802154_mlme_ops(dev); + struct ieee802154_llsec_seclevel sl; + + if (llsec_parse_seclevel(info, &sl)) + return -EINVAL; + + return ops->llsec->del_seclevel(dev, &sl); +} + +int ieee802154_llsec_del_seclevel(struct sk_buff *skb, struct genl_info *info) +{ + return ieee802154_nl_llsec_change(skb, info, llsec_del_seclevel); +} + +static int +ieee802154_nl_fill_seclevel(struct sk_buff *msg, u32 portid, u32 seq, + const struct ieee802154_llsec_seclevel *sl, + const struct net_device *dev) +{ + void *hdr; + + hdr = genlmsg_put(msg, 0, seq, &nl802154_family, NLM_F_MULTI, + IEEE802154_LLSEC_LIST_SECLEVEL); + if (!hdr) + goto out; + + if (nla_put_string(msg, IEEE802154_ATTR_DEV_NAME, dev->name) || + nla_put_u32(msg, IEEE802154_ATTR_DEV_INDEX, dev->ifindex) || + nla_put_u8(msg, IEEE802154_ATTR_LLSEC_FRAME_TYPE, sl->frame_type) || + nla_put_u8(msg, IEEE802154_ATTR_LLSEC_SECLEVELS, sl->sec_levels) || + nla_put_u8(msg, IEEE802154_ATTR_LLSEC_DEV_OVERRIDE, + sl->device_override)) + goto nla_put_failure; + + if (sl->frame_type == IEEE802154_FC_TYPE_MAC_CMD && + nla_put_u8(msg, IEEE802154_ATTR_LLSEC_CMD_FRAME_ID, + sl->cmd_frame_id)) + goto nla_put_failure; + + genlmsg_end(msg, hdr); + return 0; + +nla_put_failure: + genlmsg_cancel(msg, hdr); +out: + return -EMSGSIZE; +} + +static int llsec_iter_seclevels(struct llsec_dump_data *data) +{ + struct ieee802154_llsec_seclevel *pos; + int rc = 0, idx = 0; + + list_for_each_entry(pos, &data->table->security_levels, list) { + if (idx++ < data->s_idx) + continue; + + if (ieee802154_nl_fill_seclevel(data->skb, data->portid, + data->nlmsg_seq, pos, + data->dev)) { + rc = -EMSGSIZE; + break; + } + + data->s_idx++; + } + + return rc; +} + +int ieee802154_llsec_dump_seclevels(struct sk_buff *skb, + struct netlink_callback *cb) +{ + return ieee802154_llsec_dump_table(skb, cb, llsec_iter_seclevels); +} diff --git a/net/ieee802154/nl_policy.c b/net/ieee802154/nl_policy.c index fd7be5e45ce..3a703ab8834 100644 --- a/net/ieee802154/nl_policy.c +++ b/net/ieee802154/nl_policy.c @@ -62,5 +62,21 @@ const struct nla_policy ieee802154_policy[IEEE802154_ATTR_MAX + 1] = { [IEEE802154_ATTR_CSMA_MAX_BE] = { .type = NLA_U8, }, [IEEE802154_ATTR_FRAME_RETRIES] = { .type = NLA_S8, }, + + [IEEE802154_ATTR_LLSEC_ENABLED] = { .type = NLA_U8, }, + [IEEE802154_ATTR_LLSEC_SECLEVEL] = { .type = NLA_U8, }, + [IEEE802154_ATTR_LLSEC_KEY_MODE] = { .type = NLA_U8, }, + [IEEE802154_ATTR_LLSEC_KEY_SOURCE_SHORT] = { .type = NLA_U32, }, + [IEEE802154_ATTR_LLSEC_KEY_SOURCE_EXTENDED] = { .type = NLA_HW_ADDR, }, + [IEEE802154_ATTR_LLSEC_KEY_ID] = { .type = NLA_U8, }, + [IEEE802154_ATTR_LLSEC_FRAME_COUNTER] = { .type = NLA_U32 }, + [IEEE802154_ATTR_LLSEC_KEY_BYTES] = { .len = 16, }, + [IEEE802154_ATTR_LLSEC_KEY_USAGE_FRAME_TYPES] = { .type = NLA_U8, }, + [IEEE802154_ATTR_LLSEC_KEY_USAGE_COMMANDS] = { .len = 258 / 8 }, + [IEEE802154_ATTR_LLSEC_FRAME_TYPE] = { .type = NLA_U8, }, + [IEEE802154_ATTR_LLSEC_CMD_FRAME_ID] = { .type = NLA_U8, }, + [IEEE802154_ATTR_LLSEC_SECLEVELS] = { .type = NLA_U8, }, + [IEEE802154_ATTR_LLSEC_DEV_OVERRIDE] = { .type = NLA_U8, }, + [IEEE802154_ATTR_LLSEC_DEV_KEY_MODE] = { .type = NLA_U8, }, }; diff --git a/net/ieee802154/reassembly.c b/net/ieee802154/reassembly.c index ef2d54372b1..6f1428c4870 100644 --- a/net/ieee802154/reassembly.c +++ b/net/ieee802154/reassembly.c @@ -36,7 +36,7 @@ struct lowpan_frag_info { u8 d_offset; }; -struct lowpan_frag_info *lowpan_cb(struct sk_buff *skb) +static struct lowpan_frag_info *lowpan_cb(struct sk_buff *skb) { return (struct lowpan_frag_info *)skb->cb; } @@ -120,6 +120,8 @@ fq_find(struct net *net, const struct lowpan_frag_info *frag_info, struct inet_frag_queue *q; struct lowpan_create_arg arg; unsigned int hash; + struct netns_ieee802154_lowpan *ieee802154_lowpan = + net_ieee802154_lowpan(net); arg.tag = frag_info->d_tag; arg.d_size = frag_info->d_size; @@ -129,7 +131,7 @@ fq_find(struct net *net, const struct lowpan_frag_info *frag_info, read_lock(&lowpan_frags.lock); hash = lowpan_hash_frag(frag_info->d_tag, frag_info->d_size, src, dst); - q = inet_frag_find(&net->ieee802154_lowpan.frags, + q = inet_frag_find(&ieee802154_lowpan->frags, &lowpan_frags, &arg, hash); if (IS_ERR_OR_NULL(q)) { inet_frag_maybe_warn_overflow(q, pr_fmt()); @@ -357,6 +359,8 @@ int lowpan_frag_rcv(struct sk_buff *skb, const u8 frag_type) struct net *net = dev_net(skb->dev); struct lowpan_frag_info *frag_info = lowpan_cb(skb); struct ieee802154_addr source, dest; + struct netns_ieee802154_lowpan *ieee802154_lowpan = + net_ieee802154_lowpan(net); int err; source = mac_cb(skb)->source; @@ -366,10 +370,10 @@ int lowpan_frag_rcv(struct sk_buff *skb, const u8 frag_type) if (err < 0) goto err; - if (frag_info->d_size > net->ieee802154_lowpan.max_dsize) + if (frag_info->d_size > ieee802154_lowpan->max_dsize) goto err; - inet_frag_evictor(&net->ieee802154_lowpan.frags, &lowpan_frags, false); + inet_frag_evictor(&ieee802154_lowpan->frags, &lowpan_frags, false); fq = fq_find(net, frag_info, &source, &dest); if (fq != NULL) { @@ -436,6 +440,8 @@ static int __net_init lowpan_frags_ns_sysctl_register(struct net *net) { struct ctl_table *table; struct ctl_table_header *hdr; + struct netns_ieee802154_lowpan *ieee802154_lowpan = + net_ieee802154_lowpan(net); table = lowpan_frags_ns_ctl_table; if (!net_eq(net, &init_net)) { @@ -444,10 +450,10 @@ static int __net_init lowpan_frags_ns_sysctl_register(struct net *net) if (table == NULL) goto err_alloc; - table[0].data = &net->ieee802154_lowpan.frags.high_thresh; - table[1].data = &net->ieee802154_lowpan.frags.low_thresh; - table[2].data = &net->ieee802154_lowpan.frags.timeout; - table[3].data = &net->ieee802154_lowpan.max_dsize; + table[0].data = &ieee802154_lowpan->frags.high_thresh; + table[1].data = &ieee802154_lowpan->frags.low_thresh; + table[2].data = &ieee802154_lowpan->frags.timeout; + table[3].data = &ieee802154_lowpan->max_dsize; /* Don't export sysctls to unprivileged users */ if (net->user_ns != &init_user_ns) @@ -458,7 +464,7 @@ static int __net_init lowpan_frags_ns_sysctl_register(struct net *net) if (hdr == NULL) goto err_reg; - net->ieee802154_lowpan.sysctl.frags_hdr = hdr; + ieee802154_lowpan->sysctl.frags_hdr = hdr; return 0; err_reg: @@ -471,9 +477,11 @@ err_alloc: static void __net_exit lowpan_frags_ns_sysctl_unregister(struct net *net) { struct ctl_table *table; + struct netns_ieee802154_lowpan *ieee802154_lowpan = + net_ieee802154_lowpan(net); - table = net->ieee802154_lowpan.sysctl.frags_hdr->ctl_table_arg; - unregister_net_sysctl_table(net->ieee802154_lowpan.sysctl.frags_hdr); + table = ieee802154_lowpan->sysctl.frags_hdr->ctl_table_arg; + unregister_net_sysctl_table(ieee802154_lowpan->sysctl.frags_hdr); if (!net_eq(net, &init_net)) kfree(table); } @@ -514,20 +522,26 @@ static inline void lowpan_frags_sysctl_unregister(void) static int __net_init lowpan_frags_init_net(struct net *net) { - net->ieee802154_lowpan.frags.high_thresh = IPV6_FRAG_HIGH_THRESH; - net->ieee802154_lowpan.frags.low_thresh = IPV6_FRAG_LOW_THRESH; - net->ieee802154_lowpan.frags.timeout = IPV6_FRAG_TIMEOUT; - net->ieee802154_lowpan.max_dsize = 0xFFFF; + struct netns_ieee802154_lowpan *ieee802154_lowpan = + net_ieee802154_lowpan(net); - inet_frags_init_net(&net->ieee802154_lowpan.frags); + ieee802154_lowpan->frags.high_thresh = IPV6_FRAG_HIGH_THRESH; + ieee802154_lowpan->frags.low_thresh = IPV6_FRAG_LOW_THRESH; + ieee802154_lowpan->frags.timeout = IPV6_FRAG_TIMEOUT; + ieee802154_lowpan->max_dsize = 0xFFFF; + + inet_frags_init_net(&ieee802154_lowpan->frags); return lowpan_frags_ns_sysctl_register(net); } static void __net_exit lowpan_frags_exit_net(struct net *net) { + struct netns_ieee802154_lowpan *ieee802154_lowpan = + net_ieee802154_lowpan(net); + lowpan_frags_ns_sysctl_unregister(net); - inet_frags_exit_net(&net->ieee802154_lowpan.frags, &lowpan_frags); + inet_frags_exit_net(&ieee802154_lowpan->frags, &lowpan_frags); } static struct pernet_operations lowpan_frags_ops = { diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 8c54870db79..d5e6836cf77 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -254,7 +254,6 @@ static int inet_create(struct net *net, struct socket *sock, int protocol, struct inet_sock *inet; struct proto *answer_prot; unsigned char answer_flags; - char answer_no_check; int try_loading_module = 0; int err; @@ -312,7 +311,6 @@ lookup_protocol: sock->ops = answer->ops; answer_prot = answer->prot; - answer_no_check = answer->no_check; answer_flags = answer->flags; rcu_read_unlock(); @@ -324,7 +322,6 @@ lookup_protocol: goto out; err = 0; - sk->sk_no_check = answer_no_check; if (INET_PROTOSW_REUSE & answer_flags) sk->sk_reuse = SK_CAN_REUSE; @@ -1002,7 +999,6 @@ static struct inet_protosw inetsw_array[] = .protocol = IPPROTO_TCP, .prot = &tcp_prot, .ops = &inet_stream_ops, - .no_check = 0, .flags = INET_PROTOSW_PERMANENT | INET_PROTOSW_ICSK, }, @@ -1012,7 +1008,6 @@ static struct inet_protosw inetsw_array[] = .protocol = IPPROTO_UDP, .prot = &udp_prot, .ops = &inet_dgram_ops, - .no_check = UDP_CSUM_DEFAULT, .flags = INET_PROTOSW_PERMANENT, }, @@ -1021,7 +1016,6 @@ static struct inet_protosw inetsw_array[] = .protocol = IPPROTO_ICMP, .prot = &ping_prot, .ops = &inet_dgram_ops, - .no_check = UDP_CSUM_DEFAULT, .flags = INET_PROTOSW_REUSE, }, @@ -1030,7 +1024,6 @@ static struct inet_protosw inetsw_array[] = .protocol = IPPROTO_IP, /* wild card */ .prot = &raw_prot, .ops = &inet_sockraw_ops, - .no_check = UDP_CSUM_DEFAULT, .flags = INET_PROTOSW_REUSE, } }; @@ -1261,10 +1254,12 @@ static struct sk_buff *inet_gso_segment(struct sk_buff *skb, SKB_GSO_DODGY | SKB_GSO_TCP_ECN | SKB_GSO_GRE | + SKB_GSO_GRE_CSUM | SKB_GSO_IPIP | SKB_GSO_SIT | SKB_GSO_TCPV6 | SKB_GSO_UDP_TUNNEL | + SKB_GSO_UDP_TUNNEL_CSUM | SKB_GSO_MPLS | 0))) goto out; @@ -1476,22 +1471,20 @@ int inet_ctl_sock_create(struct sock **sk, unsigned short family, } EXPORT_SYMBOL_GPL(inet_ctl_sock_create); -unsigned long snmp_fold_field(void __percpu *mib[], int offt) +unsigned long snmp_fold_field(void __percpu *mib, int offt) { unsigned long res = 0; - int i, j; + int i; - for_each_possible_cpu(i) { - for (j = 0; j < SNMP_ARRAY_SZ; j++) - res += *(((unsigned long *) per_cpu_ptr(mib[j], i)) + offt); - } + for_each_possible_cpu(i) + res += *(((unsigned long *) per_cpu_ptr(mib, i)) + offt); return res; } EXPORT_SYMBOL_GPL(snmp_fold_field); #if BITS_PER_LONG==32 -u64 snmp_fold_field64(void __percpu *mib[], int offt, size_t syncp_offset) +u64 snmp_fold_field64(void __percpu *mib, int offt, size_t syncp_offset) { u64 res = 0; int cpu; @@ -1502,7 +1495,7 @@ u64 snmp_fold_field64(void __percpu *mib[], int offt, size_t syncp_offset) u64 v; unsigned int start; - bhptr = per_cpu_ptr(mib[0], cpu); + bhptr = per_cpu_ptr(mib, cpu); syncp = (struct u64_stats_sync *)(bhptr + syncp_offset); do { start = u64_stats_fetch_begin_irq(syncp); @@ -1516,25 +1509,6 @@ u64 snmp_fold_field64(void __percpu *mib[], int offt, size_t syncp_offset) EXPORT_SYMBOL_GPL(snmp_fold_field64); #endif -int snmp_mib_init(void __percpu *ptr[2], size_t mibsize, size_t align) -{ - BUG_ON(ptr == NULL); - ptr[0] = __alloc_percpu(mibsize, align); - if (!ptr[0]) - return -ENOMEM; - -#if SNMP_ARRAY_SZ == 2 - ptr[1] = __alloc_percpu(mibsize, align); - if (!ptr[1]) { - free_percpu(ptr[0]); - ptr[0] = NULL; - return -ENOMEM; - } -#endif - return 0; -} -EXPORT_SYMBOL_GPL(snmp_mib_init); - #ifdef CONFIG_IP_MULTICAST static const struct net_protocol igmp_protocol = { .handler = igmp_rcv, @@ -1570,40 +1544,30 @@ static __net_init int ipv4_mib_init_net(struct net *net) { int i; - if (snmp_mib_init((void __percpu **)net->mib.tcp_statistics, - sizeof(struct tcp_mib), - __alignof__(struct tcp_mib)) < 0) + net->mib.tcp_statistics = alloc_percpu(struct tcp_mib); + if (!net->mib.tcp_statistics) goto err_tcp_mib; - if (snmp_mib_init((void __percpu **)net->mib.ip_statistics, - sizeof(struct ipstats_mib), - __alignof__(struct ipstats_mib)) < 0) + net->mib.ip_statistics = alloc_percpu(struct ipstats_mib); + if (!net->mib.ip_statistics) goto err_ip_mib; for_each_possible_cpu(i) { struct ipstats_mib *af_inet_stats; - af_inet_stats = per_cpu_ptr(net->mib.ip_statistics[0], i); - u64_stats_init(&af_inet_stats->syncp); -#if SNMP_ARRAY_SZ == 2 - af_inet_stats = per_cpu_ptr(net->mib.ip_statistics[1], i); + af_inet_stats = per_cpu_ptr(net->mib.ip_statistics, i); u64_stats_init(&af_inet_stats->syncp); -#endif } - if (snmp_mib_init((void __percpu **)net->mib.net_statistics, - sizeof(struct linux_mib), - __alignof__(struct linux_mib)) < 0) + net->mib.net_statistics = alloc_percpu(struct linux_mib); + if (!net->mib.net_statistics) goto err_net_mib; - if (snmp_mib_init((void __percpu **)net->mib.udp_statistics, - sizeof(struct udp_mib), - __alignof__(struct udp_mib)) < 0) + net->mib.udp_statistics = alloc_percpu(struct udp_mib); + if (!net->mib.udp_statistics) goto err_udp_mib; - if (snmp_mib_init((void __percpu **)net->mib.udplite_statistics, - sizeof(struct udp_mib), - __alignof__(struct udp_mib)) < 0) + net->mib.udplite_statistics = alloc_percpu(struct udp_mib); + if (!net->mib.udplite_statistics) goto err_udplite_mib; - if (snmp_mib_init((void __percpu **)net->mib.icmp_statistics, - sizeof(struct icmp_mib), - __alignof__(struct icmp_mib)) < 0) + net->mib.icmp_statistics = alloc_percpu(struct icmp_mib); + if (!net->mib.icmp_statistics) goto err_icmp_mib; net->mib.icmpmsg_statistics = kzalloc(sizeof(struct icmpmsg_mib), GFP_KERNEL); @@ -1614,17 +1578,17 @@ static __net_init int ipv4_mib_init_net(struct net *net) return 0; err_icmpmsg_mib: - snmp_mib_free((void __percpu **)net->mib.icmp_statistics); + free_percpu(net->mib.icmp_statistics); err_icmp_mib: - snmp_mib_free((void __percpu **)net->mib.udplite_statistics); + free_percpu(net->mib.udplite_statistics); err_udplite_mib: - snmp_mib_free((void __percpu **)net->mib.udp_statistics); + free_percpu(net->mib.udp_statistics); err_udp_mib: - snmp_mib_free((void __percpu **)net->mib.net_statistics); + free_percpu(net->mib.net_statistics); err_net_mib: - snmp_mib_free((void __percpu **)net->mib.ip_statistics); + free_percpu(net->mib.ip_statistics); err_ip_mib: - snmp_mib_free((void __percpu **)net->mib.tcp_statistics); + free_percpu(net->mib.tcp_statistics); err_tcp_mib: return -ENOMEM; } @@ -1632,12 +1596,12 @@ err_tcp_mib: static __net_exit void ipv4_mib_exit_net(struct net *net) { kfree(net->mib.icmpmsg_statistics); - snmp_mib_free((void __percpu **)net->mib.icmp_statistics); - snmp_mib_free((void __percpu **)net->mib.udplite_statistics); - snmp_mib_free((void __percpu **)net->mib.udp_statistics); - snmp_mib_free((void __percpu **)net->mib.net_statistics); - snmp_mib_free((void __percpu **)net->mib.ip_statistics); - snmp_mib_free((void __percpu **)net->mib.tcp_statistics); + free_percpu(net->mib.icmp_statistics); + free_percpu(net->mib.udplite_statistics); + free_percpu(net->mib.udp_statistics); + free_percpu(net->mib.net_statistics); + free_percpu(net->mib.ip_statistics); + free_percpu(net->mib.tcp_statistics); } static __net_initdata struct pernet_operations ipv4_mib_ops = { @@ -1650,6 +1614,39 @@ static int __init init_ipv4_mibs(void) return register_pernet_subsys(&ipv4_mib_ops); } +static __net_init int inet_init_net(struct net *net) +{ + /* + * Set defaults for local port range + */ + seqlock_init(&net->ipv4.ip_local_ports.lock); + net->ipv4.ip_local_ports.range[0] = 32768; + net->ipv4.ip_local_ports.range[1] = 61000; + + seqlock_init(&net->ipv4.ping_group_range.lock); + /* + * Sane defaults - nobody may create ping sockets. + * Boot scripts should set this to distro-specific group. + */ + net->ipv4.ping_group_range.range[0] = make_kgid(&init_user_ns, 1); + net->ipv4.ping_group_range.range[1] = make_kgid(&init_user_ns, 0); + return 0; +} + +static __net_exit void inet_exit_net(struct net *net) +{ +} + +static __net_initdata struct pernet_operations af_inet_ops = { + .init = inet_init_net, + .exit = inet_exit_net, +}; + +static int __init init_inet_pernet_ops(void) +{ + return register_pernet_subsys(&af_inet_ops); +} + static int ipv4_proc_init(void); /* @@ -1703,13 +1700,9 @@ static int __init inet_init(void) BUILD_BUG_ON(sizeof(struct inet_skb_parm) > FIELD_SIZEOF(struct sk_buff, cb)); - sysctl_local_reserved_ports = kzalloc(65536 / 8, GFP_KERNEL); - if (!sysctl_local_reserved_ports) - goto out; - rc = proto_register(&tcp_prot, 1); if (rc) - goto out_free_reserved_ports; + goto out; rc = proto_register(&udp_prot, 1); if (rc) @@ -1794,6 +1787,9 @@ static int __init inet_init(void) if (ip_mr_init()) pr_crit("%s: Cannot init ipv4 mroute\n", __func__); #endif + + if (init_inet_pernet_ops()) + pr_crit("%s: Cannot init ipv4 inet pernet ops\n", __func__); /* * Initialise per-cpu ipv4 mibs */ @@ -1816,8 +1812,6 @@ out_unregister_udp_proto: proto_unregister(&udp_prot); out_unregister_tcp_proto: proto_unregister(&tcp_prot); -out_free_reserved_ports: - kfree(sysctl_local_reserved_ports); goto out; } diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index bdbf68bb2e2..e9449376b58 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -106,7 +106,6 @@ static const struct nla_policy ifa_ipv4_policy[IFA_MAX+1] = { #define IN4_ADDR_HSIZE (1U << IN4_ADDR_HSIZE_SHIFT) static struct hlist_head inet_addr_lst[IN4_ADDR_HSIZE]; -static DEFINE_SPINLOCK(inet_addr_hash_lock); static u32 inet_addr_hash(struct net *net, __be32 addr) { @@ -119,16 +118,14 @@ static void inet_hash_insert(struct net *net, struct in_ifaddr *ifa) { u32 hash = inet_addr_hash(net, ifa->ifa_local); - spin_lock(&inet_addr_hash_lock); + ASSERT_RTNL(); hlist_add_head_rcu(&ifa->hash, &inet_addr_lst[hash]); - spin_unlock(&inet_addr_hash_lock); } static void inet_hash_remove(struct in_ifaddr *ifa) { - spin_lock(&inet_addr_hash_lock); + ASSERT_RTNL(); hlist_del_init_rcu(&ifa->hash); - spin_unlock(&inet_addr_hash_lock); } /** @@ -830,7 +827,7 @@ static int inet_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh) ifa_existing = find_matching_ifa(ifa); if (!ifa_existing) { /* It would be best to check for !NLM_F_CREATE here but - * userspace alreay relies on not having to provide this. + * userspace already relies on not having to provide this. */ set_ifa_lifetime(ifa, valid_lft, prefered_lft); return __inet_insert_ifa(ifa, nlh, NETLINK_CB(skb).portid); diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 8a043f03c88..b10cd43a472 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -821,13 +821,13 @@ struct fib_info *fib_create_info(struct fib_config *cfg) fi = kzalloc(sizeof(*fi)+nhs*sizeof(struct fib_nh), GFP_KERNEL); if (fi == NULL) goto failure; + fib_info_cnt++; if (cfg->fc_mx) { fi->fib_metrics = kzalloc(sizeof(u32) * RTAX_MAX, GFP_KERNEL); if (!fi->fib_metrics) goto failure; } else fi->fib_metrics = (u32 *) dst_default_metrics; - fib_info_cnt++; fi->fib_net = hold_net(net); fi->fib_protocol = cfg->fc_protocol; diff --git a/net/ipv4/gre_demux.c b/net/ipv4/gre_demux.c index 250be7421ab..4e9619bca73 100644 --- a/net/ipv4/gre_demux.c +++ b/net/ipv4/gre_demux.c @@ -84,7 +84,8 @@ void gre_build_header(struct sk_buff *skb, const struct tnl_ptk_info *tpi, ptr--; } if (tpi->flags&TUNNEL_CSUM && - !(skb_shinfo(skb)->gso_type & SKB_GSO_GRE)) { + !(skb_shinfo(skb)->gso_type & + (SKB_GSO_GRE|SKB_GSO_GRE_CSUM))) { *ptr = 0; *(__sum16 *)ptr = csum_fold(skb_checksum(skb, 0, skb->len, 0)); @@ -93,28 +94,6 @@ void gre_build_header(struct sk_buff *skb, const struct tnl_ptk_info *tpi, } EXPORT_SYMBOL_GPL(gre_build_header); -static __sum16 check_checksum(struct sk_buff *skb) -{ - __sum16 csum = 0; - - switch (skb->ip_summed) { - case CHECKSUM_COMPLETE: - csum = csum_fold(skb->csum); - - if (!csum) - break; - /* Fall through. */ - - case CHECKSUM_NONE: - skb->csum = 0; - csum = __skb_checksum_complete(skb); - skb->ip_summed = CHECKSUM_COMPLETE; - break; - } - - return csum; -} - static int parse_gre_header(struct sk_buff *skb, struct tnl_ptk_info *tpi, bool *csum_err) { @@ -141,7 +120,7 @@ static int parse_gre_header(struct sk_buff *skb, struct tnl_ptk_info *tpi, options = (__be32 *)(greh + 1); if (greh->flags & GRE_CSUM) { - if (check_checksum(skb)) { + if (skb_checksum_simple_validate(skb)) { *csum_err = true; return -EINVAL; } diff --git a/net/ipv4/gre_offload.c b/net/ipv4/gre_offload.c index f1d32280cb5..24deb3928b9 100644 --- a/net/ipv4/gre_offload.c +++ b/net/ipv4/gre_offload.c @@ -42,6 +42,7 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb, SKB_GSO_DODGY | SKB_GSO_TCP_ECN | SKB_GSO_GRE | + SKB_GSO_GRE_CSUM | SKB_GSO_IPIP))) goto out; @@ -55,6 +56,8 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb, goto out; csum = !!(greh->flags & GRE_CSUM); + if (csum) + skb->encap_hdr_csum = 1; if (unlikely(!pskb_may_pull(skb, ghl))) goto out; @@ -94,10 +97,13 @@ static struct sk_buff *gre_gso_segment(struct sk_buff *skb, } } - greh = (struct gre_base_hdr *)(skb->data); + skb_reset_transport_header(skb); + + greh = (struct gre_base_hdr *) + skb_transport_header(skb); pcsum = (__be32 *)(greh + 1); *pcsum = 0; - *(__sum16 *)pcsum = csum_fold(skb_checksum(skb, 0, skb->len, 0)); + *(__sum16 *)pcsum = gso_make_checksum(skb, 0); } __skb_push(skb, tnl_hlen - ghl); diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 0134663fdbc..79c3d947a48 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -337,6 +337,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) struct sock *sk; struct inet_sock *inet; __be32 daddr, saddr; + u32 mark = IP4_REPLY_MARK(net, skb->mark); if (ip_options_echo(&icmp_param->replyopts.opt.opt, skb)) return; @@ -349,6 +350,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) icmp_param->data.icmph.checksum = 0; inet->tos = ip_hdr(skb)->tos; + sk->sk_mark = mark; daddr = ipc.addr = ip_hdr(skb)->saddr; saddr = fib_compute_spec_dst(skb); ipc.opt = NULL; @@ -364,6 +366,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) memset(&fl4, 0, sizeof(fl4)); fl4.daddr = daddr; fl4.saddr = saddr; + fl4.flowi4_mark = mark; fl4.flowi4_tos = RT_TOS(ip_hdr(skb)->tos); fl4.flowi4_proto = IPPROTO_ICMP; security_skb_classify_flow(skb, flowi4_to_flowi(&fl4)); @@ -382,7 +385,7 @@ static struct rtable *icmp_route_lookup(struct net *net, struct flowi4 *fl4, struct sk_buff *skb_in, const struct iphdr *iph, - __be32 saddr, u8 tos, + __be32 saddr, u8 tos, u32 mark, int type, int code, struct icmp_bxm *param) { @@ -394,6 +397,7 @@ static struct rtable *icmp_route_lookup(struct net *net, fl4->daddr = (param->replyopts.opt.opt.srr ? param->replyopts.opt.opt.faddr : iph->saddr); fl4->saddr = saddr; + fl4->flowi4_mark = mark; fl4->flowi4_tos = RT_TOS(tos); fl4->flowi4_proto = IPPROTO_ICMP; fl4->fl4_icmp_type = type; @@ -491,6 +495,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) struct flowi4 fl4; __be32 saddr; u8 tos; + u32 mark; struct net *net; struct sock *sk; @@ -592,6 +597,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) tos = icmp_pointers[type].error ? ((iph->tos & IPTOS_TOS_MASK) | IPTOS_PREC_INTERNETCONTROL) : iph->tos; + mark = IP4_REPLY_MARK(net, skb_in->mark); if (ip_options_echo(&icmp_param->replyopts.opt.opt, skb_in)) goto out_unlock; @@ -608,13 +614,14 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) icmp_param->skb = skb_in; icmp_param->offset = skb_network_offset(skb_in); inet_sk(sk)->tos = tos; + sk->sk_mark = mark; ipc.addr = iph->saddr; ipc.opt = &icmp_param->replyopts.opt; ipc.tx_flags = 0; ipc.ttl = 0; ipc.tos = -1; - rt = icmp_route_lookup(net, &fl4, skb_in, iph, saddr, tos, + rt = icmp_route_lookup(net, &fl4, skb_in, iph, saddr, tos, mark, type, code, icmp_param); if (IS_ERR(rt)) goto out_unlock; @@ -908,16 +915,8 @@ int icmp_rcv(struct sk_buff *skb) ICMP_INC_STATS_BH(net, ICMP_MIB_INMSGS); - switch (skb->ip_summed) { - case CHECKSUM_COMPLETE: - if (!csum_fold(skb->csum)) - break; - /* fall through */ - case CHECKSUM_NONE: - skb->csum = 0; - if (__skb_checksum_complete(skb)) - goto csum_error; - } + if (skb_checksum_simple_validate(skb)) + goto csum_error; if (!pskb_pull(skb, sizeof(*icmph))) goto error; diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 97e4d1655d2..6748d420f71 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -369,7 +369,7 @@ static struct sk_buff *igmpv3_newpack(struct net_device *dev, int size) pip->saddr = fl4.saddr; pip->protocol = IPPROTO_IGMP; pip->tot_len = 0; /* filled in later */ - ip_select_ident(skb, &rt->dst, NULL); + ip_select_ident(skb, NULL); ((u8 *)&pip[1])[0] = IPOPT_RA; ((u8 *)&pip[1])[1] = 4; ((u8 *)&pip[1])[2] = 0; @@ -714,7 +714,7 @@ static int igmp_send_report(struct in_device *in_dev, struct ip_mc_list *pmc, iph->daddr = dst; iph->saddr = fl4.saddr; iph->protocol = IPPROTO_IGMP; - ip_select_ident(skb, &rt->dst, NULL); + ip_select_ident(skb, NULL); ((u8 *)&iph[1])[0] = IPOPT_RA; ((u8 *)&iph[1])[1] = 4; ((u8 *)&iph[1])[2] = 0; @@ -988,16 +988,8 @@ int igmp_rcv(struct sk_buff *skb) if (!pskb_may_pull(skb, sizeof(struct igmphdr))) goto drop; - switch (skb->ip_summed) { - case CHECKSUM_COMPLETE: - if (!csum_fold(skb->csum)) - break; - /* fall through */ - case CHECKSUM_NONE: - skb->csum = 0; - if (__skb_checksum_complete(skb)) - goto drop; - } + if (skb_checksum_simple_validate(skb)) + goto drop; ih = igmp_hdr(skb); switch (ih->type) { diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 0d1e2cb877e..14d02ea905b 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -29,19 +29,16 @@ const char inet_csk_timer_bug_msg[] = "inet_csk BUG: unknown timer value\n"; EXPORT_SYMBOL(inet_csk_timer_bug_msg); #endif -unsigned long *sysctl_local_reserved_ports; -EXPORT_SYMBOL(sysctl_local_reserved_ports); - void inet_get_local_port_range(struct net *net, int *low, int *high) { unsigned int seq; do { - seq = read_seqbegin(&net->ipv4.sysctl_local_ports.lock); + seq = read_seqbegin(&net->ipv4.ip_local_ports.lock); - *low = net->ipv4.sysctl_local_ports.range[0]; - *high = net->ipv4.sysctl_local_ports.range[1]; - } while (read_seqretry(&net->ipv4.sysctl_local_ports.lock, seq)); + *low = net->ipv4.ip_local_ports.range[0]; + *high = net->ipv4.ip_local_ports.range[1]; + } while (read_seqretry(&net->ipv4.ip_local_ports.lock, seq)); } EXPORT_SYMBOL(inet_get_local_port_range); @@ -113,7 +110,7 @@ again: smallest_size = -1; do { - if (inet_is_reserved_local_port(rover)) + if (inet_is_local_reserved_port(net, rover)) goto next_nolock; head = &hashinfo->bhash[inet_bhashfn(net, rover, hashinfo->bhash_size)]; @@ -408,7 +405,7 @@ struct dst_entry *inet_csk_route_req(struct sock *sk, struct net *net = sock_net(sk); int flags = inet_sk_flowi_flags(sk); - flowi4_init_output(fl4, sk->sk_bound_dev_if, sk->sk_mark, + flowi4_init_output(fl4, sk->sk_bound_dev_if, ireq->ir_mark, RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE, sk->sk_protocol, flags, @@ -445,7 +442,7 @@ struct dst_entry *inet_csk_route_child_sock(struct sock *sk, rcu_read_lock(); opt = rcu_dereference(newinet->inet_opt); - flowi4_init_output(fl4, sk->sk_bound_dev_if, sk->sk_mark, + flowi4_init_output(fl4, sk->sk_bound_dev_if, inet_rsk(req)->ir_mark, RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE, sk->sk_protocol, inet_sk_flowi_flags(sk), (opt && opt->opt.srr) ? opt->opt.faddr : ireq->ir_rmt_addr, @@ -680,6 +677,8 @@ struct sock *inet_csk_clone_lock(const struct sock *sk, inet_sk(newsk)->inet_sport = htons(inet_rsk(req)->ir_num); newsk->sk_write_space = sk_stream_write_space; + newsk->sk_mark = inet_rsk(req)->ir_mark; + newicsk->icsk_retransmits = 0; newicsk->icsk_backoff = 0; newicsk->icsk_probes_out = 0; diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 8b9cf279450..43116e8c8e1 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -274,7 +274,7 @@ struct sock *__inet_lookup_established(struct net *net, const __be32 daddr, const u16 hnum, const int dif) { - INET_ADDR_COOKIE(acookie, saddr, daddr) + INET_ADDR_COOKIE(acookie, saddr, daddr); const __portpair ports = INET_COMBINED_PORTS(sport, hnum); struct sock *sk; const struct hlist_nulls_node *node; @@ -327,7 +327,7 @@ static int __inet_check_established(struct inet_timewait_death_row *death_row, __be32 daddr = inet->inet_rcv_saddr; __be32 saddr = inet->inet_daddr; int dif = sk->sk_bound_dev_if; - INET_ADDR_COOKIE(acookie, saddr, daddr) + INET_ADDR_COOKIE(acookie, saddr, daddr); const __portpair ports = INET_COMBINED_PORTS(inet->inet_dport, lport); struct net *net = sock_net(sk); unsigned int hash = inet_ehashfn(net, daddr, lport, @@ -500,7 +500,7 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, local_bh_disable(); for (i = 1; i <= remaining; i++) { port = low + (i + offset) % remaining; - if (inet_is_reserved_local_port(port)) + if (inet_is_local_reserved_port(net, port)) continue; head = &hinfo->bhash[inet_bhashfn(net, port, hinfo->bhash_size)]; diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c index 48f42446511..4ced1b9a97f 100644 --- a/net/ipv4/inetpeer.c +++ b/net/ipv4/inetpeer.c @@ -26,20 +26,7 @@ * Theory of operations. * We keep one entry for each peer IP address. The nodes contains long-living * information about the peer which doesn't depend on routes. - * At this moment this information consists only of ID field for the next - * outgoing IP packet. This field is incremented with each packet as encoded - * in inet_getid() function (include/net/inetpeer.h). - * At the moment of writing this notes identifier of IP packets is generated - * to be unpredictable using this code only for packets subjected - * (actually or potentially) to defragmentation. I.e. DF packets less than - * PMTU in size when local fragmentation is disabled use a constant ID and do - * not use this code (see ip_select_ident() in include/net/ip.h). * - * Route cache entries hold references to our nodes. - * New cache entries get references via lookup by destination IP address in - * the avl tree. The reference is grabbed only when it's needed i.e. only - * when we try to output IP packet which needs an unpredictable ID (see - * __ip_select_ident() in net/ipv4/route.c). * Nodes are removed only when reference counter goes to 0. * When it's happened the node may be removed when a sufficient amount of * time has been passed since its last use. The less-recently-used entry can @@ -62,7 +49,6 @@ * refcnt: atomically against modifications on other CPU; * usually under some other lock to prevent node disappearing * daddr: unchangeable - * ip_id_count: atomic value (no lock needed) */ static struct kmem_cache *peer_cachep __read_mostly; @@ -120,7 +106,7 @@ int inet_peer_maxttl __read_mostly = 10 * 60 * HZ; /* usual time to live: 10 min static void inetpeer_gc_worker(struct work_struct *work) { struct inet_peer *p, *n, *c; - LIST_HEAD(list); + struct list_head list; spin_lock_bh(&gc_lock); list_replace_init(&gc_list, &list); @@ -497,10 +483,6 @@ relookup: p->daddr = *daddr; atomic_set(&p->refcnt, 1); atomic_set(&p->rid, 0); - atomic_set(&p->ip_id_count, - (daddr->family == AF_INET) ? - secure_ip_id(daddr->addr.a4) : - secure_ipv6_id(daddr->addr.a6)); p->metrics[RTAX_LOCK-1] = INETPEER_METRICS_NEW; p->rate_tokens = 0; /* 60*HZ is arbitrary, but chosen enough high so that the first diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c index be8abe73bb9..3a83ce5efa8 100644 --- a/net/ipv4/ip_forward.c +++ b/net/ipv4/ip_forward.c @@ -42,12 +42,12 @@ static bool ip_may_fragment(const struct sk_buff *skb) { return unlikely((ip_hdr(skb)->frag_off & htons(IP_DF)) == 0) || - !skb->local_df; + skb->ignore_df; } static bool ip_exceeds_mtu(const struct sk_buff *skb, unsigned int mtu) { - if (skb->len <= mtu || skb->local_df) + if (skb->len <= mtu) return false; if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu) @@ -56,53 +56,6 @@ static bool ip_exceeds_mtu(const struct sk_buff *skb, unsigned int mtu) return true; } -static bool ip_gso_exceeds_dst_mtu(const struct sk_buff *skb) -{ - unsigned int mtu; - - if (skb->local_df || !skb_is_gso(skb)) - return false; - - mtu = ip_dst_mtu_maybe_forward(skb_dst(skb), true); - - /* if seglen > mtu, do software segmentation for IP fragmentation on - * output. DF bit cannot be set since ip_forward would have sent - * icmp error. - */ - return skb_gso_network_seglen(skb) > mtu; -} - -/* called if GSO skb needs to be fragmented on forward */ -static int ip_forward_finish_gso(struct sk_buff *skb) -{ - struct dst_entry *dst = skb_dst(skb); - netdev_features_t features; - struct sk_buff *segs; - int ret = 0; - - features = netif_skb_dev_features(skb, dst->dev); - segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK); - if (IS_ERR(segs)) { - kfree_skb(skb); - return -ENOMEM; - } - - consume_skb(skb); - - do { - struct sk_buff *nskb = segs->next; - int err; - - segs->next = NULL; - err = dst_output(segs); - - if (err && ret == 0) - ret = err; - segs = nskb; - } while (segs); - - return ret; -} static int ip_forward_finish(struct sk_buff *skb) { @@ -114,9 +67,6 @@ static int ip_forward_finish(struct sk_buff *skb) if (unlikely(opt->optlen)) ip_forward_options(skb); - if (ip_gso_exceeds_dst_mtu(skb)) - return ip_forward_finish_gso(skb); - return dst_output(skb); } diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index c10a3ce5cbf..ed32313e307 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -232,8 +232,9 @@ static void ip_expire(unsigned long arg) * "Fragment Reassembly Timeout" message, per RFC792. */ if (qp->user == IP_DEFRAG_AF_PACKET || - (qp->user == IP_DEFRAG_CONNTRACK_IN && - skb_rtable(head)->rt_type != RTN_LOCAL)) + ((qp->user >= IP_DEFRAG_CONNTRACK_IN) && + (qp->user <= __IP_DEFRAG_CONNTRACK_IN_END) && + (skb_rtable(head)->rt_type != RTN_LOCAL))) goto out_rcu_unlock; diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 94213c89156..c5a557a06a3 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -410,7 +410,7 @@ static int ipgre_open(struct net_device *dev) struct flowi4 fl4; struct rtable *rt; - rt = ip_route_output_gre(dev_net(dev), &fl4, + rt = ip_route_output_gre(t->net, &fl4, t->parms.iph.daddr, t->parms.iph.saddr, t->parms.o_key, @@ -434,7 +434,7 @@ static int ipgre_close(struct net_device *dev) if (ipv4_is_multicast(t->parms.iph.daddr) && t->mlink) { struct in_device *in_dev; - in_dev = inetdev_by_index(dev_net(dev), t->mlink); + in_dev = inetdev_by_index(t->net, t->mlink); if (in_dev) ip_mc_dec_group(in_dev, t->parms.iph.daddr); } @@ -478,7 +478,7 @@ static void __gre_tunnel_init(struct net_device *dev) dev->needed_headroom = LL_MAX_HEADER + sizeof(struct iphdr) + 4; dev->mtu = ETH_DATA_LEN - sizeof(struct iphdr) - 4; - dev->features |= NETIF_F_NETNS_LOCAL | GRE_FEATURES; + dev->features |= GRE_FEATURES; dev->hw_features |= GRE_FEATURES; if (!(tunnel->parms.o_flags & TUNNEL_SEQ)) { diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index f4ab72e19af..5e7aecea05c 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -364,7 +364,7 @@ int ip_options_compile(struct net *net, } if (optptr[2] <= optlen) { unsigned char *timeptr = NULL; - if (optptr[2]+3 > optptr[1]) { + if (optptr[2]+3 > optlen) { pp_ptr = optptr + 2; goto error; } @@ -376,7 +376,7 @@ int ip_options_compile(struct net *net, optptr[2] += 4; break; case IPOPT_TS_TSANDADDR: - if (optptr[2]+7 > optptr[1]) { + if (optptr[2]+7 > optlen) { pp_ptr = optptr + 2; goto error; } @@ -390,7 +390,7 @@ int ip_options_compile(struct net *net, optptr[2] += 8; break; case IPOPT_TS_PRESPEC: - if (optptr[2]+7 > optptr[1]) { + if (optptr[2]+7 > optlen) { pp_ptr = optptr + 2; goto error; } diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 1cbeba5edff..8d3b6b0e985 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -148,7 +148,7 @@ int ip_build_and_send_pkt(struct sk_buff *skb, struct sock *sk, iph->daddr = (opt && opt->opt.srr ? opt->opt.faddr : daddr); iph->saddr = saddr; iph->protocol = sk->sk_protocol; - ip_select_ident(skb, &rt->dst, sk); + ip_select_ident(skb, sk); if (opt && opt->opt.optlen) { iph->ihl += opt->opt.optlen>>2; @@ -211,6 +211,48 @@ static inline int ip_finish_output2(struct sk_buff *skb) return -EINVAL; } +static int ip_finish_output_gso(struct sk_buff *skb) +{ + netdev_features_t features; + struct sk_buff *segs; + int ret = 0; + + /* common case: locally created skb or seglen is <= mtu */ + if (((IPCB(skb)->flags & IPSKB_FORWARDED) == 0) || + skb_gso_network_seglen(skb) <= ip_skb_dst_mtu(skb)) + return ip_finish_output2(skb); + + /* Slowpath - GSO segment length is exceeding the dst MTU. + * + * This can happen in two cases: + * 1) TCP GRO packet, DF bit not set + * 2) skb arrived via virtio-net, we thus get TSO/GSO skbs directly + * from host network stack. + */ + features = netif_skb_features(skb); + segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK); + if (IS_ERR(segs)) { + kfree_skb(skb); + return -ENOMEM; + } + + consume_skb(skb); + + do { + struct sk_buff *nskb = segs->next; + int err; + + segs->next = NULL; + err = ip_fragment(segs, ip_finish_output2); + + if (err && ret == 0) + ret = err; + segs = nskb; + } while (segs); + + return ret; +} + static int ip_finish_output(struct sk_buff *skb) { #if defined(CONFIG_NETFILTER) && defined(CONFIG_XFRM) @@ -220,10 +262,13 @@ static int ip_finish_output(struct sk_buff *skb) return dst_output(skb); } #endif - if (skb->len > ip_skb_dst_mtu(skb) && !skb_is_gso(skb)) + if (skb_is_gso(skb)) + return ip_finish_output_gso(skb); + + if (skb->len > ip_skb_dst_mtu(skb)) return ip_fragment(skb, ip_finish_output2); - else - return ip_finish_output2(skb); + + return ip_finish_output2(skb); } int ip_mc_output(struct sock *sk, struct sk_buff *skb) @@ -370,7 +415,7 @@ packet_routed: skb_reset_network_header(skb); iph = ip_hdr(skb); *((__be16 *)iph) = htons((4 << 12) | (5 << 8) | (inet->tos & 0xff)); - if (ip_dont_fragment(sk, &rt->dst) && !skb->local_df) + if (ip_dont_fragment(sk, &rt->dst) && !skb->ignore_df) iph->frag_off = htons(IP_DF); else iph->frag_off = 0; @@ -385,8 +430,7 @@ packet_routed: ip_options_build(skb, &inet_opt->opt, inet->inet_daddr, rt, 0); } - ip_select_ident_more(skb, &rt->dst, sk, - (skb_shinfo(skb)->gso_segs ?: 1) - 1); + ip_select_ident_segs(skb, sk, skb_shinfo(skb)->gso_segs ?: 1); /* TODO : should we use skb->sk here instead of sk ? */ skb->priority = sk->sk_priority; @@ -456,7 +500,7 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) iph = ip_hdr(skb); mtu = ip_skb_dst_mtu(skb); - if (unlikely(((iph->frag_off & htons(IP_DF)) && !skb->local_df) || + if (unlikely(((iph->frag_off & htons(IP_DF)) && !skb->ignore_df) || (IPCB(skb)->frag_max_size && IPCB(skb)->frag_max_size > mtu))) { IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS); @@ -821,7 +865,7 @@ static int __ip_append_data(struct sock *sk, fragheaderlen = sizeof(struct iphdr) + (opt ? opt->optlen : 0); maxfraglen = ((mtu - fragheaderlen) & ~7) + fragheaderlen; - maxnonfragsize = ip_sk_local_df(sk) ? 0xFFFF : mtu; + maxnonfragsize = ip_sk_ignore_df(sk) ? 0xFFFF : mtu; if (cork->length + length > maxnonfragsize - fragheaderlen) { ip_local_error(sk, EMSGSIZE, fl4->daddr, inet->inet_dport, @@ -1144,7 +1188,7 @@ ssize_t ip_append_page(struct sock *sk, struct flowi4 *fl4, struct page *page, fragheaderlen = sizeof(struct iphdr) + (opt ? opt->optlen : 0); maxfraglen = ((mtu - fragheaderlen) & ~7) + fragheaderlen; - maxnonfragsize = ip_sk_local_df(sk) ? 0xFFFF : mtu; + maxnonfragsize = ip_sk_ignore_df(sk) ? 0xFFFF : mtu; if (cork->length + size > maxnonfragsize - fragheaderlen) { ip_local_error(sk, EMSGSIZE, fl4->daddr, inet->inet_dport, @@ -1305,10 +1349,10 @@ struct sk_buff *__ip_make_skb(struct sock *sk, * to fragment the frame generated here. No matter, what transforms * how transforms change size of the packet, it will come out. */ - skb->local_df = ip_sk_local_df(sk); + skb->ignore_df = ip_sk_ignore_df(sk); /* DF bit is set when we want to see DF on outgoing frames. - * If local_df is set too, we still allow to fragment this frame + * If ignore_df is set too, we still allow to fragment this frame * locally. */ if (inet->pmtudisc == IP_PMTUDISC_DO || inet->pmtudisc == IP_PMTUDISC_PROBE || @@ -1334,7 +1378,7 @@ struct sk_buff *__ip_make_skb(struct sock *sk, iph->ttl = ttl; iph->protocol = sk->sk_protocol; ip_copy_addrs(iph, fl4); - ip_select_ident(skb, &rt->dst, sk); + ip_select_ident(skb, sk); if (opt) { iph->ihl += opt->optlen>>2; @@ -1501,7 +1545,8 @@ void ip_send_unicast_reply(struct net *net, struct sk_buff *skb, __be32 daddr, daddr = replyopts.opt.opt.faddr; } - flowi4_init_output(&fl4, arg->bound_dev_if, 0, + flowi4_init_output(&fl4, arg->bound_dev_if, + IP4_REPLY_MARK(net, skb->mark), RT_TOS(arg->tos), RT_SCOPE_UNIVERSE, ip_hdr(skb)->protocol, ip_reply_arg_flowi_flags(arg), diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index fa5b7519765..86a00bd6684 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -395,11 +395,10 @@ static struct ip_tunnel *ip_tunnel_create(struct net *net, struct ip_tunnel_net *itn, struct ip_tunnel_parm *parms) { - struct ip_tunnel *nt, *fbt; + struct ip_tunnel *nt; struct net_device *dev; BUG_ON(!itn->fb_tunnel_dev); - fbt = netdev_priv(itn->fb_tunnel_dev); dev = __ip_tunnel_create(net, itn->fb_tunnel_dev->rtnl_link_ops, parms); if (IS_ERR(dev)) return ERR_CAST(dev); @@ -442,6 +441,8 @@ int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb, tunnel->i_seqno = ntohl(tpi->seq) + 1; } + skb_reset_network_header(skb); + err = IP_ECN_decapsulate(iph, skb); if (unlikely(err)) { if (log_ecn_error) @@ -538,9 +539,10 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, unsigned int max_headroom; /* The extra header space needed */ __be32 dst; int err; - bool connected = true; + bool connected; inner_iph = (const struct iphdr *)skb_inner_network_header(skb); + connected = (tunnel->parms.iph.daddr != 0); dst = tnl_params->daddr; if (dst == 0) { @@ -753,10 +755,8 @@ int ip_tunnel_ioctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd) if (!t && (cmd == SIOCADDTUNNEL)) { t = ip_tunnel_create(net, itn, p); - if (IS_ERR(t)) { - err = PTR_ERR(t); - break; - } + err = PTR_ERR_OR_ZERO(t); + break; } if (dev != itn->fb_tunnel_dev && cmd == SIOCCHGTUNNEL) { if (t != NULL) { @@ -880,6 +880,7 @@ int ip_tunnel_init_net(struct net *net, int ip_tnl_net_id, */ if (!IS_ERR(itn->fb_tunnel_dev)) { itn->fb_tunnel_dev->features |= NETIF_F_NETNS_LOCAL; + itn->fb_tunnel_dev->mtu = ip_tunnel_bind_dev(itn->fb_tunnel_dev); ip_tunnel_add(itn, netdev_priv(itn->fb_tunnel_dev)); } rtnl_unlock(); diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index bcf206c7900..f4c987bb7e9 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -74,7 +74,7 @@ int iptunnel_xmit(struct sock *sk, struct rtable *rt, struct sk_buff *skb, iph->daddr = dst; iph->saddr = src; iph->ttl = ttl; - __ip_select_ident(iph, &rt->dst, (skb_shinfo(skb)->gso_segs ?: 1) - 1); + __ip_select_ident(iph, skb_shinfo(skb)->gso_segs ?: 1); err = ip_local_out_sk(sk, skb); if (unlikely(net_xmit_eval(err))) @@ -135,6 +135,14 @@ struct sk_buff *iptunnel_handle_offloads(struct sk_buff *skb, return skb; } + /* If packet is not gso and we are resolving any partial checksum, + * clear encapsulation flag. This allows setting CHECKSUM_PARTIAL + * on the outer header without confusing devices that implement + * NETIF_F_IP_CSUM with encapsulation. + */ + if (csum_help) + skb->encapsulation = 0; + if (skb->ip_summed == CHECKSUM_PARTIAL && csum_help) { err = skb_checksum_help(skb); if (unlikely(err)) diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c index afcee51b90e..13ef00f1e17 100644 --- a/net/ipv4/ip_vti.c +++ b/net/ipv4/ip_vti.c @@ -239,6 +239,7 @@ static netdev_tx_t vti_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) static int vti4_err(struct sk_buff *skb, u32 info) { __be32 spi; + __u32 mark; struct xfrm_state *x; struct ip_tunnel *tunnel; struct ip_esp_hdr *esph; @@ -254,6 +255,8 @@ static int vti4_err(struct sk_buff *skb, u32 info) if (!tunnel) return -1; + mark = be32_to_cpu(tunnel->parms.o_key); + switch (protocol) { case IPPROTO_ESP: esph = (struct ip_esp_hdr *)(skb->data+(iph->ihl<<2)); @@ -281,7 +284,7 @@ static int vti4_err(struct sk_buff *skb, u32 info) return 0; } - x = xfrm_state_lookup(net, skb->mark, (const xfrm_address_t *)&iph->daddr, + x = xfrm_state_lookup(net, mark, (const xfrm_address_t *)&iph->daddr, spi, protocol, AF_INET); if (!x) return 0; diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index 812b1835146..4bc508f0db9 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -486,4 +486,5 @@ static void __exit ipip_fini(void) module_init(ipip_init); module_exit(ipip_fini); MODULE_LICENSE("GPL"); +MODULE_ALIAS_RTNL_LINK("ipip"); MODULE_ALIAS_NETDEV("tunl0"); diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index d84dc8d4c91..65bcaa78904 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -484,7 +484,7 @@ static void reg_vif_setup(struct net_device *dev) dev->type = ARPHRD_PIMREG; dev->mtu = ETH_DATA_LEN - sizeof(struct iphdr) - 8; dev->flags = IFF_NOARP; - dev->netdev_ops = ®_vif_netdev_ops, + dev->netdev_ops = ®_vif_netdev_ops; dev->destructor = free_netdev; dev->features |= NETIF_F_NETNS_LOCAL; } @@ -1663,7 +1663,7 @@ static void ip_encap(struct sk_buff *skb, __be32 saddr, __be32 daddr) iph->protocol = IPPROTO_IPIP; iph->ihl = 5; iph->tot_len = htons(skb->len); - ip_select_ident(skb, skb_dst(skb), NULL); + ip_select_ident(skb, NULL); ip_send_check(iph); memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); diff --git a/net/ipv4/netfilter/iptable_nat.c b/net/ipv4/netfilter/iptable_nat.c index ee2886126e3..f1787c04a4d 100644 --- a/net/ipv4/netfilter/iptable_nat.c +++ b/net/ipv4/netfilter/iptable_nat.c @@ -91,17 +91,9 @@ nf_nat_ipv4_fn(const struct nf_hook_ops *ops, if (nf_ct_is_untracked(ct)) return NF_ACCEPT; - nat = nfct_nat(ct); - if (!nat) { - /* NAT module was loaded late. */ - if (nf_ct_is_confirmed(ct)) - return NF_ACCEPT; - nat = nf_ct_ext_add(ct, NF_CT_EXT_NAT, GFP_ATOMIC); - if (nat == NULL) { - pr_debug("failed to add NAT extension\n"); - return NF_ACCEPT; - } - } + nat = nf_ct_nat_ext_add(ct); + if (nat == NULL) + return NF_ACCEPT; switch (ctinfo) { case IP_CT_RELATED: diff --git a/net/ipv4/netfilter/nf_defrag_ipv4.c b/net/ipv4/netfilter/nf_defrag_ipv4.c index 12e13bd82b5..b8f6381c7d0 100644 --- a/net/ipv4/netfilter/nf_defrag_ipv4.c +++ b/net/ipv4/netfilter/nf_defrag_ipv4.c @@ -22,7 +22,6 @@ #endif #include <net/netfilter/nf_conntrack_zones.h> -/* Returns new sk_buff, or NULL */ static int nf_ct_ipv4_gather_frags(struct sk_buff *skb, u_int32_t user) { int err; @@ -33,8 +32,10 @@ static int nf_ct_ipv4_gather_frags(struct sk_buff *skb, u_int32_t user) err = ip_defrag(skb, user); local_bh_enable(); - if (!err) + if (!err) { ip_send_check(ip_hdr(skb)); + skb->ignore_df = 1; + } return err; } diff --git a/net/ipv4/netfilter/nft_chain_nat_ipv4.c b/net/ipv4/netfilter/nft_chain_nat_ipv4.c index b5b256d45e6..3964157d826 100644 --- a/net/ipv4/netfilter/nft_chain_nat_ipv4.c +++ b/net/ipv4/netfilter/nft_chain_nat_ipv4.c @@ -48,15 +48,9 @@ static unsigned int nf_nat_fn(const struct nf_hook_ops *ops, NF_CT_ASSERT(!(ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET))); - nat = nfct_nat(ct); - if (nat == NULL) { - /* Conntrack module was loaded late, can't add extension. */ - if (nf_ct_is_confirmed(ct)) - return NF_ACCEPT; - nat = nf_ct_ext_add(ct, NF_CT_EXT_NAT, GFP_ATOMIC); - if (nat == NULL) - return NF_ACCEPT; - } + nat = nf_ct_nat_ext_add(ct); + if (nat == NULL) + return NF_ACCEPT; switch (ctinfo) { case IP_CT_RELATED: diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c index 8210964a9f1..044a0ddf6a7 100644 --- a/net/ipv4/ping.c +++ b/net/ipv4/ping.c @@ -236,15 +236,15 @@ exit: static void inet_get_ping_group_range_net(struct net *net, kgid_t *low, kgid_t *high) { - kgid_t *data = net->ipv4.sysctl_ping_group_range; + kgid_t *data = net->ipv4.ping_group_range.range; unsigned int seq; do { - seq = read_seqbegin(&net->ipv4.sysctl_local_ports.lock); + seq = read_seqbegin(&net->ipv4.ping_group_range.lock); *low = data[0]; *high = data[1]; - } while (read_seqretry(&net->ipv4.sysctl_local_ports.lock, seq)); + } while (read_seqretry(&net->ipv4.ping_group_range.lock, seq)); } diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index ad737fad6d8..ae0af9386f7 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -345,15 +345,15 @@ static void icmp_put(struct seq_file *seq) for (i = 0; icmpmibmap[i].name != NULL; i++) seq_printf(seq, " Out%s", icmpmibmap[i].name); seq_printf(seq, "\nIcmp: %lu %lu %lu", - snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_INMSGS), - snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_INERRORS), - snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_CSUMERRORS)); + snmp_fold_field(net->mib.icmp_statistics, ICMP_MIB_INMSGS), + snmp_fold_field(net->mib.icmp_statistics, ICMP_MIB_INERRORS), + snmp_fold_field(net->mib.icmp_statistics, ICMP_MIB_CSUMERRORS)); for (i = 0; icmpmibmap[i].name != NULL; i++) seq_printf(seq, " %lu", atomic_long_read(ptr + icmpmibmap[i].index)); seq_printf(seq, " %lu %lu", - snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_OUTMSGS), - snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_OUTERRORS)); + snmp_fold_field(net->mib.icmp_statistics, ICMP_MIB_OUTMSGS), + snmp_fold_field(net->mib.icmp_statistics, ICMP_MIB_OUTERRORS)); for (i = 0; icmpmibmap[i].name != NULL; i++) seq_printf(seq, " %lu", atomic_long_read(ptr + (icmpmibmap[i].index | 0x100))); @@ -379,7 +379,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v) BUILD_BUG_ON(offsetof(struct ipstats_mib, mibs) != 0); for (i = 0; snmp4_ipstats_list[i].name != NULL; i++) seq_printf(seq, " %llu", - snmp_fold_field64((void __percpu **)net->mib.ip_statistics, + snmp_fold_field64(net->mib.ip_statistics, snmp4_ipstats_list[i].entry, offsetof(struct ipstats_mib, syncp))); @@ -395,11 +395,11 @@ static int snmp_seq_show(struct seq_file *seq, void *v) /* MaxConn field is signed, RFC 2012 */ if (snmp4_tcp_list[i].entry == TCP_MIB_MAXCONN) seq_printf(seq, " %ld", - snmp_fold_field((void __percpu **)net->mib.tcp_statistics, + snmp_fold_field(net->mib.tcp_statistics, snmp4_tcp_list[i].entry)); else seq_printf(seq, " %lu", - snmp_fold_field((void __percpu **)net->mib.tcp_statistics, + snmp_fold_field(net->mib.tcp_statistics, snmp4_tcp_list[i].entry)); } @@ -410,7 +410,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v) seq_puts(seq, "\nUdp:"); for (i = 0; snmp4_udp_list[i].name != NULL; i++) seq_printf(seq, " %lu", - snmp_fold_field((void __percpu **)net->mib.udp_statistics, + snmp_fold_field(net->mib.udp_statistics, snmp4_udp_list[i].entry)); /* the UDP and UDP-Lite MIBs are the same */ @@ -421,7 +421,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v) seq_puts(seq, "\nUdpLite:"); for (i = 0; snmp4_udp_list[i].name != NULL; i++) seq_printf(seq, " %lu", - snmp_fold_field((void __percpu **)net->mib.udplite_statistics, + snmp_fold_field(net->mib.udplite_statistics, snmp4_udp_list[i].entry)); seq_putc(seq, '\n'); @@ -458,7 +458,7 @@ static int netstat_seq_show(struct seq_file *seq, void *v) seq_puts(seq, "\nTcpExt:"); for (i = 0; snmp4_net_list[i].name != NULL; i++) seq_printf(seq, " %lu", - snmp_fold_field((void __percpu **)net->mib.net_statistics, + snmp_fold_field(net->mib.net_statistics, snmp4_net_list[i].entry)); seq_puts(seq, "\nIpExt:"); @@ -468,7 +468,7 @@ static int netstat_seq_show(struct seq_file *seq, void *v) seq_puts(seq, "\nIpExt:"); for (i = 0; snmp4_ipextstats_list[i].name != NULL; i++) seq_printf(seq, " %llu", - snmp_fold_field64((void __percpu **)net->mib.ip_statistics, + snmp_fold_field64(net->mib.ip_statistics, snmp4_ipextstats_list[i].entry, offsetof(struct ipstats_mib, syncp))); diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index a9dbe58bdfe..2c65160565e 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -389,7 +389,7 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4, iph->check = 0; iph->tot_len = htons(length); if (!iph->id) - ip_select_ident(skb, &rt->dst, NULL); + ip_select_ident(skb, NULL); iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl); } diff --git a/net/ipv4/route.c b/net/ipv4/route.c index db1e0da871f..082239ffe34 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -89,6 +89,7 @@ #include <linux/rcupdate.h> #include <linux/times.h> #include <linux/slab.h> +#include <linux/jhash.h> #include <net/dst.h> #include <net/net_namespace.h> #include <net/protocol.h> @@ -456,39 +457,19 @@ static struct neighbour *ipv4_neigh_lookup(const struct dst_entry *dst, return neigh_create(&arp_tbl, pkey, dev); } -/* - * Peer allocation may fail only in serious out-of-memory conditions. However - * we still can generate some output. - * Random ID selection looks a bit dangerous because we have no chances to - * select ID being unique in a reasonable period of time. - * But broken packet identifier may be better than no packet at all. - */ -static void ip_select_fb_ident(struct iphdr *iph) -{ - static DEFINE_SPINLOCK(ip_fb_id_lock); - static u32 ip_fallback_id; - u32 salt; - - spin_lock_bh(&ip_fb_id_lock); - salt = secure_ip_id((__force __be32)ip_fallback_id ^ iph->daddr); - iph->id = htons(salt & 0xFFFF); - ip_fallback_id = salt; - spin_unlock_bh(&ip_fb_id_lock); -} +atomic_t *ip_idents __read_mostly; +EXPORT_SYMBOL(ip_idents); -void __ip_select_ident(struct iphdr *iph, struct dst_entry *dst, int more) +void __ip_select_ident(struct iphdr *iph, int segs) { - struct net *net = dev_net(dst->dev); - struct inet_peer *peer; + static u32 ip_idents_hashrnd __read_mostly; + u32 hash, id; - peer = inet_getpeer_v4(net->ipv4.peers, iph->daddr, 1); - if (peer) { - iph->id = htons(inet_getid(peer, more)); - inet_putpeer(peer); - return; - } + net_get_random_once(&ip_idents_hashrnd, sizeof(ip_idents_hashrnd)); - ip_select_fb_ident(iph); + hash = jhash_1word((__force u32)iph->daddr, ip_idents_hashrnd); + id = ip_idents_reserve(hash, segs); + iph->id = htons(id); } EXPORT_SYMBOL(__ip_select_ident); @@ -993,6 +974,9 @@ void ipv4_update_pmtu(struct sk_buff *skb, struct net *net, u32 mtu, struct flowi4 fl4; struct rtable *rt; + if (!mark) + mark = IP4_REPLY_MARK(net, skb->mark); + __build_flow_key(&fl4, NULL, iph, oif, RT_TOS(iph->tos), protocol, mark, flow_flags); rt = __ip_route_output_key(net, &fl4); @@ -1010,6 +994,10 @@ static void __ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu) struct rtable *rt; __build_flow_key(&fl4, sk, iph, 0, 0, 0, 0, 0); + + if (!fl4.flowi4_mark) + fl4.flowi4_mark = IP4_REPLY_MARK(sock_net(sk), skb->mark); + rt = __ip_route_output_key(sock_net(sk), &fl4); if (!IS_ERR(rt)) { __ip_rt_update_pmtu(rt, &fl4, mtu); @@ -1519,7 +1507,7 @@ static int __mkroute_input(struct sk_buff *skb, struct in_device *out_dev; unsigned int flags = 0; bool do_cache; - u32 itag; + u32 itag = 0; /* get a working reference to the output device */ out_dev = __in_dev_get_rcu(FIB_RES_DEV(*res)); @@ -2704,6 +2692,12 @@ int __init ip_rt_init(void) { int rc = 0; + ip_idents = kmalloc(IP_IDENTS_SZ * sizeof(*ip_idents), GFP_KERNEL); + if (!ip_idents) + panic("IP: failed to allocate ip_idents\n"); + + prandom_bytes(ip_idents, IP_IDENTS_SZ * sizeof(*ip_idents)); + #ifdef CONFIG_IP_ROUTE_CLASSID ip_rt_acct = __alloc_percpu(256 * sizeof(struct ip_rt_acct), __alignof__(struct ip_rt_acct)); if (!ip_rt_acct) diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index f2ed13c2125..c86624b36a6 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -303,6 +303,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, ireq->ir_rmt_port = th->source; ireq->ir_loc_addr = ip_hdr(skb)->daddr; ireq->ir_rmt_addr = ip_hdr(skb)->saddr; + ireq->ir_mark = inet_request_mark(sk, skb); ireq->ecn_ok = ecn_ok; ireq->snd_wscale = tcp_opt.snd_wscale; ireq->sack_ok = tcp_opt.sack_ok; @@ -339,7 +340,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, * hasn't changed since we received the original syn, but I see * no easy way to do this. */ - flowi4_init_output(&fl4, sk->sk_bound_dev_if, sk->sk_mark, + flowi4_init_output(&fl4, sk->sk_bound_dev_if, ireq->ir_mark, RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE, IPPROTO_TCP, inet_sk_flowi_flags(sk), (opt && opt->srr) ? opt->faddr : ireq->ir_rmt_addr, diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 44eba052b43..79a007c5255 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -45,10 +45,10 @@ static int ip_ping_group_range_max[] = { GID_T_MAX, GID_T_MAX }; /* Update system visible IP port range */ static void set_local_port_range(struct net *net, int range[2]) { - write_seqlock(&net->ipv4.sysctl_local_ports.lock); - net->ipv4.sysctl_local_ports.range[0] = range[0]; - net->ipv4.sysctl_local_ports.range[1] = range[1]; - write_sequnlock(&net->ipv4.sysctl_local_ports.lock); + write_seqlock(&net->ipv4.ip_local_ports.lock); + net->ipv4.ip_local_ports.range[0] = range[0]; + net->ipv4.ip_local_ports.range[1] = range[1]; + write_sequnlock(&net->ipv4.ip_local_ports.lock); } /* Validate changes from /proc interface. */ @@ -57,7 +57,7 @@ static int ipv4_local_port_range(struct ctl_table *table, int write, size_t *lenp, loff_t *ppos) { struct net *net = - container_of(table->data, struct net, ipv4.sysctl_local_ports.range); + container_of(table->data, struct net, ipv4.ip_local_ports.range); int ret; int range[2]; struct ctl_table tmp = { @@ -87,14 +87,14 @@ static void inet_get_ping_group_range_table(struct ctl_table *table, kgid_t *low { kgid_t *data = table->data; struct net *net = - container_of(table->data, struct net, ipv4.sysctl_ping_group_range); + container_of(table->data, struct net, ipv4.ping_group_range.range); unsigned int seq; do { - seq = read_seqbegin(&net->ipv4.sysctl_local_ports.lock); + seq = read_seqbegin(&net->ipv4.ip_local_ports.lock); *low = data[0]; *high = data[1]; - } while (read_seqretry(&net->ipv4.sysctl_local_ports.lock, seq)); + } while (read_seqretry(&net->ipv4.ip_local_ports.lock, seq)); } /* Update system visible IP port range */ @@ -102,11 +102,11 @@ static void set_ping_group_range(struct ctl_table *table, kgid_t low, kgid_t hig { kgid_t *data = table->data; struct net *net = - container_of(table->data, struct net, ipv4.sysctl_ping_group_range); - write_seqlock(&net->ipv4.sysctl_local_ports.lock); + container_of(table->data, struct net, ipv4.ping_group_range.range); + write_seqlock(&net->ipv4.ip_local_ports.lock); data[0] = low; data[1] = high; - write_sequnlock(&net->ipv4.sysctl_local_ports.lock); + write_sequnlock(&net->ipv4.ip_local_ports.lock); } /* Validate changes from /proc interface. */ @@ -437,13 +437,6 @@ static struct ctl_table ipv4_table[] = { .proc_handler = proc_dointvec }, { - .procname = "ip_local_reserved_ports", - .data = NULL, /* initialized in sysctl_ipv4_init */ - .maxlen = 65536, - .mode = 0644, - .proc_handler = proc_do_large_bitmap, - }, - { .procname = "igmp_max_memberships", .data = &sysctl_igmp_max_memberships, .maxlen = sizeof(int), @@ -805,7 +798,7 @@ static struct ctl_table ipv4_net_table[] = { }, { .procname = "ping_group_range", - .data = &init_net.ipv4.sysctl_ping_group_range, + .data = &init_net.ipv4.ping_group_range.range, .maxlen = sizeof(gid_t)*2, .mode = 0644, .proc_handler = ipv4_ping_group_range, @@ -819,12 +812,19 @@ static struct ctl_table ipv4_net_table[] = { }, { .procname = "ip_local_port_range", - .maxlen = sizeof(init_net.ipv4.sysctl_local_ports.range), - .data = &init_net.ipv4.sysctl_local_ports.range, + .maxlen = sizeof(init_net.ipv4.ip_local_ports.range), + .data = &init_net.ipv4.ip_local_ports.range, .mode = 0644, .proc_handler = ipv4_local_port_range, }, { + .procname = "ip_local_reserved_ports", + .data = &init_net.ipv4.sysctl_local_reserved_ports, + .maxlen = 65536, + .mode = 0644, + .proc_handler = proc_do_large_bitmap, + }, + { .procname = "ip_no_pmtu_disc", .data = &init_net.ipv4.sysctl_ip_no_pmtu_disc, .maxlen = sizeof(int), @@ -838,6 +838,20 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .procname = "fwmark_reflect", + .data = &init_net.ipv4.sysctl_fwmark_reflect, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .procname = "tcp_fwmark_accept", + .data = &init_net.ipv4.sysctl_tcp_fwmark_accept, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, { } }; @@ -858,26 +872,18 @@ static __net_init int ipv4_sysctl_init_net(struct net *net) table[i].data += (void *)net - (void *)&init_net; } - /* - * Sane defaults - nobody may create ping sockets. - * Boot scripts should set this to distro-specific group. - */ - net->ipv4.sysctl_ping_group_range[0] = make_kgid(&init_user_ns, 1); - net->ipv4.sysctl_ping_group_range[1] = make_kgid(&init_user_ns, 0); - - /* - * Set defaults for local port range - */ - seqlock_init(&net->ipv4.sysctl_local_ports.lock); - net->ipv4.sysctl_local_ports.range[0] = 32768; - net->ipv4.sysctl_local_ports.range[1] = 61000; - net->ipv4.ipv4_hdr = register_net_sysctl(net, "net/ipv4", table); if (net->ipv4.ipv4_hdr == NULL) goto err_reg; + net->ipv4.sysctl_local_reserved_ports = kzalloc(65536 / 8, GFP_KERNEL); + if (!net->ipv4.sysctl_local_reserved_ports) + goto err_ports; + return 0; +err_ports: + unregister_net_sysctl_table(net->ipv4.ipv4_hdr); err_reg: if (!net_eq(net, &init_net)) kfree(table); @@ -889,6 +895,7 @@ static __net_exit void ipv4_sysctl_exit_net(struct net *net) { struct ctl_table *table; + kfree(net->ipv4.sysctl_local_reserved_ports); table = net->ipv4.ipv4_hdr->ctl_table_arg; unregister_net_sysctl_table(net->ipv4.ipv4_hdr); kfree(table); @@ -902,16 +909,6 @@ static __net_initdata struct pernet_operations ipv4_sysctl_ops = { static __init int sysctl_ipv4_init(void) { struct ctl_table_header *hdr; - struct ctl_table *i; - - for (i = ipv4_table; i->procname; i++) { - if (strcmp(i->procname, "ip_local_reserved_ports") == 0) { - i->data = sysctl_local_reserved_ports; - break; - } - } - if (!i->procname) - return -EINVAL; hdr = register_net_sysctl(&init_net, "net/ipv4", ipv4_table); if (hdr == NULL) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 4bd6d52eeff..eb1dde37e67 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2916,6 +2916,14 @@ static int do_tcp_getsockopt(struct sock *sk, int level, case TCP_USER_TIMEOUT: val = jiffies_to_msecs(icsk->icsk_user_timeout); break; + + case TCP_FASTOPEN: + if (icsk->icsk_accept_queue.fastopenq != NULL) + val = icsk->icsk_accept_queue.fastopenq->max_qlen; + else + val = 0; + break; + case TCP_TIMESTAMP: val = tcp_time_stamp + tp->tsoffset; break; diff --git a/net/ipv4/tcp_bic.c b/net/ipv4/tcp_bic.c index 821846fb0a7..d5de69bc04f 100644 --- a/net/ipv4/tcp_bic.c +++ b/net/ipv4/tcp_bic.c @@ -140,13 +140,12 @@ static inline void bictcp_update(struct bictcp *ca, u32 cwnd) ca->cnt = 1; } -static void bictcp_cong_avoid(struct sock *sk, u32 ack, u32 acked, - u32 in_flight) +static void bictcp_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); struct bictcp *ca = inet_csk_ca(sk); - if (!tcp_is_cwnd_limited(sk, in_flight)) + if (!tcp_is_cwnd_limited(sk)) return; if (tp->snd_cwnd <= tp->snd_ssthresh) diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index 2b9464c93b8..7b09d8b49fa 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -276,26 +276,6 @@ int tcp_set_congestion_control(struct sock *sk, const char *name) return err; } -/* RFC2861 Check whether we are limited by application or congestion window - * This is the inverse of cwnd check in tcp_tso_should_defer - */ -bool tcp_is_cwnd_limited(const struct sock *sk, u32 in_flight) -{ - const struct tcp_sock *tp = tcp_sk(sk); - u32 left; - - if (in_flight >= tp->snd_cwnd) - return true; - - left = tp->snd_cwnd - in_flight; - if (sk_can_gso(sk) && - left * sysctl_tcp_tso_win_divisor < tp->snd_cwnd && - left < tp->xmit_size_goal_segs) - return true; - return left <= tcp_max_tso_deferred_mss(tp); -} -EXPORT_SYMBOL_GPL(tcp_is_cwnd_limited); - /* Slow start is used when congestion window is no greater than the slow start * threshold. We base on RFC2581 and also handle stretch ACKs properly. * We do not implement RFC3465 Appropriate Byte Counting (ABC) per se but @@ -337,11 +317,11 @@ EXPORT_SYMBOL_GPL(tcp_cong_avoid_ai); /* This is Jacobson's slow start and congestion avoidance. * SIGCOMM '88, p. 328. */ -void tcp_reno_cong_avoid(struct sock *sk, u32 ack, u32 acked, u32 in_flight) +void tcp_reno_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); - if (!tcp_is_cwnd_limited(sk, in_flight)) + if (!tcp_is_cwnd_limited(sk)) return; /* In "safe" area, increase. */ diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp_cubic.c index 8bf224516ba..a9bd8a4828a 100644 --- a/net/ipv4/tcp_cubic.c +++ b/net/ipv4/tcp_cubic.c @@ -304,13 +304,12 @@ static inline void bictcp_update(struct bictcp *ca, u32 cwnd) ca->cnt = 1; } -static void bictcp_cong_avoid(struct sock *sk, u32 ack, u32 acked, - u32 in_flight) +static void bictcp_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); struct bictcp *ca = inet_csk_ca(sk); - if (!tcp_is_cwnd_limited(sk, in_flight)) + if (!tcp_is_cwnd_limited(sk)) return; if (tp->snd_cwnd <= tp->snd_ssthresh) { @@ -409,7 +408,7 @@ static void bictcp_acked(struct sock *sk, u32 cnt, s32 rtt_us) ratio -= ca->delayed_ack >> ACK_RATIO_SHIFT; ratio += cnt; - ca->delayed_ack = min(ratio, ACK_RATIO_LIMIT); + ca->delayed_ack = clamp(ratio, 1U, ACK_RATIO_LIMIT); } /* Some calls are for duplicates without timetamps */ diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index f195d9316e5..62e48cf84e6 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -72,25 +72,224 @@ error: kfree(ctx); return err; } -/* Computes the fastopen cookie for the IP path. - * The path is a 128 bits long (pad with zeros for IPv4). - * - * The caller must check foc->len to determine if a valid cookie - * has been generated successfully. -*/ -void tcp_fastopen_cookie_gen(__be32 src, __be32 dst, - struct tcp_fastopen_cookie *foc) +static bool __tcp_fastopen_cookie_gen(const void *path, + struct tcp_fastopen_cookie *foc) { - __be32 path[4] = { src, dst, 0, 0 }; struct tcp_fastopen_context *ctx; + bool ok = false; tcp_fastopen_init_key_once(true); rcu_read_lock(); ctx = rcu_dereference(tcp_fastopen_ctx); if (ctx) { - crypto_cipher_encrypt_one(ctx->tfm, foc->val, (__u8 *)path); + crypto_cipher_encrypt_one(ctx->tfm, foc->val, path); foc->len = TCP_FASTOPEN_COOKIE_SIZE; + ok = true; } rcu_read_unlock(); + return ok; +} + +/* Generate the fastopen cookie by doing aes128 encryption on both + * the source and destination addresses. Pad 0s for IPv4 or IPv4-mapped-IPv6 + * addresses. For the longer IPv6 addresses use CBC-MAC. + * + * XXX (TFO) - refactor when TCP_FASTOPEN_COOKIE_SIZE != AES_BLOCK_SIZE. + */ +static bool tcp_fastopen_cookie_gen(struct request_sock *req, + struct sk_buff *syn, + struct tcp_fastopen_cookie *foc) +{ + if (req->rsk_ops->family == AF_INET) { + const struct iphdr *iph = ip_hdr(syn); + + __be32 path[4] = { iph->saddr, iph->daddr, 0, 0 }; + return __tcp_fastopen_cookie_gen(path, foc); + } + +#if IS_ENABLED(CONFIG_IPV6) + if (req->rsk_ops->family == AF_INET6) { + const struct ipv6hdr *ip6h = ipv6_hdr(syn); + struct tcp_fastopen_cookie tmp; + + if (__tcp_fastopen_cookie_gen(&ip6h->saddr, &tmp)) { + struct in6_addr *buf = (struct in6_addr *) tmp.val; + int i = 4; + + for (i = 0; i < 4; i++) + buf->s6_addr32[i] ^= ip6h->daddr.s6_addr32[i]; + return __tcp_fastopen_cookie_gen(buf, foc); + } + } +#endif + return false; +} + +static bool tcp_fastopen_create_child(struct sock *sk, + struct sk_buff *skb, + struct dst_entry *dst, + struct request_sock *req) +{ + struct tcp_sock *tp = tcp_sk(sk); + struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue; + struct sock *child; + + req->num_retrans = 0; + req->num_timeout = 0; + req->sk = NULL; + + child = inet_csk(sk)->icsk_af_ops->syn_recv_sock(sk, skb, req, NULL); + if (child == NULL) + return false; + + spin_lock(&queue->fastopenq->lock); + queue->fastopenq->qlen++; + spin_unlock(&queue->fastopenq->lock); + + /* Initialize the child socket. Have to fix some values to take + * into account the child is a Fast Open socket and is created + * only out of the bits carried in the SYN packet. + */ + tp = tcp_sk(child); + + tp->fastopen_rsk = req; + /* Do a hold on the listner sk so that if the listener is being + * closed, the child that has been accepted can live on and still + * access listen_lock. + */ + sock_hold(sk); + tcp_rsk(req)->listener = sk; + + /* RFC1323: The window in SYN & SYN/ACK segments is never + * scaled. So correct it appropriately. + */ + tp->snd_wnd = ntohs(tcp_hdr(skb)->window); + + /* Activate the retrans timer so that SYNACK can be retransmitted. + * The request socket is not added to the SYN table of the parent + * because it's been added to the accept queue directly. + */ + inet_csk_reset_xmit_timer(child, ICSK_TIME_RETRANS, + TCP_TIMEOUT_INIT, TCP_RTO_MAX); + + /* Add the child socket directly into the accept queue */ + inet_csk_reqsk_queue_add(sk, req, child); + + /* Now finish processing the fastopen child socket. */ + inet_csk(child)->icsk_af_ops->rebuild_header(child); + tcp_init_congestion_control(child); + tcp_mtup_init(child); + tcp_init_metrics(child); + tcp_init_buffer_space(child); + + /* Queue the data carried in the SYN packet. We need to first + * bump skb's refcnt because the caller will attempt to free it. + * + * XXX (TFO) - we honor a zero-payload TFO request for now, + * (any reason not to?) but no need to queue the skb since + * there is no data. How about SYN+FIN? + */ + if (TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq + 1) { + skb = skb_get(skb); + skb_dst_drop(skb); + __skb_pull(skb, tcp_hdr(skb)->doff * 4); + skb_set_owner_r(skb, child); + __skb_queue_tail(&child->sk_receive_queue, skb); + tp->syn_data_acked = 1; + } + tcp_rsk(req)->rcv_nxt = tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq; + sk->sk_data_ready(sk); + bh_unlock_sock(child); + sock_put(child); + WARN_ON(req->sk == NULL); + return true; +} +EXPORT_SYMBOL(tcp_fastopen_create_child); + +static bool tcp_fastopen_queue_check(struct sock *sk) +{ + struct fastopen_queue *fastopenq; + + /* Make sure the listener has enabled fastopen, and we don't + * exceed the max # of pending TFO requests allowed before trying + * to validating the cookie in order to avoid burning CPU cycles + * unnecessarily. + * + * XXX (TFO) - The implication of checking the max_qlen before + * processing a cookie request is that clients can't differentiate + * between qlen overflow causing Fast Open to be disabled + * temporarily vs a server not supporting Fast Open at all. + */ + fastopenq = inet_csk(sk)->icsk_accept_queue.fastopenq; + if (fastopenq == NULL || fastopenq->max_qlen == 0) + return false; + + if (fastopenq->qlen >= fastopenq->max_qlen) { + struct request_sock *req1; + spin_lock(&fastopenq->lock); + req1 = fastopenq->rskq_rst_head; + if ((req1 == NULL) || time_after(req1->expires, jiffies)) { + spin_unlock(&fastopenq->lock); + NET_INC_STATS_BH(sock_net(sk), + LINUX_MIB_TCPFASTOPENLISTENOVERFLOW); + return false; + } + fastopenq->rskq_rst_head = req1->dl_next; + fastopenq->qlen--; + spin_unlock(&fastopenq->lock); + reqsk_free(req1); + } + return true; +} + +/* Returns true if we should perform Fast Open on the SYN. The cookie (foc) + * may be updated and return the client in the SYN-ACK later. E.g., Fast Open + * cookie request (foc->len == 0). + */ +bool tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, + struct request_sock *req, + struct tcp_fastopen_cookie *foc, + struct dst_entry *dst) +{ + struct tcp_fastopen_cookie valid_foc = { .len = -1 }; + bool syn_data = TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq + 1; + + if (!((sysctl_tcp_fastopen & TFO_SERVER_ENABLE) && + (syn_data || foc->len >= 0) && + tcp_fastopen_queue_check(sk))) { + foc->len = -1; + return false; + } + + if (syn_data && (sysctl_tcp_fastopen & TFO_SERVER_COOKIE_NOT_REQD)) + goto fastopen; + + if (tcp_fastopen_cookie_gen(req, skb, &valid_foc) && + foc->len == TCP_FASTOPEN_COOKIE_SIZE && + foc->len == valid_foc.len && + !memcmp(foc->val, valid_foc.val, foc->len)) { + /* Cookie is valid. Create a (full) child socket to accept + * the data in SYN before returning a SYN-ACK to ack the + * data. If we fail to create the socket, fall back and + * ack the ISN only but includes the same cookie. + * + * Note: Data-less SYN with valid cookie is allowed to send + * data in SYN_RECV state. + */ +fastopen: + if (tcp_fastopen_create_child(sk, skb, dst, req)) { + foc->len = -1; + NET_INC_STATS_BH(sock_net(sk), + LINUX_MIB_TCPFASTOPENPASSIVE); + return true; + } + } + + NET_INC_STATS_BH(sock_net(sk), foc->len ? + LINUX_MIB_TCPFASTOPENPASSIVEFAIL : + LINUX_MIB_TCPFASTOPENCOOKIEREQD); + *foc = valid_foc; + return false; } +EXPORT_SYMBOL(tcp_try_fastopen); diff --git a/net/ipv4/tcp_highspeed.c b/net/ipv4/tcp_highspeed.c index 8b9e7bad77c..1c4908280d9 100644 --- a/net/ipv4/tcp_highspeed.c +++ b/net/ipv4/tcp_highspeed.c @@ -109,12 +109,12 @@ static void hstcp_init(struct sock *sk) tp->snd_cwnd_clamp = min_t(u32, tp->snd_cwnd_clamp, 0xffffffff/128); } -static void hstcp_cong_avoid(struct sock *sk, u32 ack, u32 acked, u32 in_flight) +static void hstcp_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); struct hstcp *ca = inet_csk_ca(sk); - if (!tcp_is_cwnd_limited(sk, in_flight)) + if (!tcp_is_cwnd_limited(sk)) return; if (tp->snd_cwnd <= tp->snd_ssthresh) diff --git a/net/ipv4/tcp_htcp.c b/net/ipv4/tcp_htcp.c index 4a194acfd92..031361311a8 100644 --- a/net/ipv4/tcp_htcp.c +++ b/net/ipv4/tcp_htcp.c @@ -227,12 +227,12 @@ static u32 htcp_recalc_ssthresh(struct sock *sk) return max((tp->snd_cwnd * ca->beta) >> 7, 2U); } -static void htcp_cong_avoid(struct sock *sk, u32 ack, u32 acked, u32 in_flight) +static void htcp_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); struct htcp *ca = inet_csk_ca(sk); - if (!tcp_is_cwnd_limited(sk, in_flight)) + if (!tcp_is_cwnd_limited(sk)) return; if (tp->snd_cwnd <= tp->snd_ssthresh) diff --git a/net/ipv4/tcp_hybla.c b/net/ipv4/tcp_hybla.c index a15a799bf76..d8f8f05a495 100644 --- a/net/ipv4/tcp_hybla.c +++ b/net/ipv4/tcp_hybla.c @@ -87,8 +87,7 @@ static inline u32 hybla_fraction(u32 odds) * o Give cwnd a new value based on the model proposed * o remember increments <1 */ -static void hybla_cong_avoid(struct sock *sk, u32 ack, u32 acked, - u32 in_flight) +static void hybla_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); struct hybla *ca = inet_csk_ca(sk); @@ -101,11 +100,11 @@ static void hybla_cong_avoid(struct sock *sk, u32 ack, u32 acked, ca->minrtt_us = tp->srtt_us; } - if (!tcp_is_cwnd_limited(sk, in_flight)) + if (!tcp_is_cwnd_limited(sk)) return; if (!ca->hybla_en) { - tcp_reno_cong_avoid(sk, ack, acked, in_flight); + tcp_reno_cong_avoid(sk, ack, acked); return; } diff --git a/net/ipv4/tcp_illinois.c b/net/ipv4/tcp_illinois.c index 863d105e301..5999b3972e6 100644 --- a/net/ipv4/tcp_illinois.c +++ b/net/ipv4/tcp_illinois.c @@ -255,8 +255,7 @@ static void tcp_illinois_state(struct sock *sk, u8 new_state) /* * Increase window in response to successful acknowledgment. */ -static void tcp_illinois_cong_avoid(struct sock *sk, u32 ack, u32 acked, - u32 in_flight) +static void tcp_illinois_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); struct illinois *ca = inet_csk_ca(sk); @@ -265,7 +264,7 @@ static void tcp_illinois_cong_avoid(struct sock *sk, u32 ack, u32 acked, update_params(sk); /* RFC2861 only increase cwnd if fully utilized */ - if (!tcp_is_cwnd_limited(sk, in_flight)) + if (!tcp_is_cwnd_limited(sk)) return; /* In slow start */ diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index d6b46eb2f94..931529d5daa 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -2684,13 +2684,12 @@ static void tcp_process_loss(struct sock *sk, int flag, bool is_dupack) bool recovered = !before(tp->snd_una, tp->high_seq); if (tp->frto) { /* F-RTO RFC5682 sec 3.1 (sack enhanced version). */ - if (flag & FLAG_ORIG_SACK_ACKED) { - /* Step 3.b. A timeout is spurious if not all data are - * lost, i.e., never-retransmitted data are (s)acked. - */ - tcp_try_undo_loss(sk, true); + /* Step 3.b. A timeout is spurious if not all data are + * lost, i.e., never-retransmitted data are (s)acked. + */ + if (tcp_try_undo_loss(sk, flag & FLAG_ORIG_SACK_ACKED)) return; - } + if (after(tp->snd_nxt, tp->high_seq) && (flag & FLAG_DATA_SACKED || is_dupack)) { tp->frto = 0; /* Loss was real: 2nd part of step 3.a */ @@ -2938,10 +2937,11 @@ static void tcp_synack_rtt_meas(struct sock *sk, const u32 synack_stamp) tcp_ack_update_rtt(sk, FLAG_SYN_ACKED, seq_rtt_us, -1L); } -static void tcp_cong_avoid(struct sock *sk, u32 ack, u32 acked, u32 in_flight) +static void tcp_cong_avoid(struct sock *sk, u32 ack, u32 acked) { const struct inet_connection_sock *icsk = inet_csk(sk); - icsk->icsk_ca_ops->cong_avoid(sk, ack, acked, in_flight); + + icsk->icsk_ca_ops->cong_avoid(sk, ack, acked); tcp_sk(sk)->snd_cwnd_stamp = tcp_time_stamp; } @@ -3364,7 +3364,6 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) u32 ack_seq = TCP_SKB_CB(skb)->seq; u32 ack = TCP_SKB_CB(skb)->ack_seq; bool is_dupack = false; - u32 prior_in_flight; u32 prior_fackets; int prior_packets = tp->packets_out; const int prior_unsacked = tp->packets_out - tp->sacked_out; @@ -3397,7 +3396,6 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) flag |= FLAG_SND_UNA_ADVANCED; prior_fackets = tp->fackets_out; - prior_in_flight = tcp_packets_in_flight(tp); /* ts_recent update must be made after we are sure that the packet * is in window. @@ -3452,7 +3450,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) /* Advance cwnd if state allows */ if (tcp_may_raise_cwnd(sk, flag)) - tcp_cong_avoid(sk, ack, acked, prior_in_flight); + tcp_cong_avoid(sk, ack, acked); if (tcp_ack_is_dubious(sk, flag)) { is_dupack = !(flag & (FLAG_SND_UNA_ADVANCED | FLAG_NOT_DUP)); @@ -4703,28 +4701,6 @@ static int tcp_prune_queue(struct sock *sk) return -1; } -/* RFC2861, slow part. Adjust cwnd, after it was not full during one rto. - * As additional protections, we do not touch cwnd in retransmission phases, - * and if application hit its sndbuf limit recently. - */ -void tcp_cwnd_application_limited(struct sock *sk) -{ - struct tcp_sock *tp = tcp_sk(sk); - - if (inet_csk(sk)->icsk_ca_state == TCP_CA_Open && - sk->sk_socket && !test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) { - /* Limited by application or receiver window. */ - u32 init_win = tcp_init_cwnd(tp, __sk_dst_get(sk)); - u32 win_used = max(tp->snd_cwnd_used, init_win); - if (win_used < tp->snd_cwnd) { - tp->snd_ssthresh = tcp_current_ssthresh(sk); - tp->snd_cwnd = (tp->snd_cwnd + win_used) >> 1; - } - tp->snd_cwnd_used = 0; - } - tp->snd_cwnd_stamp = tcp_time_stamp; -} - static bool tcp_should_expand_sndbuf(const struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 438f3b95143..77cccda1ad0 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -336,8 +336,8 @@ void tcp_v4_err(struct sk_buff *icmp_skb, u32 info) const int code = icmp_hdr(icmp_skb)->code; struct sock *sk; struct sk_buff *skb; - struct request_sock *req; - __u32 seq; + struct request_sock *fastopen; + __u32 seq, snd_una; __u32 remaining; int err; struct net *net = dev_net(icmp_skb->dev); @@ -378,12 +378,12 @@ void tcp_v4_err(struct sk_buff *icmp_skb, u32 info) icsk = inet_csk(sk); tp = tcp_sk(sk); - req = tp->fastopen_rsk; seq = ntohl(th->seq); + /* XXX (TFO) - tp->snd_una should be ISN (tcp_create_openreq_child() */ + fastopen = tp->fastopen_rsk; + snd_una = fastopen ? tcp_rsk(fastopen)->snt_isn : tp->snd_una; if (sk->sk_state != TCP_LISTEN && - !between(seq, tp->snd_una, tp->snd_nxt) && - (req == NULL || seq != tcp_rsk(req)->snt_isn)) { - /* For a Fast Open socket, allow seq to be snt_isn. */ + !between(seq, snd_una, tp->snd_nxt)) { NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS); goto out; } @@ -426,11 +426,9 @@ void tcp_v4_err(struct sk_buff *icmp_skb, u32 info) if (code != ICMP_NET_UNREACH && code != ICMP_HOST_UNREACH) break; if (seq != tp->snd_una || !icsk->icsk_retransmits || - !icsk->icsk_backoff) + !icsk->icsk_backoff || fastopen) break; - /* XXX (TFO) - revisit the following logic for TFO */ - if (sock_owned_by_user(sk)) break; @@ -462,14 +460,6 @@ void tcp_v4_err(struct sk_buff *icmp_skb, u32 info) goto out; } - /* XXX (TFO) - if it's a TFO socket and has been accepted, rather - * than following the TCP_SYN_RECV case and closing the socket, - * we ignore the ICMP error and keep trying like a fully established - * socket. Is this the right thing to do? - */ - if (req && req->sk == NULL) - goto out; - switch (sk->sk_state) { struct request_sock *req, **prev; case TCP_LISTEN: @@ -502,10 +492,13 @@ void tcp_v4_err(struct sk_buff *icmp_skb, u32 info) goto out; case TCP_SYN_SENT: - case TCP_SYN_RECV: /* Cannot happen. - It can f.e. if SYNs crossed, - or Fast Open. - */ + case TCP_SYN_RECV: + /* Only in fast or simultaneous open. If a fast open socket is + * is already accepted it is treated as a connected one below. + */ + if (fastopen && fastopen->sk == NULL) + break; + if (!sock_owned_by_user(sk)) { sk->sk_err = err; @@ -822,7 +815,8 @@ static void tcp_v4_reqsk_send_ack(struct sock *sk, struct sk_buff *skb, */ static int tcp_v4_send_synack(struct sock *sk, struct dst_entry *dst, struct request_sock *req, - u16 queue_mapping) + u16 queue_mapping, + struct tcp_fastopen_cookie *foc) { const struct inet_request_sock *ireq = inet_rsk(req); struct flowi4 fl4; @@ -833,7 +827,7 @@ static int tcp_v4_send_synack(struct sock *sk, struct dst_entry *dst, if (!dst && (dst = inet_csk_route_req(sk, &fl4, req)) == NULL) return -1; - skb = tcp_make_synack(sk, dst, req, NULL); + skb = tcp_make_synack(sk, dst, req, foc); if (skb) { __tcp_v4_send_check(skb, ireq->ir_loc_addr, ireq->ir_rmt_addr); @@ -852,7 +846,7 @@ static int tcp_v4_send_synack(struct sock *sk, struct dst_entry *dst, static int tcp_v4_rtx_synack(struct sock *sk, struct request_sock *req) { - int res = tcp_v4_send_synack(sk, NULL, req, 0); + int res = tcp_v4_send_synack(sk, NULL, req, 0, NULL); if (!res) { TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_RETRANSSEGS); @@ -1260,187 +1254,6 @@ static const struct tcp_request_sock_ops tcp_request_sock_ipv4_ops = { }; #endif -static bool tcp_fastopen_check(struct sock *sk, struct sk_buff *skb, - struct request_sock *req, - struct tcp_fastopen_cookie *foc, - struct tcp_fastopen_cookie *valid_foc) -{ - bool skip_cookie = false; - struct fastopen_queue *fastopenq; - - if (likely(!fastopen_cookie_present(foc))) { - /* See include/net/tcp.h for the meaning of these knobs */ - if ((sysctl_tcp_fastopen & TFO_SERVER_ALWAYS) || - ((sysctl_tcp_fastopen & TFO_SERVER_COOKIE_NOT_REQD) && - (TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq + 1))) - skip_cookie = true; /* no cookie to validate */ - else - return false; - } - fastopenq = inet_csk(sk)->icsk_accept_queue.fastopenq; - /* A FO option is present; bump the counter. */ - NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPFASTOPENPASSIVE); - - /* Make sure the listener has enabled fastopen, and we don't - * exceed the max # of pending TFO requests allowed before trying - * to validating the cookie in order to avoid burning CPU cycles - * unnecessarily. - * - * XXX (TFO) - The implication of checking the max_qlen before - * processing a cookie request is that clients can't differentiate - * between qlen overflow causing Fast Open to be disabled - * temporarily vs a server not supporting Fast Open at all. - */ - if ((sysctl_tcp_fastopen & TFO_SERVER_ENABLE) == 0 || - fastopenq == NULL || fastopenq->max_qlen == 0) - return false; - - if (fastopenq->qlen >= fastopenq->max_qlen) { - struct request_sock *req1; - spin_lock(&fastopenq->lock); - req1 = fastopenq->rskq_rst_head; - if ((req1 == NULL) || time_after(req1->expires, jiffies)) { - spin_unlock(&fastopenq->lock); - NET_INC_STATS_BH(sock_net(sk), - LINUX_MIB_TCPFASTOPENLISTENOVERFLOW); - /* Avoid bumping LINUX_MIB_TCPFASTOPENPASSIVEFAIL*/ - foc->len = -1; - return false; - } - fastopenq->rskq_rst_head = req1->dl_next; - fastopenq->qlen--; - spin_unlock(&fastopenq->lock); - reqsk_free(req1); - } - if (skip_cookie) { - tcp_rsk(req)->rcv_nxt = TCP_SKB_CB(skb)->end_seq; - return true; - } - - if (foc->len == TCP_FASTOPEN_COOKIE_SIZE) { - if ((sysctl_tcp_fastopen & TFO_SERVER_COOKIE_NOT_CHKED) == 0) { - tcp_fastopen_cookie_gen(ip_hdr(skb)->saddr, - ip_hdr(skb)->daddr, valid_foc); - if ((valid_foc->len != TCP_FASTOPEN_COOKIE_SIZE) || - memcmp(&foc->val[0], &valid_foc->val[0], - TCP_FASTOPEN_COOKIE_SIZE) != 0) - return false; - valid_foc->len = -1; - } - /* Acknowledge the data received from the peer. */ - tcp_rsk(req)->rcv_nxt = TCP_SKB_CB(skb)->end_seq; - return true; - } else if (foc->len == 0) { /* Client requesting a cookie */ - tcp_fastopen_cookie_gen(ip_hdr(skb)->saddr, - ip_hdr(skb)->daddr, valid_foc); - NET_INC_STATS_BH(sock_net(sk), - LINUX_MIB_TCPFASTOPENCOOKIEREQD); - } else { - /* Client sent a cookie with wrong size. Treat it - * the same as invalid and return a valid one. - */ - tcp_fastopen_cookie_gen(ip_hdr(skb)->saddr, - ip_hdr(skb)->daddr, valid_foc); - } - return false; -} - -static int tcp_v4_conn_req_fastopen(struct sock *sk, - struct sk_buff *skb, - struct sk_buff *skb_synack, - struct request_sock *req) -{ - struct tcp_sock *tp = tcp_sk(sk); - struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue; - const struct inet_request_sock *ireq = inet_rsk(req); - struct sock *child; - int err; - - req->num_retrans = 0; - req->num_timeout = 0; - req->sk = NULL; - - child = inet_csk(sk)->icsk_af_ops->syn_recv_sock(sk, skb, req, NULL); - if (child == NULL) { - NET_INC_STATS_BH(sock_net(sk), - LINUX_MIB_TCPFASTOPENPASSIVEFAIL); - kfree_skb(skb_synack); - return -1; - } - err = ip_build_and_send_pkt(skb_synack, sk, ireq->ir_loc_addr, - ireq->ir_rmt_addr, ireq->opt); - err = net_xmit_eval(err); - if (!err) - tcp_rsk(req)->snt_synack = tcp_time_stamp; - /* XXX (TFO) - is it ok to ignore error and continue? */ - - spin_lock(&queue->fastopenq->lock); - queue->fastopenq->qlen++; - spin_unlock(&queue->fastopenq->lock); - - /* Initialize the child socket. Have to fix some values to take - * into account the child is a Fast Open socket and is created - * only out of the bits carried in the SYN packet. - */ - tp = tcp_sk(child); - - tp->fastopen_rsk = req; - /* Do a hold on the listner sk so that if the listener is being - * closed, the child that has been accepted can live on and still - * access listen_lock. - */ - sock_hold(sk); - tcp_rsk(req)->listener = sk; - - /* RFC1323: The window in SYN & SYN/ACK segments is never - * scaled. So correct it appropriately. - */ - tp->snd_wnd = ntohs(tcp_hdr(skb)->window); - - /* Activate the retrans timer so that SYNACK can be retransmitted. - * The request socket is not added to the SYN table of the parent - * because it's been added to the accept queue directly. - */ - inet_csk_reset_xmit_timer(child, ICSK_TIME_RETRANS, - TCP_TIMEOUT_INIT, TCP_RTO_MAX); - - /* Add the child socket directly into the accept queue */ - inet_csk_reqsk_queue_add(sk, req, child); - - /* Now finish processing the fastopen child socket. */ - inet_csk(child)->icsk_af_ops->rebuild_header(child); - tcp_init_congestion_control(child); - tcp_mtup_init(child); - tcp_init_metrics(child); - tcp_init_buffer_space(child); - - /* Queue the data carried in the SYN packet. We need to first - * bump skb's refcnt because the caller will attempt to free it. - * - * XXX (TFO) - we honor a zero-payload TFO request for now. - * (Any reason not to?) - */ - if (TCP_SKB_CB(skb)->end_seq == TCP_SKB_CB(skb)->seq + 1) { - /* Don't queue the skb if there is no payload in SYN. - * XXX (TFO) - How about SYN+FIN? - */ - tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq; - } else { - skb = skb_get(skb); - skb_dst_drop(skb); - __skb_pull(skb, tcp_hdr(skb)->doff * 4); - skb_set_owner_r(skb, child); - __skb_queue_tail(&child->sk_receive_queue, skb); - tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq; - tp->syn_data_acked = 1; - } - sk->sk_data_ready(sk); - bh_unlock_sock(child); - sock_put(child); - WARN_ON(req->sk == NULL); - return 0; -} - int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) { struct tcp_options_received tmp_opt; @@ -1451,12 +1264,10 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) __be32 saddr = ip_hdr(skb)->saddr; __be32 daddr = ip_hdr(skb)->daddr; __u32 isn = TCP_SKB_CB(skb)->when; - bool want_cookie = false; + bool want_cookie = false, fastopen; struct flowi4 fl4; struct tcp_fastopen_cookie foc = { .len = -1 }; - struct tcp_fastopen_cookie valid_foc = { .len = -1 }; - struct sk_buff *skb_synack; - int do_fastopen; + int err; /* Never answer to SYNs send to broadcast or multicast */ if (skb_rtable(skb)->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST)) @@ -1507,6 +1318,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) ireq->ir_rmt_addr = saddr; ireq->no_srccheck = inet_sk(sk)->transparent; ireq->opt = tcp_v4_save_options(skb); + ireq->ir_mark = inet_request_mark(sk, skb); if (security_inet_conn_request(sk, skb, req)) goto drop_and_free; @@ -1555,52 +1367,24 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) isn = tcp_v4_init_sequence(skb); } - tcp_rsk(req)->snt_isn = isn; - - if (dst == NULL) { - dst = inet_csk_route_req(sk, &fl4, req); - if (dst == NULL) - goto drop_and_free; - } - do_fastopen = tcp_fastopen_check(sk, skb, req, &foc, &valid_foc); - - /* We don't call tcp_v4_send_synack() directly because we need - * to make sure a child socket can be created successfully before - * sending back synack! - * - * XXX (TFO) - Ideally one would simply call tcp_v4_send_synack() - * (or better yet, call tcp_send_synack() in the child context - * directly, but will have to fix bunch of other code first) - * after syn_recv_sock() except one will need to first fix the - * latter to remove its dependency on the current implementation - * of tcp_v4_send_synack()->tcp_select_initial_window(). - */ - skb_synack = tcp_make_synack(sk, dst, req, - fastopen_cookie_present(&valid_foc) ? &valid_foc : NULL); - - if (skb_synack) { - __tcp_v4_send_check(skb_synack, ireq->ir_loc_addr, ireq->ir_rmt_addr); - skb_set_queue_mapping(skb_synack, skb_get_queue_mapping(skb)); - } else + if (!dst && (dst = inet_csk_route_req(sk, &fl4, req)) == NULL) goto drop_and_free; - if (likely(!do_fastopen)) { - int err; - err = ip_build_and_send_pkt(skb_synack, sk, ireq->ir_loc_addr, - ireq->ir_rmt_addr, ireq->opt); - err = net_xmit_eval(err); + tcp_rsk(req)->snt_isn = isn; + tcp_rsk(req)->snt_synack = tcp_time_stamp; + tcp_openreq_init_rwin(req, sk, dst); + fastopen = !want_cookie && + tcp_try_fastopen(sk, skb, req, &foc, dst); + err = tcp_v4_send_synack(sk, dst, req, + skb_get_queue_mapping(skb), &foc); + if (!fastopen) { if (err || want_cookie) goto drop_and_free; tcp_rsk(req)->snt_synack = tcp_time_stamp; tcp_rsk(req)->listener = NULL; - /* Add the request_sock to the SYN table */ inet_csk_reqsk_queue_hash_add(sk, req, TCP_TIMEOUT_INIT); - if (fastopen_cookie_present(&foc) && foc.len != 0) - NET_INC_STATS_BH(sock_net(sk), - LINUX_MIB_TCPFASTOPENPASSIVEFAIL); - } else if (tcp_v4_conn_req_fastopen(sk, skb, skb_synack, req)) - goto drop_and_free; + } return 0; @@ -1744,28 +1528,6 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) return sk; } -static __sum16 tcp_v4_checksum_init(struct sk_buff *skb) -{ - const struct iphdr *iph = ip_hdr(skb); - - if (skb->ip_summed == CHECKSUM_COMPLETE) { - if (!tcp_v4_check(skb->len, iph->saddr, - iph->daddr, skb->csum)) { - skb->ip_summed = CHECKSUM_UNNECESSARY; - return 0; - } - } - - skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr, - skb->len, IPPROTO_TCP, 0); - - if (skb->len <= 76) { - return __skb_checksum_complete(skb); - } - return 0; -} - - /* The socket must have it's spinlock held when we get * here. * @@ -1960,7 +1722,8 @@ int tcp_v4_rcv(struct sk_buff *skb) * Packet length and doff are validated by header prediction, * provided case of th->doff==0 is eliminated. * So, we defer the checks. */ - if (!skb_csum_unnecessary(skb) && tcp_v4_checksum_init(skb)) + + if (skb_checksum_init(skb, IPPROTO_TCP, inet_compute_pseudo)) goto csum_error; th = tcp_hdr(skb); diff --git a/net/ipv4/tcp_lp.c b/net/ipv4/tcp_lp.c index c9aecae3132..1e70fa8fa79 100644 --- a/net/ipv4/tcp_lp.c +++ b/net/ipv4/tcp_lp.c @@ -115,13 +115,12 @@ static void tcp_lp_init(struct sock *sk) * Will only call newReno CA when away from inference. * From TCP-LP's paper, this will be handled in additive increasement. */ -static void tcp_lp_cong_avoid(struct sock *sk, u32 ack, u32 acked, - u32 in_flight) +static void tcp_lp_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct lp *lp = inet_csk_ca(sk); if (!(lp->flag & LP_WITHIN_INF)) - tcp_reno_cong_avoid(sk, ack, acked, in_flight); + tcp_reno_cong_avoid(sk, ack, acked); } /** diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c index dcaf72f1021..4fe04180598 100644 --- a/net/ipv4/tcp_metrics.c +++ b/net/ipv4/tcp_metrics.c @@ -1159,10 +1159,7 @@ static void __net_exit tcp_net_metrics_exit(struct net *net) tm = next; } } - if (is_vmalloc_addr(net->ipv4.tcp_metrics_hash)) - vfree(net->ipv4.tcp_metrics_hash); - else - kfree(net->ipv4.tcp_metrics_hash); + kvfree(net->ipv4.tcp_metrics_hash); } static __net_initdata struct pernet_operations tcp_net_metrics_ops = { diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 05c1b155251..e68e0d4af6c 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -362,6 +362,37 @@ void tcp_twsk_destructor(struct sock *sk) } EXPORT_SYMBOL_GPL(tcp_twsk_destructor); +void tcp_openreq_init_rwin(struct request_sock *req, + struct sock *sk, struct dst_entry *dst) +{ + struct inet_request_sock *ireq = inet_rsk(req); + struct tcp_sock *tp = tcp_sk(sk); + __u8 rcv_wscale; + int mss = dst_metric_advmss(dst); + + if (tp->rx_opt.user_mss && tp->rx_opt.user_mss < mss) + mss = tp->rx_opt.user_mss; + + /* Set this up on the first call only */ + req->window_clamp = tp->window_clamp ? : dst_metric(dst, RTAX_WINDOW); + + /* limit the window selection if the user enforce a smaller rx buffer */ + if (sk->sk_userlocks & SOCK_RCVBUF_LOCK && + (req->window_clamp > tcp_full_space(sk) || req->window_clamp == 0)) + req->window_clamp = tcp_full_space(sk); + + /* tcp_full_space because it is guaranteed to be the first packet */ + tcp_select_initial_window(tcp_full_space(sk), + mss - (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0), + &req->rcv_wnd, + &req->window_clamp, + ireq->wscale_ok, + &rcv_wscale, + dst_metric(dst, RTAX_INITRWND)); + ireq->rcv_wscale = rcv_wscale; +} +EXPORT_SYMBOL(tcp_openreq_init_rwin); + static inline void TCP_ECN_openreq_child(struct tcp_sock *tp, struct request_sock *req) { diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c index b92b81718ca..4e86c59ec7f 100644 --- a/net/ipv4/tcp_offload.c +++ b/net/ipv4/tcp_offload.c @@ -57,10 +57,12 @@ struct sk_buff *tcp_gso_segment(struct sk_buff *skb, SKB_GSO_TCP_ECN | SKB_GSO_TCPV6 | SKB_GSO_GRE | + SKB_GSO_GRE_CSUM | SKB_GSO_IPIP | SKB_GSO_SIT | SKB_GSO_MPLS | SKB_GSO_UDP_TUNNEL | + SKB_GSO_UDP_TUNNEL_CSUM | 0) || !(type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6)))) goto out; @@ -97,9 +99,7 @@ struct sk_buff *tcp_gso_segment(struct sk_buff *skb, th->check = newcheck; if (skb->ip_summed != CHECKSUM_PARTIAL) - th->check = - csum_fold(csum_partial(skb_transport_header(skb), - thlen, skb->csum)); + th->check = gso_make_checksum(skb, ~th->check); seq += mss; if (copy_destructor) { @@ -133,8 +133,7 @@ struct sk_buff *tcp_gso_segment(struct sk_buff *skb, th->check = ~csum_fold((__force __wsum)((__force u32)th->check + (__force u32)delta)); if (skb->ip_summed != CHECKSUM_PARTIAL) - th->check = csum_fold(csum_partial(skb_transport_header(skb), - thlen, skb->csum)); + th->check = gso_make_checksum(skb, ~th->check); out: return segs; } diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 025e2509398..d463c35db33 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -627,7 +627,7 @@ static unsigned int tcp_synack_options(struct sock *sk, if (unlikely(!ireq->tstamp_ok)) remaining -= TCPOLEN_SACKPERM_ALIGNED; } - if (foc != NULL) { + if (foc != NULL && foc->len >= 0) { u32 need = TCPOLEN_EXP_FASTOPEN_BASE + foc->len; need = (need + 3) & ~3U; /* Align to 32 bits */ if (remaining >= need) { @@ -878,15 +878,8 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, BUG_ON(!skb || !tcp_skb_pcount(skb)); if (clone_it) { - const struct sk_buff *fclone = skb + 1; - skb_mstamp_get(&skb->skb_mstamp); - if (unlikely(skb->fclone == SKB_FCLONE_ORIG && - fclone->fclone == SKB_FCLONE_CLONE)) - NET_INC_STATS(sock_net(sk), - LINUX_MIB_TCPSPURIOUS_RTX_HOSTQUEUES); - if (unlikely(skb_cloned(skb))) skb = pskb_copy(skb, gfp_mask); else @@ -1387,12 +1380,43 @@ unsigned int tcp_current_mss(struct sock *sk) return mss_now; } -/* Congestion window validation. (RFC2861) */ -static void tcp_cwnd_validate(struct sock *sk) +/* RFC2861, slow part. Adjust cwnd, after it was not full during one rto. + * As additional protections, we do not touch cwnd in retransmission phases, + * and if application hit its sndbuf limit recently. + */ +static void tcp_cwnd_application_limited(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); - if (tp->packets_out >= tp->snd_cwnd) { + if (inet_csk(sk)->icsk_ca_state == TCP_CA_Open && + sk->sk_socket && !test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) { + /* Limited by application or receiver window. */ + u32 init_win = tcp_init_cwnd(tp, __sk_dst_get(sk)); + u32 win_used = max(tp->snd_cwnd_used, init_win); + if (win_used < tp->snd_cwnd) { + tp->snd_ssthresh = tcp_current_ssthresh(sk); + tp->snd_cwnd = (tp->snd_cwnd + win_used) >> 1; + } + tp->snd_cwnd_used = 0; + } + tp->snd_cwnd_stamp = tcp_time_stamp; +} + +static void tcp_cwnd_validate(struct sock *sk, bool is_cwnd_limited) +{ + struct tcp_sock *tp = tcp_sk(sk); + + /* Track the maximum number of outstanding packets in each + * window, and remember whether we were cwnd-limited then. + */ + if (!before(tp->snd_una, tp->max_packets_seq) || + tp->packets_out > tp->max_packets_out) { + tp->max_packets_out = tp->packets_out; + tp->max_packets_seq = tp->snd_nxt; + tp->is_cwnd_limited = is_cwnd_limited; + } + + if (tcp_is_cwnd_limited(sk)) { /* Network is feed fully. */ tp->snd_cwnd_used = 0; tp->snd_cwnd_stamp = tcp_time_stamp; @@ -1644,7 +1668,8 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len, * * This algorithm is from John Heffner. */ -static bool tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb) +static bool tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb, + bool *is_cwnd_limited) { struct tcp_sock *tp = tcp_sk(sk); const struct inet_connection_sock *icsk = inet_csk(sk); @@ -1708,6 +1733,9 @@ static bool tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb) if (!tp->tso_deferred) tp->tso_deferred = 1 | (jiffies << 1); + if (cong_win < send_win && cong_win < skb->len) + *is_cwnd_limited = true; + return true; send_now: @@ -1868,6 +1896,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, unsigned int tso_segs, sent_pkts; int cwnd_quota; int result; + bool is_cwnd_limited = false; sent_pkts = 0; @@ -1892,6 +1921,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, cwnd_quota = tcp_cwnd_test(tp, skb); if (!cwnd_quota) { + is_cwnd_limited = true; if (push_one == 2) /* Force out a loss probe pkt. */ cwnd_quota = 1; @@ -1908,7 +1938,8 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, nonagle : TCP_NAGLE_PUSH)))) break; } else { - if (!push_one && tcp_tso_should_defer(sk, skb)) + if (!push_one && + tcp_tso_should_defer(sk, skb, &is_cwnd_limited)) break; } @@ -1975,7 +2006,7 @@ repair: /* Send one loss probe per tail loss episode. */ if (push_one != 2) tcp_schedule_loss_probe(sk); - tcp_cwnd_validate(sk); + tcp_cwnd_validate(sk, is_cwnd_limited); return false; } return (push_one == 2) || (!tp->packets_out && tcp_send_head(sk)); @@ -2039,6 +2070,25 @@ bool tcp_schedule_loss_probe(struct sock *sk) return true; } +/* Thanks to skb fast clones, we can detect if a prior transmit of + * a packet is still in a qdisc or driver queue. + * In this case, there is very little point doing a retransmit ! + * Note: This is called from BH context only. + */ +static bool skb_still_in_host_queue(const struct sock *sk, + const struct sk_buff *skb) +{ + const struct sk_buff *fclone = skb + 1; + + if (unlikely(skb->fclone == SKB_FCLONE_ORIG && + fclone->fclone == SKB_FCLONE_CLONE)) { + NET_INC_STATS_BH(sock_net(sk), + LINUX_MIB_TCPSPURIOUS_RTX_HOSTQUEUES); + return true; + } + return false; +} + /* When probe timeout (PTO) fires, send a new segment if one exists, else * retransmit the last segment. */ @@ -2064,6 +2114,9 @@ void tcp_send_loss_probe(struct sock *sk) if (WARN_ON(!skb)) goto rearm_timer; + if (skb_still_in_host_queue(sk, skb)) + goto rearm_timer; + pcount = tcp_skb_pcount(skb); if (WARN_ON(!pcount)) goto rearm_timer; @@ -2385,6 +2438,9 @@ int __tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb) min(sk->sk_wmem_queued + (sk->sk_wmem_queued >> 2), sk->sk_sndbuf)) return -EAGAIN; + if (skb_still_in_host_queue(sk, skb)) + return -EBUSY; + if (before(TCP_SKB_CB(skb)->seq, tp->snd_una)) { if (before(TCP_SKB_CB(skb)->end_seq, tp->snd_una)) BUG(); @@ -2441,8 +2497,14 @@ int __tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb) err = tcp_transmit_skb(sk, skb, 1, GFP_ATOMIC); } - if (likely(!err)) + if (likely(!err)) { TCP_SKB_CB(skb)->sacked |= TCPCB_EVER_RETRANS; + /* Update global TCP statistics. */ + TCP_INC_STATS(sock_net(sk), TCP_MIB_RETRANSSEGS); + if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN) + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPSYNRETRANS); + tp->total_retrans++; + } return err; } @@ -2452,12 +2514,6 @@ int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb) int err = __tcp_retransmit_skb(sk, skb); if (err == 0) { - /* Update global TCP statistics. */ - TCP_INC_STATS(sock_net(sk), TCP_MIB_RETRANSSEGS); - if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN) - NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPSYNRETRANS); - tp->total_retrans++; - #if FASTRETRANS_DEBUG > 0 if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS) { net_dbg_ratelimited("retrans_out leaked\n"); @@ -2478,7 +2534,7 @@ int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb) * see tcp_input.c tcp_sacktag_write_queue(). */ TCP_SKB_CB(skb)->ack_seq = tp->snd_nxt; - } else { + } else if (err != -EBUSY) { NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPRETRANSFAIL); } return err; @@ -2756,27 +2812,6 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst, if (tp->rx_opt.user_mss && tp->rx_opt.user_mss < mss) mss = tp->rx_opt.user_mss; - if (req->rcv_wnd == 0) { /* ignored for retransmitted syns */ - __u8 rcv_wscale; - /* Set this up on the first call only */ - req->window_clamp = tp->window_clamp ? : dst_metric(dst, RTAX_WINDOW); - - /* limit the window selection if the user enforce a smaller rx buffer */ - if (sk->sk_userlocks & SOCK_RCVBUF_LOCK && - (req->window_clamp > tcp_full_space(sk) || req->window_clamp == 0)) - req->window_clamp = tcp_full_space(sk); - - /* tcp_full_space because it is guaranteed to be the first packet */ - tcp_select_initial_window(tcp_full_space(sk), - mss - (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0), - &req->rcv_wnd, - &req->window_clamp, - ireq->wscale_ok, - &rcv_wscale, - dst_metric(dst, RTAX_INITRWND)); - ireq->rcv_wscale = rcv_wscale; - } - memset(&opts, 0, sizeof(opts)); #ifdef CONFIG_SYN_COOKIES if (unlikely(req->cookie_ts)) diff --git a/net/ipv4/tcp_scalable.c b/net/ipv4/tcp_scalable.c index 0ac50836da4..8250949b885 100644 --- a/net/ipv4/tcp_scalable.c +++ b/net/ipv4/tcp_scalable.c @@ -15,12 +15,11 @@ #define TCP_SCALABLE_AI_CNT 50U #define TCP_SCALABLE_MD_SCALE 3 -static void tcp_scalable_cong_avoid(struct sock *sk, u32 ack, u32 acked, - u32 in_flight) +static void tcp_scalable_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); - if (!tcp_is_cwnd_limited(sk, in_flight)) + if (!tcp_is_cwnd_limited(sk)) return; if (tp->snd_cwnd <= tp->snd_ssthresh) diff --git a/net/ipv4/tcp_vegas.c b/net/ipv4/tcp_vegas.c index 48539fff635..9a5e05f27f4 100644 --- a/net/ipv4/tcp_vegas.c +++ b/net/ipv4/tcp_vegas.c @@ -163,14 +163,13 @@ static inline u32 tcp_vegas_ssthresh(struct tcp_sock *tp) return min(tp->snd_ssthresh, tp->snd_cwnd-1); } -static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 acked, - u32 in_flight) +static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); struct vegas *vegas = inet_csk_ca(sk); if (!vegas->doing_vegas_now) { - tcp_reno_cong_avoid(sk, ack, acked, in_flight); + tcp_reno_cong_avoid(sk, ack, acked); return; } @@ -195,7 +194,7 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 acked, /* We don't have enough RTT samples to do the Vegas * calculation, so we'll behave like Reno. */ - tcp_reno_cong_avoid(sk, ack, acked, in_flight); + tcp_reno_cong_avoid(sk, ack, acked); } else { u32 rtt, diff; u64 target_cwnd; diff --git a/net/ipv4/tcp_veno.c b/net/ipv4/tcp_veno.c index 1b8e28fcd7e..27b9825753d 100644 --- a/net/ipv4/tcp_veno.c +++ b/net/ipv4/tcp_veno.c @@ -114,19 +114,18 @@ static void tcp_veno_cwnd_event(struct sock *sk, enum tcp_ca_event event) tcp_veno_init(sk); } -static void tcp_veno_cong_avoid(struct sock *sk, u32 ack, u32 acked, - u32 in_flight) +static void tcp_veno_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); struct veno *veno = inet_csk_ca(sk); if (!veno->doing_veno_now) { - tcp_reno_cong_avoid(sk, ack, acked, in_flight); + tcp_reno_cong_avoid(sk, ack, acked); return; } /* limited by applications */ - if (!tcp_is_cwnd_limited(sk, in_flight)) + if (!tcp_is_cwnd_limited(sk)) return; /* We do the Veno calculations only if we got enough rtt samples */ @@ -134,7 +133,7 @@ static void tcp_veno_cong_avoid(struct sock *sk, u32 ack, u32 acked, /* We don't have enough rtt samples to do the Veno * calculation, so we'll behave like Reno. */ - tcp_reno_cong_avoid(sk, ack, acked, in_flight); + tcp_reno_cong_avoid(sk, ack, acked); } else { u64 target_cwnd; u32 rtt; diff --git a/net/ipv4/tcp_yeah.c b/net/ipv4/tcp_yeah.c index 5ede0e72794..599b79b8eac 100644 --- a/net/ipv4/tcp_yeah.c +++ b/net/ipv4/tcp_yeah.c @@ -69,13 +69,12 @@ static void tcp_yeah_pkts_acked(struct sock *sk, u32 pkts_acked, s32 rtt_us) tcp_vegas_pkts_acked(sk, pkts_acked, rtt_us); } -static void tcp_yeah_cong_avoid(struct sock *sk, u32 ack, u32 acked, - u32 in_flight) +static void tcp_yeah_cong_avoid(struct sock *sk, u32 ack, u32 acked) { struct tcp_sock *tp = tcp_sk(sk); struct yeah *yeah = inet_csk_ca(sk); - if (!tcp_is_cwnd_limited(sk, in_flight)) + if (!tcp_is_cwnd_limited(sk)) return; if (tp->snd_cwnd <= tp->snd_ssthresh) diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 4468e1adc09..185ed3e5980 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -246,7 +246,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum, do { if (low <= snum && snum <= high && !test_bit(snum >> udptable->log, bitmap) && - !inet_is_reserved_local_port(snum)) + !inet_is_local_reserved_port(net, snum)) goto found; snum += rand; } while (snum != first); @@ -727,13 +727,12 @@ EXPORT_SYMBOL(udp_flush_pending_frames); void udp4_hwcsum(struct sk_buff *skb, __be32 src, __be32 dst) { struct udphdr *uh = udp_hdr(skb); - struct sk_buff *frags = skb_shinfo(skb)->frag_list; int offset = skb_transport_offset(skb); int len = skb->len - offset; int hlen = len; __wsum csum = 0; - if (!frags) { + if (!skb_has_frag_list(skb)) { /* * Only one fragment on the socket. */ @@ -742,15 +741,17 @@ void udp4_hwcsum(struct sk_buff *skb, __be32 src, __be32 dst) uh->check = ~csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, 0); } else { + struct sk_buff *frags; + /* * HW-checksum won't work as there are two or more * fragments on the socket so that all csums of sk_buffs * should be together */ - do { + skb_walk_frags(skb, frags) { csum = csum_add(csum, frags->csum); hlen -= frags->len; - } while ((frags = frags->next)); + } csum = skb_checksum(skb, offset, hlen, csum); skb->ip_summed = CHECKSUM_NONE; @@ -762,6 +763,43 @@ void udp4_hwcsum(struct sk_buff *skb, __be32 src, __be32 dst) } EXPORT_SYMBOL_GPL(udp4_hwcsum); +/* Function to set UDP checksum for an IPv4 UDP packet. This is intended + * for the simple case like when setting the checksum for a UDP tunnel. + */ +void udp_set_csum(bool nocheck, struct sk_buff *skb, + __be32 saddr, __be32 daddr, int len) +{ + struct udphdr *uh = udp_hdr(skb); + + if (nocheck) + uh->check = 0; + else if (skb_is_gso(skb)) + uh->check = ~udp_v4_check(len, saddr, daddr, 0); + else if (skb_dst(skb) && skb_dst(skb)->dev && + (skb_dst(skb)->dev->features & NETIF_F_V4_CSUM)) { + + BUG_ON(skb->ip_summed == CHECKSUM_PARTIAL); + + skb->ip_summed = CHECKSUM_PARTIAL; + skb->csum_start = skb_transport_header(skb) - skb->head; + skb->csum_offset = offsetof(struct udphdr, check); + uh->check = ~udp_v4_check(len, saddr, daddr, 0); + } else { + __wsum csum; + + BUG_ON(skb->ip_summed == CHECKSUM_PARTIAL); + + uh->check = 0; + csum = skb_checksum(skb, 0, len, 0); + uh->check = udp_v4_check(len, saddr, daddr, csum); + if (uh->check == 0) + uh->check = CSUM_MANGLED_0; + + skb->ip_summed = CHECKSUM_UNNECESSARY; + } +} +EXPORT_SYMBOL(udp_set_csum); + static int udp_send_skb(struct sk_buff *skb, struct flowi4 *fl4) { struct sock *sk = skb->sk; @@ -785,7 +823,7 @@ static int udp_send_skb(struct sk_buff *skb, struct flowi4 *fl4) if (is_udplite) /* UDP-Lite */ csum = udplite_csum(skb); - else if (sk->sk_no_check == UDP_CSUM_NOXMIT) { /* UDP csum disabled */ + else if (sk->sk_no_check_tx) { /* UDP csum disabled */ skb->ip_summed = CHECKSUM_NONE; goto send; @@ -1495,6 +1533,10 @@ int udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) if (skb->len > sizeof(struct udphdr) && encap_rcv != NULL) { int ret; + /* Verify checksum before giving to encap */ + if (udp_lib_checksum_complete(skb)) + goto csum_error; + ret = encap_rcv(sk, skb); if (ret <= 0) { UDP_INC_STATS_BH(sock_net(sk), @@ -1672,7 +1714,6 @@ static int __udp4_lib_mcast_deliver(struct net *net, struct sk_buff *skb, static inline int udp4_csum_init(struct sk_buff *skb, struct udphdr *uh, int proto) { - const struct iphdr *iph; int err; UDP_SKB_CB(skb)->partial_cov = 0; @@ -1684,22 +1725,8 @@ static inline int udp4_csum_init(struct sk_buff *skb, struct udphdr *uh, return err; } - iph = ip_hdr(skb); - if (uh->check == 0) { - skb->ip_summed = CHECKSUM_UNNECESSARY; - } else if (skb->ip_summed == CHECKSUM_COMPLETE) { - if (!csum_tcpudp_magic(iph->saddr, iph->daddr, skb->len, - proto, skb->csum)) - skb->ip_summed = CHECKSUM_UNNECESSARY; - } - if (!skb_csum_unnecessary(skb)) - skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr, - skb->len, proto, 0); - /* Probably, we should checksum udp header (it should be in cache - * in any case) and data in tiny packets (< rx copybreak). - */ - - return 0; + return skb_checksum_init_zero_check(skb, proto, uh->check, + inet_compute_pseudo); } /* @@ -1886,7 +1913,7 @@ static struct sock *__udp4_lib_demux_lookup(struct net *net, unsigned int hash2 = udp4_portaddr_hash(net, loc_addr, hnum); unsigned int slot2 = hash2 & udp_table.mask; struct udp_hslot *hslot2 = &udp_table.hash2[slot2]; - INET_ADDR_COOKIE(acookie, rmt_addr, loc_addr) + INET_ADDR_COOKIE(acookie, rmt_addr, loc_addr); const __portpair ports = INET_COMBINED_PORTS(rmt_port, hnum); rcu_read_lock(); @@ -1979,7 +2006,7 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, int (*push_pending_frames)(struct sock *)) { struct udp_sock *up = udp_sk(sk); - int val; + int val, valbool; int err = 0; int is_udplite = IS_UDPLITE(sk); @@ -1989,6 +2016,8 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, if (get_user(val, (int __user *)optval)) return -EFAULT; + valbool = val ? 1 : 0; + switch (optname) { case UDP_CORK: if (val != 0) { @@ -2018,6 +2047,14 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, } break; + case UDP_NO_CHECK6_TX: + up->no_check6_tx = valbool; + break; + + case UDP_NO_CHECK6_RX: + up->no_check6_rx = valbool; + break; + /* * UDP-Lite's partial checksum coverage (RFC 3828). */ @@ -2100,6 +2137,14 @@ int udp_lib_getsockopt(struct sock *sk, int level, int optname, val = up->encap_type; break; + case UDP_NO_CHECK6_TX: + val = up->no_check6_tx; + break; + + case UDP_NO_CHECK6_RX: + val = up->no_check6_rx; + break; + /* The following two cannot be changed on UDP sockets, the return is * always 0 (which corresponds to the full checksum coverage of UDP). */ case UDPLITE_SEND_CSCOV: @@ -2484,7 +2529,11 @@ struct sk_buff *skb_udp_tunnel_segment(struct sk_buff *skb, int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb); __be16 protocol = skb->protocol; netdev_features_t enc_features; - int outer_hlen; + int udp_offset, outer_hlen; + unsigned int oldlen; + bool need_csum; + + oldlen = (u16)~skb->len; if (unlikely(!pskb_may_pull(skb, tnl_hlen))) goto out; @@ -2496,6 +2545,10 @@ struct sk_buff *skb_udp_tunnel_segment(struct sk_buff *skb, skb->mac_len = skb_inner_network_offset(skb); skb->protocol = htons(ETH_P_TEB); + need_csum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM); + if (need_csum) + skb->encap_hdr_csum = 1; + /* segment inner packet. */ enc_features = skb->dev->hw_enc_features & netif_skb_features(skb); segs = skb_mac_gso_segment(skb, enc_features); @@ -2506,10 +2559,11 @@ struct sk_buff *skb_udp_tunnel_segment(struct sk_buff *skb, } outer_hlen = skb_tnl_header_len(skb); + udp_offset = outer_hlen - tnl_hlen; skb = segs; do { struct udphdr *uh; - int udp_offset = outer_hlen - tnl_hlen; + int len; skb_reset_inner_headers(skb); skb->encapsulation = 1; @@ -2520,31 +2574,20 @@ struct sk_buff *skb_udp_tunnel_segment(struct sk_buff *skb, skb_reset_mac_header(skb); skb_set_network_header(skb, mac_len); skb_set_transport_header(skb, udp_offset); + len = skb->len - udp_offset; uh = udp_hdr(skb); - uh->len = htons(skb->len - udp_offset); - - /* csum segment if tunnel sets skb with csum. */ - if (protocol == htons(ETH_P_IP) && unlikely(uh->check)) { - struct iphdr *iph = ip_hdr(skb); + uh->len = htons(len); - uh->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr, - skb->len - udp_offset, - IPPROTO_UDP, 0); - uh->check = csum_fold(skb_checksum(skb, udp_offset, - skb->len - udp_offset, 0)); - if (uh->check == 0) - uh->check = CSUM_MANGLED_0; + if (need_csum) { + __be32 delta = htonl(oldlen + len); - } else if (protocol == htons(ETH_P_IPV6)) { - struct ipv6hdr *ipv6h = ipv6_hdr(skb); - u32 len = skb->len - udp_offset; + uh->check = ~csum_fold((__force __wsum) + ((__force u32)uh->check + + (__force u32)delta)); + uh->check = gso_make_checksum(skb, ~uh->check); - uh->check = ~csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr, - len, IPPROTO_UDP, 0); - uh->check = csum_fold(skb_checksum(skb, udp_offset, len, 0)); if (uh->check == 0) uh->check = CSUM_MANGLED_0; - skb->ip_summed = CHECKSUM_NONE; } skb->protocol = protocol; diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index 88b4023ecfc..7b184011017 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -56,7 +56,8 @@ static struct sk_buff *udp4_ufo_fragment(struct sk_buff *skb, __wsum csum; if (skb->encapsulation && - skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL) { + (skb_shinfo(skb)->gso_type & + (SKB_GSO_UDP_TUNNEL|SKB_GSO_UDP_TUNNEL_CSUM))) { segs = skb_udp_tunnel_segment(skb, features); goto out; } @@ -71,8 +72,10 @@ static struct sk_buff *udp4_ufo_fragment(struct sk_buff *skb, if (unlikely(type & ~(SKB_GSO_UDP | SKB_GSO_DODGY | SKB_GSO_UDP_TUNNEL | + SKB_GSO_UDP_TUNNEL_CSUM | SKB_GSO_IPIP | - SKB_GSO_GRE | SKB_GSO_MPLS) || + SKB_GSO_GRE | SKB_GSO_GRE_CSUM | + SKB_GSO_MPLS) || !(type & (SKB_GSO_UDP)))) goto out; diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c index 2c46acd4cc3..3b3efbda48e 100644 --- a/net/ipv4/udplite.c +++ b/net/ipv4/udplite.c @@ -70,7 +70,6 @@ static struct inet_protosw udplite4_protosw = { .protocol = IPPROTO_UDPLITE, .prot = &udplite_prot, .ops = &inet_dgram_ops, - .no_check = 0, /* must checksum (RFC 3828) */ .flags = INET_PROTOSW_PERMANENT, }; diff --git a/net/ipv4/xfrm4_mode_tunnel.c b/net/ipv4/xfrm4_mode_tunnel.c index 05f2b484954..91771a7c802 100644 --- a/net/ipv4/xfrm4_mode_tunnel.c +++ b/net/ipv4/xfrm4_mode_tunnel.c @@ -58,12 +58,12 @@ static int xfrm4_mode_tunnel_output(struct xfrm_state *x, struct sk_buff *skb) top_iph->frag_off = (flags & XFRM_STATE_NOPMTUDISC) ? 0 : (XFRM_MODE_SKB_CB(skb)->frag_off & htons(IP_DF)); - ip_select_ident(skb, dst->child, NULL); top_iph->ttl = ip4_dst_hoplimit(dst->child); top_iph->saddr = x->props.saddr.a4; top_iph->daddr = x->id.daddr.a4; + ip_select_ident(skb, NULL); return 0; } diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c index 40e701f2e1e..d5f6bd9a210 100644 --- a/net/ipv4/xfrm4_output.c +++ b/net/ipv4/xfrm4_output.c @@ -25,7 +25,7 @@ static int xfrm4_tunnel_check_size(struct sk_buff *skb) if (IPCB(skb)->flags & IPSKB_XFRM_TUNNEL_SIZE) goto out; - if (!(ip_hdr(skb)->frag_off & htons(IP_DF)) || skb->local_df) + if (!(ip_hdr(skb)->frag_off & htons(IP_DF)) || skb->ignore_df) goto out; mtu = dst_mtu(skb_dst(skb)); @@ -62,10 +62,7 @@ int xfrm4_prepare_output(struct xfrm_state *x, struct sk_buff *skb) if (err) return err; - memset(IPCB(skb), 0, sizeof(*IPCB(skb))); - IPCB(skb)->flags |= IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED; - - skb->protocol = htons(ETH_P_IP); + IPCB(skb)->flags |= IPSKB_XFRM_TUNNEL_SIZE; return x->outer_mode->output2(x, skb); } @@ -73,27 +70,34 @@ EXPORT_SYMBOL(xfrm4_prepare_output); int xfrm4_output_finish(struct sk_buff *skb) { + memset(IPCB(skb), 0, sizeof(*IPCB(skb))); + skb->protocol = htons(ETH_P_IP); + +#ifdef CONFIG_NETFILTER + IPCB(skb)->flags |= IPSKB_XFRM_TRANSFORMED; +#endif + + return xfrm_output(skb); +} + +static int __xfrm4_output(struct sk_buff *skb) +{ + struct xfrm_state *x = skb_dst(skb)->xfrm; + #ifdef CONFIG_NETFILTER - if (!skb_dst(skb)->xfrm) { + if (!x) { IPCB(skb)->flags |= IPSKB_REROUTED; return dst_output(skb); } - - IPCB(skb)->flags |= IPSKB_XFRM_TRANSFORMED; #endif - skb->protocol = htons(ETH_P_IP); - return xfrm_output(skb); + return x->outer_mode->afinfo->output_finish(skb); } int xfrm4_output(struct sock *sk, struct sk_buff *skb) { - struct dst_entry *dst = skb_dst(skb); - struct xfrm_state *x = dst->xfrm; - return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, skb, - NULL, dst->dev, - x->outer_mode->afinfo->output_finish, + NULL, skb_dst(skb)->dev, __xfrm4_output, !(IPCB(skb)->flags & IPSKB_REROUTED)); } diff --git a/net/ipv4/xfrm4_protocol.c b/net/ipv4/xfrm4_protocol.c index 7f7b243e813..a2ce0101eaa 100644 --- a/net/ipv4/xfrm4_protocol.c +++ b/net/ipv4/xfrm4_protocol.c @@ -50,8 +50,12 @@ int xfrm4_rcv_cb(struct sk_buff *skb, u8 protocol, int err) { int ret; struct xfrm4_protocol *handler; + struct xfrm4_protocol __rcu **head = proto_handlers(protocol); - for_each_protocol_rcu(*proto_handlers(protocol), handler) + if (!head) + return 0; + + for_each_protocol_rcu(*head, handler) if ((ret = handler->cb_handler(skb, err)) <= 0) return ret; @@ -64,15 +68,20 @@ int xfrm4_rcv_encap(struct sk_buff *skb, int nexthdr, __be32 spi, { int ret; struct xfrm4_protocol *handler; + struct xfrm4_protocol __rcu **head = proto_handlers(nexthdr); XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4 = NULL; XFRM_SPI_SKB_CB(skb)->family = AF_INET; XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct iphdr, daddr); - for_each_protocol_rcu(*proto_handlers(nexthdr), handler) + if (!head) + goto out; + + for_each_protocol_rcu(*head, handler) if ((ret = handler->input_handler(skb, nexthdr, spi, encap_type)) != -EINVAL) return ret; +out: icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); kfree_skb(skb); @@ -208,6 +217,9 @@ int xfrm4_protocol_register(struct xfrm4_protocol *handler, int ret = -EEXIST; int priority = handler->priority; + if (!proto_handlers(protocol) || !netproto(protocol)) + return -EINVAL; + mutex_lock(&xfrm4_protocol_mutex); if (!rcu_dereference_protected(*proto_handlers(protocol), @@ -250,6 +262,9 @@ int xfrm4_protocol_deregister(struct xfrm4_protocol *handler, struct xfrm4_protocol *t; int ret = -ENOENT; + if (!proto_handlers(protocol) || !netproto(protocol)) + return -EINVAL; + mutex_lock(&xfrm4_protocol_mutex); for (pprev = proto_handlers(protocol); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 6c7fa0853fc..5667b3003af 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -275,19 +275,14 @@ static int snmp6_alloc_dev(struct inet6_dev *idev) { int i; - if (snmp_mib_init((void __percpu **)idev->stats.ipv6, - sizeof(struct ipstats_mib), - __alignof__(struct ipstats_mib)) < 0) + idev->stats.ipv6 = alloc_percpu(struct ipstats_mib); + if (!idev->stats.ipv6) goto err_ip; for_each_possible_cpu(i) { struct ipstats_mib *addrconf_stats; - addrconf_stats = per_cpu_ptr(idev->stats.ipv6[0], i); + addrconf_stats = per_cpu_ptr(idev->stats.ipv6, i); u64_stats_init(&addrconf_stats->syncp); -#if SNMP_ARRAY_SZ == 2 - addrconf_stats = per_cpu_ptr(idev->stats.ipv6[1], i); - u64_stats_init(&addrconf_stats->syncp); -#endif } @@ -305,7 +300,7 @@ static int snmp6_alloc_dev(struct inet6_dev *idev) err_icmpmsg: kfree(idev->stats.icmpv6dev); err_icmp: - snmp_mib_free((void __percpu **)idev->stats.ipv6); + free_percpu(idev->stats.ipv6); err_ip: return -ENOMEM; } @@ -2504,8 +2499,8 @@ static int inet6_addr_add(struct net *net, int ifindex, return PTR_ERR(ifp); } -static int inet6_addr_del(struct net *net, int ifindex, const struct in6_addr *pfx, - unsigned int plen) +static int inet6_addr_del(struct net *net, int ifindex, u32 ifa_flags, + const struct in6_addr *pfx, unsigned int plen) { struct inet6_ifaddr *ifp; struct inet6_dev *idev; @@ -2528,7 +2523,12 @@ static int inet6_addr_del(struct net *net, int ifindex, const struct in6_addr *p in6_ifa_hold(ifp); read_unlock_bh(&idev->lock); + if (!(ifp->flags & IFA_F_TEMPORARY) && + (ifa_flags & IFA_F_MANAGETEMPADDR)) + manage_tempaddrs(idev, ifp, 0, 0, false, + jiffies); ipv6_del_addr(ifp); + addrconf_verify_rtnl(); return 0; } } @@ -2568,7 +2568,7 @@ int addrconf_del_ifaddr(struct net *net, void __user *arg) return -EFAULT; rtnl_lock(); - err = inet6_addr_del(net, ireq.ifr6_ifindex, &ireq.ifr6_addr, + err = inet6_addr_del(net, ireq.ifr6_ifindex, 0, &ireq.ifr6_addr, ireq.ifr6_prefixlen); rtnl_unlock(); return err; @@ -2813,18 +2813,6 @@ static void addrconf_gre_config(struct net_device *dev) } #endif -static inline int -ipv6_inherit_linklocal(struct inet6_dev *idev, struct net_device *link_dev) -{ - struct in6_addr lladdr; - - if (!ipv6_get_lladdr(link_dev, &lladdr, IFA_F_TENTATIVE)) { - addrconf_add_linklocal(idev, &lladdr); - return 0; - } - return -1; -} - static int addrconf_notify(struct notifier_block *this, unsigned long event, void *ptr) { @@ -3743,6 +3731,7 @@ inet6_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh) struct ifaddrmsg *ifm; struct nlattr *tb[IFA_MAX+1]; struct in6_addr *pfx, *peer_pfx; + u32 ifa_flags; int err; err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFA_MAX, ifa_ipv6_policy); @@ -3754,7 +3743,13 @@ inet6_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh) if (pfx == NULL) return -EINVAL; - return inet6_addr_del(net, ifm->ifa_index, pfx, ifm->ifa_prefixlen); + ifa_flags = tb[IFA_FLAGS] ? nla_get_u32(tb[IFA_FLAGS]) : ifm->ifa_flags; + + /* We ignore other flags so far. */ + ifa_flags &= IFA_F_MANAGETEMPADDR; + + return inet6_addr_del(net, ifm->ifa_index, ifa_flags, pfx, + ifm->ifa_prefixlen); } static int inet6_addr_modify(struct inet6_ifaddr *ifp, u32 ifa_flags, @@ -4363,7 +4358,7 @@ static inline void __snmp6_fill_statsdev(u64 *stats, atomic_long_t *mib, memset(&stats[items], 0, pad); } -static inline void __snmp6_fill_stats64(u64 *stats, void __percpu **mib, +static inline void __snmp6_fill_stats64(u64 *stats, void __percpu *mib, int items, int bytes, size_t syncpoff) { int i; @@ -4383,7 +4378,7 @@ static void snmp6_fill_stats(u64 *stats, struct inet6_dev *idev, int attrtype, { switch (attrtype) { case IFLA_INET6_STATS: - __snmp6_fill_stats64(stats, (void __percpu **)idev->stats.ipv6, + __snmp6_fill_stats64(stats, idev->stats.ipv6, IPSTATS_MIB_MAX, bytes, offsetof(struct ipstats_mib, syncp)); break; case IFLA_INET6_ICMP6STATS: diff --git a/net/ipv6/addrconf_core.c b/net/ipv6/addrconf_core.c index 4c11cbcf830..e6960457f62 100644 --- a/net/ipv6/addrconf_core.c +++ b/net/ipv6/addrconf_core.c @@ -123,7 +123,7 @@ static void snmp6_free_dev(struct inet6_dev *idev) { kfree(idev->stats.icmpv6msgdev); kfree(idev->stats.icmpv6dev); - snmp_mib_free((void __percpu **)idev->stats.ipv6); + free_percpu(idev->stats.ipv6); } /* Nobody refers to this device, we may destroy it. */ diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index d935889f100..7cb4392690d 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -106,7 +106,6 @@ static int inet6_create(struct net *net, struct socket *sock, int protocol, struct inet_protosw *answer; struct proto *answer_prot; unsigned char answer_flags; - char answer_no_check; int try_loading_module = 0; int err; @@ -162,7 +161,6 @@ lookup_protocol: sock->ops = answer->ops; answer_prot = answer->prot; - answer_no_check = answer->no_check; answer_flags = answer->flags; rcu_read_unlock(); @@ -176,7 +174,6 @@ lookup_protocol: sock_init_data(sock, sk); err = 0; - sk->sk_no_check = answer_no_check; if (INET_PROTOSW_REUSE & answer_flags) sk->sk_reuse = SK_CAN_REUSE; @@ -715,33 +712,25 @@ static int __net_init ipv6_init_mibs(struct net *net) { int i; - if (snmp_mib_init((void __percpu **)net->mib.udp_stats_in6, - sizeof(struct udp_mib), - __alignof__(struct udp_mib)) < 0) + net->mib.udp_stats_in6 = alloc_percpu(struct udp_mib); + if (!net->mib.udp_stats_in6) return -ENOMEM; - if (snmp_mib_init((void __percpu **)net->mib.udplite_stats_in6, - sizeof(struct udp_mib), - __alignof__(struct udp_mib)) < 0) + net->mib.udplite_stats_in6 = alloc_percpu(struct udp_mib); + if (!net->mib.udplite_stats_in6) goto err_udplite_mib; - if (snmp_mib_init((void __percpu **)net->mib.ipv6_statistics, - sizeof(struct ipstats_mib), - __alignof__(struct ipstats_mib)) < 0) + net->mib.ipv6_statistics = alloc_percpu(struct ipstats_mib); + if (!net->mib.ipv6_statistics) goto err_ip_mib; for_each_possible_cpu(i) { struct ipstats_mib *af_inet6_stats; - af_inet6_stats = per_cpu_ptr(net->mib.ipv6_statistics[0], i); + af_inet6_stats = per_cpu_ptr(net->mib.ipv6_statistics, i); u64_stats_init(&af_inet6_stats->syncp); -#if SNMP_ARRAY_SZ == 2 - af_inet6_stats = per_cpu_ptr(net->mib.ipv6_statistics[1], i); - u64_stats_init(&af_inet6_stats->syncp); -#endif } - if (snmp_mib_init((void __percpu **)net->mib.icmpv6_statistics, - sizeof(struct icmpv6_mib), - __alignof__(struct icmpv6_mib)) < 0) + net->mib.icmpv6_statistics = alloc_percpu(struct icmpv6_mib); + if (!net->mib.icmpv6_statistics) goto err_icmp_mib; net->mib.icmpv6msg_statistics = kzalloc(sizeof(struct icmpv6msg_mib), GFP_KERNEL); @@ -750,22 +739,22 @@ static int __net_init ipv6_init_mibs(struct net *net) return 0; err_icmpmsg_mib: - snmp_mib_free((void __percpu **)net->mib.icmpv6_statistics); + free_percpu(net->mib.icmpv6_statistics); err_icmp_mib: - snmp_mib_free((void __percpu **)net->mib.ipv6_statistics); + free_percpu(net->mib.ipv6_statistics); err_ip_mib: - snmp_mib_free((void __percpu **)net->mib.udplite_stats_in6); + free_percpu(net->mib.udplite_stats_in6); err_udplite_mib: - snmp_mib_free((void __percpu **)net->mib.udp_stats_in6); + free_percpu(net->mib.udp_stats_in6); return -ENOMEM; } static void ipv6_cleanup_mibs(struct net *net) { - snmp_mib_free((void __percpu **)net->mib.udp_stats_in6); - snmp_mib_free((void __percpu **)net->mib.udplite_stats_in6); - snmp_mib_free((void __percpu **)net->mib.ipv6_statistics); - snmp_mib_free((void __percpu **)net->mib.icmpv6_statistics); + free_percpu(net->mib.udp_stats_in6); + free_percpu(net->mib.udplite_stats_in6); + free_percpu(net->mib.ipv6_statistics); + free_percpu(net->mib.icmpv6_statistics); kfree(net->mib.icmpv6msg_statistics); } diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 7b326529e6a..f6c84a6eb23 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -400,6 +400,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) int len; int hlimit; int err = 0; + u32 mark = IP6_REPLY_MARK(net, skb->mark); if ((u8 *)hdr < skb->head || (skb_network_header(skb) + sizeof(*hdr)) > skb_tail_pointer(skb)) @@ -466,6 +467,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) fl6.daddr = hdr->saddr; if (saddr) fl6.saddr = *saddr; + fl6.flowi6_mark = mark; fl6.flowi6_oif = iif; fl6.fl6_icmp_type = type; fl6.fl6_icmp_code = code; @@ -474,6 +476,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) sk = icmpv6_xmit_lock(net); if (sk == NULL) return; + sk->sk_mark = mark; np = inet6_sk(sk); if (!icmpv6_xrlim_allow(sk, type, &fl6)) @@ -493,12 +496,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) if (IS_ERR(dst)) goto out; - if (ipv6_addr_is_multicast(&fl6.daddr)) - hlimit = np->mcast_hops; - else - hlimit = np->hop_limit; - if (hlimit < 0) - hlimit = ip6_dst_hoplimit(dst); + hlimit = ip6_sk_dst_hoplimit(np, &fl6, dst); msg.skb = skb; msg.offset = skb_network_offset(skb); @@ -556,6 +554,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb) int err = 0; int hlimit; u8 tclass; + u32 mark = IP6_REPLY_MARK(net, skb->mark); saddr = &ipv6_hdr(skb)->daddr; @@ -574,11 +573,13 @@ static void icmpv6_echo_reply(struct sk_buff *skb) fl6.saddr = *saddr; fl6.flowi6_oif = skb->dev->ifindex; fl6.fl6_icmp_type = ICMPV6_ECHO_REPLY; + fl6.flowi6_mark = mark; security_skb_classify_flow(skb, flowi6_to_flowi(&fl6)); sk = icmpv6_xmit_lock(net); if (sk == NULL) return; + sk->sk_mark = mark; np = inet6_sk(sk); if (!fl6.flowi6_oif && ipv6_addr_is_multicast(&fl6.daddr)) @@ -593,12 +594,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb) if (IS_ERR(dst)) goto out; - if (ipv6_addr_is_multicast(&fl6.daddr)) - hlimit = np->mcast_hops; - else - hlimit = np->hop_limit; - if (hlimit < 0) - hlimit = ip6_dst_hoplimit(dst); + hlimit = ip6_sk_dst_hoplimit(np, &fl6, dst); idev = __in6_dev_get(skb->dev); @@ -702,22 +698,11 @@ static int icmpv6_rcv(struct sk_buff *skb) saddr = &ipv6_hdr(skb)->saddr; daddr = &ipv6_hdr(skb)->daddr; - /* Perform checksum. */ - switch (skb->ip_summed) { - case CHECKSUM_COMPLETE: - if (!csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_ICMPV6, - skb->csum)) - break; - /* fall through */ - case CHECKSUM_NONE: - skb->csum = ~csum_unfold(csum_ipv6_magic(saddr, daddr, skb->len, - IPPROTO_ICMPV6, 0)); - if (__skb_checksum_complete(skb)) { - LIMIT_NETDEBUG(KERN_DEBUG - "ICMPv6 checksum failed [%pI6c > %pI6c]\n", - saddr, daddr); - goto csum_error; - } + if (skb_checksum_validate(skb, IPPROTO_ICMPV6, ip6_compute_pseudo)) { + LIMIT_NETDEBUG(KERN_DEBUG + "ICMPv6 checksum failed [%pI6c > %pI6c]\n", + saddr, daddr); + goto csum_error; } if (!pskb_pull(skb, sizeof(*hdr))) diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c index d4ade34ab37..a245e5ddffb 100644 --- a/net/ipv6/inet6_connection_sock.c +++ b/net/ipv6/inet6_connection_sock.c @@ -81,7 +81,7 @@ struct dst_entry *inet6_csk_route_req(struct sock *sk, final_p = fl6_update_dst(fl6, np->opt, &final); fl6->saddr = ireq->ir_v6_loc_addr; fl6->flowi6_oif = ireq->ir_iif; - fl6->flowi6_mark = sk->sk_mark; + fl6->flowi6_mark = ireq->ir_mark; fl6->fl6_dport = ireq->ir_rmt_port; fl6->fl6_sport = htons(ireq->ir_num); security_req_classify_flow(req, flowi6_to_flowi(fl6)); diff --git a/net/ipv6/ip6_checksum.c b/net/ipv6/ip6_checksum.c index ee7a97f510c..9a4d7322fb2 100644 --- a/net/ipv6/ip6_checksum.c +++ b/net/ipv6/ip6_checksum.c @@ -75,25 +75,50 @@ int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh, int proto) return err; } - if (uh->check == 0) { - /* RFC 2460 section 8.1 says that we SHOULD log - this error. Well, it is reasonable. - */ - LIMIT_NETDEBUG(KERN_INFO "IPv6: udp checksum is 0 for [%pI6c]:%u->[%pI6c]:%u\n", - &ipv6_hdr(skb)->saddr, ntohs(uh->source), - &ipv6_hdr(skb)->daddr, ntohs(uh->dest)); - return 1; - } - if (skb->ip_summed == CHECKSUM_COMPLETE && - !csum_ipv6_magic(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, - skb->len, proto, skb->csum)) - skb->ip_summed = CHECKSUM_UNNECESSARY; + /* To support RFC 6936 (allow zero checksum in UDP/IPV6 for tunnels) + * we accept a checksum of zero here. When we find the socket + * for the UDP packet we'll check if that socket allows zero checksum + * for IPv6 (set by socket option). + */ + return skb_checksum_init_zero_check(skb, proto, uh->check, + ip6_compute_pseudo); +} +EXPORT_SYMBOL(udp6_csum_init); + +/* Function to set UDP checksum for an IPv6 UDP packet. This is intended + * for the simple case like when setting the checksum for a UDP tunnel. + */ +void udp6_set_csum(bool nocheck, struct sk_buff *skb, + const struct in6_addr *saddr, + const struct in6_addr *daddr, int len) +{ + struct udphdr *uh = udp_hdr(skb); + + if (nocheck) + uh->check = 0; + else if (skb_is_gso(skb)) + uh->check = ~udp_v6_check(len, saddr, daddr, 0); + else if (skb_dst(skb) && skb_dst(skb)->dev && + (skb_dst(skb)->dev->features & NETIF_F_IPV6_CSUM)) { - if (!skb_csum_unnecessary(skb)) - skb->csum = ~csum_unfold(csum_ipv6_magic(&ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, - skb->len, proto, 0)); + BUG_ON(skb->ip_summed == CHECKSUM_PARTIAL); - return 0; + skb->ip_summed = CHECKSUM_PARTIAL; + skb->csum_start = skb_transport_header(skb) - skb->head; + skb->csum_offset = offsetof(struct udphdr, check); + uh->check = ~udp_v6_check(len, saddr, daddr, 0); + } else { + __wsum csum; + + BUG_ON(skb->ip_summed == CHECKSUM_PARTIAL); + + uh->check = 0; + csum = skb_checksum(skb, 0, len, 0); + uh->check = udp_v6_check(len, saddr, daddr, csum); + if (uh->check == 0) + uh->check = CSUM_MANGLED_0; + + skb->ip_summed = CHECKSUM_UNNECESSARY; + } } -EXPORT_SYMBOL(udp6_csum_init); +EXPORT_SYMBOL(udp6_set_csum); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 34e0ded5c14..cb4459bd1d2 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -71,8 +71,7 @@ static DEFINE_RWLOCK(fib6_walker_lock); #define FWS_INIT FWS_L #endif -static void fib6_prune_clones(struct net *net, struct fib6_node *fn, - struct rt6_info *rt); +static void fib6_prune_clones(struct net *net, struct fib6_node *fn); static struct rt6_info *fib6_find_prefix(struct net *net, struct fib6_node *fn); static struct fib6_node *fib6_repair_tree(struct net *net, struct fib6_node *fn); static int fib6_walk(struct fib6_walker_t *w); @@ -941,7 +940,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nl_info *info, if (!err) { fib6_start_gc(info->nl_net, rt); if (!(rt->rt6i_flags & RTF_CACHE)) - fib6_prune_clones(info->nl_net, pn, rt); + fib6_prune_clones(info->nl_net, pn); } out: @@ -1375,7 +1374,7 @@ int fib6_del(struct rt6_info *rt, struct nl_info *info) pn = pn->parent; } #endif - fib6_prune_clones(info->nl_net, pn, rt); + fib6_prune_clones(info->nl_net, pn); } /* @@ -1459,7 +1458,7 @@ static int fib6_walk_continue(struct fib6_walker_t *w) if (w->skip) { w->skip--; - continue; + goto skip; } err = w->func(w); @@ -1469,6 +1468,7 @@ static int fib6_walk_continue(struct fib6_walker_t *w) w->count++; continue; } +skip: w->state = FWS_U; case FWS_U: if (fn == w->root) @@ -1600,10 +1600,9 @@ static int fib6_prune_clone(struct rt6_info *rt, void *arg) return 0; } -static void fib6_prune_clones(struct net *net, struct fib6_node *fn, - struct rt6_info *rt) +static void fib6_prune_clones(struct net *net, struct fib6_node *fn) { - fib6_clean_tree(net, fn, fib6_prune_clone, 1, rt); + fib6_clean_tree(net, fn, fib6_prune_clone, 1, NULL); } /* diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c index 0961b5ef866..4052694c6f2 100644 --- a/net/ipv6/ip6_flowlabel.c +++ b/net/ipv6/ip6_flowlabel.c @@ -26,7 +26,6 @@ #include <net/sock.h> #include <net/ipv6.h> -#include <net/addrconf.h> #include <net/rawv6.h> #include <net/transp_v6.h> diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index 9d921462b57..3873181ed85 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -72,6 +72,7 @@ struct ip6gre_net { }; static struct rtnl_link_ops ip6gre_link_ops __read_mostly; +static struct rtnl_link_ops ip6gre_tap_ops __read_mostly; static int ip6gre_tunnel_init(struct net_device *dev); static void ip6gre_tunnel_setup(struct net_device *dev); static void ip6gre_tunnel_link(struct ip6gre_net *ign, struct ip6_tnl *t); @@ -353,10 +354,10 @@ failed_free: static void ip6gre_tunnel_uninit(struct net_device *dev) { - struct net *net = dev_net(dev); - struct ip6gre_net *ign = net_generic(net, ip6gre_net_id); + struct ip6_tnl *t = netdev_priv(dev); + struct ip6gre_net *ign = net_generic(t->net, ip6gre_net_id); - ip6gre_tunnel_unlink(ign, netdev_priv(dev)); + ip6gre_tunnel_unlink(ign, t); dev_put(dev); } @@ -467,17 +468,7 @@ static int ip6gre_rcv(struct sk_buff *skb) goto drop; if (flags&GRE_CSUM) { - switch (skb->ip_summed) { - case CHECKSUM_COMPLETE: - csum = csum_fold(skb->csum); - if (!csum) - break; - /* fall through */ - case CHECKSUM_NONE: - skb->csum = 0; - csum = __skb_checksum_complete(skb); - skb->ip_summed = CHECKSUM_COMPLETE; - } + csum = skb_checksum_simple_validate(skb); offset += 4; } if (flags&GRE_KEY) { @@ -611,8 +602,8 @@ static netdev_tx_t ip6gre_xmit2(struct sk_buff *skb, int encap_limit, __u32 *pmtu) { - struct net *net = dev_net(dev); struct ip6_tnl *tunnel = netdev_priv(dev); + struct net *net = tunnel->net; struct net_device *tdev; /* Device to other host */ struct ipv6hdr *ipv6h; /* Our new IP header */ unsigned int max_headroom = 0; /* The extra header space needed */ @@ -979,7 +970,7 @@ static void ip6gre_tnl_link_config(struct ip6_tnl *t, int set_mtu) int strict = (ipv6_addr_type(&p->raddr) & (IPV6_ADDR_MULTICAST|IPV6_ADDR_LINKLOCAL)); - struct rt6_info *rt = rt6_lookup(dev_net(dev), + struct rt6_info *rt = rt6_lookup(t->net, &p->raddr, &p->laddr, p->link, strict); @@ -1063,13 +1054,12 @@ static int ip6gre_tunnel_ioctl(struct net_device *dev, int err = 0; struct ip6_tnl_parm2 p; struct __ip6_tnl_parm p1; - struct ip6_tnl *t; - struct net *net = dev_net(dev); + struct ip6_tnl *t = netdev_priv(dev); + struct net *net = t->net; struct ip6gre_net *ign = net_generic(net, ip6gre_net_id); switch (cmd) { case SIOCGETTUNNEL: - t = NULL; if (dev == ign->fb_tunnel_dev) { if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p))) { err = -EFAULT; @@ -1077,9 +1067,9 @@ static int ip6gre_tunnel_ioctl(struct net_device *dev, } ip6gre_tnl_parm_from_user(&p1, &p); t = ip6gre_tunnel_locate(net, &p1, 0); + if (t == NULL) + t = netdev_priv(dev); } - if (t == NULL) - t = netdev_priv(dev); memset(&p, 0, sizeof(p)); ip6gre_tnl_parm_to_user(&p, &t->parms); if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p))) @@ -1242,7 +1232,6 @@ static void ip6gre_tunnel_setup(struct net_device *dev) dev->flags |= IFF_NOARP; dev->iflink = 0; dev->addr_len = sizeof(struct in6_addr); - dev->features |= NETIF_F_NETNS_LOCAL; dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; } @@ -1297,11 +1286,17 @@ static struct inet6_protocol ip6gre_protocol __read_mostly = { .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, }; -static void ip6gre_destroy_tunnels(struct ip6gre_net *ign, - struct list_head *head) +static void ip6gre_destroy_tunnels(struct net *net, struct list_head *head) { + struct ip6gre_net *ign = net_generic(net, ip6gre_net_id); + struct net_device *dev, *aux; int prio; + for_each_netdev_safe(net, dev, aux) + if (dev->rtnl_link_ops == &ip6gre_link_ops || + dev->rtnl_link_ops == &ip6gre_tap_ops) + unregister_netdevice_queue(dev, head); + for (prio = 0; prio < 4; prio++) { int h; for (h = 0; h < HASH_SIZE; h++) { @@ -1310,7 +1305,12 @@ static void ip6gre_destroy_tunnels(struct ip6gre_net *ign, t = rtnl_dereference(ign->tunnels[prio][h]); while (t != NULL) { - unregister_netdevice_queue(t->dev, head); + /* If dev is in the same netns, it has already + * been added to the list by the previous loop. + */ + if (!net_eq(dev_net(t->dev), net)) + unregister_netdevice_queue(t->dev, + head); t = rtnl_dereference(t->next); } } @@ -1329,6 +1329,11 @@ static int __net_init ip6gre_init_net(struct net *net) goto err_alloc_dev; } dev_net_set(ign->fb_tunnel_dev, net); + /* FB netdevice is special: we have one, and only one per netns. + * Allowing to move it to another netns is clearly unsafe. + */ + ign->fb_tunnel_dev->features |= NETIF_F_NETNS_LOCAL; + ip6gre_fb_tunnel_init(ign->fb_tunnel_dev); ign->fb_tunnel_dev->rtnl_link_ops = &ip6gre_link_ops; @@ -1349,12 +1354,10 @@ err_alloc_dev: static void __net_exit ip6gre_exit_net(struct net *net) { - struct ip6gre_net *ign; LIST_HEAD(list); - ign = net_generic(net, ip6gre_net_id); rtnl_lock(); - ip6gre_destroy_tunnels(ign, &list); + ip6gre_destroy_tunnels(net, &list); unregister_netdevice_many(&list); rtnl_unlock(); } @@ -1531,15 +1534,14 @@ out: static int ip6gre_changelink(struct net_device *dev, struct nlattr *tb[], struct nlattr *data[]) { - struct ip6_tnl *t, *nt; - struct net *net = dev_net(dev); + struct ip6_tnl *t, *nt = netdev_priv(dev); + struct net *net = nt->net; struct ip6gre_net *ign = net_generic(net, ip6gre_net_id); struct __ip6_tnl_parm p; if (dev == ign->fb_tunnel_dev) return -EINVAL; - nt = netdev_priv(dev); ip6gre_netlink_parms(data, &p); t = ip6gre_tunnel_locate(net, &p, 0); diff --git a/net/ipv6/ip6_offload.c b/net/ipv6/ip6_offload.c index 59f95affceb..65eda2a8af4 100644 --- a/net/ipv6/ip6_offload.c +++ b/net/ipv6/ip6_offload.c @@ -97,9 +97,11 @@ static struct sk_buff *ipv6_gso_segment(struct sk_buff *skb, SKB_GSO_DODGY | SKB_GSO_TCP_ECN | SKB_GSO_GRE | + SKB_GSO_GRE_CSUM | SKB_GSO_IPIP | SKB_GSO_SIT | SKB_GSO_UDP_TUNNEL | + SKB_GSO_UDP_TUNNEL_CSUM | SKB_GSO_MPLS | SKB_GSO_TCPV6 | 0))) @@ -196,7 +198,6 @@ static struct sk_buff **ipv6_gro_receive(struct sk_buff **head, unsigned int off; u16 flush = 1; int proto; - __wsum csum; off = skb_gro_offset(skb); hlen = off + sizeof(*iph); @@ -264,13 +265,10 @@ static struct sk_buff **ipv6_gro_receive(struct sk_buff **head, NAPI_GRO_CB(skb)->flush |= flush; - csum = skb->csum; - skb_postpull_rcsum(skb, iph, skb_network_header_len(skb)); + skb_gro_postpull_rcsum(skb, iph, nlen); pp = ops->callbacks.gro_receive(head, skb); - skb->csum = csum; - out_unlock: rcu_read_unlock(); diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 40e7581374f..cb9df0eb402 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -219,7 +219,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6, skb->mark = sk->sk_mark; mtu = dst_mtu(dst); - if ((skb->len <= mtu) || skb->local_df || skb_is_gso(skb)) { + if ((skb->len <= mtu) || skb->ignore_df || skb_is_gso(skb)) { IP6_UPD_PO_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_OUT, skb->len); return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, skb, NULL, @@ -344,12 +344,16 @@ static unsigned int ip6_dst_mtu_forward(const struct dst_entry *dst) static bool ip6_pkt_too_big(const struct sk_buff *skb, unsigned int mtu) { - if (skb->len <= mtu || skb->local_df) + if (skb->len <= mtu) return false; + /* ipv6 conntrack defrag sets max_frag_size + ignore_df */ if (IP6CB(skb)->frag_max_size && IP6CB(skb)->frag_max_size > mtu) return true; + if (skb->ignore_df) + return false; + if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu) return false; @@ -533,6 +537,18 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from) skb_copy_secmark(to, from); } +static void ipv6_select_ident(struct frag_hdr *fhdr, struct rt6_info *rt) +{ + static u32 ip6_idents_hashrnd __read_mostly; + u32 hash, id; + + net_get_random_once(&ip6_idents_hashrnd, sizeof(ip6_idents_hashrnd)); + + hash = __ipv6_addr_jhash(&rt->rt6i_dst.addr, ip6_idents_hashrnd); + id = ip_idents_reserve(hash, 1); + fhdr->identification = htonl(id); +} + int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) { struct sk_buff *frag; @@ -555,7 +571,7 @@ int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) /* We must not fragment if the socket is set to force MTU discovery * or if the skb it not generated by a local socket. */ - if (unlikely(!skb->local_df && skb->len > mtu) || + if (unlikely(!skb->ignore_df && skb->len > mtu) || (IP6CB(skb)->frag_max_size && IP6CB(skb)->frag_max_size > mtu)) { if (skb->sk && dst_allfrag(skb_dst(skb))) @@ -1225,12 +1241,12 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to, unsigned int maxnonfragsize, headersize; headersize = sizeof(struct ipv6hdr) + - (opt ? opt->tot_len : 0) + + (opt ? opt->opt_flen + opt->opt_nflen : 0) + (dst_allfrag(&rt->dst) ? sizeof(struct frag_hdr) : 0) + rt->rt6i_nfheader_len; - if (ip6_sk_local_df(sk)) + if (ip6_sk_ignore_df(sk)) maxnonfragsize = sizeof(struct ipv6hdr) + IPV6_MAXPLEN; else maxnonfragsize = mtu; @@ -1540,7 +1556,7 @@ int ip6_push_pending_frames(struct sock *sk) } /* Allow local fragmentation. */ - skb->local_df = ip6_sk_local_df(sk); + skb->ignore_df = ip6_sk_ignore_df(sk); *final_dst = fl6->daddr; __skb_pull(skb, skb_network_header_len(skb)); diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index b05b609f69d..afa08245836 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -61,6 +61,7 @@ MODULE_AUTHOR("Ville Nuorvala"); MODULE_DESCRIPTION("IPv6 tunneling device"); MODULE_LICENSE("GPL"); +MODULE_ALIAS_RTNL_LINK("ip6tnl"); MODULE_ALIAS_NETDEV("ip6tnl0"); #ifdef IP6_TNL_DEBUG @@ -1557,7 +1558,7 @@ static int ip6_tnl_validate(struct nlattr *tb[], struct nlattr *data[]) { u8 proto; - if (!data) + if (!data || !data[IFLA_IPTUN_PROTO]) return 0; proto = nla_get_u8(data[IFLA_IPTUN_PROTO]); diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index b7c0f827140..9aaa6bb229e 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -511,6 +511,7 @@ static int vti6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, u8 type, u8 code, int offset, __be32 info) { __be32 spi; + __u32 mark; struct xfrm_state *x; struct ip6_tnl *t; struct ip_esp_hdr *esph; @@ -524,6 +525,8 @@ static int vti6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, if (!t) return -1; + mark = be32_to_cpu(t->parms.o_key); + switch (protocol) { case IPPROTO_ESP: esph = (struct ip_esp_hdr *)(skb->data + offset); @@ -545,7 +548,7 @@ static int vti6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, type != NDISC_REDIRECT) return 0; - x = xfrm_state_lookup(net, skb->mark, (const xfrm_address_t *)&iph->daddr, + x = xfrm_state_lookup(net, mark, (const xfrm_address_t *)&iph->daddr, spi, protocol, AF_INET6); if (!x) return 0; @@ -792,15 +795,12 @@ static const struct net_device_ops vti6_netdev_ops = { **/ static void vti6_dev_setup(struct net_device *dev) { - struct ip6_tnl *t; - dev->netdev_ops = &vti6_netdev_ops; dev->destructor = vti6_dev_free; dev->type = ARPHRD_TUNNEL6; dev->hard_header_len = LL_MAX_HEADER + sizeof(struct ipv6hdr); dev->mtu = ETH_DATA_LEN; - t = netdev_priv(dev); dev->flags |= IFF_NOARP; dev->addr_len = sizeof(struct in6_addr); dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; @@ -1097,7 +1097,6 @@ static int __init vti6_tunnel_init(void) err = xfrm6_protocol_register(&vti_esp6_protocol, IPPROTO_ESP); if (err < 0) { - unregister_pernet_device(&vti6_net_ops); pr_err("%s: can't register vti6 protocol\n", __func__); goto out; @@ -1106,7 +1105,6 @@ static int __init vti6_tunnel_init(void) err = xfrm6_protocol_register(&vti_ah6_protocol, IPPROTO_AH); if (err < 0) { xfrm6_protocol_deregister(&vti_esp6_protocol, IPPROTO_ESP); - unregister_pernet_device(&vti6_net_ops); pr_err("%s: can't register vti6 protocol\n", __func__); goto out; @@ -1116,7 +1114,6 @@ static int __init vti6_tunnel_init(void) if (err < 0) { xfrm6_protocol_deregister(&vti_ah6_protocol, IPPROTO_AH); xfrm6_protocol_deregister(&vti_esp6_protocol, IPPROTO_ESP); - unregister_pernet_device(&vti6_net_ops); pr_err("%s: can't register vti6 protocol\n", __func__); goto out; diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index 8659067da28..8250474ab7d 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -1633,7 +1633,7 @@ struct sock *mroute6_socket(struct net *net, struct sk_buff *skb) { struct mr6_table *mrt; struct flowi6 fl6 = { - .flowi6_iif = skb->skb_iif, + .flowi6_iif = skb->skb_iif ? : LOOPBACK_IFINDEX, .flowi6_oif = skb->dev->ifindex, .flowi6_mark = skb->mark, }; diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 09a22f4f36c..ca8d4ea48a5 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -851,7 +851,7 @@ out: static void ndisc_recv_na(struct sk_buff *skb) { struct nd_msg *msg = (struct nd_msg *)skb_transport_header(skb); - const struct in6_addr *saddr = &ipv6_hdr(skb)->saddr; + struct in6_addr *saddr = &ipv6_hdr(skb)->saddr; const struct in6_addr *daddr = &ipv6_hdr(skb)->daddr; u8 *lladdr = NULL; u32 ndoptlen = skb_tail_pointer(skb) - (skb_transport_header(skb) + @@ -944,10 +944,7 @@ static void ndisc_recv_na(struct sk_buff *skb) /* * Change: router to host */ - struct rt6_info *rt; - rt = rt6_get_dflt_router(saddr, dev); - if (rt) - ip6_del_rt(rt); + rt6_clean_tohost(dev_net(dev), saddr); } out: diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index 95f3f1da0d7..d38e6a8d8b9 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -30,13 +30,15 @@ int ip6_route_me_harder(struct sk_buff *skb) .daddr = iph->daddr, .saddr = iph->saddr, }; + int err; dst = ip6_route_output(net, skb->sk, &fl6); - if (dst->error) { + err = dst->error; + if (err) { IP6_INC_STATS(net, ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES); LIMIT_NETDEBUG(KERN_DEBUG "ip6_route_me_harder: No more route.\n"); dst_release(dst); - return dst->error; + return err; } /* Drop old route. */ diff --git a/net/ipv6/netfilter/ip6t_rpfilter.c b/net/ipv6/netfilter/ip6t_rpfilter.c index e0983f3648a..790e0c6b19e 100644 --- a/net/ipv6/netfilter/ip6t_rpfilter.c +++ b/net/ipv6/netfilter/ip6t_rpfilter.c @@ -33,6 +33,7 @@ static bool rpfilter_lookup_reverse6(const struct sk_buff *skb, struct ipv6hdr *iph = ipv6_hdr(skb); bool ret = false; struct flowi6 fl6 = { + .flowi6_iif = LOOPBACK_IFINDEX, .flowlabel = (* (__be32 *) iph) & IPV6_FLOWINFO_MASK, .flowi6_proto = iph->nexthdr, .daddr = iph->saddr, diff --git a/net/ipv6/netfilter/ip6table_nat.c b/net/ipv6/netfilter/ip6table_nat.c index 84c7f33d0cf..387d8b8fc18 100644 --- a/net/ipv6/netfilter/ip6table_nat.c +++ b/net/ipv6/netfilter/ip6table_nat.c @@ -90,17 +90,9 @@ nf_nat_ipv6_fn(const struct nf_hook_ops *ops, if (nf_ct_is_untracked(ct)) return NF_ACCEPT; - nat = nfct_nat(ct); - if (!nat) { - /* NAT module was loaded late. */ - if (nf_ct_is_confirmed(ct)) - return NF_ACCEPT; - nat = nf_ct_ext_add(ct, NF_CT_EXT_NAT, GFP_ATOMIC); - if (nat == NULL) { - pr_debug("failed to add NAT extension\n"); - return NF_ACCEPT; - } - } + nat = nf_ct_nat_ext_add(ct); + if (nat == NULL) + return NF_ACCEPT; switch (ctinfo) { case IP_CT_RELATED: diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index 767ab8da821..0d5279fd852 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -451,7 +451,7 @@ nf_ct_frag6_reasm(struct frag_queue *fq, struct net_device *dev) } sub_frag_mem_limit(&fq->q, head->truesize); - head->local_df = 1; + head->ignore_df = 1; head->next = NULL; head->dev = dev; head->tstamp = fq->q.stamp; diff --git a/net/ipv6/netfilter/nft_chain_nat_ipv6.c b/net/ipv6/netfilter/nft_chain_nat_ipv6.c index 9c3297a768f..d189fcb437f 100644 --- a/net/ipv6/netfilter/nft_chain_nat_ipv6.c +++ b/net/ipv6/netfilter/nft_chain_nat_ipv6.c @@ -47,15 +47,9 @@ static unsigned int nf_nat_ipv6_fn(const struct nf_hook_ops *ops, if (ct == NULL || nf_ct_is_untracked(ct)) return NF_ACCEPT; - nat = nfct_nat(ct); - if (nat == NULL) { - /* Conntrack module was loaded late, can't add extension. */ - if (nf_ct_is_confirmed(ct)) - return NF_ACCEPT; - nat = nf_ct_ext_add(ct, NF_CT_EXT_NAT, GFP_ATOMIC); - if (nat == NULL) - return NF_ACCEPT; - } + nat = nf_ct_nat_ext_add(ct); + if (nat == NULL) + return NF_ACCEPT; switch (ctinfo) { case IP_CT_RELATED: diff --git a/net/ipv6/output_core.c b/net/ipv6/output_core.c index 6313abd53c9..ffa029305a0 100644 --- a/net/ipv6/output_core.c +++ b/net/ipv6/output_core.c @@ -8,37 +8,6 @@ #include <net/addrconf.h> #include <net/secure_seq.h> -void ipv6_select_ident(struct frag_hdr *fhdr, struct rt6_info *rt) -{ - static atomic_t ipv6_fragmentation_id; - struct in6_addr addr; - int old, new; - -#if IS_ENABLED(CONFIG_IPV6) - struct inet_peer *peer; - struct net *net; - - net = dev_net(rt->dst.dev); - peer = inet_getpeer_v6(net->ipv6.peers, &rt->rt6i_dst.addr, 1); - if (peer) { - fhdr->identification = htonl(inet_getid(peer, 0)); - inet_putpeer(peer); - return; - } -#endif - do { - old = atomic_read(&ipv6_fragmentation_id); - new = old + 1; - if (!new) - new = 1; - } while (atomic_cmpxchg(&ipv6_fragmentation_id, old, new) != old); - - addr = rt->rt6i_dst.addr; - addr.s6_addr32[0] ^= (__force __be32)new; - fhdr->identification = htonl(secure_ipv6_id(addr.s6_addr32)); -} -EXPORT_SYMBOL(ipv6_select_ident); - int ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr) { u16 offset = sizeof(struct ipv6hdr); diff --git a/net/ipv6/ping.c b/net/ipv6/ping.c index bda74291c3e..5b7a1ed2aba 100644 --- a/net/ipv6/ping.c +++ b/net/ipv6/ping.c @@ -51,7 +51,6 @@ static struct inet_protosw pingv6_protosw = { .protocol = IPPROTO_ICMPV6, .prot = &pingv6_prot, .ops = &inet6_dgram_ops, - .no_check = UDP_CSUM_DEFAULT, .flags = INET_PROTOSW_REUSE, }; @@ -168,12 +167,7 @@ int ping_v6_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, pfh.wcheck = 0; pfh.family = AF_INET6; - if (ipv6_addr_is_multicast(&fl6.daddr)) - hlimit = np->mcast_hops; - else - hlimit = np->hop_limit; - if (hlimit < 0) - hlimit = ip6_dst_hoplimit(dst); + hlimit = ip6_sk_dst_hoplimit(np, &fl6, dst); lock_sock(sk); err = ip6_append_data(sk, ping_getfrag, &pfh, len, diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c index 091d066a57b..3317440ea34 100644 --- a/net/ipv6/proc.c +++ b/net/ipv6/proc.c @@ -186,7 +186,7 @@ static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, atomic_long_t *smib) /* can be called either with percpu mib (pcpumib != NULL), * or shared one (smib != NULL) */ -static void snmp6_seq_show_item(struct seq_file *seq, void __percpu **pcpumib, +static void snmp6_seq_show_item(struct seq_file *seq, void __percpu *pcpumib, atomic_long_t *smib, const struct snmp_mib *itemlist) { @@ -201,7 +201,7 @@ static void snmp6_seq_show_item(struct seq_file *seq, void __percpu **pcpumib, } } -static void snmp6_seq_show_item64(struct seq_file *seq, void __percpu **mib, +static void snmp6_seq_show_item64(struct seq_file *seq, void __percpu *mib, const struct snmp_mib *itemlist, size_t syncpoff) { int i; @@ -215,14 +215,14 @@ static int snmp6_seq_show(struct seq_file *seq, void *v) { struct net *net = (struct net *)seq->private; - snmp6_seq_show_item64(seq, (void __percpu **)net->mib.ipv6_statistics, + snmp6_seq_show_item64(seq, net->mib.ipv6_statistics, snmp6_ipstats_list, offsetof(struct ipstats_mib, syncp)); - snmp6_seq_show_item(seq, (void __percpu **)net->mib.icmpv6_statistics, + snmp6_seq_show_item(seq, net->mib.icmpv6_statistics, NULL, snmp6_icmp6_list); snmp6_seq_show_icmpv6msg(seq, net->mib.icmpv6msg_statistics->mibs); - snmp6_seq_show_item(seq, (void __percpu **)net->mib.udp_stats_in6, + snmp6_seq_show_item(seq, net->mib.udp_stats_in6, NULL, snmp6_udp6_list); - snmp6_seq_show_item(seq, (void __percpu **)net->mib.udplite_stats_in6, + snmp6_seq_show_item(seq, net->mib.udplite_stats_in6, NULL, snmp6_udplite6_list); return 0; } @@ -245,7 +245,7 @@ static int snmp6_dev_seq_show(struct seq_file *seq, void *v) struct inet6_dev *idev = (struct inet6_dev *)seq->private; seq_printf(seq, "%-32s\t%u\n", "ifIndex", idev->dev->ifindex); - snmp6_seq_show_item64(seq, (void __percpu **)idev->stats.ipv6, + snmp6_seq_show_item64(seq, idev->stats.ipv6, snmp6_ipstats_list, offsetof(struct ipstats_mib, syncp)); snmp6_seq_show_item(seq, NULL, idev->stats.icmpv6dev->mibs, snmp6_icmp6_list); diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 1f29996e368..b2dc60b0c76 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -873,14 +873,8 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk, err = PTR_ERR(dst); goto out; } - if (hlimit < 0) { - if (ipv6_addr_is_multicast(&fl6.daddr)) - hlimit = np->mcast_hops; - else - hlimit = np->hop_limit; - if (hlimit < 0) - hlimit = ip6_dst_hoplimit(dst); - } + if (hlimit < 0) + hlimit = ip6_sk_dst_hoplimit(np, &fl6, dst); if (tclass < 0) tclass = np->tclass; @@ -1328,7 +1322,6 @@ static struct inet_protosw rawv6_protosw = { .protocol = IPPROTO_IP, /* wild card */ .prot = &rawv6_prot, .ops = &inet6_sockraw_ops, - .no_check = UDP_CSUM_DEFAULT, .flags = INET_PROTOSW_REUSE, }; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 4011617cca6..f23fbd28a50 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1176,7 +1176,7 @@ void ip6_update_pmtu(struct sk_buff *skb, struct net *net, __be32 mtu, memset(&fl6, 0, sizeof(fl6)); fl6.flowi6_oif = oif; - fl6.flowi6_mark = mark; + fl6.flowi6_mark = mark ? mark : IP6_REPLY_MARK(net, skb->mark); fl6.daddr = iph->daddr; fl6.saddr = iph->saddr; fl6.flowlabel = ip6_flowinfo(iph); @@ -1273,6 +1273,7 @@ void ip6_redirect(struct sk_buff *skb, struct net *net, int oif, u32 mark) struct flowi6 fl6; memset(&fl6, 0, sizeof(fl6)); + fl6.flowi6_iif = LOOPBACK_IFINDEX; fl6.flowi6_oif = oif; fl6.flowi6_mark = mark; fl6.daddr = iph->daddr; @@ -1294,6 +1295,7 @@ void ip6_redirect_no_header(struct sk_buff *skb, struct net *net, int oif, struct flowi6 fl6; memset(&fl6, 0, sizeof(fl6)); + fl6.flowi6_iif = LOOPBACK_IFINDEX; fl6.flowi6_oif = oif; fl6.flowi6_mark = mark; fl6.daddr = msg->dest; @@ -1453,7 +1455,7 @@ static int ip6_dst_gc(struct dst_ops *ops) goto out; net->ipv6.ip6_rt_gc_expire++; - fib6_run_gc(net->ipv6.ip6_rt_gc_expire, net, entries > rt_max_size); + fib6_run_gc(net->ipv6.ip6_rt_gc_expire, net, true); entries = dst_entries_get_slow(ops); if (entries < ops->gc_thresh) net->ipv6.ip6_rt_gc_expire = rt_gc_timeout>>1; @@ -2232,6 +2234,27 @@ void rt6_remove_prefsrc(struct inet6_ifaddr *ifp) fib6_clean_all(net, fib6_remove_prefsrc, &adni); } +#define RTF_RA_ROUTER (RTF_ADDRCONF | RTF_DEFAULT | RTF_GATEWAY) +#define RTF_CACHE_GATEWAY (RTF_GATEWAY | RTF_CACHE) + +/* Remove routers and update dst entries when gateway turn into host. */ +static int fib6_clean_tohost(struct rt6_info *rt, void *arg) +{ + struct in6_addr *gateway = (struct in6_addr *)arg; + + if ((((rt->rt6i_flags & RTF_RA_ROUTER) == RTF_RA_ROUTER) || + ((rt->rt6i_flags & RTF_CACHE_GATEWAY) == RTF_CACHE_GATEWAY)) && + ipv6_addr_equal(gateway, &rt->rt6i_gateway)) { + return -1; + } + return 0; +} + +void rt6_clean_tohost(struct net *net, struct in6_addr *gateway) +{ + fib6_clean_all(net, fib6_clean_tohost, gateway); +} + struct arg_dev_net { struct net_device *dev; struct net *net; @@ -2707,6 +2730,9 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh) if (tb[RTA_OIF]) oif = nla_get_u32(tb[RTA_OIF]); + if (tb[RTA_MARK]) + fl6.flowi6_mark = nla_get_u32(tb[RTA_MARK]); + if (iif) { struct net_device *dev; int flags = 0; diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index e5a453ca302..f4380041f5e 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -1828,4 +1828,5 @@ xfrm_tunnel_failed: module_init(sit_init); module_exit(sit_cleanup); MODULE_LICENSE("GPL"); +MODULE_ALIAS_RTNL_LINK("sit"); MODULE_ALIAS_NETDEV("sit0"); diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c index bb53a5e73c1..a822b880689 100644 --- a/net/ipv6/syncookies.c +++ b/net/ipv6/syncookies.c @@ -216,6 +216,8 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) ipv6_addr_type(&ireq->ir_v6_rmt_addr) & IPV6_ADDR_LINKLOCAL) ireq->ir_iif = inet6_iif(skb); + ireq->ir_mark = inet_request_mark(sk, skb); + req->expires = 0UL; req->num_retrans = 0; ireq->ecn_ok = ecn_ok; @@ -242,7 +244,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) final_p = fl6_update_dst(&fl6, np->opt, &final); fl6.saddr = ireq->ir_v6_loc_addr; fl6.flowi6_oif = sk->sk_bound_dev_if; - fl6.flowi6_mark = sk->sk_mark; + fl6.flowi6_mark = ireq->ir_mark; fl6.fl6_dport = ireq->ir_rmt_port; fl6.fl6_sport = inet_sk(sk)->inet_sport; security_req_classify_flow(req, flowi6_to_flowi(&fl6)); diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c index 7f405a16882..058f3eca2e5 100644 --- a/net/ipv6/sysctl_net_ipv6.c +++ b/net/ipv6/sysctl_net_ipv6.c @@ -38,6 +38,13 @@ static struct ctl_table ipv6_table_template[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "fwmark_reflect", + .data = &init_net.ipv6.sysctl.fwmark_reflect, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index e289830ed6e..229239ad96b 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -340,7 +340,8 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, struct sock *sk; int err; struct tcp_sock *tp; - __u32 seq; + struct request_sock *fastopen; + __u32 seq, snd_una; struct net *net = dev_net(skb->dev); sk = inet6_lookup(net, &tcp_hashinfo, &hdr->daddr, @@ -371,8 +372,11 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, tp = tcp_sk(sk); seq = ntohl(th->seq); + /* XXX (TFO) - tp->snd_una should be ISN (tcp_create_openreq_child() */ + fastopen = tp->fastopen_rsk; + snd_una = fastopen ? tcp_rsk(fastopen)->snt_isn : tp->snd_una; if (sk->sk_state != TCP_LISTEN && - !between(seq, tp->snd_una, tp->snd_nxt)) { + !between(seq, snd_una, tp->snd_nxt)) { NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS); goto out; } @@ -436,8 +440,13 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, goto out; case TCP_SYN_SENT: - case TCP_SYN_RECV: /* Cannot happen. - It can, it SYNs are crossed. --ANK */ + case TCP_SYN_RECV: + /* Only in fast or simultaneous open. If a fast open socket is + * is already accepted it is treated as a connected one below. + */ + if (fastopen && fastopen->sk == NULL) + break; + if (!sock_owned_by_user(sk)) { sk->sk_err = err; sk->sk_error_report(sk); /* Wake people up to see the error (see connect in sock.c) */ @@ -463,7 +472,8 @@ out: static int tcp_v6_send_synack(struct sock *sk, struct dst_entry *dst, struct flowi6 *fl6, struct request_sock *req, - u16 queue_mapping) + u16 queue_mapping, + struct tcp_fastopen_cookie *foc) { struct inet_request_sock *ireq = inet_rsk(req); struct ipv6_pinfo *np = inet6_sk(sk); @@ -474,7 +484,7 @@ static int tcp_v6_send_synack(struct sock *sk, struct dst_entry *dst, if (!dst && (dst = inet6_csk_route_req(sk, fl6, req)) == NULL) goto done; - skb = tcp_make_synack(sk, dst, req, NULL); + skb = tcp_make_synack(sk, dst, req, foc); if (skb) { __tcp_v6_send_check(skb, &ireq->ir_v6_loc_addr, @@ -498,7 +508,7 @@ static int tcp_v6_rtx_synack(struct sock *sk, struct request_sock *req) struct flowi6 fl6; int res; - res = tcp_v6_send_synack(sk, NULL, &fl6, req, 0); + res = tcp_v6_send_synack(sk, NULL, &fl6, req, 0, NULL); if (!res) { TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_RETRANSSEGS); NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPSYNRETRANS); @@ -802,6 +812,7 @@ static void tcp_v6_send_response(struct sk_buff *skb, u32 seq, u32 ack, u32 win, fl6.flowi6_oif = inet6_iif(skb); else fl6.flowi6_oif = oif; + fl6.flowi6_mark = IP6_REPLY_MARK(net, skb->mark); fl6.fl6_dport = t1->dest; fl6.fl6_sport = t1->source; security_skb_classify_flow(skb, flowi6_to_flowi(&fl6)); @@ -917,7 +928,12 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb) static void tcp_v6_reqsk_send_ack(struct sock *sk, struct sk_buff *skb, struct request_sock *req) { - tcp_v6_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, + /* sk->sk_state == TCP_LISTEN -> for regular TCP_SYN_RECV + * sk->sk_state == TCP_SYN_RECV -> for Fast Open. + */ + tcp_v6_send_ack(skb, (sk->sk_state == TCP_LISTEN) ? + tcp_rsk(req)->snt_isn + 1 : tcp_sk(sk)->snd_nxt, + tcp_rsk(req)->rcv_nxt, req->rcv_wnd, tcp_time_stamp, req->ts_recent, sk->sk_bound_dev_if, tcp_v6_md5_do_lookup(sk, &ipv6_hdr(skb)->daddr), 0, 0); @@ -969,8 +985,10 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) struct tcp_sock *tp = tcp_sk(sk); __u32 isn = TCP_SKB_CB(skb)->when; struct dst_entry *dst = NULL; + struct tcp_fastopen_cookie foc = { .len = -1 }; + bool want_cookie = false, fastopen; struct flowi6 fl6; - bool want_cookie = false; + int err; if (skb->protocol == htons(ETH_P_IP)) return tcp_v4_conn_request(sk, skb); @@ -1001,7 +1019,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) tcp_clear_options(&tmp_opt); tmp_opt.mss_clamp = IPV6_MIN_MTU - sizeof(struct tcphdr) - sizeof(struct ipv6hdr); tmp_opt.user_mss = tp->rx_opt.user_mss; - tcp_parse_options(skb, &tmp_opt, 0, NULL); + tcp_parse_options(skb, &tmp_opt, 0, want_cookie ? NULL : &foc); if (want_cookie && !tmp_opt.saw_tstamp) tcp_clear_options(&tmp_opt); @@ -1016,6 +1034,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) TCP_ECN_create_request(req, skb, sock_net(sk)); ireq->ir_iif = sk->sk_bound_dev_if; + ireq->ir_mark = inet_request_mark(sk, skb); /* So that link locals have meaning */ if (!sk->sk_bound_dev_if && @@ -1074,19 +1093,27 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) isn = tcp_v6_init_sequence(skb); } have_isn: - tcp_rsk(req)->snt_isn = isn; if (security_inet_conn_request(sk, skb, req)) goto drop_and_release; - if (tcp_v6_send_synack(sk, dst, &fl6, req, - skb_get_queue_mapping(skb)) || - want_cookie) + if (!dst && (dst = inet6_csk_route_req(sk, &fl6, req)) == NULL) goto drop_and_free; + tcp_rsk(req)->snt_isn = isn; tcp_rsk(req)->snt_synack = tcp_time_stamp; - tcp_rsk(req)->listener = NULL; - inet6_csk_reqsk_queue_hash_add(sk, req, TCP_TIMEOUT_INIT); + tcp_openreq_init_rwin(req, sk, dst); + fastopen = !want_cookie && + tcp_try_fastopen(sk, skb, req, &foc, dst); + err = tcp_v6_send_synack(sk, dst, &fl6, req, + skb_get_queue_mapping(skb), &foc); + if (!fastopen) { + if (err || want_cookie) + goto drop_and_free; + + tcp_rsk(req)->listener = NULL; + inet6_csk_reqsk_queue_hash_add(sk, req, TCP_TIMEOUT_INIT); + } return 0; drop_and_release: @@ -1294,25 +1321,6 @@ out: return NULL; } -static __sum16 tcp_v6_checksum_init(struct sk_buff *skb) -{ - if (skb->ip_summed == CHECKSUM_COMPLETE) { - if (!tcp_v6_check(skb->len, &ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, skb->csum)) { - skb->ip_summed = CHECKSUM_UNNECESSARY; - return 0; - } - } - - skb->csum = ~csum_unfold(tcp_v6_check(skb->len, - &ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, 0)); - - if (skb->len <= 76) - return __skb_checksum_complete(skb); - return 0; -} - /* The socket must have it's spinlock held when we get * here. * @@ -1486,7 +1494,7 @@ static int tcp_v6_rcv(struct sk_buff *skb) if (!pskb_may_pull(skb, th->doff*4)) goto discard_it; - if (!skb_csum_unnecessary(skb) && tcp_v6_checksum_init(skb)) + if (skb_checksum_init(skb, IPPROTO_TCP, ip6_compute_pseudo)) goto csum_error; th = tcp_hdr(skb); @@ -1779,6 +1787,7 @@ static void get_tcp6_sock(struct seq_file *seq, struct sock *sp, int i) const struct inet_sock *inet = inet_sk(sp); const struct tcp_sock *tp = tcp_sk(sp); const struct inet_connection_sock *icsk = inet_csk(sp); + struct fastopen_queue *fastopenq = icsk->icsk_accept_queue.fastopenq; dest = &sp->sk_v6_daddr; src = &sp->sk_v6_rcv_saddr; @@ -1821,7 +1830,9 @@ static void get_tcp6_sock(struct seq_file *seq, struct sock *sp, int i) jiffies_to_clock_t(icsk->icsk_ack.ato), (icsk->icsk_ack.quick << 1) | icsk->icsk_ack.pingpong, tp->snd_cwnd, - tcp_in_initial_slowstart(tp) ? -1 : tp->snd_ssthresh + sp->sk_state == TCP_LISTEN ? + (fastopenq ? fastopenq->max_qlen : 0) : + (tcp_in_initial_slowstart(tp) ? -1 : tp->snd_ssthresh) ); } @@ -1981,7 +1992,6 @@ static struct inet_protosw tcpv6_protosw = { .protocol = IPPROTO_TCP, .prot = &tcpv6_prot, .ops = &inet6_stream_ops, - .no_check = 0, .flags = INET_PROTOSW_PERMANENT | INET_PROTOSW_ICSK, }; diff --git a/net/ipv6/tcpv6_offload.c b/net/ipv6/tcpv6_offload.c index 0d78132ff18..8517d3cd1ae 100644 --- a/net/ipv6/tcpv6_offload.c +++ b/net/ipv6/tcpv6_offload.c @@ -42,7 +42,7 @@ static struct sk_buff **tcp6_gro_receive(struct sk_buff **head, if (NAPI_GRO_CB(skb)->flush) goto skip_csum; - wsum = skb->csum; + wsum = NAPI_GRO_CB(skb)->csum; switch (skb->ip_summed) { case CHECKSUM_NONE: diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 1e586d92260..95c83479928 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -634,6 +634,10 @@ int udpv6_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) if (skb->len > sizeof(struct udphdr) && encap_rcv != NULL) { int ret; + /* Verify checksum before giving to encap */ + if (udp_lib_checksum_complete(skb)) + goto csum_error; + ret = encap_rcv(sk, skb); if (ret <= 0) { UDP_INC_STATS_BH(sock_net(sk), @@ -701,17 +705,16 @@ static struct sock *udp_v6_mcast_next(struct net *net, struct sock *sk, int dif) { struct hlist_nulls_node *node; - struct sock *s = sk; unsigned short num = ntohs(loc_port); - sk_nulls_for_each_from(s, node) { - struct inet_sock *inet = inet_sk(s); + sk_nulls_for_each_from(sk, node) { + struct inet_sock *inet = inet_sk(sk); - if (!net_eq(sock_net(s), net)) + if (!net_eq(sock_net(sk), net)) continue; - if (udp_sk(s)->udp_port_hash == num && - s->sk_family == PF_INET6) { + if (udp_sk(sk)->udp_port_hash == num && + sk->sk_family == PF_INET6) { if (inet->inet_dport) { if (inet->inet_dport != rmt_port) continue; @@ -720,16 +723,16 @@ static struct sock *udp_v6_mcast_next(struct net *net, struct sock *sk, !ipv6_addr_equal(&sk->sk_v6_daddr, rmt_addr)) continue; - if (s->sk_bound_dev_if && s->sk_bound_dev_if != dif) + if (sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif) continue; if (!ipv6_addr_any(&sk->sk_v6_rcv_saddr)) { if (!ipv6_addr_equal(&sk->sk_v6_rcv_saddr, loc_addr)) continue; } - if (!inet6_mc_check(s, loc_addr, rmt_addr)) + if (!inet6_mc_check(sk, loc_addr, rmt_addr)) continue; - return s; + return sk; } } return NULL; @@ -760,6 +763,17 @@ static void flush_stack(struct sock **stack, unsigned int count, if (unlikely(skb1)) kfree_skb(skb1); } + +static void udp6_csum_zero_error(struct sk_buff *skb) +{ + /* RFC 2460 section 8.1 says that we SHOULD log + * this error. Well, it is reasonable. + */ + LIMIT_NETDEBUG(KERN_INFO "IPv6: udp checksum is 0 for [%pI6c]:%u->[%pI6c]:%u\n", + &ipv6_hdr(skb)->saddr, ntohs(udp_hdr(skb)->source), + &ipv6_hdr(skb)->daddr, ntohs(udp_hdr(skb)->dest)); +} + /* * Note: called only from the BH handler context, * so we don't need to lock the hashes. @@ -779,7 +793,12 @@ static int __udp6_lib_mcast_deliver(struct net *net, struct sk_buff *skb, dif = inet6_iif(skb); sk = udp_v6_mcast_next(net, sk, uh->dest, daddr, uh->source, saddr, dif); while (sk) { - stack[count++] = sk; + /* If zero checksum and no_check is not on for + * the socket then skip it. + */ + if (uh->check || udp_sk(sk)->no_check6_rx) + stack[count++] = sk; + sk = udp_v6_mcast_next(net, sk_nulls_next(sk), uh->dest, daddr, uh->source, saddr, dif); if (unlikely(count == ARRAY_SIZE(stack))) { @@ -867,6 +886,12 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, if (sk != NULL) { int ret; + if (!uh->check && !udp_sk(sk)->no_check6_rx) { + sock_put(sk); + udp6_csum_zero_error(skb); + goto csum_error; + } + ret = udpv6_queue_rcv_skb(sk, skb); sock_put(sk); @@ -879,6 +904,11 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, return 0; } + if (!uh->check) { + udp6_csum_zero_error(skb); + goto csum_error; + } + if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) goto discard; @@ -1006,7 +1036,10 @@ static int udp_v6_push_pending_frames(struct sock *sk) if (is_udplite) csum = udplite_csum_outgoing(sk, skb); - else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */ + else if (up->no_check6_tx) { /* UDP csum disabled */ + skb->ip_summed = CHECKSUM_NONE; + goto send; + } else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */ udp6_hwcsum_outgoing(sk, skb, &fl6->saddr, &fl6->daddr, up->len); goto send; @@ -1232,14 +1265,8 @@ do_udp_sendmsg: goto out; } - if (hlimit < 0) { - if (ipv6_addr_is_multicast(&fl6.daddr)) - hlimit = np->mcast_hops; - else - hlimit = np->hop_limit; - if (hlimit < 0) - hlimit = ip6_dst_hoplimit(dst); - } + if (hlimit < 0) + hlimit = ip6_sk_dst_hoplimit(np, &fl6, dst); if (tclass < 0) tclass = np->tclass; @@ -1479,7 +1506,6 @@ static struct inet_protosw udpv6_protosw = { .protocol = IPPROTO_UDP, .prot = &udpv6_prot, .ops = &inet6_dgram_ops, - .no_check = UDP_CSUM_DEFAULT, .flags = INET_PROTOSW_PERMANENT, }; diff --git a/net/ipv6/udp_offload.c b/net/ipv6/udp_offload.c index b261ee8b83f..0ae3d98f83e 100644 --- a/net/ipv6/udp_offload.c +++ b/net/ipv6/udp_offload.c @@ -63,7 +63,9 @@ static struct sk_buff *udp6_ufo_fragment(struct sk_buff *skb, if (unlikely(type & ~(SKB_GSO_UDP | SKB_GSO_DODGY | SKB_GSO_UDP_TUNNEL | + SKB_GSO_UDP_TUNNEL_CSUM | SKB_GSO_GRE | + SKB_GSO_GRE_CSUM | SKB_GSO_IPIP | SKB_GSO_SIT | SKB_GSO_MPLS) || @@ -76,7 +78,8 @@ static struct sk_buff *udp6_ufo_fragment(struct sk_buff *skb, goto out; } - if (skb->encapsulation && skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL) + if (skb->encapsulation && skb_shinfo(skb)->gso_type & + (SKB_GSO_UDP_TUNNEL|SKB_GSO_UDP_TUNNEL_CSUM)) segs = skb_udp_tunnel_segment(skb, features); else { /* Do software UFO. Complete and fill in the UDP checksum as HW cannot diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c index dfcc4be4689..9cf097e206e 100644 --- a/net/ipv6/udplite.c +++ b/net/ipv6/udplite.c @@ -64,7 +64,6 @@ static struct inet_protosw udplite6_protosw = { .protocol = IPPROTO_UDPLITE, .prot = &udplitev6_prot, .ops = &inet6_dgram_ops, - .no_check = 0, .flags = INET_PROTOSW_PERMANENT, }; diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c index 19ef329bdbf..433672d07d0 100644 --- a/net/ipv6/xfrm6_output.c +++ b/net/ipv6/xfrm6_output.c @@ -78,7 +78,7 @@ static int xfrm6_tunnel_check_size(struct sk_buff *skb) if (mtu < IPV6_MIN_MTU) mtu = IPV6_MIN_MTU; - if (!skb->local_df && skb->len > mtu) { + if (!skb->ignore_df && skb->len > mtu) { skb->dev = dst->dev; if (xfrm6_local_dontfrag(skb)) @@ -114,13 +114,7 @@ int xfrm6_prepare_output(struct xfrm_state *x, struct sk_buff *skb) if (err) return err; - memset(IP6CB(skb), 0, sizeof(*IP6CB(skb))); -#ifdef CONFIG_NETFILTER - IP6CB(skb)->flags |= IP6SKB_XFRM_TRANSFORMED; -#endif - - skb->protocol = htons(ETH_P_IPV6); - skb->local_df = 1; + skb->ignore_df = 1; return x->outer_mode->output2(x, skb); } @@ -128,11 +122,13 @@ EXPORT_SYMBOL(xfrm6_prepare_output); int xfrm6_output_finish(struct sk_buff *skb) { + memset(IP6CB(skb), 0, sizeof(*IP6CB(skb))); + skb->protocol = htons(ETH_P_IPV6); + #ifdef CONFIG_NETFILTER IP6CB(skb)->flags |= IP6SKB_XFRM_TRANSFORMED; #endif - skb->protocol = htons(ETH_P_IPV6); return xfrm_output(skb); } @@ -142,6 +138,13 @@ static int __xfrm6_output(struct sk_buff *skb) struct xfrm_state *x = dst->xfrm; int mtu; +#ifdef CONFIG_NETFILTER + if (!x) { + IP6CB(skb)->flags |= IP6SKB_REROUTED; + return dst_output(skb); + } +#endif + if (skb->protocol == htons(ETH_P_IPV6)) mtu = ip6_skb_dst_mtu(skb); else @@ -150,7 +153,7 @@ static int __xfrm6_output(struct sk_buff *skb) if (skb->len > mtu && xfrm6_local_dontfrag(skb)) { xfrm6_local_rxpmtu(skb, mtu); return -EMSGSIZE; - } else if (!skb->local_df && skb->len > mtu && skb->sk) { + } else if (!skb->ignore_df && skb->len > mtu && skb->sk) { xfrm_local_error(skb, mtu); return -EMSGSIZE; } @@ -165,6 +168,7 @@ static int __xfrm6_output(struct sk_buff *skb) int xfrm6_output(struct sock *sk, struct sk_buff *skb) { - return NF_HOOK(NFPROTO_IPV6, NF_INET_POST_ROUTING, skb, NULL, - skb_dst(skb)->dev, __xfrm6_output); + return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING, skb, + NULL, skb_dst(skb)->dev, __xfrm6_output, + !(IP6CB(skb)->flags & IP6SKB_REROUTED)); } diff --git a/net/ipv6/xfrm6_protocol.c b/net/ipv6/xfrm6_protocol.c index 6ab989c486f..54d13f8dbba 100644 --- a/net/ipv6/xfrm6_protocol.c +++ b/net/ipv6/xfrm6_protocol.c @@ -50,6 +50,10 @@ int xfrm6_rcv_cb(struct sk_buff *skb, u8 protocol, int err) { int ret; struct xfrm6_protocol *handler; + struct xfrm6_protocol __rcu **head = proto_handlers(protocol); + + if (!head) + return 0; for_each_protocol_rcu(*proto_handlers(protocol), handler) if ((ret = handler->cb_handler(skb, err)) <= 0) @@ -184,10 +188,12 @@ int xfrm6_protocol_register(struct xfrm6_protocol *handler, struct xfrm6_protocol __rcu **pprev; struct xfrm6_protocol *t; bool add_netproto = false; - int ret = -EEXIST; int priority = handler->priority; + if (!proto_handlers(protocol) || !netproto(protocol)) + return -EINVAL; + mutex_lock(&xfrm6_protocol_mutex); if (!rcu_dereference_protected(*proto_handlers(protocol), @@ -230,6 +236,9 @@ int xfrm6_protocol_deregister(struct xfrm6_protocol *handler, struct xfrm6_protocol *t; int ret = -ENOENT; + if (!proto_handlers(protocol) || !netproto(protocol)) + return -EINVAL; + mutex_lock(&xfrm6_protocol_mutex); for (pprev = proto_handlers(protocol); diff --git a/net/ipx/af_ipx.c b/net/ipx/af_ipx.c index 41e4e93cb3a..91729b807c7 100644 --- a/net/ipx/af_ipx.c +++ b/net/ipx/af_ipx.c @@ -1353,7 +1353,7 @@ static int ipx_create(struct net *net, struct socket *sock, int protocol, sk_refcnt_debug_inc(sk); sock_init_data(sock, sk); - sk->sk_no_check = 1; /* Checksum off by default */ + sk->sk_no_check_tx = 1; /* Checksum off by default */ sock->ops = &ipx_dgram_ops; rc = 0; out: diff --git a/net/ipx/ipx_route.c b/net/ipx/ipx_route.c index c1f03185c5e..67e7ad3d46b 100644 --- a/net/ipx/ipx_route.c +++ b/net/ipx/ipx_route.c @@ -236,7 +236,8 @@ int ipxrtr_route_packet(struct sock *sk, struct sockaddr_ipx *usipx, } /* Apply checksum. Not allowed on 802.3 links. */ - if (sk->sk_no_check || intrfc->if_dlink_type == htons(IPX_FRAME_8023)) + if (sk->sk_no_check_tx || + intrfc->if_dlink_type == htons(IPX_FRAME_8023)) ipx->ipx_checksum = htons(0xFFFF); else ipx->ipx_checksum = ipx_cksum(ipx, len + sizeof(struct ipxhdr)); diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index 01e77b0ae07..7a95fa4a3de 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -682,6 +682,18 @@ struct sock *iucv_accept_dequeue(struct sock *parent, struct socket *newsock) return NULL; } +static void __iucv_auto_name(struct iucv_sock *iucv) +{ + char name[12]; + + sprintf(name, "%08x", atomic_inc_return(&iucv_sk_list.autobind_name)); + while (__iucv_get_sock_by_name(name)) { + sprintf(name, "%08x", + atomic_inc_return(&iucv_sk_list.autobind_name)); + } + memcpy(iucv->src_name, name, 8); +} + /* Bind an unbound socket */ static int iucv_sock_bind(struct socket *sock, struct sockaddr *addr, int addr_len) @@ -724,8 +736,12 @@ static int iucv_sock_bind(struct socket *sock, struct sockaddr *addr, rcu_read_lock(); for_each_netdev_rcu(&init_net, dev) { if (!memcmp(dev->perm_addr, uid, 8)) { - memcpy(iucv->src_name, sa->siucv_name, 8); memcpy(iucv->src_user_id, sa->siucv_user_id, 8); + /* Check for unitialized siucv_name */ + if (strncmp(sa->siucv_name, " ", 8) == 0) + __iucv_auto_name(iucv); + else + memcpy(iucv->src_name, sa->siucv_name, 8); sk->sk_bound_dev_if = dev->ifindex; iucv->hs_dev = dev; dev_hold(dev); @@ -763,7 +779,6 @@ done: static int iucv_sock_autobind(struct sock *sk) { struct iucv_sock *iucv = iucv_sk(sk); - char name[12]; int err = 0; if (unlikely(!pr_iucv)) @@ -772,17 +787,9 @@ static int iucv_sock_autobind(struct sock *sk) memcpy(iucv->src_user_id, iucv_userid, 8); write_lock_bh(&iucv_sk_list.lock); - - sprintf(name, "%08x", atomic_inc_return(&iucv_sk_list.autobind_name)); - while (__iucv_get_sock_by_name(name)) { - sprintf(name, "%08x", - atomic_inc_return(&iucv_sk_list.autobind_name)); - } - + __iucv_auto_name(iucv); write_unlock_bh(&iucv_sk_list.lock); - memcpy(&iucv->src_name, name, 8); - if (!iucv->msglimit) iucv->msglimit = IUCV_QUEUELEN_DEFAULT; @@ -1830,7 +1837,7 @@ static void iucv_callback_txdone(struct iucv_path *path, spin_lock_irqsave(&list->lock, flags); while (list_skb != (struct sk_buff *)list) { - if (msg->tag != IUCV_SKB_CB(list_skb)->tag) { + if (msg->tag == IUCV_SKB_CB(list_skb)->tag) { this = list_skb; break; } @@ -1936,11 +1943,10 @@ static int afiucv_hs_callback_syn(struct sock *sk, struct sk_buff *skb) sk_acceptq_is_full(sk) || !nsk) { /* error on server socket - connection refused */ - if (nsk) - sk_free(nsk); afiucv_swap_src_dest(skb); trans_hdr->flags = AF_IUCV_FLAG_SYN | AF_IUCV_FLAG_FIN; err = dev_queue_xmit(skb); + iucv_sock_kill(nsk); bh_unlock_sock(sk); goto out; } diff --git a/net/key/af_key.c b/net/key/af_key.c index f3c83073afc..ba2a2f95911 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -1476,9 +1476,7 @@ static int pfkey_add(struct sock *sk, struct sk_buff *skb, const struct sadb_msg else err = xfrm_state_update(x); - xfrm_audit_state_add(x, err ? 0 : 1, - audit_get_loginuid(current), - audit_get_sessionid(current), 0); + xfrm_audit_state_add(x, err ? 0 : 1, true); if (err < 0) { x->km.state = XFRM_STATE_DEAD; @@ -1532,9 +1530,7 @@ static int pfkey_delete(struct sock *sk, struct sk_buff *skb, const struct sadb_ c.event = XFRM_MSG_DELSA; km_state_notify(x, &c); out: - xfrm_audit_state_delete(x, err ? 0 : 1, - audit_get_loginuid(current), - audit_get_sessionid(current), 0); + xfrm_audit_state_delete(x, err ? 0 : 1, true); xfrm_state_put(x); return err; @@ -1726,17 +1722,13 @@ static int pfkey_flush(struct sock *sk, struct sk_buff *skb, const struct sadb_m struct net *net = sock_net(sk); unsigned int proto; struct km_event c; - struct xfrm_audit audit_info; int err, err2; proto = pfkey_satype2proto(hdr->sadb_msg_satype); if (proto == 0) return -EINVAL; - audit_info.loginuid = audit_get_loginuid(current); - audit_info.sessionid = audit_get_sessionid(current); - audit_info.secid = 0; - err = xfrm_state_flush(net, proto, &audit_info); + err = xfrm_state_flush(net, proto, true); err2 = unicast_flush_resp(sk, hdr); if (err || err2) { if (err == -ESRCH) /* empty table - go quietly */ @@ -2288,9 +2280,7 @@ static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, const struct sadb_ err = xfrm_policy_insert(pol->sadb_x_policy_dir-1, xp, hdr->sadb_msg_type != SADB_X_SPDUPDATE); - xfrm_audit_policy_add(xp, err ? 0 : 1, - audit_get_loginuid(current), - audit_get_sessionid(current), 0); + xfrm_audit_policy_add(xp, err ? 0 : 1, true); if (err) goto out; @@ -2372,9 +2362,7 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, const struct sa if (xp == NULL) return -ENOENT; - xfrm_audit_policy_delete(xp, err ? 0 : 1, - audit_get_loginuid(current), - audit_get_sessionid(current), 0); + xfrm_audit_policy_delete(xp, err ? 0 : 1, true); if (err) goto out; @@ -2553,7 +2541,7 @@ static int pfkey_migrate(struct sock *sk, struct sk_buff *skb, sel.sport_mask = htons(0xffff); /* set destination address info of selector */ - sa = ext_hdrs[SADB_EXT_ADDRESS_DST - 1], + sa = ext_hdrs[SADB_EXT_ADDRESS_DST - 1]; pfkey_sadb_addr2xfrm_addr(sa, &sel.daddr); sel.prefixlen_d = sa->sadb_address_prefixlen; sel.proto = pfkey_proto_to_xfrm(sa->sadb_address_proto); @@ -2622,9 +2610,7 @@ static int pfkey_spdget(struct sock *sk, struct sk_buff *skb, const struct sadb_ return -ENOENT; if (delete) { - xfrm_audit_policy_delete(xp, err ? 0 : 1, - audit_get_loginuid(current), - audit_get_sessionid(current), 0); + xfrm_audit_policy_delete(xp, err ? 0 : 1, true); if (err) goto out; @@ -2733,13 +2719,9 @@ static int pfkey_spdflush(struct sock *sk, struct sk_buff *skb, const struct sad { struct net *net = sock_net(sk); struct km_event c; - struct xfrm_audit audit_info; int err, err2; - audit_info.loginuid = audit_get_loginuid(current); - audit_info.sessionid = audit_get_sessionid(current); - audit_info.secid = 0; - err = xfrm_policy_flush(net, XFRM_POLICY_TYPE_MAIN, &audit_info); + err = xfrm_policy_flush(net, XFRM_POLICY_TYPE_MAIN, true); err2 = unicast_flush_resp(sk, hdr); if (err || err2) { if (err == -ESRCH) /* empty table - old silent behavior */ diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index a4e37d7158d..bea25904320 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -495,52 +495,6 @@ out: spin_unlock_bh(&session->reorder_q.lock); } -static inline int l2tp_verify_udp_checksum(struct sock *sk, - struct sk_buff *skb) -{ - struct udphdr *uh = udp_hdr(skb); - u16 ulen = ntohs(uh->len); - __wsum psum; - - if (sk->sk_no_check || skb_csum_unnecessary(skb)) - return 0; - -#if IS_ENABLED(CONFIG_IPV6) - if (sk->sk_family == PF_INET6 && !l2tp_tunnel(sk)->v4mapped) { - if (!uh->check) { - LIMIT_NETDEBUG(KERN_INFO "L2TP: IPv6: checksum is 0\n"); - return 1; - } - if ((skb->ip_summed == CHECKSUM_COMPLETE) && - !csum_ipv6_magic(&ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, ulen, - IPPROTO_UDP, skb->csum)) { - skb->ip_summed = CHECKSUM_UNNECESSARY; - return 0; - } - skb->csum = ~csum_unfold(csum_ipv6_magic(&ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, - skb->len, IPPROTO_UDP, - 0)); - } else -#endif - { - struct inet_sock *inet; - if (!uh->check) - return 0; - inet = inet_sk(sk); - psum = csum_tcpudp_nofold(inet->inet_saddr, inet->inet_daddr, - ulen, IPPROTO_UDP, 0); - - if ((skb->ip_summed == CHECKSUM_COMPLETE) && - !csum_fold(csum_add(psum, skb->csum))) - return 0; - skb->csum = psum; - } - - return __skb_checksum_complete(skb); -} - static int l2tp_seq_check_rx_window(struct l2tp_session *session, u32 nr) { u32 nws; @@ -895,8 +849,7 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb, u16 version; int length; - if (tunnel->sock && l2tp_verify_udp_checksum(tunnel->sock, skb)) - goto discard_bad_csum; + /* UDP has verifed checksum */ /* UDP always verifies the packet length. */ __skb_pull(skb, sizeof(struct udphdr)); @@ -979,14 +932,6 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb, return 0; -discard_bad_csum: - LIMIT_NETDEBUG("%s: UDP: bad checksum\n", tunnel->name); - UDP_INC_STATS_USER(tunnel->l2tp_net, UDP_MIB_INERRORS, 0); - atomic_long_inc(&tunnel->stats.rx_errors); - kfree_skb(skb); - - return 0; - error: /* Put UDP header back */ __skb_push(skb, sizeof(struct udphdr)); @@ -1128,7 +1073,7 @@ static int l2tp_xmit_core(struct l2tp_session *session, struct sk_buff *skb, } /* Queue the packet to IP for output */ - skb->local_df = 1; + skb->ignore_df = 1; #if IS_ENABLED(CONFIG_IPV6) if (tunnel->sock->sk_family == PF_INET6 && !tunnel->v4mapped) error = inet6_csk_xmit(tunnel->sock, skb, NULL); @@ -1150,31 +1095,6 @@ static int l2tp_xmit_core(struct l2tp_session *session, struct sk_buff *skb, return 0; } -#if IS_ENABLED(CONFIG_IPV6) -static void l2tp_xmit_ipv6_csum(struct sock *sk, struct sk_buff *skb, - int udp_len) -{ - struct ipv6_pinfo *np = inet6_sk(sk); - struct udphdr *uh = udp_hdr(skb); - - if (!skb_dst(skb) || !skb_dst(skb)->dev || - !(skb_dst(skb)->dev->features & NETIF_F_IPV6_CSUM)) { - __wsum csum = skb_checksum(skb, 0, udp_len, 0); - skb->ip_summed = CHECKSUM_UNNECESSARY; - uh->check = csum_ipv6_magic(&np->saddr, &sk->sk_v6_daddr, udp_len, - IPPROTO_UDP, csum); - if (uh->check == 0) - uh->check = CSUM_MANGLED_0; - } else { - skb->ip_summed = CHECKSUM_PARTIAL; - skb->csum_start = skb_transport_header(skb) - skb->head; - skb->csum_offset = offsetof(struct udphdr, check); - uh->check = ~csum_ipv6_magic(&np->saddr, &sk->sk_v6_daddr, - udp_len, IPPROTO_UDP, 0); - } -} -#endif - /* If caller requires the skb to have a ppp header, the header must be * inserted in the skb data before calling this function. */ @@ -1186,7 +1106,6 @@ int l2tp_xmit_skb(struct l2tp_session *session, struct sk_buff *skb, int hdr_len struct flowi *fl; struct udphdr *uh; struct inet_sock *inet; - __wsum csum; int headroom; int uhlen = (tunnel->encap == L2TP_ENCAPTYPE_UDP) ? sizeof(struct udphdr) : 0; int udp_len; @@ -1235,33 +1154,17 @@ int l2tp_xmit_skb(struct l2tp_session *session, struct sk_buff *skb, int hdr_len uh->dest = inet->inet_dport; udp_len = uhlen + hdr_len + data_len; uh->len = htons(udp_len); - uh->check = 0; /* Calculate UDP checksum if configured to do so */ #if IS_ENABLED(CONFIG_IPV6) if (sk->sk_family == PF_INET6 && !tunnel->v4mapped) - l2tp_xmit_ipv6_csum(sk, skb, udp_len); + udp6_set_csum(udp_get_no_check6_tx(sk), + skb, &inet6_sk(sk)->saddr, + &sk->sk_v6_daddr, udp_len); else #endif - if (sk->sk_no_check == UDP_CSUM_NOXMIT) - skb->ip_summed = CHECKSUM_NONE; - else if ((skb_dst(skb) && skb_dst(skb)->dev) && - (!(skb_dst(skb)->dev->features & NETIF_F_V4_CSUM))) { - skb->ip_summed = CHECKSUM_COMPLETE; - csum = skb_checksum(skb, 0, udp_len, 0); - uh->check = csum_tcpudp_magic(inet->inet_saddr, - inet->inet_daddr, - udp_len, IPPROTO_UDP, csum); - if (uh->check == 0) - uh->check = CSUM_MANGLED_0; - } else { - skb->ip_summed = CHECKSUM_PARTIAL; - skb->csum_start = skb_transport_header(skb) - skb->head; - skb->csum_offset = offsetof(struct udphdr, check); - uh->check = ~csum_tcpudp_magic(inet->inet_saddr, - inet->inet_daddr, - udp_len, IPPROTO_UDP, 0); - } + udp_set_csum(sk->sk_no_check_tx, skb, inet->inet_saddr, + inet->inet_daddr, udp_len); break; case L2TP_ENCAPTYPE_IP: @@ -1490,6 +1393,11 @@ static int l2tp_tunnel_sock_create(struct net *net, sizeof(udp6_addr), 0); if (err < 0) goto out; + + if (cfg->udp6_zero_tx_checksums) + udp_set_no_check6_tx(sock->sk, true); + if (cfg->udp6_zero_rx_checksums) + udp_set_no_check6_rx(sock->sk, true); } else #endif { @@ -1518,7 +1426,7 @@ static int l2tp_tunnel_sock_create(struct net *net, } if (!cfg->use_udp_checksums) - sock->sk->sk_no_check = UDP_CSUM_NOXMIT; + sock->sk->sk_no_check_tx = 1; break; diff --git a/net/l2tp/l2tp_core.h b/net/l2tp/l2tp_core.h index 3f93ccd6ba9..68aa9ffd4ae 100644 --- a/net/l2tp/l2tp_core.h +++ b/net/l2tp/l2tp_core.h @@ -162,7 +162,9 @@ struct l2tp_tunnel_cfg { #endif u16 local_udp_port; u16 peer_udp_port; - unsigned int use_udp_checksums:1; + unsigned int use_udp_checksums:1, + udp6_zero_tx_checksums:1, + udp6_zero_rx_checksums:1; }; struct l2tp_tunnel { diff --git a/net/l2tp/l2tp_ip.c b/net/l2tp/l2tp_ip.c index 3397fe6897c..369a9822488 100644 --- a/net/l2tp/l2tp_ip.c +++ b/net/l2tp/l2tp_ip.c @@ -606,7 +606,6 @@ static struct inet_protosw l2tp_ip_protosw = { .protocol = IPPROTO_L2TP, .prot = &l2tp_ip_prot, .ops = &l2tp_ip_ops, - .no_check = 0, }; static struct net_protocol l2tp_ip_protocol __read_mostly = { diff --git a/net/l2tp/l2tp_ip6.c b/net/l2tp/l2tp_ip6.c index 7704ea9502f..f3f98a156ce 100644 --- a/net/l2tp/l2tp_ip6.c +++ b/net/l2tp/l2tp_ip6.c @@ -605,14 +605,8 @@ static int l2tp_ip6_sendmsg(struct kiocb *iocb, struct sock *sk, goto out; } - if (hlimit < 0) { - if (ipv6_addr_is_multicast(&fl6.daddr)) - hlimit = np->mcast_hops; - else - hlimit = np->hop_limit; - if (hlimit < 0) - hlimit = ip6_dst_hoplimit(dst); - } + if (hlimit < 0) + hlimit = ip6_sk_dst_hoplimit(np, &fl6, dst); if (tclass < 0) tclass = np->tclass; @@ -761,7 +755,6 @@ static struct inet_protosw l2tp_ip6_protosw = { .protocol = IPPROTO_L2TP, .prot = &l2tp_ip6_prot, .ops = &l2tp_ip6_ops, - .no_check = 0, }; static struct inet6_protocol l2tp_ip6_protocol __read_mostly = { diff --git a/net/l2tp/l2tp_netlink.c b/net/l2tp/l2tp_netlink.c index bd7387adea9..0ac907adb2f 100644 --- a/net/l2tp/l2tp_netlink.c +++ b/net/l2tp/l2tp_netlink.c @@ -161,6 +161,13 @@ static int l2tp_nl_cmd_tunnel_create(struct sk_buff *skb, struct genl_info *info cfg.peer_udp_port = nla_get_u16(info->attrs[L2TP_ATTR_UDP_DPORT]); if (info->attrs[L2TP_ATTR_UDP_CSUM]) cfg.use_udp_checksums = nla_get_flag(info->attrs[L2TP_ATTR_UDP_CSUM]); + +#if IS_ENABLED(CONFIG_IPV6) + if (info->attrs[L2TP_ATTR_UDP_ZERO_CSUM6_TX]) + cfg.udp6_zero_tx_checksums = nla_get_flag(info->attrs[L2TP_ATTR_UDP_ZERO_CSUM6_TX]); + if (info->attrs[L2TP_ATTR_UDP_ZERO_CSUM6_RX]) + cfg.udp6_zero_rx_checksums = nla_get_flag(info->attrs[L2TP_ATTR_UDP_ZERO_CSUM6_RX]); +#endif } if (info->attrs[L2TP_ATTR_DEBUG]) @@ -297,8 +304,7 @@ static int l2tp_nl_tunnel_send(struct sk_buff *skb, u32 portid, u32 seq, int fla case L2TP_ENCAPTYPE_UDP: if (nla_put_u16(skb, L2TP_ATTR_UDP_SPORT, ntohs(inet->inet_sport)) || nla_put_u16(skb, L2TP_ATTR_UDP_DPORT, ntohs(inet->inet_dport)) || - nla_put_u8(skb, L2TP_ATTR_UDP_CSUM, - (sk->sk_no_check != UDP_CSUM_NOXMIT))) + nla_put_u8(skb, L2TP_ATTR_UDP_CSUM, !sk->sk_no_check_tx)) goto nla_put_failure; /* NOBREAK */ case L2TP_ENCAPTYPE_IP: diff --git a/net/mac802154/Kconfig b/net/mac802154/Kconfig index b33dd76d430..1818a99b308 100644 --- a/net/mac802154/Kconfig +++ b/net/mac802154/Kconfig @@ -2,6 +2,10 @@ config MAC802154 tristate "Generic IEEE 802.15.4 Soft Networking Stack (mac802154)" depends on IEEE802154 select CRC_CCITT + select CRYPTO_AUTHENC + select CRYPTO_CCM + select CRYPTO_CTR + select CRYPTO_AES ---help--- This option enables the hardware independent IEEE 802.15.4 networking stack for SoftMAC devices (the ones implementing diff --git a/net/mac802154/Makefile b/net/mac802154/Makefile index 15d62df5218..9723d6f3f3e 100644 --- a/net/mac802154/Makefile +++ b/net/mac802154/Makefile @@ -1,4 +1,5 @@ obj-$(CONFIG_MAC802154) += mac802154.o -mac802154-objs := ieee802154_dev.o rx.o tx.o mac_cmd.o mib.o monitor.o wpan.o +mac802154-objs := ieee802154_dev.o rx.o tx.o mac_cmd.o mib.o \ + monitor.o wpan.o llsec.o ccflags-y += -D__CHECK_ENDIAN__ diff --git a/net/mac802154/llsec.c b/net/mac802154/llsec.c new file mode 100644 index 00000000000..e4a25589ec1 --- /dev/null +++ b/net/mac802154/llsec.c @@ -0,0 +1,1069 @@ +/* + * Copyright (C) 2014 Fraunhofer ITWM + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * Written by: + * Phoebe Buckheister <phoebe.buckheister@itwm.fraunhofer.de> + */ + +#include <linux/err.h> +#include <linux/bug.h> +#include <linux/completion.h> +#include <net/ieee802154.h> +#include <crypto/algapi.h> + +#include "mac802154.h" +#include "llsec.h" + +static void llsec_key_put(struct mac802154_llsec_key *key); +static bool llsec_key_id_equal(const struct ieee802154_llsec_key_id *a, + const struct ieee802154_llsec_key_id *b); + +static void llsec_dev_free(struct mac802154_llsec_device *dev); + +void mac802154_llsec_init(struct mac802154_llsec *sec) +{ + memset(sec, 0, sizeof(*sec)); + + memset(&sec->params.default_key_source, 0xFF, IEEE802154_ADDR_LEN); + + INIT_LIST_HEAD(&sec->table.security_levels); + INIT_LIST_HEAD(&sec->table.devices); + INIT_LIST_HEAD(&sec->table.keys); + hash_init(sec->devices_short); + hash_init(sec->devices_hw); + rwlock_init(&sec->lock); +} + +void mac802154_llsec_destroy(struct mac802154_llsec *sec) +{ + struct ieee802154_llsec_seclevel *sl, *sn; + struct ieee802154_llsec_device *dev, *dn; + struct ieee802154_llsec_key_entry *key, *kn; + + list_for_each_entry_safe(sl, sn, &sec->table.security_levels, list) { + struct mac802154_llsec_seclevel *msl; + + msl = container_of(sl, struct mac802154_llsec_seclevel, level); + list_del(&sl->list); + kfree(msl); + } + + list_for_each_entry_safe(dev, dn, &sec->table.devices, list) { + struct mac802154_llsec_device *mdev; + + mdev = container_of(dev, struct mac802154_llsec_device, dev); + list_del(&dev->list); + llsec_dev_free(mdev); + } + + list_for_each_entry_safe(key, kn, &sec->table.keys, list) { + struct mac802154_llsec_key *mkey; + + mkey = container_of(key->key, struct mac802154_llsec_key, key); + list_del(&key->list); + llsec_key_put(mkey); + kfree(key); + } +} + + + +int mac802154_llsec_get_params(struct mac802154_llsec *sec, + struct ieee802154_llsec_params *params) +{ + read_lock_bh(&sec->lock); + *params = sec->params; + read_unlock_bh(&sec->lock); + + return 0; +} + +int mac802154_llsec_set_params(struct mac802154_llsec *sec, + const struct ieee802154_llsec_params *params, + int changed) +{ + write_lock_bh(&sec->lock); + + if (changed & IEEE802154_LLSEC_PARAM_ENABLED) + sec->params.enabled = params->enabled; + if (changed & IEEE802154_LLSEC_PARAM_FRAME_COUNTER) + sec->params.frame_counter = params->frame_counter; + if (changed & IEEE802154_LLSEC_PARAM_OUT_LEVEL) + sec->params.out_level = params->out_level; + if (changed & IEEE802154_LLSEC_PARAM_OUT_KEY) + sec->params.out_key = params->out_key; + if (changed & IEEE802154_LLSEC_PARAM_KEY_SOURCE) + sec->params.default_key_source = params->default_key_source; + if (changed & IEEE802154_LLSEC_PARAM_PAN_ID) + sec->params.pan_id = params->pan_id; + if (changed & IEEE802154_LLSEC_PARAM_HWADDR) + sec->params.hwaddr = params->hwaddr; + if (changed & IEEE802154_LLSEC_PARAM_COORD_HWADDR) + sec->params.coord_hwaddr = params->coord_hwaddr; + if (changed & IEEE802154_LLSEC_PARAM_COORD_SHORTADDR) + sec->params.coord_shortaddr = params->coord_shortaddr; + + write_unlock_bh(&sec->lock); + + return 0; +} + + + +static struct mac802154_llsec_key* +llsec_key_alloc(const struct ieee802154_llsec_key *template) +{ + const int authsizes[3] = { 4, 8, 16 }; + struct mac802154_llsec_key *key; + int i; + + key = kzalloc(sizeof(*key), GFP_KERNEL); + if (!key) + return NULL; + + kref_init(&key->ref); + key->key = *template; + + BUILD_BUG_ON(ARRAY_SIZE(authsizes) != ARRAY_SIZE(key->tfm)); + + for (i = 0; i < ARRAY_SIZE(key->tfm); i++) { + key->tfm[i] = crypto_alloc_aead("ccm(aes)", 0, + CRYPTO_ALG_ASYNC); + if (!key->tfm[i]) + goto err_tfm; + if (crypto_aead_setkey(key->tfm[i], template->key, + IEEE802154_LLSEC_KEY_SIZE)) + goto err_tfm; + if (crypto_aead_setauthsize(key->tfm[i], authsizes[i])) + goto err_tfm; + } + + key->tfm0 = crypto_alloc_blkcipher("ctr(aes)", 0, CRYPTO_ALG_ASYNC); + if (!key->tfm0) + goto err_tfm; + + if (crypto_blkcipher_setkey(key->tfm0, template->key, + IEEE802154_LLSEC_KEY_SIZE)) + goto err_tfm0; + + return key; + +err_tfm0: + crypto_free_blkcipher(key->tfm0); +err_tfm: + for (i = 0; i < ARRAY_SIZE(key->tfm); i++) + if (key->tfm[i]) + crypto_free_aead(key->tfm[i]); + + kfree(key); + return NULL; +} + +static void llsec_key_release(struct kref *ref) +{ + struct mac802154_llsec_key *key; + int i; + + key = container_of(ref, struct mac802154_llsec_key, ref); + + for (i = 0; i < ARRAY_SIZE(key->tfm); i++) + crypto_free_aead(key->tfm[i]); + + crypto_free_blkcipher(key->tfm0); + kfree(key); +} + +static struct mac802154_llsec_key* +llsec_key_get(struct mac802154_llsec_key *key) +{ + kref_get(&key->ref); + return key; +} + +static void llsec_key_put(struct mac802154_llsec_key *key) +{ + kref_put(&key->ref, llsec_key_release); +} + +static bool llsec_key_id_equal(const struct ieee802154_llsec_key_id *a, + const struct ieee802154_llsec_key_id *b) +{ + if (a->mode != b->mode) + return false; + + if (a->mode == IEEE802154_SCF_KEY_IMPLICIT) + return ieee802154_addr_equal(&a->device_addr, &b->device_addr); + + if (a->id != b->id) + return false; + + switch (a->mode) { + case IEEE802154_SCF_KEY_INDEX: + return true; + case IEEE802154_SCF_KEY_SHORT_INDEX: + return a->short_source == b->short_source; + case IEEE802154_SCF_KEY_HW_INDEX: + return a->extended_source == b->extended_source; + } + + return false; +} + +int mac802154_llsec_key_add(struct mac802154_llsec *sec, + const struct ieee802154_llsec_key_id *id, + const struct ieee802154_llsec_key *key) +{ + struct mac802154_llsec_key *mkey = NULL; + struct ieee802154_llsec_key_entry *pos, *new; + + if (!(key->frame_types & (1 << IEEE802154_FC_TYPE_MAC_CMD)) && + key->cmd_frame_ids) + return -EINVAL; + + list_for_each_entry(pos, &sec->table.keys, list) { + if (llsec_key_id_equal(&pos->id, id)) + return -EEXIST; + + if (memcmp(pos->key->key, key->key, + IEEE802154_LLSEC_KEY_SIZE)) + continue; + + mkey = container_of(pos->key, struct mac802154_llsec_key, key); + + /* Don't allow multiple instances of the same AES key to have + * different allowed frame types/command frame ids, as this is + * not possible in the 802.15.4 PIB. + */ + if (pos->key->frame_types != key->frame_types || + pos->key->cmd_frame_ids != key->cmd_frame_ids) + return -EEXIST; + + break; + } + + new = kzalloc(sizeof(*new), GFP_KERNEL); + if (!new) + return -ENOMEM; + + if (!mkey) + mkey = llsec_key_alloc(key); + else + mkey = llsec_key_get(mkey); + + if (!mkey) + goto fail; + + new->id = *id; + new->key = &mkey->key; + + list_add_rcu(&new->list, &sec->table.keys); + + return 0; + +fail: + kfree(new); + return -ENOMEM; +} + +int mac802154_llsec_key_del(struct mac802154_llsec *sec, + const struct ieee802154_llsec_key_id *key) +{ + struct ieee802154_llsec_key_entry *pos; + + list_for_each_entry(pos, &sec->table.keys, list) { + struct mac802154_llsec_key *mkey; + + mkey = container_of(pos->key, struct mac802154_llsec_key, key); + + if (llsec_key_id_equal(&pos->id, key)) { + llsec_key_put(mkey); + return 0; + } + } + + return -ENOENT; +} + + + +static bool llsec_dev_use_shortaddr(__le16 short_addr) +{ + return short_addr != cpu_to_le16(IEEE802154_ADDR_UNDEF) && + short_addr != cpu_to_le16(0xffff); +} + +static u32 llsec_dev_hash_short(__le16 short_addr, __le16 pan_id) +{ + return ((__force u16) short_addr) << 16 | (__force u16) pan_id; +} + +static u64 llsec_dev_hash_long(__le64 hwaddr) +{ + return (__force u64) hwaddr; +} + +static struct mac802154_llsec_device* +llsec_dev_find_short(struct mac802154_llsec *sec, __le16 short_addr, + __le16 pan_id) +{ + struct mac802154_llsec_device *dev; + u32 key = llsec_dev_hash_short(short_addr, pan_id); + + hash_for_each_possible_rcu(sec->devices_short, dev, bucket_s, key) { + if (dev->dev.short_addr == short_addr && + dev->dev.pan_id == pan_id) + return dev; + } + + return NULL; +} + +static struct mac802154_llsec_device* +llsec_dev_find_long(struct mac802154_llsec *sec, __le64 hwaddr) +{ + struct mac802154_llsec_device *dev; + u64 key = llsec_dev_hash_long(hwaddr); + + hash_for_each_possible_rcu(sec->devices_hw, dev, bucket_hw, key) { + if (dev->dev.hwaddr == hwaddr) + return dev; + } + + return NULL; +} + +static void llsec_dev_free(struct mac802154_llsec_device *dev) +{ + struct ieee802154_llsec_device_key *pos, *pn; + struct mac802154_llsec_device_key *devkey; + + list_for_each_entry_safe(pos, pn, &dev->dev.keys, list) { + devkey = container_of(pos, struct mac802154_llsec_device_key, + devkey); + + list_del(&pos->list); + kfree(devkey); + } + + kfree(dev); +} + +int mac802154_llsec_dev_add(struct mac802154_llsec *sec, + const struct ieee802154_llsec_device *dev) +{ + struct mac802154_llsec_device *entry; + u32 skey = llsec_dev_hash_short(dev->short_addr, dev->pan_id); + u64 hwkey = llsec_dev_hash_long(dev->hwaddr); + + BUILD_BUG_ON(sizeof(hwkey) != IEEE802154_ADDR_LEN); + + if ((llsec_dev_use_shortaddr(dev->short_addr) && + llsec_dev_find_short(sec, dev->short_addr, dev->pan_id)) || + llsec_dev_find_long(sec, dev->hwaddr)) + return -EEXIST; + + entry = kmalloc(sizeof(*entry), GFP_KERNEL); + if (!entry) + return -ENOMEM; + + entry->dev = *dev; + spin_lock_init(&entry->lock); + INIT_LIST_HEAD(&entry->dev.keys); + + if (llsec_dev_use_shortaddr(dev->short_addr)) + hash_add_rcu(sec->devices_short, &entry->bucket_s, skey); + else + INIT_HLIST_NODE(&entry->bucket_s); + + hash_add_rcu(sec->devices_hw, &entry->bucket_hw, hwkey); + list_add_tail_rcu(&entry->dev.list, &sec->table.devices); + + return 0; +} + +static void llsec_dev_free_rcu(struct rcu_head *rcu) +{ + llsec_dev_free(container_of(rcu, struct mac802154_llsec_device, rcu)); +} + +int mac802154_llsec_dev_del(struct mac802154_llsec *sec, __le64 device_addr) +{ + struct mac802154_llsec_device *pos; + + pos = llsec_dev_find_long(sec, device_addr); + if (!pos) + return -ENOENT; + + hash_del_rcu(&pos->bucket_s); + hash_del_rcu(&pos->bucket_hw); + call_rcu(&pos->rcu, llsec_dev_free_rcu); + + return 0; +} + + + +static struct mac802154_llsec_device_key* +llsec_devkey_find(struct mac802154_llsec_device *dev, + const struct ieee802154_llsec_key_id *key) +{ + struct ieee802154_llsec_device_key *devkey; + + list_for_each_entry_rcu(devkey, &dev->dev.keys, list) { + if (!llsec_key_id_equal(key, &devkey->key_id)) + continue; + + return container_of(devkey, struct mac802154_llsec_device_key, + devkey); + } + + return NULL; +} + +int mac802154_llsec_devkey_add(struct mac802154_llsec *sec, + __le64 dev_addr, + const struct ieee802154_llsec_device_key *key) +{ + struct mac802154_llsec_device *dev; + struct mac802154_llsec_device_key *devkey; + + dev = llsec_dev_find_long(sec, dev_addr); + + if (!dev) + return -ENOENT; + + if (llsec_devkey_find(dev, &key->key_id)) + return -EEXIST; + + devkey = kmalloc(sizeof(*devkey), GFP_KERNEL); + if (!devkey) + return -ENOMEM; + + devkey->devkey = *key; + list_add_tail_rcu(&devkey->devkey.list, &dev->dev.keys); + return 0; +} + +int mac802154_llsec_devkey_del(struct mac802154_llsec *sec, + __le64 dev_addr, + const struct ieee802154_llsec_device_key *key) +{ + struct mac802154_llsec_device *dev; + struct mac802154_llsec_device_key *devkey; + + dev = llsec_dev_find_long(sec, dev_addr); + + if (!dev) + return -ENOENT; + + devkey = llsec_devkey_find(dev, &key->key_id); + if (!devkey) + return -ENOENT; + + list_del_rcu(&devkey->devkey.list); + kfree_rcu(devkey, rcu); + return 0; +} + + + +static struct mac802154_llsec_seclevel* +llsec_find_seclevel(const struct mac802154_llsec *sec, + const struct ieee802154_llsec_seclevel *sl) +{ + struct ieee802154_llsec_seclevel *pos; + + list_for_each_entry(pos, &sec->table.security_levels, list) { + if (pos->frame_type != sl->frame_type || + (pos->frame_type == IEEE802154_FC_TYPE_MAC_CMD && + pos->cmd_frame_id != sl->cmd_frame_id) || + pos->device_override != sl->device_override || + pos->sec_levels != sl->sec_levels) + continue; + + return container_of(pos, struct mac802154_llsec_seclevel, + level); + } + + return NULL; +} + +int mac802154_llsec_seclevel_add(struct mac802154_llsec *sec, + const struct ieee802154_llsec_seclevel *sl) +{ + struct mac802154_llsec_seclevel *entry; + + if (llsec_find_seclevel(sec, sl)) + return -EEXIST; + + entry = kmalloc(sizeof(*entry), GFP_KERNEL); + if (!entry) + return -ENOMEM; + + entry->level = *sl; + + list_add_tail_rcu(&entry->level.list, &sec->table.security_levels); + + return 0; +} + +int mac802154_llsec_seclevel_del(struct mac802154_llsec *sec, + const struct ieee802154_llsec_seclevel *sl) +{ + struct mac802154_llsec_seclevel *pos; + + pos = llsec_find_seclevel(sec, sl); + if (!pos) + return -ENOENT; + + list_del_rcu(&pos->level.list); + kfree_rcu(pos, rcu); + + return 0; +} + + + +static int llsec_recover_addr(struct mac802154_llsec *sec, + struct ieee802154_addr *addr) +{ + __le16 caddr = sec->params.coord_shortaddr; + addr->pan_id = sec->params.pan_id; + + if (caddr == cpu_to_le16(IEEE802154_ADDR_BROADCAST)) { + return -EINVAL; + } else if (caddr == cpu_to_le16(IEEE802154_ADDR_UNDEF)) { + addr->extended_addr = sec->params.coord_hwaddr; + addr->mode = IEEE802154_ADDR_LONG; + } else { + addr->short_addr = sec->params.coord_shortaddr; + addr->mode = IEEE802154_ADDR_SHORT; + } + + return 0; +} + +static struct mac802154_llsec_key* +llsec_lookup_key(struct mac802154_llsec *sec, + const struct ieee802154_hdr *hdr, + const struct ieee802154_addr *addr, + struct ieee802154_llsec_key_id *key_id) +{ + struct ieee802154_addr devaddr = *addr; + u8 key_id_mode = hdr->sec.key_id_mode; + struct ieee802154_llsec_key_entry *key_entry; + struct mac802154_llsec_key *key; + + if (key_id_mode == IEEE802154_SCF_KEY_IMPLICIT && + devaddr.mode == IEEE802154_ADDR_NONE) { + if (hdr->fc.type == IEEE802154_FC_TYPE_BEACON) { + devaddr.extended_addr = sec->params.coord_hwaddr; + devaddr.mode = IEEE802154_ADDR_LONG; + } else if (llsec_recover_addr(sec, &devaddr) < 0) { + return NULL; + } + } + + list_for_each_entry_rcu(key_entry, &sec->table.keys, list) { + const struct ieee802154_llsec_key_id *id = &key_entry->id; + + if (!(key_entry->key->frame_types & BIT(hdr->fc.type))) + continue; + + if (id->mode != key_id_mode) + continue; + + if (key_id_mode == IEEE802154_SCF_KEY_IMPLICIT) { + if (ieee802154_addr_equal(&devaddr, &id->device_addr)) + goto found; + } else { + if (id->id != hdr->sec.key_id) + continue; + + if ((key_id_mode == IEEE802154_SCF_KEY_INDEX) || + (key_id_mode == IEEE802154_SCF_KEY_SHORT_INDEX && + id->short_source == hdr->sec.short_src) || + (key_id_mode == IEEE802154_SCF_KEY_HW_INDEX && + id->extended_source == hdr->sec.extended_src)) + goto found; + } + } + + return NULL; + +found: + key = container_of(key_entry->key, struct mac802154_llsec_key, key); + if (key_id) + *key_id = key_entry->id; + return llsec_key_get(key); +} + + +static void llsec_geniv(u8 iv[16], __le64 addr, + const struct ieee802154_sechdr *sec) +{ + __be64 addr_bytes = (__force __be64) swab64((__force u64) addr); + __be32 frame_counter = (__force __be32) swab32((__force u32) sec->frame_counter); + + iv[0] = 1; /* L' = L - 1 = 1 */ + memcpy(iv + 1, &addr_bytes, sizeof(addr_bytes)); + memcpy(iv + 9, &frame_counter, sizeof(frame_counter)); + iv[13] = sec->level; + iv[14] = 0; + iv[15] = 1; +} + +static int +llsec_do_encrypt_unauth(struct sk_buff *skb, const struct mac802154_llsec *sec, + const struct ieee802154_hdr *hdr, + struct mac802154_llsec_key *key) +{ + u8 iv[16]; + struct scatterlist src; + struct blkcipher_desc req = { + .tfm = key->tfm0, + .info = iv, + .flags = 0, + }; + + llsec_geniv(iv, sec->params.hwaddr, &hdr->sec); + sg_init_one(&src, skb->data, skb->len); + return crypto_blkcipher_encrypt_iv(&req, &src, &src, skb->len); +} + +static struct crypto_aead* +llsec_tfm_by_len(struct mac802154_llsec_key *key, int authlen) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(key->tfm); i++) + if (crypto_aead_authsize(key->tfm[i]) == authlen) + return key->tfm[i]; + + BUG(); +} + +static int +llsec_do_encrypt_auth(struct sk_buff *skb, const struct mac802154_llsec *sec, + const struct ieee802154_hdr *hdr, + struct mac802154_llsec_key *key) +{ + u8 iv[16]; + unsigned char *data; + int authlen, assoclen, datalen, rc; + struct scatterlist src, assoc[2], dst[2]; + struct aead_request *req; + + authlen = ieee802154_sechdr_authtag_len(&hdr->sec); + llsec_geniv(iv, sec->params.hwaddr, &hdr->sec); + + req = aead_request_alloc(llsec_tfm_by_len(key, authlen), GFP_ATOMIC); + if (!req) + return -ENOMEM; + + sg_init_table(assoc, 2); + sg_set_buf(&assoc[0], skb_mac_header(skb), skb->mac_len); + assoclen = skb->mac_len; + + data = skb_mac_header(skb) + skb->mac_len; + datalen = skb_tail_pointer(skb) - data; + + if (hdr->sec.level & IEEE802154_SCF_SECLEVEL_ENC) { + sg_set_buf(&assoc[1], data, 0); + } else { + sg_set_buf(&assoc[1], data, datalen); + assoclen += datalen; + datalen = 0; + } + + sg_init_one(&src, data, datalen); + + sg_init_table(dst, 2); + sg_set_buf(&dst[0], data, datalen); + sg_set_buf(&dst[1], skb_put(skb, authlen), authlen); + + aead_request_set_callback(req, 0, NULL, NULL); + aead_request_set_assoc(req, assoc, assoclen); + aead_request_set_crypt(req, &src, dst, datalen, iv); + + rc = crypto_aead_encrypt(req); + + kfree(req); + + return rc; +} + +static int llsec_do_encrypt(struct sk_buff *skb, + const struct mac802154_llsec *sec, + const struct ieee802154_hdr *hdr, + struct mac802154_llsec_key *key) +{ + if (hdr->sec.level == IEEE802154_SCF_SECLEVEL_ENC) + return llsec_do_encrypt_unauth(skb, sec, hdr, key); + else + return llsec_do_encrypt_auth(skb, sec, hdr, key); +} + +int mac802154_llsec_encrypt(struct mac802154_llsec *sec, struct sk_buff *skb) +{ + struct ieee802154_hdr hdr; + int rc, authlen, hlen; + struct mac802154_llsec_key *key; + u32 frame_ctr; + + hlen = ieee802154_hdr_pull(skb, &hdr); + + if (hlen < 0 || hdr.fc.type != IEEE802154_FC_TYPE_DATA) + return -EINVAL; + + if (!hdr.fc.security_enabled || hdr.sec.level == 0) { + skb_push(skb, hlen); + return 0; + } + + authlen = ieee802154_sechdr_authtag_len(&hdr.sec); + + if (skb->len + hlen + authlen + IEEE802154_MFR_SIZE > IEEE802154_MTU) + return -EMSGSIZE; + + rcu_read_lock(); + + read_lock_bh(&sec->lock); + + if (!sec->params.enabled) { + rc = -EINVAL; + goto fail_read; + } + + key = llsec_lookup_key(sec, &hdr, &hdr.dest, NULL); + if (!key) { + rc = -ENOKEY; + goto fail_read; + } + + read_unlock_bh(&sec->lock); + + write_lock_bh(&sec->lock); + + frame_ctr = be32_to_cpu(sec->params.frame_counter); + hdr.sec.frame_counter = cpu_to_le32(frame_ctr); + if (frame_ctr == 0xFFFFFFFF) { + write_unlock_bh(&sec->lock); + llsec_key_put(key); + rc = -EOVERFLOW; + goto fail; + } + + sec->params.frame_counter = cpu_to_be32(frame_ctr + 1); + + write_unlock_bh(&sec->lock); + + rcu_read_unlock(); + + skb->mac_len = ieee802154_hdr_push(skb, &hdr); + skb_reset_mac_header(skb); + + rc = llsec_do_encrypt(skb, sec, &hdr, key); + llsec_key_put(key); + + return rc; + +fail_read: + read_unlock_bh(&sec->lock); +fail: + rcu_read_unlock(); + return rc; +} + + + +static struct mac802154_llsec_device* +llsec_lookup_dev(struct mac802154_llsec *sec, + const struct ieee802154_addr *addr) +{ + struct ieee802154_addr devaddr = *addr; + struct mac802154_llsec_device *dev = NULL; + + if (devaddr.mode == IEEE802154_ADDR_NONE && + llsec_recover_addr(sec, &devaddr) < 0) + return NULL; + + if (devaddr.mode == IEEE802154_ADDR_SHORT) { + u32 key = llsec_dev_hash_short(devaddr.short_addr, + devaddr.pan_id); + + hash_for_each_possible_rcu(sec->devices_short, dev, + bucket_s, key) { + if (dev->dev.pan_id == devaddr.pan_id && + dev->dev.short_addr == devaddr.short_addr) + return dev; + } + } else { + u64 key = llsec_dev_hash_long(devaddr.extended_addr); + + hash_for_each_possible_rcu(sec->devices_hw, dev, + bucket_hw, key) { + if (dev->dev.hwaddr == devaddr.extended_addr) + return dev; + } + } + + return NULL; +} + +static int +llsec_lookup_seclevel(const struct mac802154_llsec *sec, + u8 frame_type, u8 cmd_frame_id, + struct ieee802154_llsec_seclevel *rlevel) +{ + struct ieee802154_llsec_seclevel *level; + + list_for_each_entry_rcu(level, &sec->table.security_levels, list) { + if (level->frame_type == frame_type && + (frame_type != IEEE802154_FC_TYPE_MAC_CMD || + level->cmd_frame_id == cmd_frame_id)) { + *rlevel = *level; + return 0; + } + } + + return -EINVAL; +} + +static int +llsec_do_decrypt_unauth(struct sk_buff *skb, const struct mac802154_llsec *sec, + const struct ieee802154_hdr *hdr, + struct mac802154_llsec_key *key, __le64 dev_addr) +{ + u8 iv[16]; + unsigned char *data; + int datalen; + struct scatterlist src; + struct blkcipher_desc req = { + .tfm = key->tfm0, + .info = iv, + .flags = 0, + }; + + llsec_geniv(iv, dev_addr, &hdr->sec); + data = skb_mac_header(skb) + skb->mac_len; + datalen = skb_tail_pointer(skb) - data; + + sg_init_one(&src, data, datalen); + + return crypto_blkcipher_decrypt_iv(&req, &src, &src, datalen); +} + +static int +llsec_do_decrypt_auth(struct sk_buff *skb, const struct mac802154_llsec *sec, + const struct ieee802154_hdr *hdr, + struct mac802154_llsec_key *key, __le64 dev_addr) +{ + u8 iv[16]; + unsigned char *data; + int authlen, datalen, assoclen, rc; + struct scatterlist src, assoc[2]; + struct aead_request *req; + + authlen = ieee802154_sechdr_authtag_len(&hdr->sec); + llsec_geniv(iv, dev_addr, &hdr->sec); + + req = aead_request_alloc(llsec_tfm_by_len(key, authlen), GFP_ATOMIC); + if (!req) + return -ENOMEM; + + sg_init_table(assoc, 2); + sg_set_buf(&assoc[0], skb_mac_header(skb), skb->mac_len); + assoclen = skb->mac_len; + + data = skb_mac_header(skb) + skb->mac_len; + datalen = skb_tail_pointer(skb) - data; + + if (hdr->sec.level & IEEE802154_SCF_SECLEVEL_ENC) { + sg_set_buf(&assoc[1], data, 0); + } else { + sg_set_buf(&assoc[1], data, datalen - authlen); + assoclen += datalen - authlen; + data += datalen - authlen; + datalen = authlen; + } + + sg_init_one(&src, data, datalen); + + aead_request_set_callback(req, 0, NULL, NULL); + aead_request_set_assoc(req, assoc, assoclen); + aead_request_set_crypt(req, &src, &src, datalen, iv); + + rc = crypto_aead_decrypt(req); + + kfree(req); + skb_trim(skb, skb->len - authlen); + + return rc; +} + +static int +llsec_do_decrypt(struct sk_buff *skb, const struct mac802154_llsec *sec, + const struct ieee802154_hdr *hdr, + struct mac802154_llsec_key *key, __le64 dev_addr) +{ + if (hdr->sec.level == IEEE802154_SCF_SECLEVEL_ENC) + return llsec_do_decrypt_unauth(skb, sec, hdr, key, dev_addr); + else + return llsec_do_decrypt_auth(skb, sec, hdr, key, dev_addr); +} + +static int +llsec_update_devkey_record(struct mac802154_llsec_device *dev, + const struct ieee802154_llsec_key_id *in_key) +{ + struct mac802154_llsec_device_key *devkey; + + devkey = llsec_devkey_find(dev, in_key); + + if (!devkey) { + struct mac802154_llsec_device_key *next; + + next = kzalloc(sizeof(*devkey), GFP_ATOMIC); + if (!next) + return -ENOMEM; + + next->devkey.key_id = *in_key; + + spin_lock_bh(&dev->lock); + + devkey = llsec_devkey_find(dev, in_key); + if (!devkey) + list_add_rcu(&next->devkey.list, &dev->dev.keys); + else + kfree(next); + + spin_unlock_bh(&dev->lock); + } + + return 0; +} + +static int +llsec_update_devkey_info(struct mac802154_llsec_device *dev, + const struct ieee802154_llsec_key_id *in_key, + u32 frame_counter) +{ + struct mac802154_llsec_device_key *devkey = NULL; + + if (dev->dev.key_mode == IEEE802154_LLSEC_DEVKEY_RESTRICT) { + devkey = llsec_devkey_find(dev, in_key); + if (!devkey) + return -ENOENT; + } + + if (dev->dev.key_mode == IEEE802154_LLSEC_DEVKEY_RECORD) { + int rc = llsec_update_devkey_record(dev, in_key); + + if (rc < 0) + return rc; + } + + spin_lock_bh(&dev->lock); + + if ((!devkey && frame_counter < dev->dev.frame_counter) || + (devkey && frame_counter < devkey->devkey.frame_counter)) { + spin_unlock_bh(&dev->lock); + return -EINVAL; + } + + if (devkey) + devkey->devkey.frame_counter = frame_counter + 1; + else + dev->dev.frame_counter = frame_counter + 1; + + spin_unlock_bh(&dev->lock); + + return 0; +} + +int mac802154_llsec_decrypt(struct mac802154_llsec *sec, struct sk_buff *skb) +{ + struct ieee802154_hdr hdr; + struct mac802154_llsec_key *key; + struct ieee802154_llsec_key_id key_id; + struct mac802154_llsec_device *dev; + struct ieee802154_llsec_seclevel seclevel; + int err; + __le64 dev_addr; + u32 frame_ctr; + + if (ieee802154_hdr_peek(skb, &hdr) < 0) + return -EINVAL; + if (!hdr.fc.security_enabled) + return 0; + if (hdr.fc.version == 0) + return -EINVAL; + + read_lock_bh(&sec->lock); + if (!sec->params.enabled) { + read_unlock_bh(&sec->lock); + return -EINVAL; + } + read_unlock_bh(&sec->lock); + + rcu_read_lock(); + + key = llsec_lookup_key(sec, &hdr, &hdr.source, &key_id); + if (!key) { + err = -ENOKEY; + goto fail; + } + + dev = llsec_lookup_dev(sec, &hdr.source); + if (!dev) { + err = -EINVAL; + goto fail_dev; + } + + if (llsec_lookup_seclevel(sec, hdr.fc.type, 0, &seclevel) < 0) { + err = -EINVAL; + goto fail_dev; + } + + if (!(seclevel.sec_levels & BIT(hdr.sec.level)) && + (hdr.sec.level == 0 && seclevel.device_override && + !dev->dev.seclevel_exempt)) { + err = -EINVAL; + goto fail_dev; + } + + frame_ctr = le32_to_cpu(hdr.sec.frame_counter); + + if (frame_ctr == 0xffffffff) { + err = -EOVERFLOW; + goto fail_dev; + } + + err = llsec_update_devkey_info(dev, &key_id, frame_ctr); + if (err) + goto fail_dev; + + dev_addr = dev->dev.hwaddr; + + rcu_read_unlock(); + + err = llsec_do_decrypt(skb, sec, &hdr, key, dev_addr); + llsec_key_put(key); + return err; + +fail_dev: + llsec_key_put(key); +fail: + rcu_read_unlock(); + return err; +} diff --git a/net/mac802154/llsec.h b/net/mac802154/llsec.h new file mode 100644 index 00000000000..950578e1d7b --- /dev/null +++ b/net/mac802154/llsec.h @@ -0,0 +1,108 @@ +/* + * Copyright (C) 2014 Fraunhofer ITWM + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * Written by: + * Phoebe Buckheister <phoebe.buckheister@itwm.fraunhofer.de> + */ + +#ifndef MAC802154_LLSEC_H +#define MAC802154_LLSEC_H + +#include <linux/slab.h> +#include <linux/hashtable.h> +#include <linux/crypto.h> +#include <linux/kref.h> +#include <linux/spinlock.h> +#include <net/af_ieee802154.h> +#include <net/ieee802154_netdev.h> + +struct mac802154_llsec_key { + struct ieee802154_llsec_key key; + + /* one tfm for each authsize (4/8/16) */ + struct crypto_aead *tfm[3]; + struct crypto_blkcipher *tfm0; + + struct kref ref; +}; + +struct mac802154_llsec_device_key { + struct ieee802154_llsec_device_key devkey; + + struct rcu_head rcu; +}; + +struct mac802154_llsec_device { + struct ieee802154_llsec_device dev; + + struct hlist_node bucket_s; + struct hlist_node bucket_hw; + + /* protects dev.frame_counter and the elements of dev.keys */ + spinlock_t lock; + + struct rcu_head rcu; +}; + +struct mac802154_llsec_seclevel { + struct ieee802154_llsec_seclevel level; + + struct rcu_head rcu; +}; + +struct mac802154_llsec { + struct ieee802154_llsec_params params; + struct ieee802154_llsec_table table; + + DECLARE_HASHTABLE(devices_short, 6); + DECLARE_HASHTABLE(devices_hw, 6); + + /* protects params, all other fields are fine with RCU */ + rwlock_t lock; +}; + +void mac802154_llsec_init(struct mac802154_llsec *sec); +void mac802154_llsec_destroy(struct mac802154_llsec *sec); + +int mac802154_llsec_get_params(struct mac802154_llsec *sec, + struct ieee802154_llsec_params *params); +int mac802154_llsec_set_params(struct mac802154_llsec *sec, + const struct ieee802154_llsec_params *params, + int changed); + +int mac802154_llsec_key_add(struct mac802154_llsec *sec, + const struct ieee802154_llsec_key_id *id, + const struct ieee802154_llsec_key *key); +int mac802154_llsec_key_del(struct mac802154_llsec *sec, + const struct ieee802154_llsec_key_id *key); + +int mac802154_llsec_dev_add(struct mac802154_llsec *sec, + const struct ieee802154_llsec_device *dev); +int mac802154_llsec_dev_del(struct mac802154_llsec *sec, + __le64 device_addr); + +int mac802154_llsec_devkey_add(struct mac802154_llsec *sec, + __le64 dev_addr, + const struct ieee802154_llsec_device_key *key); +int mac802154_llsec_devkey_del(struct mac802154_llsec *sec, + __le64 dev_addr, + const struct ieee802154_llsec_device_key *key); + +int mac802154_llsec_seclevel_add(struct mac802154_llsec *sec, + const struct ieee802154_llsec_seclevel *sl); +int mac802154_llsec_seclevel_del(struct mac802154_llsec *sec, + const struct ieee802154_llsec_seclevel *sl); + +int mac802154_llsec_encrypt(struct mac802154_llsec *sec, struct sk_buff *skb); +int mac802154_llsec_decrypt(struct mac802154_llsec *sec, struct sk_buff *skb); + +#endif /* MAC802154_LLSEC_H */ diff --git a/net/mac802154/mac802154.h b/net/mac802154/mac802154.h index 28ef59c566e..762a6f849c6 100644 --- a/net/mac802154/mac802154.h +++ b/net/mac802154/mac802154.h @@ -23,8 +23,12 @@ #ifndef MAC802154_H #define MAC802154_H +#include <linux/mutex.h> +#include <net/mac802154.h> #include <net/ieee802154_netdev.h> +#include "llsec.h" + /* mac802154 device private data */ struct mac802154_priv { struct ieee802154_dev hw; @@ -90,6 +94,13 @@ struct mac802154_sub_if_data { u8 bsn; /* MAC DSN field */ u8 dsn; + + /* protects sec from concurrent access by netlink. access by + * encrypt/decrypt/header_create safe without additional protection. + */ + struct mutex sec_mtx; + + struct mac802154_llsec sec; }; #define mac802154_to_priv(_hw) container_of(_hw, struct mac802154_priv, hw) @@ -125,4 +136,37 @@ int mac802154_set_mac_params(struct net_device *dev, void mac802154_get_mac_params(struct net_device *dev, struct ieee802154_mac_params *params); +int mac802154_get_params(struct net_device *dev, + struct ieee802154_llsec_params *params); +int mac802154_set_params(struct net_device *dev, + const struct ieee802154_llsec_params *params, + int changed); + +int mac802154_add_key(struct net_device *dev, + const struct ieee802154_llsec_key_id *id, + const struct ieee802154_llsec_key *key); +int mac802154_del_key(struct net_device *dev, + const struct ieee802154_llsec_key_id *id); + +int mac802154_add_dev(struct net_device *dev, + const struct ieee802154_llsec_device *llsec_dev); +int mac802154_del_dev(struct net_device *dev, __le64 dev_addr); + +int mac802154_add_devkey(struct net_device *dev, + __le64 device_addr, + const struct ieee802154_llsec_device_key *key); +int mac802154_del_devkey(struct net_device *dev, + __le64 device_addr, + const struct ieee802154_llsec_device_key *key); + +int mac802154_add_seclevel(struct net_device *dev, + const struct ieee802154_llsec_seclevel *sl); +int mac802154_del_seclevel(struct net_device *dev, + const struct ieee802154_llsec_seclevel *sl); + +void mac802154_lock_table(struct net_device *dev); +void mac802154_get_table(struct net_device *dev, + struct ieee802154_llsec_table **t); +void mac802154_unlock_table(struct net_device *dev); + #endif /* MAC802154_H */ diff --git a/net/mac802154/mac_cmd.c b/net/mac802154/mac_cmd.c index d40c0928bc6..bf809131eef 100644 --- a/net/mac802154/mac_cmd.c +++ b/net/mac802154/mac_cmd.c @@ -40,6 +40,9 @@ static int mac802154_mlme_start_req(struct net_device *dev, u8 pan_coord, u8 blx, u8 coord_realign) { + struct ieee802154_mlme_ops *ops = ieee802154_mlme_ops(dev); + int rc = 0; + BUG_ON(addr->mode != IEEE802154_ADDR_SHORT); mac802154_dev_set_pan_id(dev, addr->pan_id); @@ -47,12 +50,31 @@ static int mac802154_mlme_start_req(struct net_device *dev, mac802154_dev_set_ieee_addr(dev); mac802154_dev_set_page_channel(dev, page, channel); + if (ops->llsec) { + struct ieee802154_llsec_params params; + int changed = 0; + + params.coord_shortaddr = addr->short_addr; + changed |= IEEE802154_LLSEC_PARAM_COORD_SHORTADDR; + + params.pan_id = addr->pan_id; + changed |= IEEE802154_LLSEC_PARAM_PAN_ID; + + params.hwaddr = ieee802154_devaddr_from_raw(dev->dev_addr); + changed |= IEEE802154_LLSEC_PARAM_HWADDR; + + params.coord_hwaddr = params.hwaddr; + changed |= IEEE802154_LLSEC_PARAM_COORD_HWADDR; + + rc = ops->llsec->set_params(dev, ¶ms, changed); + } + /* FIXME: add validation for unused parameters to be sane * for SoftMAC */ ieee802154_nl_start_confirm(dev, IEEE802154_SUCCESS); - return 0; + return rc; } static struct wpan_phy *mac802154_get_phy(const struct net_device *dev) @@ -64,6 +86,22 @@ static struct wpan_phy *mac802154_get_phy(const struct net_device *dev) return to_phy(get_device(&priv->hw->phy->dev)); } +static struct ieee802154_llsec_ops mac802154_llsec_ops = { + .get_params = mac802154_get_params, + .set_params = mac802154_set_params, + .add_key = mac802154_add_key, + .del_key = mac802154_del_key, + .add_dev = mac802154_add_dev, + .del_dev = mac802154_del_dev, + .add_devkey = mac802154_add_devkey, + .del_devkey = mac802154_del_devkey, + .add_seclevel = mac802154_add_seclevel, + .del_seclevel = mac802154_del_seclevel, + .lock_table = mac802154_lock_table, + .get_table = mac802154_get_table, + .unlock_table = mac802154_unlock_table, +}; + struct ieee802154_reduced_mlme_ops mac802154_mlme_reduced = { .get_phy = mac802154_get_phy, }; @@ -75,6 +113,8 @@ struct ieee802154_mlme_ops mac802154_mlme_wpan = { .get_short_addr = mac802154_dev_get_short_addr, .get_dsn = mac802154_dev_get_dsn, + .llsec = &mac802154_llsec_ops, + .set_mac_params = mac802154_set_mac_params, .get_mac_params = mac802154_get_mac_params, }; diff --git a/net/mac802154/mib.c b/net/mac802154/mib.c index f0991f2344d..15aa2f2b03a 100644 --- a/net/mac802154/mib.c +++ b/net/mac802154/mib.c @@ -213,3 +213,190 @@ void mac802154_dev_set_page_channel(struct net_device *dev, u8 page, u8 chan) } else mutex_unlock(&priv->hw->phy->pib_lock); } + + +int mac802154_get_params(struct net_device *dev, + struct ieee802154_llsec_params *params) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + int res; + + BUG_ON(dev->type != ARPHRD_IEEE802154); + + mutex_lock(&priv->sec_mtx); + res = mac802154_llsec_get_params(&priv->sec, params); + mutex_unlock(&priv->sec_mtx); + + return res; +} + +int mac802154_set_params(struct net_device *dev, + const struct ieee802154_llsec_params *params, + int changed) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + int res; + + BUG_ON(dev->type != ARPHRD_IEEE802154); + + mutex_lock(&priv->sec_mtx); + res = mac802154_llsec_set_params(&priv->sec, params, changed); + mutex_unlock(&priv->sec_mtx); + + return res; +} + + +int mac802154_add_key(struct net_device *dev, + const struct ieee802154_llsec_key_id *id, + const struct ieee802154_llsec_key *key) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + int res; + + BUG_ON(dev->type != ARPHRD_IEEE802154); + + mutex_lock(&priv->sec_mtx); + res = mac802154_llsec_key_add(&priv->sec, id, key); + mutex_unlock(&priv->sec_mtx); + + return res; +} + +int mac802154_del_key(struct net_device *dev, + const struct ieee802154_llsec_key_id *id) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + int res; + + BUG_ON(dev->type != ARPHRD_IEEE802154); + + mutex_lock(&priv->sec_mtx); + res = mac802154_llsec_key_del(&priv->sec, id); + mutex_unlock(&priv->sec_mtx); + + return res; +} + + +int mac802154_add_dev(struct net_device *dev, + const struct ieee802154_llsec_device *llsec_dev) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + int res; + + BUG_ON(dev->type != ARPHRD_IEEE802154); + + mutex_lock(&priv->sec_mtx); + res = mac802154_llsec_dev_add(&priv->sec, llsec_dev); + mutex_unlock(&priv->sec_mtx); + + return res; +} + +int mac802154_del_dev(struct net_device *dev, __le64 dev_addr) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + int res; + + BUG_ON(dev->type != ARPHRD_IEEE802154); + + mutex_lock(&priv->sec_mtx); + res = mac802154_llsec_dev_del(&priv->sec, dev_addr); + mutex_unlock(&priv->sec_mtx); + + return res; +} + + +int mac802154_add_devkey(struct net_device *dev, + __le64 device_addr, + const struct ieee802154_llsec_device_key *key) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + int res; + + BUG_ON(dev->type != ARPHRD_IEEE802154); + + mutex_lock(&priv->sec_mtx); + res = mac802154_llsec_devkey_add(&priv->sec, device_addr, key); + mutex_unlock(&priv->sec_mtx); + + return res; +} + +int mac802154_del_devkey(struct net_device *dev, + __le64 device_addr, + const struct ieee802154_llsec_device_key *key) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + int res; + + BUG_ON(dev->type != ARPHRD_IEEE802154); + + mutex_lock(&priv->sec_mtx); + res = mac802154_llsec_devkey_del(&priv->sec, device_addr, key); + mutex_unlock(&priv->sec_mtx); + + return res; +} + + +int mac802154_add_seclevel(struct net_device *dev, + const struct ieee802154_llsec_seclevel *sl) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + int res; + + BUG_ON(dev->type != ARPHRD_IEEE802154); + + mutex_lock(&priv->sec_mtx); + res = mac802154_llsec_seclevel_add(&priv->sec, sl); + mutex_unlock(&priv->sec_mtx); + + return res; +} + +int mac802154_del_seclevel(struct net_device *dev, + const struct ieee802154_llsec_seclevel *sl) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + int res; + + BUG_ON(dev->type != ARPHRD_IEEE802154); + + mutex_lock(&priv->sec_mtx); + res = mac802154_llsec_seclevel_del(&priv->sec, sl); + mutex_unlock(&priv->sec_mtx); + + return res; +} + + +void mac802154_lock_table(struct net_device *dev) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + + BUG_ON(dev->type != ARPHRD_IEEE802154); + + mutex_lock(&priv->sec_mtx); +} + +void mac802154_get_table(struct net_device *dev, + struct ieee802154_llsec_table **t) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + + BUG_ON(dev->type != ARPHRD_IEEE802154); + + *t = &priv->sec.table; +} + +void mac802154_unlock_table(struct net_device *dev) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + + BUG_ON(dev->type != ARPHRD_IEEE802154); + + mutex_unlock(&priv->sec_mtx); +} diff --git a/net/mac802154/rx.c b/net/mac802154/rx.c index 03855b0677c..0597b96dc9b 100644 --- a/net/mac802154/rx.c +++ b/net/mac802154/rx.c @@ -59,8 +59,6 @@ mac802154_subif_rx(struct ieee802154_dev *hw, struct sk_buff *skb, u8 lqi) skb->protocol = htons(ETH_P_IEEE802154); skb_reset_mac_header(skb); - BUILD_BUG_ON(sizeof(struct ieee802154_mac_cb) > sizeof(skb->cb)); - if (!(priv->hw.flags & IEEE802154_HW_OMIT_CKSUM)) { u16 crc; diff --git a/net/mac802154/wpan.c b/net/mac802154/wpan.c index 1df7a6a5738..23bc91cf99c 100644 --- a/net/mac802154/wpan.c +++ b/net/mac802154/wpan.c @@ -35,6 +35,28 @@ #include "mac802154.h" +static int mac802154_wpan_update_llsec(struct net_device *dev) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + struct ieee802154_mlme_ops *ops = ieee802154_mlme_ops(dev); + int rc = 0; + + if (ops->llsec) { + struct ieee802154_llsec_params params; + int changed = 0; + + params.pan_id = priv->pan_id; + changed |= IEEE802154_LLSEC_PARAM_PAN_ID; + + params.hwaddr = priv->extended_addr; + changed |= IEEE802154_LLSEC_PARAM_HWADDR; + + rc = ops->llsec->set_params(dev, ¶ms, changed); + } + + return rc; +} + static int mac802154_wpan_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { @@ -81,7 +103,7 @@ mac802154_wpan_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) priv->pan_id = cpu_to_le16(sa->addr.pan_id); priv->short_addr = cpu_to_le16(sa->addr.short_addr); - err = 0; + err = mac802154_wpan_update_llsec(dev); break; } @@ -99,7 +121,7 @@ static int mac802154_wpan_mac_addr(struct net_device *dev, void *p) /* FIXME: validate addr */ memcpy(dev->dev_addr, addr->sa_data, dev->addr_len); mac802154_dev_set_ieee_addr(dev); - return 0; + return mac802154_wpan_update_llsec(dev); } int mac802154_set_mac_params(struct net_device *dev, @@ -124,7 +146,7 @@ void mac802154_get_mac_params(struct net_device *dev, mutex_unlock(&priv->hw->slaves_mtx); } -int mac802154_wpan_open(struct net_device *dev) +static int mac802154_wpan_open(struct net_device *dev) { int rc; struct mac802154_sub_if_data *priv = netdev_priv(dev); @@ -183,6 +205,38 @@ out: return rc; } +static int mac802154_set_header_security(struct mac802154_sub_if_data *priv, + struct ieee802154_hdr *hdr, + const struct ieee802154_mac_cb *cb) +{ + struct ieee802154_llsec_params params; + u8 level; + + mac802154_llsec_get_params(&priv->sec, ¶ms); + + if (!params.enabled && cb->secen_override && cb->secen) + return -EINVAL; + if (!params.enabled || + (cb->secen_override && !cb->secen) || + !params.out_level) + return 0; + if (cb->seclevel_override && !cb->seclevel) + return -EINVAL; + + level = cb->seclevel_override ? cb->seclevel : params.out_level; + + hdr->fc.security_enabled = 1; + hdr->sec.level = level; + hdr->sec.key_id_mode = params.out_key.mode; + if (params.out_key.mode == IEEE802154_SCF_KEY_SHORT_INDEX) + hdr->sec.short_src = params.out_key.short_source; + else if (params.out_key.mode == IEEE802154_SCF_KEY_HW_INDEX) + hdr->sec.extended_src = params.out_key.extended_source; + hdr->sec.key_id = params.out_key.id; + + return 0; +} + static int mac802154_header_create(struct sk_buff *skb, struct net_device *dev, unsigned short type, @@ -192,15 +246,20 @@ static int mac802154_header_create(struct sk_buff *skb, { struct ieee802154_hdr hdr; struct mac802154_sub_if_data *priv = netdev_priv(dev); + struct ieee802154_mac_cb *cb = mac_cb(skb); int hlen; if (!daddr) return -EINVAL; memset(&hdr.fc, 0, sizeof(hdr.fc)); - hdr.fc.type = mac_cb_type(skb); - hdr.fc.security_enabled = mac_cb_is_secen(skb); - hdr.fc.ack_request = mac_cb_is_ackreq(skb); + hdr.fc.type = cb->type; + hdr.fc.security_enabled = cb->secen; + hdr.fc.ack_request = cb->ackreq; + hdr.seq = ieee802154_mlme_ops(dev)->get_dsn(dev); + + if (mac802154_set_header_security(priv, &hdr, cb) < 0) + return -EINVAL; if (!saddr) { spin_lock_bh(&priv->mib_lock); @@ -231,7 +290,7 @@ static int mac802154_header_create(struct sk_buff *skb, skb_reset_mac_header(skb); skb->mac_len = hlen; - if (hlen + len + 2 > dev->mtu) + if (len > ieee802154_max_payload(&hdr)) return -EMSGSIZE; return hlen; @@ -257,6 +316,7 @@ mac802154_wpan_xmit(struct sk_buff *skb, struct net_device *dev) { struct mac802154_sub_if_data *priv; u8 chan, page; + int rc; priv = netdev_priv(dev); @@ -272,6 +332,13 @@ mac802154_wpan_xmit(struct sk_buff *skb, struct net_device *dev) return NETDEV_TX_OK; } + rc = mac802154_llsec_encrypt(&priv->sec, skb); + if (rc) { + pr_warn("encryption failed: %i\n", rc); + kfree_skb(skb); + return NETDEV_TX_OK; + } + skb->skb_iif = dev->ifindex; dev->stats.tx_packets++; dev->stats.tx_bytes += skb->len; @@ -292,6 +359,15 @@ static const struct net_device_ops mac802154_wpan_ops = { .ndo_set_mac_address = mac802154_wpan_mac_addr, }; +static void mac802154_wpan_free(struct net_device *dev) +{ + struct mac802154_sub_if_data *priv = netdev_priv(dev); + + mac802154_llsec_destroy(&priv->sec); + + free_netdev(dev); +} + void mac802154_wpan_setup(struct net_device *dev) { struct mac802154_sub_if_data *priv; @@ -301,14 +377,14 @@ void mac802154_wpan_setup(struct net_device *dev) dev->hard_header_len = MAC802154_FRAME_HARD_HEADER_LEN; dev->header_ops = &mac802154_header_ops; - dev->needed_tailroom = 2; /* FCS */ + dev->needed_tailroom = 2 + 16; /* FCS + MIC */ dev->mtu = IEEE802154_MTU; dev->tx_queue_len = 300; dev->type = ARPHRD_IEEE802154; dev->flags = IFF_NOARP | IFF_BROADCAST; dev->watchdog_timeo = 0; - dev->destructor = free_netdev; + dev->destructor = mac802154_wpan_free; dev->netdev_ops = &mac802154_wpan_ops; dev->ml_priv = &mac802154_mlme_wpan; @@ -319,6 +395,7 @@ void mac802154_wpan_setup(struct net_device *dev) priv->page = 0; spin_lock_init(&priv->mib_lock); + mutex_init(&priv->sec_mtx); get_random_bytes(&priv->bsn, 1); get_random_bytes(&priv->dsn, 1); @@ -331,6 +408,8 @@ void mac802154_wpan_setup(struct net_device *dev) priv->pan_id = cpu_to_le16(IEEE802154_PANID_BROADCAST); priv->short_addr = cpu_to_le16(IEEE802154_ADDR_BROADCAST); + + mac802154_llsec_init(&priv->sec); } static int mac802154_process_data(struct net_device *dev, struct sk_buff *skb) @@ -339,9 +418,11 @@ static int mac802154_process_data(struct net_device *dev, struct sk_buff *skb) } static int -mac802154_subif_frame(struct mac802154_sub_if_data *sdata, struct sk_buff *skb) +mac802154_subif_frame(struct mac802154_sub_if_data *sdata, struct sk_buff *skb, + const struct ieee802154_hdr *hdr) { __le16 span, sshort; + int rc; pr_debug("getting packet via slave interface %s\n", sdata->dev->name); @@ -388,15 +469,21 @@ mac802154_subif_frame(struct mac802154_sub_if_data *sdata, struct sk_buff *skb) skb->dev = sdata->dev; + rc = mac802154_llsec_decrypt(&sdata->sec, skb); + if (rc) { + pr_debug("decryption failed: %i\n", rc); + return NET_RX_DROP; + } + sdata->dev->stats.rx_packets++; sdata->dev->stats.rx_bytes += skb->len; - switch (mac_cb_type(skb)) { + switch (mac_cb(skb)->type) { case IEEE802154_FC_TYPE_DATA: return mac802154_process_data(sdata->dev, skb); default: pr_warn("ieee802154: bad frame received (type = %d)\n", - mac_cb_type(skb)); + mac_cb(skb)->type); kfree_skb(skb); return NET_RX_DROP; } @@ -419,62 +506,58 @@ static void mac802154_print_addr(const char *name, } } -static int mac802154_parse_frame_start(struct sk_buff *skb) +static int mac802154_parse_frame_start(struct sk_buff *skb, + struct ieee802154_hdr *hdr) { int hlen; - struct ieee802154_hdr hdr; + struct ieee802154_mac_cb *cb = mac_cb_init(skb); - hlen = ieee802154_hdr_pull(skb, &hdr); + hlen = ieee802154_hdr_pull(skb, hdr); if (hlen < 0) return -EINVAL; skb->mac_len = hlen; - pr_debug("fc: %04x dsn: %02x\n", le16_to_cpup((__le16 *)&hdr.fc), - hdr.seq); - - mac_cb(skb)->flags = hdr.fc.type; + pr_debug("fc: %04x dsn: %02x\n", le16_to_cpup((__le16 *)&hdr->fc), + hdr->seq); - if (hdr.fc.ack_request) - mac_cb(skb)->flags |= MAC_CB_FLAG_ACKREQ; - if (hdr.fc.security_enabled) - mac_cb(skb)->flags |= MAC_CB_FLAG_SECEN; + cb->type = hdr->fc.type; + cb->ackreq = hdr->fc.ack_request; + cb->secen = hdr->fc.security_enabled; - mac802154_print_addr("destination", &hdr.dest); - mac802154_print_addr("source", &hdr.source); + mac802154_print_addr("destination", &hdr->dest); + mac802154_print_addr("source", &hdr->source); - mac_cb(skb)->source = hdr.source; - mac_cb(skb)->dest = hdr.dest; + cb->source = hdr->source; + cb->dest = hdr->dest; - if (hdr.fc.security_enabled) { + if (hdr->fc.security_enabled) { u64 key; - pr_debug("seclevel %i\n", hdr.sec.level); + pr_debug("seclevel %i\n", hdr->sec.level); - switch (hdr.sec.key_id_mode) { + switch (hdr->sec.key_id_mode) { case IEEE802154_SCF_KEY_IMPLICIT: pr_debug("implicit key\n"); break; case IEEE802154_SCF_KEY_INDEX: - pr_debug("key %02x\n", hdr.sec.key_id); + pr_debug("key %02x\n", hdr->sec.key_id); break; case IEEE802154_SCF_KEY_SHORT_INDEX: pr_debug("key %04x:%04x %02x\n", - le32_to_cpu(hdr.sec.short_src) >> 16, - le32_to_cpu(hdr.sec.short_src) & 0xffff, - hdr.sec.key_id); + le32_to_cpu(hdr->sec.short_src) >> 16, + le32_to_cpu(hdr->sec.short_src) & 0xffff, + hdr->sec.key_id); break; case IEEE802154_SCF_KEY_HW_INDEX: - key = swab64((__force u64) hdr.sec.extended_src); + key = swab64((__force u64) hdr->sec.extended_src); pr_debug("key source %8phC %02x\n", &key, - hdr.sec.key_id); + hdr->sec.key_id); break; } - - return -EINVAL; } return 0; @@ -485,8 +568,9 @@ void mac802154_wpans_rx(struct mac802154_priv *priv, struct sk_buff *skb) int ret; struct sk_buff *sskb; struct mac802154_sub_if_data *sdata; + struct ieee802154_hdr hdr; - ret = mac802154_parse_frame_start(skb); + ret = mac802154_parse_frame_start(skb, &hdr); if (ret) { pr_debug("got invalid frame\n"); return; @@ -499,7 +583,7 @@ void mac802154_wpans_rx(struct mac802154_priv *priv, struct sk_buff *skb) sskb = skb_clone(skb, GFP_ATOMIC); if (sskb) - mac802154_subif_frame(sdata, sskb); + mac802154_subif_frame(sdata, sskb, &hdr); } rcu_read_unlock(); } diff --git a/net/mpls/mpls_gso.c b/net/mpls/mpls_gso.c index 851cd880b0c..6b38d083e1c 100644 --- a/net/mpls/mpls_gso.c +++ b/net/mpls/mpls_gso.c @@ -33,6 +33,7 @@ static struct sk_buff *mpls_gso_segment(struct sk_buff *skb, SKB_GSO_DODGY | SKB_GSO_TCP_ECN | SKB_GSO_GRE | + SKB_GSO_GRE_CSUM | SKB_GSO_IPIP | SKB_GSO_MPLS))) goto out; diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c index 117208321f1..ec8114fae50 100644 --- a/net/netfilter/ipset/ip_set_core.c +++ b/net/netfilter/ipset/ip_set_core.c @@ -271,10 +271,7 @@ ip_set_free(void *members) { pr_debug("%p: free with %s\n", members, is_vmalloc_addr(members) ? "vfree" : "kfree"); - if (is_vmalloc_addr(members)) - vfree(members); - else - kfree(members); + kvfree(members); } EXPORT_SYMBOL_GPL(ip_set_free); diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 4f26ee46b51..e6836755c45 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -97,7 +97,7 @@ const char *ip_vs_proto_name(unsigned int proto) return "ICMPv6"; #endif default: - sprintf(buf, "IP_%d", proto); + sprintf(buf, "IP_%u", proto); return buf; } } @@ -1392,15 +1392,19 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum) if (ipip) { __be32 info = ic->un.gateway; + __u8 type = ic->type; + __u8 code = ic->code; /* Update the MTU */ if (ic->type == ICMP_DEST_UNREACH && ic->code == ICMP_FRAG_NEEDED) { struct ip_vs_dest *dest = cp->dest; u32 mtu = ntohs(ic->un.frag.mtu); + __be16 frag_off = cih->frag_off; /* Strip outer IP and ICMP, go to IPIP header */ - __skb_pull(skb, ihl + sizeof(_icmph)); + if (pskb_pull(skb, ihl + sizeof(_icmph)) == NULL) + goto ignore_ipip; offset2 -= ihl + sizeof(_icmph); skb_reset_network_header(skb); IP_VS_DBG(12, "ICMP for IPIP %pI4->%pI4: mtu=%u\n", @@ -1408,7 +1412,7 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum) ipv4_update_pmtu(skb, dev_net(skb->dev), mtu, 0, 0, 0, 0); /* Client uses PMTUD? */ - if (!(cih->frag_off & htons(IP_DF))) + if (!(frag_off & htons(IP_DF))) goto ignore_ipip; /* Prefer the resulting PMTU */ if (dest) { @@ -1427,12 +1431,13 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum) /* Strip outer IP, ICMP and IPIP, go to IP header of * original request. */ - __skb_pull(skb, offset2); + if (pskb_pull(skb, offset2) == NULL) + goto ignore_ipip; skb_reset_network_header(skb); IP_VS_DBG(12, "Sending ICMP for %pI4->%pI4: t=%u, c=%u, i=%u\n", &ip_hdr(skb)->saddr, &ip_hdr(skb)->daddr, - ic->type, ic->code, ntohl(info)); - icmp_send(skb, ic->type, ic->code, info); + type, code, ntohl(info)); + icmp_send(skb, type, code, info); /* ICMP can be shorter but anyways, account it */ ip_vs_out_stats(cp, skb); diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index c47444e4cf8..73ba1cc7a88 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -562,7 +562,7 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ip_send_check(iph); /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; ip_vs_send_or_cont(NFPROTO_IPV4, skb, cp, 0); rcu_read_unlock(); @@ -590,7 +590,7 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, goto tx_error; /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; ip_vs_send_or_cont(NFPROTO_IPV6, skb, cp, 0); rcu_read_unlock(); @@ -684,7 +684,7 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, MTU problem. */ /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; rc = ip_vs_nat_send_or_cont(NFPROTO_IPV4, skb, cp, local); rcu_read_unlock(); @@ -774,7 +774,7 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, MTU problem. */ /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; rc = ip_vs_nat_send_or_cont(NFPROTO_IPV6, skb, cp, local); rcu_read_unlock(); @@ -883,10 +883,10 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, iph->daddr = cp->daddr.ip; iph->saddr = saddr; iph->ttl = old_iph->ttl; - ip_select_ident(skb, &rt->dst, NULL); + ip_select_ident(skb, NULL); /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; ret = ip_vs_tunnel_xmit_prepare(skb, cp); if (ret == NF_ACCEPT) @@ -974,7 +974,7 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, iph->hop_limit = old_iph->hop_limit; /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; ret = ip_vs_tunnel_xmit_prepare(skb, cp); if (ret == NF_ACCEPT) @@ -1023,7 +1023,7 @@ ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ip_send_check(ip_hdr(skb)); /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; ip_vs_send_or_cont(NFPROTO_IPV4, skb, cp, 0); rcu_read_unlock(); @@ -1060,7 +1060,7 @@ ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, } /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; ip_vs_send_or_cont(NFPROTO_IPV6, skb, cp, 0); rcu_read_unlock(); @@ -1157,7 +1157,7 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ip_vs_nat_icmp(skb, pp, cp, 0); /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; rc = ip_vs_nat_send_or_cont(NFPROTO_IPV4, skb, cp, local); rcu_read_unlock(); @@ -1249,7 +1249,7 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, ip_vs_nat_icmp_v6(skb, pp, cp, 0); /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; rc = ip_vs_nat_send_or_cont(NFPROTO_IPV6, skb, cp, local); rcu_read_unlock(); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index ccc46fa5edb..58579634427 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -1336,6 +1336,9 @@ ctnetlink_setup_nat(struct nf_conn *ct, const struct nlattr * const cda[]) #ifdef CONFIG_NF_NAT_NEEDED int ret; + if (!cda[CTA_NAT_DST] && !cda[CTA_NAT_SRC]) + return 0; + ret = ctnetlink_parse_nat_setup(ct, NF_NAT_MANIP_DST, cda[CTA_NAT_DST]); if (ret < 0) diff --git a/net/netfilter/nf_nat_core.c b/net/netfilter/nf_nat_core.c index 52ca952b802..09096a670c4 100644 --- a/net/netfilter/nf_nat_core.c +++ b/net/netfilter/nf_nat_core.c @@ -358,6 +358,19 @@ out: rcu_read_unlock(); } +struct nf_conn_nat *nf_ct_nat_ext_add(struct nf_conn *ct) +{ + struct nf_conn_nat *nat = nfct_nat(ct); + if (nat) + return nat; + + if (!nf_ct_is_confirmed(ct)) + nat = nf_ct_ext_add(ct, NF_CT_EXT_NAT, GFP_ATOMIC); + + return nat; +} +EXPORT_SYMBOL_GPL(nf_ct_nat_ext_add); + unsigned int nf_nat_setup_info(struct nf_conn *ct, const struct nf_nat_range *range, @@ -368,14 +381,9 @@ nf_nat_setup_info(struct nf_conn *ct, struct nf_conn_nat *nat; /* nat helper or nfctnetlink also setup binding */ - nat = nfct_nat(ct); - if (!nat) { - nat = nf_ct_ext_add(ct, NF_CT_EXT_NAT, GFP_ATOMIC); - if (nat == NULL) { - pr_debug("failed to add NAT extension\n"); - return NF_ACCEPT; - } - } + nat = nf_ct_nat_ext_add(ct); + if (nat == NULL) + return NF_ACCEPT; NF_CT_ASSERT(maniptype == NF_NAT_MANIP_SRC || maniptype == NF_NAT_MANIP_DST); diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 3fd159db9f0..624e083125b 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -88,6 +88,45 @@ nf_tables_afinfo_lookup(struct net *net, int family, bool autoload) return ERR_PTR(-EAFNOSUPPORT); } +static void nft_ctx_init(struct nft_ctx *ctx, + const struct sk_buff *skb, + const struct nlmsghdr *nlh, + struct nft_af_info *afi, + struct nft_table *table, + struct nft_chain *chain, + const struct nlattr * const *nla) +{ + ctx->net = sock_net(skb->sk); + ctx->afi = afi; + ctx->table = table; + ctx->chain = chain; + ctx->nla = nla; + ctx->portid = NETLINK_CB(skb).portid; + ctx->report = nlmsg_report(nlh); + ctx->seq = nlh->nlmsg_seq; +} + +static struct nft_trans *nft_trans_alloc(struct nft_ctx *ctx, int msg_type, + u32 size) +{ + struct nft_trans *trans; + + trans = kzalloc(sizeof(struct nft_trans) + size, GFP_KERNEL); + if (trans == NULL) + return NULL; + + trans->msg_type = msg_type; + trans->ctx = *ctx; + + return trans; +} + +static void nft_trans_destroy(struct nft_trans *trans) +{ + list_del(&trans->list); + kfree(trans); +} + /* * Tables */ @@ -197,20 +236,13 @@ nla_put_failure: return -1; } -static int nf_tables_table_notify(const struct sk_buff *oskb, - const struct nlmsghdr *nlh, - const struct nft_table *table, - int event, int family) +static int nf_tables_table_notify(const struct nft_ctx *ctx, int event) { struct sk_buff *skb; - u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; - u32 seq = nlh ? nlh->nlmsg_seq : 0; - struct net *net = oskb ? sock_net(oskb->sk) : &init_net; - bool report; int err; - report = nlh ? nlmsg_report(nlh) : false; - if (!report && !nfnetlink_has_listeners(net, NFNLGRP_NFTABLES)) + if (!ctx->report && + !nfnetlink_has_listeners(ctx->net, NFNLGRP_NFTABLES)) return 0; err = -ENOBUFS; @@ -218,18 +250,20 @@ static int nf_tables_table_notify(const struct sk_buff *oskb, if (skb == NULL) goto err; - err = nf_tables_fill_table_info(skb, portid, seq, event, 0, - family, table); + err = nf_tables_fill_table_info(skb, ctx->portid, ctx->seq, event, 0, + ctx->afi->family, ctx->table); if (err < 0) { kfree_skb(skb); goto err; } - err = nfnetlink_send(skb, net, portid, NFNLGRP_NFTABLES, report, - GFP_KERNEL); + err = nfnetlink_send(skb, ctx->net, ctx->portid, NFNLGRP_NFTABLES, + ctx->report, GFP_KERNEL); err: - if (err < 0) - nfnetlink_set_err(net, portid, NFNLGRP_NFTABLES, err); + if (err < 0) { + nfnetlink_set_err(ctx->net, ctx->portid, NFNLGRP_NFTABLES, + err); + } return err; } @@ -269,6 +303,9 @@ done: return skb->len; } +/* Internal table flags */ +#define NFT_TABLE_INACTIVE (1 << 15) + static int nf_tables_gettable(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) @@ -295,6 +332,8 @@ static int nf_tables_gettable(struct sock *nlsk, struct sk_buff *skb, table = nf_tables_table_lookup(afi, nla[NFTA_TABLE_NAME]); if (IS_ERR(table)) return PTR_ERR(table); + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb2) @@ -343,7 +382,7 @@ err: return err; } -static int nf_tables_table_disable(const struct nft_af_info *afi, +static void nf_tables_table_disable(const struct nft_af_info *afi, struct nft_table *table) { struct nft_chain *chain; @@ -353,45 +392,63 @@ static int nf_tables_table_disable(const struct nft_af_info *afi, nf_unregister_hooks(nft_base_chain(chain)->ops, afi->nops); } - - return 0; } -static int nf_tables_updtable(struct sock *nlsk, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const nla[], - struct nft_af_info *afi, struct nft_table *table) +static int nf_tables_updtable(struct nft_ctx *ctx) { - const struct nfgenmsg *nfmsg = nlmsg_data(nlh); - int family = nfmsg->nfgen_family, ret = 0; + struct nft_trans *trans; + u32 flags; + int ret = 0; - if (nla[NFTA_TABLE_FLAGS]) { - u32 flags; + if (!ctx->nla[NFTA_TABLE_FLAGS]) + return 0; - flags = ntohl(nla_get_be32(nla[NFTA_TABLE_FLAGS])); - if (flags & ~NFT_TABLE_F_DORMANT) - return -EINVAL; + flags = ntohl(nla_get_be32(ctx->nla[NFTA_TABLE_FLAGS])); + if (flags & ~NFT_TABLE_F_DORMANT) + return -EINVAL; + + trans = nft_trans_alloc(ctx, NFT_MSG_NEWTABLE, + sizeof(struct nft_trans_table)); + if (trans == NULL) + return -ENOMEM; - if ((flags & NFT_TABLE_F_DORMANT) && - !(table->flags & NFT_TABLE_F_DORMANT)) { - ret = nf_tables_table_disable(afi, table); - if (ret >= 0) - table->flags |= NFT_TABLE_F_DORMANT; - } else if (!(flags & NFT_TABLE_F_DORMANT) && - table->flags & NFT_TABLE_F_DORMANT) { - ret = nf_tables_table_enable(afi, table); - if (ret >= 0) - table->flags &= ~NFT_TABLE_F_DORMANT; + if ((flags & NFT_TABLE_F_DORMANT) && + !(ctx->table->flags & NFT_TABLE_F_DORMANT)) { + nft_trans_table_enable(trans) = false; + } else if (!(flags & NFT_TABLE_F_DORMANT) && + ctx->table->flags & NFT_TABLE_F_DORMANT) { + ret = nf_tables_table_enable(ctx->afi, ctx->table); + if (ret >= 0) { + ctx->table->flags &= ~NFT_TABLE_F_DORMANT; + nft_trans_table_enable(trans) = true; } - if (ret < 0) - goto err; } + if (ret < 0) + goto err; - nf_tables_table_notify(skb, nlh, table, NFT_MSG_NEWTABLE, family); + nft_trans_table_update(trans) = true; + list_add_tail(&trans->list, &ctx->net->nft.commit_list); + return 0; err: + nft_trans_destroy(trans); return ret; } +static int nft_trans_table_add(struct nft_ctx *ctx, int msg_type) +{ + struct nft_trans *trans; + + trans = nft_trans_alloc(ctx, msg_type, sizeof(struct nft_trans_table)); + if (trans == NULL) + return -ENOMEM; + + if (msg_type == NFT_MSG_NEWTABLE) + ctx->table->flags |= NFT_TABLE_INACTIVE; + + list_add_tail(&trans->list, &ctx->net->nft.commit_list); + return 0; +} + static int nf_tables_newtable(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) @@ -403,6 +460,8 @@ static int nf_tables_newtable(struct sock *nlsk, struct sk_buff *skb, struct net *net = sock_net(skb->sk); int family = nfmsg->nfgen_family; u32 flags = 0; + struct nft_ctx ctx; + int err; afi = nf_tables_afinfo_lookup(net, family, true); if (IS_ERR(afi)) @@ -417,11 +476,15 @@ static int nf_tables_newtable(struct sock *nlsk, struct sk_buff *skb, } if (table != NULL) { + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; if (nlh->nlmsg_flags & NLM_F_EXCL) return -EEXIST; if (nlh->nlmsg_flags & NLM_F_REPLACE) return -EOPNOTSUPP; - return nf_tables_updtable(nlsk, skb, nlh, nla, afi, table); + + nft_ctx_init(&ctx, skb, nlh, afi, table, NULL, nla); + return nf_tables_updtable(&ctx); } if (nla[NFTA_TABLE_FLAGS]) { @@ -444,8 +507,14 @@ static int nf_tables_newtable(struct sock *nlsk, struct sk_buff *skb, INIT_LIST_HEAD(&table->sets); table->flags = flags; + nft_ctx_init(&ctx, skb, nlh, afi, table, NULL, nla); + err = nft_trans_table_add(&ctx, NFT_MSG_NEWTABLE); + if (err < 0) { + kfree(table); + module_put(afi->owner); + return err; + } list_add_tail(&table->list, &afi->tables); - nf_tables_table_notify(skb, nlh, table, NFT_MSG_NEWTABLE, family); return 0; } @@ -457,7 +526,8 @@ static int nf_tables_deltable(struct sock *nlsk, struct sk_buff *skb, struct nft_af_info *afi; struct nft_table *table; struct net *net = sock_net(skb->sk); - int family = nfmsg->nfgen_family; + int family = nfmsg->nfgen_family, err; + struct nft_ctx ctx; afi = nf_tables_afinfo_lookup(net, family, false); if (IS_ERR(afi)) @@ -466,17 +536,28 @@ static int nf_tables_deltable(struct sock *nlsk, struct sk_buff *skb, table = nf_tables_table_lookup(afi, nla[NFTA_TABLE_NAME]); if (IS_ERR(table)) return PTR_ERR(table); - - if (!list_empty(&table->chains) || !list_empty(&table->sets)) + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; + if (table->use > 0) return -EBUSY; + nft_ctx_init(&ctx, skb, nlh, afi, table, NULL, nla); + err = nft_trans_table_add(&ctx, NFT_MSG_DELTABLE); + if (err < 0) + return err; + list_del(&table->list); - nf_tables_table_notify(skb, nlh, table, NFT_MSG_DELTABLE, family); - kfree(table); - module_put(afi->owner); return 0; } +static void nf_tables_table_destroy(struct nft_ctx *ctx) +{ + BUG_ON(ctx->table->use > 0); + + kfree(ctx->table); + module_put(ctx->afi->owner); +} + int nft_register_chain_type(const struct nf_chain_type *ctype) { int err = 0; @@ -541,7 +622,7 @@ static const struct nla_policy nft_chain_policy[NFTA_CHAIN_MAX + 1] = { .len = NFT_CHAIN_MAXNAMELEN - 1 }, [NFTA_CHAIN_HOOK] = { .type = NLA_NESTED }, [NFTA_CHAIN_POLICY] = { .type = NLA_U32 }, - [NFTA_CHAIN_TYPE] = { .type = NLA_NUL_STRING }, + [NFTA_CHAIN_TYPE] = { .type = NLA_STRING }, [NFTA_CHAIN_COUNTERS] = { .type = NLA_NESTED }, }; @@ -637,21 +718,13 @@ nla_put_failure: return -1; } -static int nf_tables_chain_notify(const struct sk_buff *oskb, - const struct nlmsghdr *nlh, - const struct nft_table *table, - const struct nft_chain *chain, - int event, int family) +static int nf_tables_chain_notify(const struct nft_ctx *ctx, int event) { struct sk_buff *skb; - u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; - struct net *net = oskb ? sock_net(oskb->sk) : &init_net; - u32 seq = nlh ? nlh->nlmsg_seq : 0; - bool report; int err; - report = nlh ? nlmsg_report(nlh) : false; - if (!report && !nfnetlink_has_listeners(net, NFNLGRP_NFTABLES)) + if (!ctx->report && + !nfnetlink_has_listeners(ctx->net, NFNLGRP_NFTABLES)) return 0; err = -ENOBUFS; @@ -659,18 +732,21 @@ static int nf_tables_chain_notify(const struct sk_buff *oskb, if (skb == NULL) goto err; - err = nf_tables_fill_chain_info(skb, portid, seq, event, 0, family, - table, chain); + err = nf_tables_fill_chain_info(skb, ctx->portid, ctx->seq, event, 0, + ctx->afi->family, ctx->table, + ctx->chain); if (err < 0) { kfree_skb(skb); goto err; } - err = nfnetlink_send(skb, net, portid, NFNLGRP_NFTABLES, report, - GFP_KERNEL); + err = nfnetlink_send(skb, ctx->net, ctx->portid, NFNLGRP_NFTABLES, + ctx->report, GFP_KERNEL); err: - if (err < 0) - nfnetlink_set_err(net, portid, NFNLGRP_NFTABLES, err); + if (err < 0) { + nfnetlink_set_err(ctx->net, ctx->portid, NFNLGRP_NFTABLES, + err); + } return err; } @@ -740,10 +816,14 @@ static int nf_tables_getchain(struct sock *nlsk, struct sk_buff *skb, table = nf_tables_table_lookup(afi, nla[NFTA_CHAIN_TABLE]); if (IS_ERR(table)) return PTR_ERR(table); + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; chain = nf_tables_chain_lookup(table, nla[NFTA_CHAIN_NAME]); if (IS_ERR(chain)) return PTR_ERR(chain); + if (chain->flags & NFT_CHAIN_INACTIVE) + return -ENOENT; skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb2) @@ -767,8 +847,7 @@ static const struct nla_policy nft_counter_policy[NFTA_COUNTER_MAX + 1] = { [NFTA_COUNTER_BYTES] = { .type = NLA_U64 }, }; -static int -nf_tables_counters(struct nft_base_chain *chain, const struct nlattr *attr) +static struct nft_stats __percpu *nft_stats_alloc(const struct nlattr *attr) { struct nlattr *tb[NFTA_COUNTER_MAX+1]; struct nft_stats __percpu *newstats; @@ -777,14 +856,14 @@ nf_tables_counters(struct nft_base_chain *chain, const struct nlattr *attr) err = nla_parse_nested(tb, NFTA_COUNTER_MAX, attr, nft_counter_policy); if (err < 0) - return err; + return ERR_PTR(err); if (!tb[NFTA_COUNTER_BYTES] || !tb[NFTA_COUNTER_PACKETS]) - return -EINVAL; + return ERR_PTR(-EINVAL); newstats = alloc_percpu(struct nft_stats); if (newstats == NULL) - return -ENOMEM; + return ERR_PTR(-ENOMEM); /* Restore old counters on this cpu, no problem. Per-cpu statistics * are not exposed to userspace. @@ -793,6 +872,12 @@ nf_tables_counters(struct nft_base_chain *chain, const struct nlattr *attr) stats->bytes = be64_to_cpu(nla_get_be64(tb[NFTA_COUNTER_BYTES])); stats->pkts = be64_to_cpu(nla_get_be64(tb[NFTA_COUNTER_PACKETS])); + return newstats; +} + +static void nft_chain_stats_replace(struct nft_base_chain *chain, + struct nft_stats __percpu *newstats) +{ if (chain->stats) { struct nft_stats __percpu *oldstats = nft_dereference(chain->stats); @@ -802,17 +887,43 @@ nf_tables_counters(struct nft_base_chain *chain, const struct nlattr *attr) free_percpu(oldstats); } else rcu_assign_pointer(chain->stats, newstats); +} + +static int nft_trans_chain_add(struct nft_ctx *ctx, int msg_type) +{ + struct nft_trans *trans; + trans = nft_trans_alloc(ctx, msg_type, sizeof(struct nft_trans_chain)); + if (trans == NULL) + return -ENOMEM; + + if (msg_type == NFT_MSG_NEWCHAIN) + ctx->chain->flags |= NFT_CHAIN_INACTIVE; + + list_add_tail(&trans->list, &ctx->net->nft.commit_list); return 0; } +static void nf_tables_chain_destroy(struct nft_chain *chain) +{ + BUG_ON(chain->use > 0); + + if (chain->flags & NFT_BASE_CHAIN) { + module_put(nft_base_chain(chain)->type->owner); + free_percpu(nft_base_chain(chain)->stats); + kfree(nft_base_chain(chain)); + } else { + kfree(chain); + } +} + static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); const struct nlattr * uninitialized_var(name); - const struct nft_af_info *afi; + struct nft_af_info *afi; struct nft_table *table; struct nft_chain *chain; struct nft_base_chain *basechain = NULL; @@ -822,8 +933,10 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, u8 policy = NF_ACCEPT; u64 handle = 0; unsigned int i; + struct nft_stats __percpu *stats; int err; bool create; + struct nft_ctx ctx; create = nlh->nlmsg_flags & NLM_F_CREATE ? true : false; @@ -869,6 +982,11 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, } if (chain != NULL) { + struct nft_stats *stats = NULL; + struct nft_trans *trans; + + if (chain->flags & NFT_CHAIN_INACTIVE) + return -ENOENT; if (nlh->nlmsg_flags & NLM_F_EXCL) return -EEXIST; if (nlh->nlmsg_flags & NLM_F_REPLACE) @@ -882,19 +1000,31 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, if (!(chain->flags & NFT_BASE_CHAIN)) return -EOPNOTSUPP; - err = nf_tables_counters(nft_base_chain(chain), - nla[NFTA_CHAIN_COUNTERS]); - if (err < 0) - return err; + stats = nft_stats_alloc(nla[NFTA_CHAIN_COUNTERS]); + if (IS_ERR(stats)) + return PTR_ERR(stats); } - if (nla[NFTA_CHAIN_POLICY]) - nft_base_chain(chain)->policy = policy; + nft_ctx_init(&ctx, skb, nlh, afi, table, chain, nla); + trans = nft_trans_alloc(&ctx, NFT_MSG_NEWCHAIN, + sizeof(struct nft_trans_chain)); + if (trans == NULL) + return -ENOMEM; + + nft_trans_chain_stats(trans) = stats; + nft_trans_chain_update(trans) = true; - if (nla[NFTA_CHAIN_HANDLE] && name) - nla_strlcpy(chain->name, name, NFT_CHAIN_MAXNAMELEN); + if (nla[NFTA_CHAIN_POLICY]) + nft_trans_chain_policy(trans) = policy; + else + nft_trans_chain_policy(trans) = -1; - goto notify; + if (nla[NFTA_CHAIN_HANDLE] && name) { + nla_strlcpy(nft_trans_chain_name(trans), name, + NFT_CHAIN_MAXNAMELEN); + } + list_add_tail(&trans->list, &net->nft.commit_list); + return 0; } if (table->use == UINT_MAX) @@ -939,23 +1069,21 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, return -ENOMEM; if (nla[NFTA_CHAIN_COUNTERS]) { - err = nf_tables_counters(basechain, - nla[NFTA_CHAIN_COUNTERS]); - if (err < 0) { + stats = nft_stats_alloc(nla[NFTA_CHAIN_COUNTERS]); + if (IS_ERR(stats)) { module_put(type->owner); kfree(basechain); - return err; + return PTR_ERR(stats); } + basechain->stats = stats; } else { - struct nft_stats __percpu *newstats; - - newstats = alloc_percpu(struct nft_stats); - if (newstats == NULL) { + stats = alloc_percpu(struct nft_stats); + if (IS_ERR(stats)) { module_put(type->owner); kfree(basechain); - return -ENOMEM; + return PTR_ERR(stats); } - rcu_assign_pointer(basechain->stats, newstats); + rcu_assign_pointer(basechain->stats, stats); } basechain->type = type; @@ -992,31 +1120,27 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, if (!(table->flags & NFT_TABLE_F_DORMANT) && chain->flags & NFT_BASE_CHAIN) { err = nf_register_hooks(nft_base_chain(chain)->ops, afi->nops); - if (err < 0) { - module_put(basechain->type->owner); - free_percpu(basechain->stats); - kfree(basechain); - return err; - } + if (err < 0) + goto err1; } - list_add_tail(&chain->list, &table->chains); - table->use++; -notify: - nf_tables_chain_notify(skb, nlh, table, chain, NFT_MSG_NEWCHAIN, - family); - return 0; -} -static void nf_tables_chain_destroy(struct nft_chain *chain) -{ - BUG_ON(chain->use > 0); + nft_ctx_init(&ctx, skb, nlh, afi, table, chain, nla); + err = nft_trans_chain_add(&ctx, NFT_MSG_NEWCHAIN); + if (err < 0) + goto err2; - if (chain->flags & NFT_BASE_CHAIN) { - module_put(nft_base_chain(chain)->type->owner); - free_percpu(nft_base_chain(chain)->stats); - kfree(nft_base_chain(chain)); - } else - kfree(chain); + table->use++; + list_add_tail(&chain->list, &table->chains); + return 0; +err2: + if (!(table->flags & NFT_TABLE_F_DORMANT) && + chain->flags & NFT_BASE_CHAIN) { + nf_unregister_hooks(nft_base_chain(chain)->ops, + afi->nops); + } +err1: + nf_tables_chain_destroy(chain); + return err; } static int nf_tables_delchain(struct sock *nlsk, struct sk_buff *skb, @@ -1024,11 +1148,13 @@ static int nf_tables_delchain(struct sock *nlsk, struct sk_buff *skb, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); - const struct nft_af_info *afi; + struct nft_af_info *afi; struct nft_table *table; struct nft_chain *chain; struct net *net = sock_net(skb->sk); int family = nfmsg->nfgen_family; + struct nft_ctx ctx; + int err; afi = nf_tables_afinfo_lookup(net, family, false); if (IS_ERR(afi)) @@ -1037,48 +1163,27 @@ static int nf_tables_delchain(struct sock *nlsk, struct sk_buff *skb, table = nf_tables_table_lookup(afi, nla[NFTA_CHAIN_TABLE]); if (IS_ERR(table)) return PTR_ERR(table); + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; chain = nf_tables_chain_lookup(table, nla[NFTA_CHAIN_NAME]); if (IS_ERR(chain)) return PTR_ERR(chain); - - if (!list_empty(&chain->rules) || chain->use > 0) + if (chain->flags & NFT_CHAIN_INACTIVE) + return -ENOENT; + if (chain->use > 0) return -EBUSY; - list_del(&chain->list); - table->use--; - - if (!(table->flags & NFT_TABLE_F_DORMANT) && - chain->flags & NFT_BASE_CHAIN) - nf_unregister_hooks(nft_base_chain(chain)->ops, afi->nops); - - nf_tables_chain_notify(skb, nlh, table, chain, NFT_MSG_DELCHAIN, - family); - - /* Make sure all rule references are gone before this is released */ - synchronize_rcu(); + nft_ctx_init(&ctx, skb, nlh, afi, table, chain, nla); + err = nft_trans_chain_add(&ctx, NFT_MSG_DELCHAIN); + if (err < 0) + return err; - nf_tables_chain_destroy(chain); + table->use--; + list_del(&chain->list); return 0; } -static void nft_ctx_init(struct nft_ctx *ctx, - const struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nft_af_info *afi, - const struct nft_table *table, - const struct nft_chain *chain, - const struct nlattr * const *nla) -{ - ctx->net = sock_net(skb->sk); - ctx->skb = skb; - ctx->nlh = nlh; - ctx->afi = afi; - ctx->table = table; - ctx->chain = chain; - ctx->nla = nla; -} - /* * Expressions */ @@ -1093,7 +1198,10 @@ static void nft_ctx_init(struct nft_ctx *ctx, int nft_register_expr(struct nft_expr_type *type) { nfnl_lock(NFNL_SUBSYS_NFTABLES); - list_add_tail(&type->list, &nf_tables_expressions); + if (type->family == NFPROTO_UNSPEC) + list_add_tail(&type->list, &nf_tables_expressions); + else + list_add(&type->list, &nf_tables_expressions); nfnl_unlock(NFNL_SUBSYS_NFTABLES); return 0; } @@ -1361,22 +1469,15 @@ nla_put_failure: return -1; } -static int nf_tables_rule_notify(const struct sk_buff *oskb, - const struct nlmsghdr *nlh, - const struct nft_table *table, - const struct nft_chain *chain, +static int nf_tables_rule_notify(const struct nft_ctx *ctx, const struct nft_rule *rule, - int event, u32 flags, int family) + int event) { struct sk_buff *skb; - u32 portid = NETLINK_CB(oskb).portid; - struct net *net = oskb ? sock_net(oskb->sk) : &init_net; - u32 seq = nlh->nlmsg_seq; - bool report; int err; - report = nlmsg_report(nlh); - if (!report && !nfnetlink_has_listeners(net, NFNLGRP_NFTABLES)) + if (!ctx->report && + !nfnetlink_has_listeners(ctx->net, NFNLGRP_NFTABLES)) return 0; err = -ENOBUFS; @@ -1384,18 +1485,21 @@ static int nf_tables_rule_notify(const struct sk_buff *oskb, if (skb == NULL) goto err; - err = nf_tables_fill_rule_info(skb, portid, seq, event, flags, - family, table, chain, rule); + err = nf_tables_fill_rule_info(skb, ctx->portid, ctx->seq, event, 0, + ctx->afi->family, ctx->table, + ctx->chain, rule); if (err < 0) { kfree_skb(skb); goto err; } - err = nfnetlink_send(skb, net, portid, NFNLGRP_NFTABLES, report, - GFP_KERNEL); + err = nfnetlink_send(skb, ctx->net, ctx->portid, NFNLGRP_NFTABLES, + ctx->report, GFP_KERNEL); err: - if (err < 0) - nfnetlink_set_err(net, portid, NFNLGRP_NFTABLES, err); + if (err < 0) { + nfnetlink_set_err(ctx->net, ctx->portid, NFNLGRP_NFTABLES, + err); + } return err; } @@ -1511,10 +1615,14 @@ static int nf_tables_getrule(struct sock *nlsk, struct sk_buff *skb, table = nf_tables_table_lookup(afi, nla[NFTA_RULE_TABLE]); if (IS_ERR(table)) return PTR_ERR(table); + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; chain = nf_tables_chain_lookup(table, nla[NFTA_RULE_CHAIN]); if (IS_ERR(chain)) return PTR_ERR(chain); + if (chain->flags & NFT_CHAIN_INACTIVE) + return -ENOENT; rule = nf_tables_rule_lookup(chain, nla[NFTA_RULE_HANDLE]); if (IS_ERR(rule)) @@ -1554,37 +1662,36 @@ static void nf_tables_rule_destroy(const struct nft_ctx *ctx, kfree(rule); } -#define NFT_RULE_MAXEXPRS 128 - -static struct nft_expr_info *info; - -static struct nft_rule_trans * -nf_tables_trans_add(struct nft_ctx *ctx, struct nft_rule *rule) +static struct nft_trans *nft_trans_rule_add(struct nft_ctx *ctx, int msg_type, + struct nft_rule *rule) { - struct nft_rule_trans *rupd; + struct nft_trans *trans; - rupd = kmalloc(sizeof(struct nft_rule_trans), GFP_KERNEL); - if (rupd == NULL) - return NULL; + trans = nft_trans_alloc(ctx, msg_type, sizeof(struct nft_trans_rule)); + if (trans == NULL) + return NULL; - rupd->ctx = *ctx; - rupd->rule = rule; - list_add_tail(&rupd->list, &ctx->net->nft.commit_list); + nft_trans_rule(trans) = rule; + list_add_tail(&trans->list, &ctx->net->nft.commit_list); - return rupd; + return trans; } +#define NFT_RULE_MAXEXPRS 128 + +static struct nft_expr_info *info; + static int nf_tables_newrule(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); - const struct nft_af_info *afi; + struct nft_af_info *afi; struct net *net = sock_net(skb->sk); struct nft_table *table; struct nft_chain *chain; struct nft_rule *rule, *old_rule = NULL; - struct nft_rule_trans *repl = NULL; + struct nft_trans *trans = NULL; struct nft_expr *expr; struct nft_ctx ctx; struct nlattr *tmp; @@ -1682,8 +1789,9 @@ static int nf_tables_newrule(struct sock *nlsk, struct sk_buff *skb, if (nlh->nlmsg_flags & NLM_F_REPLACE) { if (nft_rule_is_active_next(net, old_rule)) { - repl = nf_tables_trans_add(&ctx, old_rule); - if (repl == NULL) { + trans = nft_trans_rule_add(&ctx, NFT_MSG_NEWRULE, + old_rule); + if (trans == NULL) { err = -ENOMEM; goto err2; } @@ -1705,19 +1813,19 @@ static int nf_tables_newrule(struct sock *nlsk, struct sk_buff *skb, list_add_rcu(&rule->list, &chain->rules); } - if (nf_tables_trans_add(&ctx, rule) == NULL) { + if (nft_trans_rule_add(&ctx, NFT_MSG_NEWRULE, rule) == NULL) { err = -ENOMEM; goto err3; } + chain->use++; return 0; err3: list_del_rcu(&rule->list); - if (repl) { - list_del_rcu(&repl->rule->list); - list_del(&repl->list); - nft_rule_clear(net, repl->rule); - kfree(repl); + if (trans) { + list_del_rcu(&nft_trans_rule(trans)->list); + nft_rule_clear(net, nft_trans_rule(trans)); + nft_trans_destroy(trans); } err2: nf_tables_rule_destroy(&ctx, rule); @@ -1734,9 +1842,10 @@ nf_tables_delrule_one(struct nft_ctx *ctx, struct nft_rule *rule) { /* You cannot delete the same rule twice */ if (nft_rule_is_active_next(ctx->net, rule)) { - if (nf_tables_trans_add(ctx, rule) == NULL) + if (nft_trans_rule_add(ctx, NFT_MSG_DELRULE, rule) == NULL) return -ENOMEM; nft_rule_disactivate_next(ctx->net, rule); + ctx->chain->use--; return 0; } return -ENOENT; @@ -1760,9 +1869,9 @@ static int nf_tables_delrule(struct sock *nlsk, struct sk_buff *skb, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); - const struct nft_af_info *afi; + struct nft_af_info *afi; struct net *net = sock_net(skb->sk); - const struct nft_table *table; + struct nft_table *table; struct nft_chain *chain = NULL; struct nft_rule *rule; int family = nfmsg->nfgen_family, err = 0; @@ -1775,6 +1884,8 @@ static int nf_tables_delrule(struct sock *nlsk, struct sk_buff *skb, table = nf_tables_table_lookup(afi, nla[NFTA_RULE_TABLE]); if (IS_ERR(table)) return PTR_ERR(table); + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; if (nla[NFTA_RULE_CHAIN]) { chain = nf_tables_chain_lookup(table, nla[NFTA_RULE_CHAIN]); @@ -1807,88 +1918,6 @@ static int nf_tables_delrule(struct sock *nlsk, struct sk_buff *skb, return err; } -static int nf_tables_commit(struct sk_buff *skb) -{ - struct net *net = sock_net(skb->sk); - struct nft_rule_trans *rupd, *tmp; - - /* Bump generation counter, invalidate any dump in progress */ - net->nft.genctr++; - - /* A new generation has just started */ - net->nft.gencursor = gencursor_next(net); - - /* Make sure all packets have left the previous generation before - * purging old rules. - */ - synchronize_rcu(); - - list_for_each_entry_safe(rupd, tmp, &net->nft.commit_list, list) { - /* This rule was inactive in the past and just became active. - * Clear the next bit of the genmask since its meaning has - * changed, now it is the future. - */ - if (nft_rule_is_active(net, rupd->rule)) { - nft_rule_clear(net, rupd->rule); - nf_tables_rule_notify(skb, rupd->ctx.nlh, - rupd->ctx.table, rupd->ctx.chain, - rupd->rule, NFT_MSG_NEWRULE, 0, - rupd->ctx.afi->family); - list_del(&rupd->list); - kfree(rupd); - continue; - } - - /* This rule is in the past, get rid of it */ - list_del_rcu(&rupd->rule->list); - nf_tables_rule_notify(skb, rupd->ctx.nlh, - rupd->ctx.table, rupd->ctx.chain, - rupd->rule, NFT_MSG_DELRULE, 0, - rupd->ctx.afi->family); - } - - /* Make sure we don't see any packet traversing old rules */ - synchronize_rcu(); - - /* Now we can safely release unused old rules */ - list_for_each_entry_safe(rupd, tmp, &net->nft.commit_list, list) { - nf_tables_rule_destroy(&rupd->ctx, rupd->rule); - list_del(&rupd->list); - kfree(rupd); - } - - return 0; -} - -static int nf_tables_abort(struct sk_buff *skb) -{ - struct net *net = sock_net(skb->sk); - struct nft_rule_trans *rupd, *tmp; - - list_for_each_entry_safe(rupd, tmp, &net->nft.commit_list, list) { - if (!nft_rule_is_active_next(net, rupd->rule)) { - nft_rule_clear(net, rupd->rule); - list_del(&rupd->list); - kfree(rupd); - continue; - } - - /* This rule is inactive, get rid of it */ - list_del_rcu(&rupd->rule->list); - } - - /* Make sure we don't see any packet accessing aborted rules */ - synchronize_rcu(); - - list_for_each_entry_safe(rupd, tmp, &net->nft.commit_list, list) { - nf_tables_rule_destroy(&rupd->ctx, rupd->rule); - list_del(&rupd->list); - kfree(rupd); - } - - return 0; -} - /* * Sets */ @@ -1912,9 +1941,18 @@ void nft_unregister_set(struct nft_set_ops *ops) } EXPORT_SYMBOL_GPL(nft_unregister_set); -static const struct nft_set_ops *nft_select_set_ops(const struct nlattr * const nla[]) +/* + * Select a set implementation based on the data characteristics and the + * given policy. The total memory use might not be known if no size is + * given, in that case the amount of memory per element is used. + */ +static const struct nft_set_ops * +nft_select_set_ops(const struct nlattr * const nla[], + const struct nft_set_desc *desc, + enum nft_set_policies policy) { - const struct nft_set_ops *ops; + const struct nft_set_ops *ops, *bops; + struct nft_set_estimate est, best; u32 features; #ifdef CONFIG_MODULES @@ -1932,15 +1970,45 @@ static const struct nft_set_ops *nft_select_set_ops(const struct nlattr * const features &= NFT_SET_INTERVAL | NFT_SET_MAP; } - // FIXME: implement selection properly + bops = NULL; + best.size = ~0; + best.class = ~0; + list_for_each_entry(ops, &nf_tables_set_ops, list) { if ((ops->features & features) != features) continue; + if (!ops->estimate(desc, features, &est)) + continue; + + switch (policy) { + case NFT_SET_POL_PERFORMANCE: + if (est.class < best.class) + break; + if (est.class == best.class && est.size < best.size) + break; + continue; + case NFT_SET_POL_MEMORY: + if (est.size < best.size) + break; + if (est.size == best.size && est.class < best.class) + break; + continue; + default: + break; + } + if (!try_module_get(ops->owner)) continue; - return ops; + if (bops != NULL) + module_put(bops->owner); + + bops = ops; + best = est; } + if (bops != NULL) + return bops; + return ERR_PTR(-EOPNOTSUPP); } @@ -1953,6 +2021,13 @@ static const struct nla_policy nft_set_policy[NFTA_SET_MAX + 1] = { [NFTA_SET_KEY_LEN] = { .type = NLA_U32 }, [NFTA_SET_DATA_TYPE] = { .type = NLA_U32 }, [NFTA_SET_DATA_LEN] = { .type = NLA_U32 }, + [NFTA_SET_POLICY] = { .type = NLA_U32 }, + [NFTA_SET_DESC] = { .type = NLA_NESTED }, + [NFTA_SET_ID] = { .type = NLA_U32 }, +}; + +static const struct nla_policy nft_set_desc_policy[NFTA_SET_DESC_MAX + 1] = { + [NFTA_SET_DESC_SIZE] = { .type = NLA_U32 }, }; static int nft_ctx_init_from_setattr(struct nft_ctx *ctx, @@ -1962,8 +2037,8 @@ static int nft_ctx_init_from_setattr(struct nft_ctx *ctx, { struct net *net = sock_net(skb->sk); const struct nfgenmsg *nfmsg = nlmsg_data(nlh); - const struct nft_af_info *afi = NULL; - const struct nft_table *table = NULL; + struct nft_af_info *afi = NULL; + struct nft_table *table = NULL; if (nfmsg->nfgen_family != NFPROTO_UNSPEC) { afi = nf_tables_afinfo_lookup(net, nfmsg->nfgen_family, false); @@ -1978,6 +2053,8 @@ static int nft_ctx_init_from_setattr(struct nft_ctx *ctx, table = nf_tables_table_lookup(afi, nla[NFTA_SET_TABLE]); if (IS_ERR(table)) return PTR_ERR(table); + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; } nft_ctx_init(ctx, skb, nlh, afi, table, NULL, nla); @@ -1999,13 +2076,27 @@ struct nft_set *nf_tables_set_lookup(const struct nft_table *table, return ERR_PTR(-ENOENT); } +struct nft_set *nf_tables_set_lookup_byid(const struct net *net, + const struct nlattr *nla) +{ + struct nft_trans *trans; + u32 id = ntohl(nla_get_be32(nla)); + + list_for_each_entry(trans, &net->nft.commit_list, list) { + if (trans->msg_type == NFT_MSG_NEWSET && + id == nft_trans_set_id(trans)) + return nft_trans_set(trans); + } + return ERR_PTR(-ENOENT); +} + static int nf_tables_set_alloc_name(struct nft_ctx *ctx, struct nft_set *set, const char *name) { const struct nft_set *i; const char *p; unsigned long *inuse; - unsigned int n = 0; + unsigned int n = 0, min = 0; p = strnchr(name, IFNAMSIZ, '%'); if (p != NULL) { @@ -2015,23 +2106,28 @@ static int nf_tables_set_alloc_name(struct nft_ctx *ctx, struct nft_set *set, inuse = (unsigned long *)get_zeroed_page(GFP_KERNEL); if (inuse == NULL) return -ENOMEM; - +cont: list_for_each_entry(i, &ctx->table->sets, list) { int tmp; if (!sscanf(i->name, name, &tmp)) continue; - if (tmp < 0 || tmp >= BITS_PER_BYTE * PAGE_SIZE) + if (tmp < min || tmp >= min + BITS_PER_BYTE * PAGE_SIZE) continue; - set_bit(tmp, inuse); + set_bit(tmp - min, inuse); } n = find_first_zero_bit(inuse, BITS_PER_BYTE * PAGE_SIZE); + if (n >= BITS_PER_BYTE * PAGE_SIZE) { + min += BITS_PER_BYTE * PAGE_SIZE; + memset(inuse, 0, PAGE_SIZE); + goto cont; + } free_page((unsigned long)inuse); } - snprintf(set->name, sizeof(set->name), name, n); + snprintf(set->name, sizeof(set->name), name, min + n); list_for_each_entry(i, &ctx->table->sets, list) { if (!strcmp(set->name, i->name)) return -ENFILE; @@ -2044,8 +2140,9 @@ static int nf_tables_fill_set(struct sk_buff *skb, const struct nft_ctx *ctx, { struct nfgenmsg *nfmsg; struct nlmsghdr *nlh; - u32 portid = NETLINK_CB(ctx->skb).portid; - u32 seq = ctx->nlh->nlmsg_seq; + struct nlattr *desc; + u32 portid = ctx->portid; + u32 seq = ctx->seq; event |= NFNL_SUBSYS_NFTABLES << 8; nlh = nlmsg_put(skb, portid, seq, event, sizeof(struct nfgenmsg), @@ -2077,6 +2174,14 @@ static int nf_tables_fill_set(struct sk_buff *skb, const struct nft_ctx *ctx, goto nla_put_failure; } + desc = nla_nest_start(skb, NFTA_SET_DESC); + if (desc == NULL) + goto nla_put_failure; + if (set->size && + nla_put_be32(skb, NFTA_SET_DESC_SIZE, htonl(set->size))) + goto nla_put_failure; + nla_nest_end(skb, desc); + return nlmsg_end(skb, nlh); nla_put_failure: @@ -2086,19 +2191,18 @@ nla_put_failure: static int nf_tables_set_notify(const struct nft_ctx *ctx, const struct nft_set *set, - int event) + int event, gfp_t gfp_flags) { struct sk_buff *skb; - u32 portid = NETLINK_CB(ctx->skb).portid; - bool report; + u32 portid = ctx->portid; int err; - report = nlmsg_report(ctx->nlh); - if (!report && !nfnetlink_has_listeners(ctx->net, NFNLGRP_NFTABLES)) + if (!ctx->report && + !nfnetlink_has_listeners(ctx->net, NFNLGRP_NFTABLES)) return 0; err = -ENOBUFS; - skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + skb = nlmsg_new(NLMSG_GOODSIZE, gfp_flags); if (skb == NULL) goto err; @@ -2108,8 +2212,8 @@ static int nf_tables_set_notify(const struct nft_ctx *ctx, goto err; } - err = nfnetlink_send(skb, ctx->net, portid, NFNLGRP_NFTABLES, report, - GFP_KERNEL); + err = nfnetlink_send(skb, ctx->net, portid, NFNLGRP_NFTABLES, + ctx->report, gfp_flags); err: if (err < 0) nfnetlink_set_err(ctx->net, portid, NFNLGRP_NFTABLES, err); @@ -2183,7 +2287,7 @@ static int nf_tables_dump_sets_all(struct nft_ctx *ctx, struct sk_buff *skb, { const struct nft_set *set; unsigned int idx, s_idx = cb->args[0]; - const struct nft_af_info *afi; + struct nft_af_info *afi; struct nft_table *table, *cur_table = (struct nft_table *)cb->args[2]; struct net *net = sock_net(skb->sk); int cur_family = cb->args[3]; @@ -2260,6 +2364,8 @@ static int nf_tables_dump_sets(struct sk_buff *skb, struct netlink_callback *cb) return ret; } +#define NFT_SET_INACTIVE (1 << 15) /* Internal set flag */ + static int nf_tables_getset(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) @@ -2289,6 +2395,8 @@ static int nf_tables_getset(struct sock *nlsk, struct sk_buff *skb, set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_NAME]); if (IS_ERR(set)) return PTR_ERR(set); + if (set->flags & NFT_SET_INACTIVE) + return -ENOENT; skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (skb2 == NULL) @@ -2305,13 +2413,50 @@ err: return err; } +static int nf_tables_set_desc_parse(const struct nft_ctx *ctx, + struct nft_set_desc *desc, + const struct nlattr *nla) +{ + struct nlattr *da[NFTA_SET_DESC_MAX + 1]; + int err; + + err = nla_parse_nested(da, NFTA_SET_DESC_MAX, nla, nft_set_desc_policy); + if (err < 0) + return err; + + if (da[NFTA_SET_DESC_SIZE] != NULL) + desc->size = ntohl(nla_get_be32(da[NFTA_SET_DESC_SIZE])); + + return 0; +} + +static int nft_trans_set_add(struct nft_ctx *ctx, int msg_type, + struct nft_set *set) +{ + struct nft_trans *trans; + + trans = nft_trans_alloc(ctx, msg_type, sizeof(struct nft_trans_set)); + if (trans == NULL) + return -ENOMEM; + + if (msg_type == NFT_MSG_NEWSET && ctx->nla[NFTA_SET_ID] != NULL) { + nft_trans_set_id(trans) = + ntohl(nla_get_be32(ctx->nla[NFTA_SET_ID])); + set->flags |= NFT_SET_INACTIVE; + } + nft_trans_set(trans) = set; + list_add_tail(&trans->list, &ctx->net->nft.commit_list); + + return 0; +} + static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); const struct nft_set_ops *ops; - const struct nft_af_info *afi; + struct nft_af_info *afi; struct net *net = sock_net(skb->sk); struct nft_table *table; struct nft_set *set; @@ -2319,14 +2464,18 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, char name[IFNAMSIZ]; unsigned int size; bool create; - u32 ktype, klen, dlen, dtype, flags; + u32 ktype, dtype, flags, policy; + struct nft_set_desc desc; int err; if (nla[NFTA_SET_TABLE] == NULL || nla[NFTA_SET_NAME] == NULL || - nla[NFTA_SET_KEY_LEN] == NULL) + nla[NFTA_SET_KEY_LEN] == NULL || + nla[NFTA_SET_ID] == NULL) return -EINVAL; + memset(&desc, 0, sizeof(desc)); + ktype = NFT_DATA_VALUE; if (nla[NFTA_SET_KEY_TYPE] != NULL) { ktype = ntohl(nla_get_be32(nla[NFTA_SET_KEY_TYPE])); @@ -2334,8 +2483,8 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, return -EINVAL; } - klen = ntohl(nla_get_be32(nla[NFTA_SET_KEY_LEN])); - if (klen == 0 || klen > FIELD_SIZEOF(struct nft_data, data)) + desc.klen = ntohl(nla_get_be32(nla[NFTA_SET_KEY_LEN])); + if (desc.klen == 0 || desc.klen > FIELD_SIZEOF(struct nft_data, data)) return -EINVAL; flags = 0; @@ -2347,7 +2496,6 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, } dtype = 0; - dlen = 0; if (nla[NFTA_SET_DATA_TYPE] != NULL) { if (!(flags & NFT_SET_MAP)) return -EINVAL; @@ -2360,15 +2508,25 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, if (dtype != NFT_DATA_VERDICT) { if (nla[NFTA_SET_DATA_LEN] == NULL) return -EINVAL; - dlen = ntohl(nla_get_be32(nla[NFTA_SET_DATA_LEN])); - if (dlen == 0 || - dlen > FIELD_SIZEOF(struct nft_data, data)) + desc.dlen = ntohl(nla_get_be32(nla[NFTA_SET_DATA_LEN])); + if (desc.dlen == 0 || + desc.dlen > FIELD_SIZEOF(struct nft_data, data)) return -EINVAL; } else - dlen = sizeof(struct nft_data); + desc.dlen = sizeof(struct nft_data); } else if (flags & NFT_SET_MAP) return -EINVAL; + policy = NFT_SET_POL_PERFORMANCE; + if (nla[NFTA_SET_POLICY] != NULL) + policy = ntohl(nla_get_be32(nla[NFTA_SET_POLICY])); + + if (nla[NFTA_SET_DESC] != NULL) { + err = nf_tables_set_desc_parse(&ctx, &desc, nla[NFTA_SET_DESC]); + if (err < 0) + return err; + } + create = nlh->nlmsg_flags & NLM_F_CREATE ? true : false; afi = nf_tables_afinfo_lookup(net, nfmsg->nfgen_family, create); @@ -2399,7 +2557,7 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, if (!(nlh->nlmsg_flags & NLM_F_CREATE)) return -ENOENT; - ops = nft_select_set_ops(nla); + ops = nft_select_set_ops(nla, &desc, policy); if (IS_ERR(ops)) return PTR_ERR(ops); @@ -2420,17 +2578,22 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, INIT_LIST_HEAD(&set->bindings); set->ops = ops; set->ktype = ktype; - set->klen = klen; + set->klen = desc.klen; set->dtype = dtype; - set->dlen = dlen; + set->dlen = desc.dlen; set->flags = flags; + set->size = desc.size; + + err = ops->init(set, &desc, nla); + if (err < 0) + goto err2; - err = ops->init(set, nla); + err = nft_trans_set_add(&ctx, NFT_MSG_NEWSET, set); if (err < 0) goto err2; list_add_tail(&set->list, &table->sets); - nf_tables_set_notify(&ctx, set, NFT_MSG_NEWSET); + table->use++; return 0; err2: @@ -2440,16 +2603,20 @@ err1: return err; } -static void nf_tables_set_destroy(const struct nft_ctx *ctx, struct nft_set *set) +static void nft_set_destroy(struct nft_set *set) { - list_del(&set->list); - nf_tables_set_notify(ctx, set, NFT_MSG_DELSET); - set->ops->destroy(set); module_put(set->ops->owner); kfree(set); } +static void nf_tables_set_destroy(const struct nft_ctx *ctx, struct nft_set *set) +{ + list_del(&set->list); + nf_tables_set_notify(ctx, set, NFT_MSG_DELSET, GFP_ATOMIC); + nft_set_destroy(set); +} + static int nf_tables_delset(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) @@ -2471,10 +2638,17 @@ static int nf_tables_delset(struct sock *nlsk, struct sk_buff *skb, set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_NAME]); if (IS_ERR(set)) return PTR_ERR(set); + if (set->flags & NFT_SET_INACTIVE) + return -ENOENT; if (!list_empty(&set->bindings)) return -EBUSY; - nf_tables_set_destroy(&ctx, set); + err = nft_trans_set_add(&ctx, NFT_MSG_DELSET, set); + if (err < 0) + return err; + + list_del(&set->list); + ctx.table->use--; return 0; } @@ -2534,7 +2708,8 @@ void nf_tables_unbind_set(const struct nft_ctx *ctx, struct nft_set *set, { list_del(&binding->list); - if (list_empty(&set->bindings) && set->flags & NFT_SET_ANONYMOUS) + if (list_empty(&set->bindings) && set->flags & NFT_SET_ANONYMOUS && + !(set->flags & NFT_SET_INACTIVE)) nf_tables_set_destroy(ctx, set); } @@ -2552,16 +2727,18 @@ static const struct nla_policy nft_set_elem_list_policy[NFTA_SET_ELEM_LIST_MAX + [NFTA_SET_ELEM_LIST_TABLE] = { .type = NLA_STRING }, [NFTA_SET_ELEM_LIST_SET] = { .type = NLA_STRING }, [NFTA_SET_ELEM_LIST_ELEMENTS] = { .type = NLA_NESTED }, + [NFTA_SET_ELEM_LIST_SET_ID] = { .type = NLA_U32 }, }; static int nft_ctx_init_from_elemattr(struct nft_ctx *ctx, const struct sk_buff *skb, const struct nlmsghdr *nlh, - const struct nlattr * const nla[]) + const struct nlattr * const nla[], + bool trans) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); - const struct nft_af_info *afi; - const struct nft_table *table; + struct nft_af_info *afi; + struct nft_table *table; struct net *net = sock_net(skb->sk); afi = nf_tables_afinfo_lookup(net, nfmsg->nfgen_family, false); @@ -2571,6 +2748,8 @@ static int nft_ctx_init_from_elemattr(struct nft_ctx *ctx, table = nf_tables_table_lookup(afi, nla[NFTA_SET_ELEM_LIST_TABLE]); if (IS_ERR(table)) return PTR_ERR(table); + if (!trans && (table->flags & NFT_TABLE_INACTIVE)) + return -ENOENT; nft_ctx_init(ctx, skb, nlh, afi, table, NULL, nla); return 0; @@ -2644,13 +2823,16 @@ static int nf_tables_dump_set(struct sk_buff *skb, struct netlink_callback *cb) if (err < 0) return err; - err = nft_ctx_init_from_elemattr(&ctx, cb->skb, cb->nlh, (void *)nla); + err = nft_ctx_init_from_elemattr(&ctx, cb->skb, cb->nlh, (void *)nla, + false); if (err < 0) return err; set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); if (IS_ERR(set)) return PTR_ERR(set); + if (set->flags & NFT_SET_INACTIVE) + return -ENOENT; event = NFT_MSG_NEWSETELEM; event |= NFNL_SUBSYS_NFTABLES << 8; @@ -2707,13 +2889,15 @@ static int nf_tables_getsetelem(struct sock *nlsk, struct sk_buff *skb, struct nft_ctx ctx; int err; - err = nft_ctx_init_from_elemattr(&ctx, skb, nlh, nla); + err = nft_ctx_init_from_elemattr(&ctx, skb, nlh, nla, false); if (err < 0) return err; set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); if (IS_ERR(set)) return PTR_ERR(set); + if (set->flags & NFT_SET_INACTIVE) + return -ENOENT; if (nlh->nlmsg_flags & NLM_F_DUMP) { struct netlink_dump_control c = { @@ -2724,7 +2908,98 @@ static int nf_tables_getsetelem(struct sock *nlsk, struct sk_buff *skb, return -EOPNOTSUPP; } -static int nft_add_set_elem(const struct nft_ctx *ctx, struct nft_set *set, +static int nf_tables_fill_setelem_info(struct sk_buff *skb, + const struct nft_ctx *ctx, u32 seq, + u32 portid, int event, u16 flags, + const struct nft_set *set, + const struct nft_set_elem *elem) +{ + struct nfgenmsg *nfmsg; + struct nlmsghdr *nlh; + struct nlattr *nest; + int err; + + event |= NFNL_SUBSYS_NFTABLES << 8; + nlh = nlmsg_put(skb, portid, seq, event, sizeof(struct nfgenmsg), + flags); + if (nlh == NULL) + goto nla_put_failure; + + nfmsg = nlmsg_data(nlh); + nfmsg->nfgen_family = ctx->afi->family; + nfmsg->version = NFNETLINK_V0; + nfmsg->res_id = 0; + + if (nla_put_string(skb, NFTA_SET_TABLE, ctx->table->name)) + goto nla_put_failure; + if (nla_put_string(skb, NFTA_SET_NAME, set->name)) + goto nla_put_failure; + + nest = nla_nest_start(skb, NFTA_SET_ELEM_LIST_ELEMENTS); + if (nest == NULL) + goto nla_put_failure; + + err = nf_tables_fill_setelem(skb, set, elem); + if (err < 0) + goto nla_put_failure; + + nla_nest_end(skb, nest); + + return nlmsg_end(skb, nlh); + +nla_put_failure: + nlmsg_trim(skb, nlh); + return -1; +} + +static int nf_tables_setelem_notify(const struct nft_ctx *ctx, + const struct nft_set *set, + const struct nft_set_elem *elem, + int event, u16 flags) +{ + struct net *net = ctx->net; + u32 portid = ctx->portid; + struct sk_buff *skb; + int err; + + if (!ctx->report && !nfnetlink_has_listeners(net, NFNLGRP_NFTABLES)) + return 0; + + err = -ENOBUFS; + skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + if (skb == NULL) + goto err; + + err = nf_tables_fill_setelem_info(skb, ctx, 0, portid, event, flags, + set, elem); + if (err < 0) { + kfree_skb(skb); + goto err; + } + + err = nfnetlink_send(skb, net, portid, NFNLGRP_NFTABLES, ctx->report, + GFP_KERNEL); +err: + if (err < 0) + nfnetlink_set_err(net, portid, NFNLGRP_NFTABLES, err); + return err; +} + +static struct nft_trans *nft_trans_elem_alloc(struct nft_ctx *ctx, + int msg_type, + struct nft_set *set) +{ + struct nft_trans *trans; + + trans = nft_trans_alloc(ctx, msg_type, sizeof(struct nft_trans_elem)); + if (trans == NULL) + return NULL; + + nft_trans_elem_set(trans) = set; + return trans; +} + +static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set, const struct nlattr *attr) { struct nlattr *nla[NFTA_SET_ELEM_MAX + 1]; @@ -2732,8 +3007,12 @@ static int nft_add_set_elem(const struct nft_ctx *ctx, struct nft_set *set, struct nft_set_elem elem; struct nft_set_binding *binding; enum nft_registers dreg; + struct nft_trans *trans; int err; + if (set->size && set->nelems == set->size) + return -ENFILE; + err = nla_parse_nested(nla, NFTA_SET_ELEM_MAX, attr, nft_set_elem_policy); if (err < 0) @@ -2786,7 +3065,7 @@ static int nft_add_set_elem(const struct nft_ctx *ctx, struct nft_set *set, struct nft_ctx bind_ctx = { .afi = ctx->afi, .table = ctx->table, - .chain = binding->chain, + .chain = (struct nft_chain *)binding->chain, }; err = nft_validate_data_load(&bind_ctx, dreg, @@ -2796,12 +3075,20 @@ static int nft_add_set_elem(const struct nft_ctx *ctx, struct nft_set *set, } } + trans = nft_trans_elem_alloc(ctx, NFT_MSG_NEWSETELEM, set); + if (trans == NULL) + goto err3; + err = set->ops->insert(set, &elem); if (err < 0) - goto err3; + goto err4; + nft_trans_elem(trans) = elem; + list_add_tail(&trans->list, &ctx->net->nft.commit_list); return 0; +err4: + kfree(trans); err3: if (nla[NFTA_SET_ELEM_DATA] != NULL) nft_data_uninit(&elem.data, d2.type); @@ -2815,35 +3102,46 @@ static int nf_tables_newsetelem(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { + struct net *net = sock_net(skb->sk); const struct nlattr *attr; struct nft_set *set; struct nft_ctx ctx; - int rem, err; + int rem, err = 0; - err = nft_ctx_init_from_elemattr(&ctx, skb, nlh, nla); + err = nft_ctx_init_from_elemattr(&ctx, skb, nlh, nla, true); if (err < 0) return err; set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); - if (IS_ERR(set)) - return PTR_ERR(set); + if (IS_ERR(set)) { + if (nla[NFTA_SET_ELEM_LIST_SET_ID]) { + set = nf_tables_set_lookup_byid(net, + nla[NFTA_SET_ELEM_LIST_SET_ID]); + } + if (IS_ERR(set)) + return PTR_ERR(set); + } + if (!list_empty(&set->bindings) && set->flags & NFT_SET_CONSTANT) return -EBUSY; nla_for_each_nested(attr, nla[NFTA_SET_ELEM_LIST_ELEMENTS], rem) { err = nft_add_set_elem(&ctx, set, attr); if (err < 0) - return err; + break; + + set->nelems++; } - return 0; + return err; } -static int nft_del_setelem(const struct nft_ctx *ctx, struct nft_set *set, +static int nft_del_setelem(struct nft_ctx *ctx, struct nft_set *set, const struct nlattr *attr) { struct nlattr *nla[NFTA_SET_ELEM_MAX + 1]; struct nft_data_desc desc; struct nft_set_elem elem; + struct nft_trans *trans; int err; err = nla_parse_nested(nla, NFTA_SET_ELEM_MAX, attr, @@ -2867,7 +3165,12 @@ static int nft_del_setelem(const struct nft_ctx *ctx, struct nft_set *set, if (err < 0) goto err2; - set->ops->remove(set, &elem); + trans = nft_trans_elem_alloc(ctx, NFT_MSG_DELSETELEM, set); + if (trans == NULL) + goto err2; + + nft_trans_elem(trans) = elem; + list_add_tail(&trans->list, &ctx->net->nft.commit_list); nft_data_uninit(&elem.key, NFT_DATA_VALUE); if (set->flags & NFT_SET_MAP) @@ -2886,9 +3189,9 @@ static int nf_tables_delsetelem(struct sock *nlsk, struct sk_buff *skb, const struct nlattr *attr; struct nft_set *set; struct nft_ctx ctx; - int rem, err; + int rem, err = 0; - err = nft_ctx_init_from_elemattr(&ctx, skb, nlh, nla); + err = nft_ctx_init_from_elemattr(&ctx, skb, nlh, nla, false); if (err < 0) return err; @@ -2901,14 +3204,16 @@ static int nf_tables_delsetelem(struct sock *nlsk, struct sk_buff *skb, nla_for_each_nested(attr, nla[NFTA_SET_ELEM_LIST_ELEMENTS], rem) { err = nft_del_setelem(&ctx, set, attr); if (err < 0) - return err; + break; + + set->nelems--; } - return 0; + return err; } static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { [NFT_MSG_NEWTABLE] = { - .call = nf_tables_newtable, + .call_batch = nf_tables_newtable, .attr_count = NFTA_TABLE_MAX, .policy = nft_table_policy, }, @@ -2918,12 +3223,12 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { .policy = nft_table_policy, }, [NFT_MSG_DELTABLE] = { - .call = nf_tables_deltable, + .call_batch = nf_tables_deltable, .attr_count = NFTA_TABLE_MAX, .policy = nft_table_policy, }, [NFT_MSG_NEWCHAIN] = { - .call = nf_tables_newchain, + .call_batch = nf_tables_newchain, .attr_count = NFTA_CHAIN_MAX, .policy = nft_chain_policy, }, @@ -2933,7 +3238,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { .policy = nft_chain_policy, }, [NFT_MSG_DELCHAIN] = { - .call = nf_tables_delchain, + .call_batch = nf_tables_delchain, .attr_count = NFTA_CHAIN_MAX, .policy = nft_chain_policy, }, @@ -2953,7 +3258,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { .policy = nft_rule_policy, }, [NFT_MSG_NEWSET] = { - .call = nf_tables_newset, + .call_batch = nf_tables_newset, .attr_count = NFTA_SET_MAX, .policy = nft_set_policy, }, @@ -2963,12 +3268,12 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { .policy = nft_set_policy, }, [NFT_MSG_DELSET] = { - .call = nf_tables_delset, + .call_batch = nf_tables_delset, .attr_count = NFTA_SET_MAX, .policy = nft_set_policy, }, [NFT_MSG_NEWSETELEM] = { - .call = nf_tables_newsetelem, + .call_batch = nf_tables_newsetelem, .attr_count = NFTA_SET_ELEM_LIST_MAX, .policy = nft_set_elem_list_policy, }, @@ -2978,12 +3283,282 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { .policy = nft_set_elem_list_policy, }, [NFT_MSG_DELSETELEM] = { - .call = nf_tables_delsetelem, + .call_batch = nf_tables_delsetelem, .attr_count = NFTA_SET_ELEM_LIST_MAX, .policy = nft_set_elem_list_policy, }, }; +static void nft_chain_commit_update(struct nft_trans *trans) +{ + struct nft_base_chain *basechain; + + if (nft_trans_chain_name(trans)[0]) + strcpy(trans->ctx.chain->name, nft_trans_chain_name(trans)); + + if (!(trans->ctx.chain->flags & NFT_BASE_CHAIN)) + return; + + basechain = nft_base_chain(trans->ctx.chain); + nft_chain_stats_replace(basechain, nft_trans_chain_stats(trans)); + + switch (nft_trans_chain_policy(trans)) { + case NF_DROP: + case NF_ACCEPT: + basechain->policy = nft_trans_chain_policy(trans); + break; + } +} + +/* Schedule objects for release via rcu to make sure no packets are accesing + * removed rules. + */ +static void nf_tables_commit_release_rcu(struct rcu_head *rt) +{ + struct nft_trans *trans = container_of(rt, struct nft_trans, rcu_head); + + switch (trans->msg_type) { + case NFT_MSG_DELTABLE: + nf_tables_table_destroy(&trans->ctx); + break; + case NFT_MSG_DELCHAIN: + nf_tables_chain_destroy(trans->ctx.chain); + break; + case NFT_MSG_DELRULE: + nf_tables_rule_destroy(&trans->ctx, nft_trans_rule(trans)); + break; + case NFT_MSG_DELSET: + nft_set_destroy(nft_trans_set(trans)); + break; + } + kfree(trans); +} + +static int nf_tables_commit(struct sk_buff *skb) +{ + struct net *net = sock_net(skb->sk); + struct nft_trans *trans, *next; + struct nft_set *set; + + /* Bump generation counter, invalidate any dump in progress */ + net->nft.genctr++; + + /* A new generation has just started */ + net->nft.gencursor = gencursor_next(net); + + /* Make sure all packets have left the previous generation before + * purging old rules. + */ + synchronize_rcu(); + + list_for_each_entry_safe(trans, next, &net->nft.commit_list, list) { + switch (trans->msg_type) { + case NFT_MSG_NEWTABLE: + if (nft_trans_table_update(trans)) { + if (!nft_trans_table_enable(trans)) { + nf_tables_table_disable(trans->ctx.afi, + trans->ctx.table); + trans->ctx.table->flags |= NFT_TABLE_F_DORMANT; + } + } else { + trans->ctx.table->flags &= ~NFT_TABLE_INACTIVE; + } + nf_tables_table_notify(&trans->ctx, NFT_MSG_NEWTABLE); + nft_trans_destroy(trans); + break; + case NFT_MSG_DELTABLE: + nf_tables_table_notify(&trans->ctx, NFT_MSG_DELTABLE); + break; + case NFT_MSG_NEWCHAIN: + if (nft_trans_chain_update(trans)) + nft_chain_commit_update(trans); + else + trans->ctx.chain->flags &= ~NFT_CHAIN_INACTIVE; + + nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN); + nft_trans_destroy(trans); + break; + case NFT_MSG_DELCHAIN: + nf_tables_chain_notify(&trans->ctx, NFT_MSG_DELCHAIN); + if (!(trans->ctx.table->flags & NFT_TABLE_F_DORMANT) && + trans->ctx.chain->flags & NFT_BASE_CHAIN) { + nf_unregister_hooks(nft_base_chain(trans->ctx.chain)->ops, + trans->ctx.afi->nops); + } + break; + case NFT_MSG_NEWRULE: + nft_rule_clear(trans->ctx.net, nft_trans_rule(trans)); + nf_tables_rule_notify(&trans->ctx, + nft_trans_rule(trans), + NFT_MSG_NEWRULE); + nft_trans_destroy(trans); + break; + case NFT_MSG_DELRULE: + list_del_rcu(&nft_trans_rule(trans)->list); + nf_tables_rule_notify(&trans->ctx, + nft_trans_rule(trans), + NFT_MSG_DELRULE); + break; + case NFT_MSG_NEWSET: + nft_trans_set(trans)->flags &= ~NFT_SET_INACTIVE; + /* This avoids hitting -EBUSY when deleting the table + * from the transaction. + */ + if (nft_trans_set(trans)->flags & NFT_SET_ANONYMOUS && + !list_empty(&nft_trans_set(trans)->bindings)) + trans->ctx.table->use--; + + nf_tables_set_notify(&trans->ctx, nft_trans_set(trans), + NFT_MSG_NEWSET, GFP_KERNEL); + nft_trans_destroy(trans); + break; + case NFT_MSG_DELSET: + nf_tables_set_notify(&trans->ctx, nft_trans_set(trans), + NFT_MSG_DELSET, GFP_KERNEL); + break; + case NFT_MSG_NEWSETELEM: + nf_tables_setelem_notify(&trans->ctx, + nft_trans_elem_set(trans), + &nft_trans_elem(trans), + NFT_MSG_NEWSETELEM, 0); + nft_trans_destroy(trans); + break; + case NFT_MSG_DELSETELEM: + nf_tables_setelem_notify(&trans->ctx, + nft_trans_elem_set(trans), + &nft_trans_elem(trans), + NFT_MSG_DELSETELEM, 0); + set = nft_trans_elem_set(trans); + set->ops->get(set, &nft_trans_elem(trans)); + set->ops->remove(set, &nft_trans_elem(trans)); + nft_trans_destroy(trans); + break; + } + } + + list_for_each_entry_safe(trans, next, &net->nft.commit_list, list) { + list_del(&trans->list); + trans->ctx.nla = NULL; + call_rcu(&trans->rcu_head, nf_tables_commit_release_rcu); + } + + return 0; +} + +/* Schedule objects for release via rcu to make sure no packets are accesing + * aborted rules. + */ +static void nf_tables_abort_release_rcu(struct rcu_head *rt) +{ + struct nft_trans *trans = container_of(rt, struct nft_trans, rcu_head); + + switch (trans->msg_type) { + case NFT_MSG_NEWTABLE: + nf_tables_table_destroy(&trans->ctx); + break; + case NFT_MSG_NEWCHAIN: + nf_tables_chain_destroy(trans->ctx.chain); + break; + case NFT_MSG_NEWRULE: + nf_tables_rule_destroy(&trans->ctx, nft_trans_rule(trans)); + break; + case NFT_MSG_NEWSET: + nft_set_destroy(nft_trans_set(trans)); + break; + } + kfree(trans); +} + +static int nf_tables_abort(struct sk_buff *skb) +{ + struct net *net = sock_net(skb->sk); + struct nft_trans *trans, *next; + struct nft_set *set; + + list_for_each_entry_safe(trans, next, &net->nft.commit_list, list) { + switch (trans->msg_type) { + case NFT_MSG_NEWTABLE: + if (nft_trans_table_update(trans)) { + if (nft_trans_table_enable(trans)) { + nf_tables_table_disable(trans->ctx.afi, + trans->ctx.table); + trans->ctx.table->flags |= NFT_TABLE_F_DORMANT; + } + nft_trans_destroy(trans); + } else { + list_del(&trans->ctx.table->list); + } + break; + case NFT_MSG_DELTABLE: + list_add_tail(&trans->ctx.table->list, + &trans->ctx.afi->tables); + nft_trans_destroy(trans); + break; + case NFT_MSG_NEWCHAIN: + if (nft_trans_chain_update(trans)) { + if (nft_trans_chain_stats(trans)) + free_percpu(nft_trans_chain_stats(trans)); + + nft_trans_destroy(trans); + } else { + trans->ctx.table->use--; + list_del(&trans->ctx.chain->list); + if (!(trans->ctx.table->flags & NFT_TABLE_F_DORMANT) && + trans->ctx.chain->flags & NFT_BASE_CHAIN) { + nf_unregister_hooks(nft_base_chain(trans->ctx.chain)->ops, + trans->ctx.afi->nops); + } + } + break; + case NFT_MSG_DELCHAIN: + trans->ctx.table->use++; + list_add_tail(&trans->ctx.chain->list, + &trans->ctx.table->chains); + nft_trans_destroy(trans); + break; + case NFT_MSG_NEWRULE: + trans->ctx.chain->use--; + list_del_rcu(&nft_trans_rule(trans)->list); + break; + case NFT_MSG_DELRULE: + trans->ctx.chain->use++; + nft_rule_clear(trans->ctx.net, nft_trans_rule(trans)); + nft_trans_destroy(trans); + break; + case NFT_MSG_NEWSET: + trans->ctx.table->use--; + list_del(&nft_trans_set(trans)->list); + break; + case NFT_MSG_DELSET: + trans->ctx.table->use++; + list_add_tail(&nft_trans_set(trans)->list, + &trans->ctx.table->sets); + nft_trans_destroy(trans); + break; + case NFT_MSG_NEWSETELEM: + nft_trans_elem_set(trans)->nelems--; + set = nft_trans_elem_set(trans); + set->ops->get(set, &nft_trans_elem(trans)); + set->ops->remove(set, &nft_trans_elem(trans)); + nft_trans_destroy(trans); + break; + case NFT_MSG_DELSETELEM: + nft_trans_elem_set(trans)->nelems++; + nft_trans_destroy(trans); + break; + } + } + + list_for_each_entry_safe_reverse(trans, next, + &net->nft.commit_list, list) { + list_del(&trans->list); + trans->ctx.nla = NULL; + call_rcu(&trans->rcu_head, nf_tables_abort_release_rcu); + } + + return 0; +} + static const struct nfnetlink_subsystem nf_tables_subsys = { .name = "nf_tables", .subsys_id = NFNL_SUBSYS_NFTABLES, diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c index 804105391b9..345acfb1720 100644 --- a/net/netfilter/nf_tables_core.c +++ b/net/netfilter/nf_tables_core.c @@ -66,20 +66,6 @@ struct nft_jumpstack { int rulenum; }; -static inline void -nft_chain_stats(const struct nft_chain *this, const struct nft_pktinfo *pkt, - struct nft_jumpstack *jumpstack, unsigned int stackptr) -{ - struct nft_stats __percpu *stats; - const struct nft_chain *chain = stackptr ? jumpstack[0].chain : this; - - rcu_read_lock_bh(); - stats = rcu_dereference(nft_base_chain(chain)->stats); - __this_cpu_inc(stats->pkts); - __this_cpu_add(stats->bytes, pkt->skb->len); - rcu_read_unlock_bh(); -} - enum nft_trace { NFT_TRACE_RULE, NFT_TRACE_RETURN, @@ -117,13 +103,14 @@ static void nft_trace_packet(const struct nft_pktinfo *pkt, unsigned int nft_do_chain(struct nft_pktinfo *pkt, const struct nf_hook_ops *ops) { - const struct nft_chain *chain = ops->priv; + const struct nft_chain *chain = ops->priv, *basechain = chain; const struct nft_rule *rule; const struct nft_expr *expr, *last; struct nft_data data[NFT_REG_MAX + 1]; unsigned int stackptr = 0; struct nft_jumpstack jumpstack[NFT_JUMP_STACK_SIZE]; - int rulenum = 0; + struct nft_stats __percpu *stats; + int rulenum; /* * Cache cursor to avoid problems in case that the cursor is updated * while traversing the ruleset. @@ -131,6 +118,7 @@ nft_do_chain(struct nft_pktinfo *pkt, const struct nf_hook_ops *ops) unsigned int gencursor = ACCESS_ONCE(chain->net->nft.gencursor); do_chain: + rulenum = 0; rule = list_entry(&chain->rules, struct nft_rule, list); next_rule: data[NFT_REG_VERDICT].verdict = NFT_CONTINUE; @@ -156,8 +144,10 @@ next_rule: switch (data[NFT_REG_VERDICT].verdict) { case NFT_BREAK: data[NFT_REG_VERDICT].verdict = NFT_CONTINUE; - /* fall through */ + continue; case NFT_CONTINUE: + if (unlikely(pkt->skb->nf_trace)) + nft_trace_packet(pkt, chain, rulenum, NFT_TRACE_RULE); continue; } break; @@ -183,37 +173,44 @@ next_rule: jumpstack[stackptr].rule = rule; jumpstack[stackptr].rulenum = rulenum; stackptr++; - /* fall through */ + chain = data[NFT_REG_VERDICT].chain; + goto do_chain; case NFT_GOTO: + if (unlikely(pkt->skb->nf_trace)) + nft_trace_packet(pkt, chain, rulenum, NFT_TRACE_RULE); + chain = data[NFT_REG_VERDICT].chain; goto do_chain; case NFT_RETURN: if (unlikely(pkt->skb->nf_trace)) nft_trace_packet(pkt, chain, rulenum, NFT_TRACE_RETURN); - - /* fall through */ + break; case NFT_CONTINUE: + if (unlikely(pkt->skb->nf_trace && !(chain->flags & NFT_BASE_CHAIN))) + nft_trace_packet(pkt, chain, ++rulenum, NFT_TRACE_RETURN); break; default: WARN_ON(1); } if (stackptr > 0) { - if (unlikely(pkt->skb->nf_trace)) - nft_trace_packet(pkt, chain, ++rulenum, NFT_TRACE_RETURN); - stackptr--; chain = jumpstack[stackptr].chain; rule = jumpstack[stackptr].rule; rulenum = jumpstack[stackptr].rulenum; goto next_rule; } - nft_chain_stats(chain, pkt, jumpstack, stackptr); if (unlikely(pkt->skb->nf_trace)) - nft_trace_packet(pkt, chain, ++rulenum, NFT_TRACE_POLICY); + nft_trace_packet(pkt, basechain, -1, NFT_TRACE_POLICY); + + rcu_read_lock_bh(); + stats = rcu_dereference(nft_base_chain(basechain)->stats); + __this_cpu_inc(stats->pkts); + __this_cpu_add(stats->bytes, pkt->skb->len); + rcu_read_unlock_bh(); - return nft_base_chain(chain)->policy; + return nft_base_chain(basechain)->policy; } EXPORT_SYMBOL_GPL(nft_do_chain); diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index e8138da4c14..c138b8fbe28 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -256,15 +256,15 @@ replay: #endif { nfnl_unlock(subsys_id); - kfree_skb(nskb); - return netlink_ack(skb, nlh, -EOPNOTSUPP); + netlink_ack(skb, nlh, -EOPNOTSUPP); + return kfree_skb(nskb); } } if (!ss->commit || !ss->abort) { nfnl_unlock(subsys_id); - kfree_skb(nskb); - return netlink_ack(skb, nlh, -EOPNOTSUPP); + netlink_ack(skb, nlh, -EOPNOTSUPP); + return kfree_skb(skb); } while (skb->len >= nlmsg_total_size(0)) { @@ -368,14 +368,13 @@ done: static void nfnetlink_rcv(struct sk_buff *skb) { struct nlmsghdr *nlh = nlmsg_hdr(skb); - struct net *net = sock_net(skb->sk); int msglen; if (nlh->nlmsg_len < NLMSG_HDRLEN || skb->len < nlh->nlmsg_len) return; - if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) { + if (!netlink_net_capable(skb, CAP_NET_ADMIN)) { netlink_ack(skb, nlh, -EPERM); return; } @@ -400,19 +399,17 @@ static void nfnetlink_rcv(struct sk_buff *skb) } #ifdef CONFIG_MODULES -static void nfnetlink_bind(int group) +static int nfnetlink_bind(int group) { const struct nfnetlink_subsystem *ss; int type = nfnl_group2type[group]; rcu_read_lock(); ss = nfnetlink_get_subsys(type); - if (!ss) { - rcu_read_unlock(); - request_module("nfnetlink-subsys-%d", type); - return; - } rcu_read_unlock(); + if (!ss) + request_module("nfnetlink-subsys-%d", type); + return 0; } #endif diff --git a/net/netfilter/nfnetlink_acct.c b/net/netfilter/nfnetlink_acct.c index c7b6d466a66..54af9853e2c 100644 --- a/net/netfilter/nfnetlink_acct.c +++ b/net/netfilter/nfnetlink_acct.c @@ -32,18 +32,24 @@ static LIST_HEAD(nfnl_acct_list); struct nf_acct { atomic64_t pkts; atomic64_t bytes; + unsigned long flags; struct list_head head; atomic_t refcnt; char name[NFACCT_NAME_MAX]; struct rcu_head rcu_head; + char data[0]; }; +#define NFACCT_F_QUOTA (NFACCT_F_QUOTA_PKTS | NFACCT_F_QUOTA_BYTES) + static int nfnl_acct_new(struct sock *nfnl, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const tb[]) { struct nf_acct *nfacct, *matching = NULL; char *acct_name; + unsigned int size = 0; + u32 flags = 0; if (!tb[NFACCT_NAME]) return -EINVAL; @@ -68,15 +74,38 @@ nfnl_acct_new(struct sock *nfnl, struct sk_buff *skb, /* reset counters if you request a replacement. */ atomic64_set(&matching->pkts, 0); atomic64_set(&matching->bytes, 0); + smp_mb__before_clear_bit(); + /* reset overquota flag if quota is enabled. */ + if ((matching->flags & NFACCT_F_QUOTA)) + clear_bit(NFACCT_F_OVERQUOTA, &matching->flags); return 0; } return -EBUSY; } - nfacct = kzalloc(sizeof(struct nf_acct), GFP_KERNEL); + if (tb[NFACCT_FLAGS]) { + flags = ntohl(nla_get_be32(tb[NFACCT_FLAGS])); + if (flags & ~NFACCT_F_QUOTA) + return -EOPNOTSUPP; + if ((flags & NFACCT_F_QUOTA) == NFACCT_F_QUOTA) + return -EINVAL; + if (flags & NFACCT_F_OVERQUOTA) + return -EINVAL; + + size += sizeof(u64); + } + + nfacct = kzalloc(sizeof(struct nf_acct) + size, GFP_KERNEL); if (nfacct == NULL) return -ENOMEM; + if (flags & NFACCT_F_QUOTA) { + u64 *quota = (u64 *)nfacct->data; + + *quota = be64_to_cpu(nla_get_be64(tb[NFACCT_QUOTA])); + nfacct->flags = flags; + } + strncpy(nfacct->name, nla_data(tb[NFACCT_NAME]), NFACCT_NAME_MAX); if (tb[NFACCT_BYTES]) { @@ -117,6 +146,9 @@ nfnl_acct_fill_info(struct sk_buff *skb, u32 portid, u32 seq, u32 type, if (type == NFNL_MSG_ACCT_GET_CTRZERO) { pkts = atomic64_xchg(&acct->pkts, 0); bytes = atomic64_xchg(&acct->bytes, 0); + smp_mb__before_clear_bit(); + if (acct->flags & NFACCT_F_QUOTA) + clear_bit(NFACCT_F_OVERQUOTA, &acct->flags); } else { pkts = atomic64_read(&acct->pkts); bytes = atomic64_read(&acct->bytes); @@ -125,7 +157,13 @@ nfnl_acct_fill_info(struct sk_buff *skb, u32 portid, u32 seq, u32 type, nla_put_be64(skb, NFACCT_BYTES, cpu_to_be64(bytes)) || nla_put_be32(skb, NFACCT_USE, htonl(atomic_read(&acct->refcnt)))) goto nla_put_failure; + if (acct->flags & NFACCT_F_QUOTA) { + u64 *quota = (u64 *)acct->data; + if (nla_put_be32(skb, NFACCT_FLAGS, htonl(acct->flags)) || + nla_put_be64(skb, NFACCT_QUOTA, cpu_to_be64(*quota))) + goto nla_put_failure; + } nlmsg_end(skb, nlh); return skb->len; @@ -270,6 +308,8 @@ static const struct nla_policy nfnl_acct_policy[NFACCT_MAX+1] = { [NFACCT_NAME] = { .type = NLA_NUL_STRING, .len = NFACCT_NAME_MAX-1 }, [NFACCT_BYTES] = { .type = NLA_U64 }, [NFACCT_PKTS] = { .type = NLA_U64 }, + [NFACCT_FLAGS] = { .type = NLA_U32 }, + [NFACCT_QUOTA] = { .type = NLA_U64 }, }; static const struct nfnl_callback nfnl_acct_cb[NFNL_MSG_ACCT_MAX] = { @@ -336,6 +376,50 @@ void nfnl_acct_update(const struct sk_buff *skb, struct nf_acct *nfacct) } EXPORT_SYMBOL_GPL(nfnl_acct_update); +static void nfnl_overquota_report(struct nf_acct *nfacct) +{ + int ret; + struct sk_buff *skb; + + skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC); + if (skb == NULL) + return; + + ret = nfnl_acct_fill_info(skb, 0, 0, NFNL_MSG_ACCT_OVERQUOTA, 0, + nfacct); + if (ret <= 0) { + kfree_skb(skb); + return; + } + netlink_broadcast(init_net.nfnl, skb, 0, NFNLGRP_ACCT_QUOTA, + GFP_ATOMIC); +} + +int nfnl_acct_overquota(const struct sk_buff *skb, struct nf_acct *nfacct) +{ + u64 now; + u64 *quota; + int ret = NFACCT_UNDERQUOTA; + + /* no place here if we don't have a quota */ + if (!(nfacct->flags & NFACCT_F_QUOTA)) + return NFACCT_NO_QUOTA; + + quota = (u64 *)nfacct->data; + now = (nfacct->flags & NFACCT_F_QUOTA_PKTS) ? + atomic64_read(&nfacct->pkts) : atomic64_read(&nfacct->bytes); + + ret = now > *quota; + + if (now >= *quota && + !test_and_set_bit(NFACCT_F_OVERQUOTA, &nfacct->flags)) { + nfnl_overquota_report(nfacct); + } + + return ret; +} +EXPORT_SYMBOL_GPL(nfnl_acct_overquota); + static int __init nfnl_acct_init(void) { int ret; diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c index bd0d41e6934..cc560301624 100644 --- a/net/netfilter/nft_ct.c +++ b/net/netfilter/nft_ct.c @@ -215,22 +215,14 @@ static void nft_ct_l3proto_module_put(uint8_t family) nf_ct_l3proto_module_put(family); } -static int nft_ct_init_validate_get(const struct nft_expr *expr, - const struct nlattr * const tb[]) +static int nft_ct_get_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) { struct nft_ct *priv = nft_expr_priv(expr); + int err; - if (tb[NFTA_CT_DIRECTION] != NULL) { - priv->dir = nla_get_u8(tb[NFTA_CT_DIRECTION]); - switch (priv->dir) { - case IP_CT_DIR_ORIGINAL: - case IP_CT_DIR_REPLY: - break; - default: - return -EINVAL; - } - } - + priv->key = ntohl(nla_get_be32(tb[NFTA_CT_KEY])); switch (priv->key) { case NFT_CT_STATE: case NFT_CT_DIRECTION: @@ -262,55 +254,55 @@ static int nft_ct_init_validate_get(const struct nft_expr *expr, return -EOPNOTSUPP; } - return 0; -} - -static int nft_ct_init_validate_set(uint32_t key) -{ - switch (key) { - case NFT_CT_MARK: - break; - default: - return -EOPNOTSUPP; + if (tb[NFTA_CT_DIRECTION] != NULL) { + priv->dir = nla_get_u8(tb[NFTA_CT_DIRECTION]); + switch (priv->dir) { + case IP_CT_DIR_ORIGINAL: + case IP_CT_DIR_REPLY: + break; + default: + return -EINVAL; + } } + priv->dreg = ntohl(nla_get_be32(tb[NFTA_CT_DREG])); + err = nft_validate_output_register(priv->dreg); + if (err < 0) + return err; + + err = nft_validate_data_load(ctx, priv->dreg, NULL, NFT_DATA_VALUE); + if (err < 0) + return err; + + err = nft_ct_l3proto_try_module_get(ctx->afi->family); + if (err < 0) + return err; + return 0; } -static int nft_ct_init(const struct nft_ctx *ctx, - const struct nft_expr *expr, - const struct nlattr * const tb[]) +static int nft_ct_set_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) { struct nft_ct *priv = nft_expr_priv(expr); int err; priv->key = ntohl(nla_get_be32(tb[NFTA_CT_KEY])); - - if (tb[NFTA_CT_DREG]) { - err = nft_ct_init_validate_get(expr, tb); - if (err < 0) - return err; - - priv->dreg = ntohl(nla_get_be32(tb[NFTA_CT_DREG])); - err = nft_validate_output_register(priv->dreg); - if (err < 0) - return err; - - err = nft_validate_data_load(ctx, priv->dreg, NULL, - NFT_DATA_VALUE); - if (err < 0) - return err; - } else { - err = nft_ct_init_validate_set(priv->key); - if (err < 0) - return err; - - priv->sreg = ntohl(nla_get_be32(tb[NFTA_CT_SREG])); - err = nft_validate_input_register(priv->sreg); - if (err < 0) - return err; + switch (priv->key) { +#ifdef CONFIG_NF_CONNTRACK_MARK + case NFT_CT_MARK: + break; +#endif + default: + return -EOPNOTSUPP; } + priv->sreg = ntohl(nla_get_be32(tb[NFTA_CT_SREG])); + err = nft_validate_input_register(priv->sreg); + if (err < 0) + return err; + err = nft_ct_l3proto_try_module_get(ctx->afi->family); if (err < 0) return err; @@ -370,7 +362,7 @@ static const struct nft_expr_ops nft_ct_get_ops = { .type = &nft_ct_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_ct)), .eval = nft_ct_get_eval, - .init = nft_ct_init, + .init = nft_ct_get_init, .destroy = nft_ct_destroy, .dump = nft_ct_get_dump, }; @@ -379,7 +371,7 @@ static const struct nft_expr_ops nft_ct_set_ops = { .type = &nft_ct_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_ct)), .eval = nft_ct_set_eval, - .init = nft_ct_init, + .init = nft_ct_set_init, .destroy = nft_ct_destroy, .dump = nft_ct_set_dump, }; diff --git a/net/netfilter/nft_hash.c b/net/netfilter/nft_hash.c index 3b1ad876d6b..4080ed6a072 100644 --- a/net/netfilter/nft_hash.c +++ b/net/netfilter/nft_hash.c @@ -12,6 +12,7 @@ #include <linux/init.h> #include <linux/module.h> #include <linux/list.h> +#include <linux/log2.h> #include <linux/jhash.h> #include <linux/netlink.h> #include <linux/vmalloc.h> @@ -19,7 +20,7 @@ #include <linux/netfilter/nf_tables.h> #include <net/netfilter/nf_tables.h> -#define NFT_HASH_MIN_SIZE 4 +#define NFT_HASH_MIN_SIZE 4UL struct nft_hash { struct nft_hash_table __rcu *tbl; @@ -27,7 +28,6 @@ struct nft_hash { struct nft_hash_table { unsigned int size; - unsigned int elements; struct nft_hash_elem __rcu *buckets[]; }; @@ -76,10 +76,12 @@ static bool nft_hash_lookup(const struct nft_set *set, static void nft_hash_tbl_free(const struct nft_hash_table *tbl) { - if (is_vmalloc_addr(tbl)) - vfree(tbl); - else - kfree(tbl); + kvfree(tbl); +} + +static unsigned int nft_hash_tbl_size(unsigned int nelem) +{ + return max(roundup_pow_of_two(nelem * 4 / 3), NFT_HASH_MIN_SIZE); } static struct nft_hash_table *nft_hash_tbl_alloc(unsigned int nbuckets) @@ -161,7 +163,6 @@ static int nft_hash_tbl_expand(const struct nft_set *set, struct nft_hash *priv) break; } } - ntbl->elements = tbl->elements; /* Publish new table */ rcu_assign_pointer(priv->tbl, ntbl); @@ -201,7 +202,6 @@ static int nft_hash_tbl_shrink(const struct nft_set *set, struct nft_hash *priv) ; RCU_INIT_POINTER(*pprev, tbl->buckets[i + ntbl->size]); } - ntbl->elements = tbl->elements; /* Publish new table */ rcu_assign_pointer(priv->tbl, ntbl); @@ -237,10 +237,9 @@ static int nft_hash_insert(const struct nft_set *set, h = nft_hash_data(&he->key, tbl->size, set->klen); RCU_INIT_POINTER(he->next, tbl->buckets[h]); rcu_assign_pointer(tbl->buckets[h], he); - tbl->elements++; /* Expand table when exceeding 75% load */ - if (tbl->elements > tbl->size / 4 * 3) + if (set->nelems + 1 > tbl->size / 4 * 3) nft_hash_tbl_expand(set, priv); return 0; @@ -268,10 +267,9 @@ static void nft_hash_remove(const struct nft_set *set, RCU_INIT_POINTER(*pprev, he->next); synchronize_rcu(); kfree(he); - tbl->elements--; /* Shrink table beneath 30% load */ - if (tbl->elements < tbl->size * 3 / 10 && + if (set->nelems - 1 < tbl->size * 3 / 10 && tbl->size > NFT_HASH_MIN_SIZE) nft_hash_tbl_shrink(set, priv); } @@ -335,17 +333,23 @@ static unsigned int nft_hash_privsize(const struct nlattr * const nla[]) } static int nft_hash_init(const struct nft_set *set, + const struct nft_set_desc *desc, const struct nlattr * const tb[]) { struct nft_hash *priv = nft_set_priv(set); struct nft_hash_table *tbl; + unsigned int size; if (unlikely(!nft_hash_rnd_initted)) { get_random_bytes(&nft_hash_rnd, 4); nft_hash_rnd_initted = true; } - tbl = nft_hash_tbl_alloc(NFT_HASH_MIN_SIZE); + size = NFT_HASH_MIN_SIZE; + if (desc->size) + size = nft_hash_tbl_size(desc->size); + + tbl = nft_hash_tbl_alloc(size); if (tbl == NULL) return -ENOMEM; RCU_INIT_POINTER(priv->tbl, tbl); @@ -369,8 +373,37 @@ static void nft_hash_destroy(const struct nft_set *set) kfree(tbl); } +static bool nft_hash_estimate(const struct nft_set_desc *desc, u32 features, + struct nft_set_estimate *est) +{ + unsigned int esize; + + esize = sizeof(struct nft_hash_elem); + if (features & NFT_SET_MAP) + esize += FIELD_SIZEOF(struct nft_hash_elem, data[0]); + + if (desc->size) { + est->size = sizeof(struct nft_hash) + + nft_hash_tbl_size(desc->size) * + sizeof(struct nft_hash_elem *) + + desc->size * esize; + } else { + /* Resizing happens when the load drops below 30% or goes + * above 75%. The average of 52.5% load (approximated by 50%) + * is used for the size estimation of the hash buckets, + * meaning we calculate two buckets per element. + */ + est->size = esize + 2 * sizeof(struct nft_hash_elem *); + } + + est->class = NFT_SET_CLASS_O_1; + + return true; +} + static struct nft_set_ops nft_hash_ops __read_mostly = { .privsize = nft_hash_privsize, + .estimate = nft_hash_estimate, .init = nft_hash_init, .destroy = nft_hash_destroy, .get = nft_hash_get, diff --git a/net/netfilter/nft_lookup.c b/net/netfilter/nft_lookup.c index 7fd2bea8aa2..6404a726d17 100644 --- a/net/netfilter/nft_lookup.c +++ b/net/netfilter/nft_lookup.c @@ -56,8 +56,14 @@ static int nft_lookup_init(const struct nft_ctx *ctx, return -EINVAL; set = nf_tables_set_lookup(ctx->table, tb[NFTA_LOOKUP_SET]); - if (IS_ERR(set)) - return PTR_ERR(set); + if (IS_ERR(set)) { + if (tb[NFTA_LOOKUP_SET_ID]) { + set = nf_tables_set_lookup_byid(ctx->net, + tb[NFTA_LOOKUP_SET_ID]); + } + if (IS_ERR(set)) + return PTR_ERR(set); + } priv->sreg = ntohl(nla_get_be32(tb[NFTA_LOOKUP_SREG])); err = nft_validate_input_register(priv->sreg); diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c index 425cf39af89..852b178c6ae 100644 --- a/net/netfilter/nft_meta.c +++ b/net/netfilter/nft_meta.c @@ -18,18 +18,11 @@ #include <net/sock.h> #include <net/tcp_states.h> /* for TCP_TIME_WAIT */ #include <net/netfilter/nf_tables.h> +#include <net/netfilter/nft_meta.h> -struct nft_meta { - enum nft_meta_keys key:8; - union { - enum nft_registers dreg:8; - enum nft_registers sreg:8; - }; -}; - -static void nft_meta_get_eval(const struct nft_expr *expr, - struct nft_data data[NFT_REG_MAX + 1], - const struct nft_pktinfo *pkt) +void nft_meta_get_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt) { const struct nft_meta *priv = nft_expr_priv(expr); const struct sk_buff *skb = pkt->skb; @@ -140,10 +133,11 @@ static void nft_meta_get_eval(const struct nft_expr *expr, err: data[NFT_REG_VERDICT].verdict = NFT_BREAK; } +EXPORT_SYMBOL_GPL(nft_meta_get_eval); -static void nft_meta_set_eval(const struct nft_expr *expr, - struct nft_data data[NFT_REG_MAX + 1], - const struct nft_pktinfo *pkt) +void nft_meta_set_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt) { const struct nft_meta *meta = nft_expr_priv(expr); struct sk_buff *skb = pkt->skb; @@ -163,28 +157,24 @@ static void nft_meta_set_eval(const struct nft_expr *expr, WARN_ON(1); } } +EXPORT_SYMBOL_GPL(nft_meta_set_eval); -static const struct nla_policy nft_meta_policy[NFTA_META_MAX + 1] = { +const struct nla_policy nft_meta_policy[NFTA_META_MAX + 1] = { [NFTA_META_DREG] = { .type = NLA_U32 }, [NFTA_META_KEY] = { .type = NLA_U32 }, [NFTA_META_SREG] = { .type = NLA_U32 }, }; +EXPORT_SYMBOL_GPL(nft_meta_policy); -static int nft_meta_init_validate_set(uint32_t key) +int nft_meta_get_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) { - switch (key) { - case NFT_META_MARK: - case NFT_META_PRIORITY: - case NFT_META_NFTRACE: - return 0; - default: - return -EOPNOTSUPP; - } -} + struct nft_meta *priv = nft_expr_priv(expr); + int err; -static int nft_meta_init_validate_get(uint32_t key) -{ - switch (key) { + priv->key = ntohl(nla_get_be32(tb[NFTA_META_KEY])); + switch (priv->key) { case NFT_META_LEN: case NFT_META_PROTOCOL: case NFT_META_NFPROTO: @@ -205,39 +195,41 @@ static int nft_meta_init_validate_get(uint32_t key) #ifdef CONFIG_NETWORK_SECMARK case NFT_META_SECMARK: #endif - return 0; + break; default: return -EOPNOTSUPP; } + priv->dreg = ntohl(nla_get_be32(tb[NFTA_META_DREG])); + err = nft_validate_output_register(priv->dreg); + if (err < 0) + return err; + + err = nft_validate_data_load(ctx, priv->dreg, NULL, NFT_DATA_VALUE); + if (err < 0) + return err; + + return 0; } +EXPORT_SYMBOL_GPL(nft_meta_get_init); -static int nft_meta_init(const struct nft_ctx *ctx, const struct nft_expr *expr, - const struct nlattr * const tb[]) +int nft_meta_set_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) { struct nft_meta *priv = nft_expr_priv(expr); int err; priv->key = ntohl(nla_get_be32(tb[NFTA_META_KEY])); - - if (tb[NFTA_META_DREG]) { - err = nft_meta_init_validate_get(priv->key); - if (err < 0) - return err; - - priv->dreg = ntohl(nla_get_be32(tb[NFTA_META_DREG])); - err = nft_validate_output_register(priv->dreg); - if (err < 0) - return err; - - return nft_validate_data_load(ctx, priv->dreg, NULL, - NFT_DATA_VALUE); + switch (priv->key) { + case NFT_META_MARK: + case NFT_META_PRIORITY: + case NFT_META_NFTRACE: + break; + default: + return -EOPNOTSUPP; } - err = nft_meta_init_validate_set(priv->key); - if (err < 0) - return err; - priv->sreg = ntohl(nla_get_be32(tb[NFTA_META_SREG])); err = nft_validate_input_register(priv->sreg); if (err < 0) @@ -245,9 +237,10 @@ static int nft_meta_init(const struct nft_ctx *ctx, const struct nft_expr *expr, return 0; } +EXPORT_SYMBOL_GPL(nft_meta_set_init); -static int nft_meta_get_dump(struct sk_buff *skb, - const struct nft_expr *expr) +int nft_meta_get_dump(struct sk_buff *skb, + const struct nft_expr *expr) { const struct nft_meta *priv = nft_expr_priv(expr); @@ -260,9 +253,10 @@ static int nft_meta_get_dump(struct sk_buff *skb, nla_put_failure: return -1; } +EXPORT_SYMBOL_GPL(nft_meta_get_dump); -static int nft_meta_set_dump(struct sk_buff *skb, - const struct nft_expr *expr) +int nft_meta_set_dump(struct sk_buff *skb, + const struct nft_expr *expr) { const struct nft_meta *priv = nft_expr_priv(expr); @@ -276,13 +270,14 @@ static int nft_meta_set_dump(struct sk_buff *skb, nla_put_failure: return -1; } +EXPORT_SYMBOL_GPL(nft_meta_set_dump); static struct nft_expr_type nft_meta_type; static const struct nft_expr_ops nft_meta_get_ops = { .type = &nft_meta_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_meta)), .eval = nft_meta_get_eval, - .init = nft_meta_init, + .init = nft_meta_get_init, .dump = nft_meta_get_dump, }; @@ -290,7 +285,7 @@ static const struct nft_expr_ops nft_meta_set_ops = { .type = &nft_meta_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_meta)), .eval = nft_meta_set_eval, - .init = nft_meta_init, + .init = nft_meta_set_init, .dump = nft_meta_set_dump, }; diff --git a/net/netfilter/nft_rbtree.c b/net/netfilter/nft_rbtree.c index e21d69d1350..e1836ff8819 100644 --- a/net/netfilter/nft_rbtree.c +++ b/net/netfilter/nft_rbtree.c @@ -18,6 +18,8 @@ #include <linux/netfilter/nf_tables.h> #include <net/netfilter/nf_tables.h> +static DEFINE_SPINLOCK(nft_rbtree_lock); + struct nft_rbtree { struct rb_root root; }; @@ -38,6 +40,7 @@ static bool nft_rbtree_lookup(const struct nft_set *set, const struct rb_node *parent = priv->root.rb_node; int d; + spin_lock_bh(&nft_rbtree_lock); while (parent != NULL) { rbe = rb_entry(parent, struct nft_rbtree_elem, node); @@ -53,6 +56,8 @@ found: goto out; if (set->flags & NFT_SET_MAP) nft_data_copy(data, rbe->data); + + spin_unlock_bh(&nft_rbtree_lock); return true; } } @@ -62,6 +67,7 @@ found: goto found; } out: + spin_unlock_bh(&nft_rbtree_lock); return false; } @@ -124,9 +130,12 @@ static int nft_rbtree_insert(const struct nft_set *set, !(rbe->flags & NFT_SET_ELEM_INTERVAL_END)) nft_data_copy(rbe->data, &elem->data); + spin_lock_bh(&nft_rbtree_lock); err = __nft_rbtree_insert(set, rbe); if (err < 0) kfree(rbe); + + spin_unlock_bh(&nft_rbtree_lock); return err; } @@ -136,7 +145,9 @@ static void nft_rbtree_remove(const struct nft_set *set, struct nft_rbtree *priv = nft_set_priv(set); struct nft_rbtree_elem *rbe = elem->cookie; + spin_lock_bh(&nft_rbtree_lock); rb_erase(&rbe->node, &priv->root); + spin_unlock_bh(&nft_rbtree_lock); kfree(rbe); } @@ -147,6 +158,7 @@ static int nft_rbtree_get(const struct nft_set *set, struct nft_set_elem *elem) struct nft_rbtree_elem *rbe; int d; + spin_lock_bh(&nft_rbtree_lock); while (parent != NULL) { rbe = rb_entry(parent, struct nft_rbtree_elem, node); @@ -161,9 +173,11 @@ static int nft_rbtree_get(const struct nft_set *set, struct nft_set_elem *elem) !(rbe->flags & NFT_SET_ELEM_INTERVAL_END)) nft_data_copy(&elem->data, rbe->data); elem->flags = rbe->flags; + spin_unlock_bh(&nft_rbtree_lock); return 0; } } + spin_unlock_bh(&nft_rbtree_lock); return -ENOENT; } @@ -176,6 +190,7 @@ static void nft_rbtree_walk(const struct nft_ctx *ctx, struct nft_set_elem elem; struct rb_node *node; + spin_lock_bh(&nft_rbtree_lock); for (node = rb_first(&priv->root); node != NULL; node = rb_next(node)) { if (iter->count < iter->skip) goto cont; @@ -188,11 +203,14 @@ static void nft_rbtree_walk(const struct nft_ctx *ctx, elem.flags = rbe->flags; iter->err = iter->fn(ctx, set, iter, &elem); - if (iter->err < 0) + if (iter->err < 0) { + spin_unlock_bh(&nft_rbtree_lock); return; + } cont: iter->count++; } + spin_unlock_bh(&nft_rbtree_lock); } static unsigned int nft_rbtree_privsize(const struct nlattr * const nla[]) @@ -201,6 +219,7 @@ static unsigned int nft_rbtree_privsize(const struct nlattr * const nla[]) } static int nft_rbtree_init(const struct nft_set *set, + const struct nft_set_desc *desc, const struct nlattr * const nla[]) { struct nft_rbtree *priv = nft_set_priv(set); @@ -215,15 +234,37 @@ static void nft_rbtree_destroy(const struct nft_set *set) struct nft_rbtree_elem *rbe; struct rb_node *node; + spin_lock_bh(&nft_rbtree_lock); while ((node = priv->root.rb_node) != NULL) { rb_erase(node, &priv->root); rbe = rb_entry(node, struct nft_rbtree_elem, node); nft_rbtree_elem_destroy(set, rbe); } + spin_unlock_bh(&nft_rbtree_lock); +} + +static bool nft_rbtree_estimate(const struct nft_set_desc *desc, u32 features, + struct nft_set_estimate *est) +{ + unsigned int nsize; + + nsize = sizeof(struct nft_rbtree_elem); + if (features & NFT_SET_MAP) + nsize += FIELD_SIZEOF(struct nft_rbtree_elem, data[0]); + + if (desc->size) + est->size = sizeof(struct nft_rbtree) + desc->size * nsize; + else + est->size = nsize; + + est->class = NFT_SET_CLASS_O_LOG_N; + + return true; } static struct nft_set_ops nft_rbtree_ops __read_mostly = { .privsize = nft_rbtree_privsize, + .estimate = nft_rbtree_estimate, .init = nft_rbtree_init, .destroy = nft_rbtree_destroy, .insert = nft_rbtree_insert, diff --git a/net/netfilter/xt_bpf.c b/net/netfilter/xt_bpf.c index 12d4da8e6c7..bbffdbdaf60 100644 --- a/net/netfilter/xt_bpf.c +++ b/net/netfilter/xt_bpf.c @@ -23,10 +23,11 @@ MODULE_ALIAS("ip6t_bpf"); static int bpf_mt_check(const struct xt_mtchk_param *par) { struct xt_bpf_info *info = par->matchinfo; - struct sock_fprog program; + struct sock_fprog_kern program; program.len = info->bpf_program_num_elem; - program.filter = (struct sock_filter __user *) info->bpf_program; + program.filter = info->bpf_program; + if (sk_unattached_filter_create(&info->filter, &program)) { pr_info("bpf: check failed: parse error\n"); return -EINVAL; diff --git a/net/netfilter/xt_nfacct.c b/net/netfilter/xt_nfacct.c index b3be0ef21f1..8c646ed9c92 100644 --- a/net/netfilter/xt_nfacct.c +++ b/net/netfilter/xt_nfacct.c @@ -21,11 +21,14 @@ MODULE_ALIAS("ip6t_nfacct"); static bool nfacct_mt(const struct sk_buff *skb, struct xt_action_param *par) { + int overquota; const struct xt_nfacct_match_info *info = par->targinfo; nfnl_acct_update(skb, info->nfacct); - return true; + overquota = nfnl_acct_overquota(skb, info->nfacct); + + return overquota == NFACCT_UNDERQUOTA ? false : true; } static int diff --git a/net/netfilter/xt_recent.c b/net/netfilter/xt_recent.c index 1e657cf715c..a9faae89f95 100644 --- a/net/netfilter/xt_recent.c +++ b/net/netfilter/xt_recent.c @@ -313,10 +313,7 @@ out: static void recent_table_free(void *addr) { - if (is_vmalloc_addr(addr)) - vfree(addr); - else - kfree(addr); + kvfree(addr); } static int recent_mt_check(const struct xt_mtchk_param *par, diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 894cda0206b..15c731f03fa 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1206,7 +1206,8 @@ static int netlink_create(struct net *net, struct socket *sock, int protocol, struct module *module = NULL; struct mutex *cb_mutex; struct netlink_sock *nlk; - void (*bind)(int group); + int (*bind)(int group); + void (*unbind)(int group); int err = 0; sock->state = SS_UNCONNECTED; @@ -1232,6 +1233,7 @@ static int netlink_create(struct net *net, struct socket *sock, int protocol, err = -EPROTONOSUPPORT; cb_mutex = nl_table[protocol].cb_mutex; bind = nl_table[protocol].bind; + unbind = nl_table[protocol].unbind; netlink_unlock_table(); if (err < 0) @@ -1248,6 +1250,7 @@ static int netlink_create(struct net *net, struct socket *sock, int protocol, nlk = nlk_sk(sock->sk); nlk->module = module; nlk->netlink_bind = bind; + nlk->netlink_unbind = unbind; out: return err; @@ -1301,6 +1304,7 @@ static int netlink_release(struct socket *sock) kfree_rcu(old, rcu); nl_table[sk->sk_protocol].module = NULL; nl_table[sk->sk_protocol].bind = NULL; + nl_table[sk->sk_protocol].unbind = NULL; nl_table[sk->sk_protocol].flags = 0; nl_table[sk->sk_protocol].registered = 0; } @@ -1360,7 +1364,74 @@ retry: return err; } -static inline int netlink_capable(const struct socket *sock, unsigned int flag) +/** + * __netlink_ns_capable - General netlink message capability test + * @nsp: NETLINK_CB of the socket buffer holding a netlink command from userspace. + * @user_ns: The user namespace of the capability to use + * @cap: The capability to use + * + * Test to see if the opener of the socket we received the message + * from had when the netlink socket was created and the sender of the + * message has has the capability @cap in the user namespace @user_ns. + */ +bool __netlink_ns_capable(const struct netlink_skb_parms *nsp, + struct user_namespace *user_ns, int cap) +{ + return ((nsp->flags & NETLINK_SKB_DST) || + file_ns_capable(nsp->sk->sk_socket->file, user_ns, cap)) && + ns_capable(user_ns, cap); +} +EXPORT_SYMBOL(__netlink_ns_capable); + +/** + * netlink_ns_capable - General netlink message capability test + * @skb: socket buffer holding a netlink command from userspace + * @user_ns: The user namespace of the capability to use + * @cap: The capability to use + * + * Test to see if the opener of the socket we received the message + * from had when the netlink socket was created and the sender of the + * message has has the capability @cap in the user namespace @user_ns. + */ +bool netlink_ns_capable(const struct sk_buff *skb, + struct user_namespace *user_ns, int cap) +{ + return __netlink_ns_capable(&NETLINK_CB(skb), user_ns, cap); +} +EXPORT_SYMBOL(netlink_ns_capable); + +/** + * netlink_capable - Netlink global message capability test + * @skb: socket buffer holding a netlink command from userspace + * @cap: The capability to use + * + * Test to see if the opener of the socket we received the message + * from had when the netlink socket was created and the sender of the + * message has has the capability @cap in all user namespaces. + */ +bool netlink_capable(const struct sk_buff *skb, int cap) +{ + return netlink_ns_capable(skb, &init_user_ns, cap); +} +EXPORT_SYMBOL(netlink_capable); + +/** + * netlink_net_capable - Netlink network namespace message capability test + * @skb: socket buffer holding a netlink command from userspace + * @cap: The capability to use + * + * Test to see if the opener of the socket we received the message + * from had when the netlink socket was created and the sender of the + * message has has the capability @cap over the network namespace of + * the socket we received the message from. + */ +bool netlink_net_capable(const struct sk_buff *skb, int cap) +{ + return netlink_ns_capable(skb, sock_net(skb->sk)->user_ns, cap); +} +EXPORT_SYMBOL(netlink_net_capable); + +static inline int netlink_allowed(const struct socket *sock, unsigned int flag) { return (nl_table[sock->sk->sk_protocol].flags & flag) || ns_capable(sock_net(sock->sk)->user_ns, CAP_NET_ADMIN); @@ -1411,6 +1482,19 @@ static int netlink_realloc_groups(struct sock *sk) return err; } +static void netlink_unbind(int group, long unsigned int groups, + struct netlink_sock *nlk) +{ + int undo; + + if (!nlk->netlink_unbind) + return; + + for (undo = 0; undo < group; undo++) + if (test_bit(group, &groups)) + nlk->netlink_unbind(undo); +} + static int netlink_bind(struct socket *sock, struct sockaddr *addr, int addr_len) { @@ -1419,6 +1503,7 @@ static int netlink_bind(struct socket *sock, struct sockaddr *addr, struct netlink_sock *nlk = nlk_sk(sk); struct sockaddr_nl *nladdr = (struct sockaddr_nl *)addr; int err; + long unsigned int groups = nladdr->nl_groups; if (addr_len < sizeof(struct sockaddr_nl)) return -EINVAL; @@ -1427,45 +1512,53 @@ static int netlink_bind(struct socket *sock, struct sockaddr *addr, return -EINVAL; /* Only superuser is allowed to listen multicasts */ - if (nladdr->nl_groups) { - if (!netlink_capable(sock, NL_CFG_F_NONROOT_RECV)) + if (groups) { + if (!netlink_allowed(sock, NL_CFG_F_NONROOT_RECV)) return -EPERM; err = netlink_realloc_groups(sk); if (err) return err; } - if (nlk->portid) { + if (nlk->portid) if (nladdr->nl_pid != nlk->portid) return -EINVAL; - } else { + + if (nlk->netlink_bind && groups) { + int group; + + for (group = 0; group < nlk->ngroups; group++) { + if (!test_bit(group, &groups)) + continue; + err = nlk->netlink_bind(group); + if (!err) + continue; + netlink_unbind(group, groups, nlk); + return err; + } + } + + if (!nlk->portid) { err = nladdr->nl_pid ? netlink_insert(sk, net, nladdr->nl_pid) : netlink_autobind(sock); - if (err) + if (err) { + netlink_unbind(nlk->ngroups - 1, groups, nlk); return err; + } } - if (!nladdr->nl_groups && (nlk->groups == NULL || !(u32)nlk->groups[0])) + if (!groups && (nlk->groups == NULL || !(u32)nlk->groups[0])) return 0; netlink_table_grab(); netlink_update_subscriptions(sk, nlk->subscriptions + - hweight32(nladdr->nl_groups) - + hweight32(groups) - hweight32(nlk->groups[0])); - nlk->groups[0] = (nlk->groups[0] & ~0xffffffffUL) | nladdr->nl_groups; + nlk->groups[0] = (nlk->groups[0] & ~0xffffffffUL) | groups; netlink_update_listeners(sk); netlink_table_ungrab(); - if (nlk->netlink_bind && nlk->groups[0]) { - int i; - - for (i = 0; i < nlk->ngroups; i++) { - if (test_bit(i, nlk->groups)) - nlk->netlink_bind(i); - } - } - return 0; } @@ -1490,7 +1583,7 @@ static int netlink_connect(struct socket *sock, struct sockaddr *addr, return -EINVAL; if ((nladdr->nl_groups || nladdr->nl_pid) && - !netlink_capable(sock, NL_CFG_F_NONROOT_SEND)) + !netlink_allowed(sock, NL_CFG_F_NONROOT_SEND)) return -EPERM; if (!nlk->portid) @@ -2096,20 +2189,24 @@ static int netlink_setsockopt(struct socket *sock, int level, int optname, break; case NETLINK_ADD_MEMBERSHIP: case NETLINK_DROP_MEMBERSHIP: { - if (!netlink_capable(sock, NL_CFG_F_NONROOT_RECV)) + if (!netlink_allowed(sock, NL_CFG_F_NONROOT_RECV)) return -EPERM; err = netlink_realloc_groups(sk); if (err) return err; if (!val || val - 1 >= nlk->ngroups) return -EINVAL; + if (optname == NETLINK_ADD_MEMBERSHIP && nlk->netlink_bind) { + err = nlk->netlink_bind(val); + if (err) + return err; + } netlink_table_grab(); netlink_update_socket_mc(nlk, val, optname == NETLINK_ADD_MEMBERSHIP); netlink_table_ungrab(); - - if (nlk->netlink_bind) - nlk->netlink_bind(val); + if (optname == NETLINK_DROP_MEMBERSHIP && nlk->netlink_unbind) + nlk->netlink_unbind(val); err = 0; break; @@ -2228,6 +2325,7 @@ static int netlink_sendmsg(struct kiocb *kiocb, struct socket *sock, struct sk_buff *skb; int err; struct scm_cookie scm; + u32 netlink_skb_flags = 0; if (msg->msg_flags&MSG_OOB) return -EOPNOTSUPP; @@ -2247,8 +2345,9 @@ static int netlink_sendmsg(struct kiocb *kiocb, struct socket *sock, dst_group = ffs(addr->nl_groups); err = -EPERM; if ((dst_group || dst_portid) && - !netlink_capable(sock, NL_CFG_F_NONROOT_SEND)) + !netlink_allowed(sock, NL_CFG_F_NONROOT_SEND)) goto out; + netlink_skb_flags |= NETLINK_SKB_DST; } else { dst_portid = nlk->dst_portid; dst_group = nlk->dst_group; @@ -2278,6 +2377,7 @@ static int netlink_sendmsg(struct kiocb *kiocb, struct socket *sock, NETLINK_CB(skb).portid = nlk->portid; NETLINK_CB(skb).dst_group = dst_group; NETLINK_CB(skb).creds = siocb->scm->creds; + NETLINK_CB(skb).flags = netlink_skb_flags; err = -EFAULT; if (memcpy_fromiovec(skb_put(skb, len), msg->msg_iov, len)) { diff --git a/net/netlink/af_netlink.h b/net/netlink/af_netlink.h index ed13a790b00..0b59d441f5b 100644 --- a/net/netlink/af_netlink.h +++ b/net/netlink/af_netlink.h @@ -38,7 +38,8 @@ struct netlink_sock { struct mutex *cb_mutex; struct mutex cb_def_mutex; void (*netlink_rcv)(struct sk_buff *skb); - void (*netlink_bind)(int group); + int (*netlink_bind)(int group); + void (*netlink_unbind)(int group); struct module *module; #ifdef CONFIG_NETLINK_MMAP struct mutex pg_vec_lock; @@ -74,7 +75,8 @@ struct netlink_table { unsigned int groups; struct mutex *cb_mutex; struct module *module; - void (*bind)(int group); + int (*bind)(int group); + void (*unbind)(int group); bool (*compare)(struct net *net, struct sock *sock); int registered; }; diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index b1dcdb932a8..76393f2f4b2 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -317,7 +317,7 @@ static void genl_unregister_mc_groups(struct genl_family *family) } } -static int genl_validate_ops(struct genl_family *family) +static int genl_validate_ops(const struct genl_family *family) { const struct genl_ops *ops = family->ops; unsigned int n_ops = family->n_ops; @@ -337,10 +337,6 @@ static int genl_validate_ops(struct genl_family *family) return -EINVAL; } - /* family is not registered yet, so no locking needed */ - family->ops = ops; - family->n_ops = n_ops; - return 0; } @@ -561,7 +557,7 @@ static int genl_family_rcv_msg(struct genl_family *family, return -EOPNOTSUPP; if ((ops->flags & GENL_ADMIN_PERM) && - !capable(CAP_NET_ADMIN)) + !netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; if ((nlh->nlmsg_flags & NLM_F_DUMP) == NLM_F_DUMP) { diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index 2c77e7b1a91..c36856a457c 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -134,8 +134,8 @@ static int set_eth_addr(struct sk_buff *skb, skb_postpull_rcsum(skb, eth_hdr(skb), ETH_ALEN * 2); - memcpy(eth_hdr(skb)->h_source, eth_key->eth_src, ETH_ALEN); - memcpy(eth_hdr(skb)->h_dest, eth_key->eth_dst, ETH_ALEN); + ether_addr_copy(eth_hdr(skb)->h_source, eth_key->eth_src); + ether_addr_copy(eth_hdr(skb)->h_dest, eth_key->eth_dst); ovs_skb_postpush_rcsum(skb, eth_hdr(skb), ETH_ALEN * 2); diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index a3276e3c4fe..0d407bca81e 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -44,11 +44,11 @@ #include <linux/netfilter_ipv4.h> #include <linux/inetdevice.h> #include <linux/list.h> -#include <linux/lockdep.h> #include <linux/openvswitch.h> #include <linux/rculist.h> #include <linux/dmi.h> -#include <linux/workqueue.h> +#include <linux/genetlink.h> +#include <net/genetlink.h> #include <net/genetlink.h> #include <net/net_namespace.h> #include <net/netns/generic.h> @@ -62,6 +62,31 @@ int ovs_net_id __read_mostly; +static struct genl_family dp_packet_genl_family; +static struct genl_family dp_flow_genl_family; +static struct genl_family dp_datapath_genl_family; + +static struct genl_multicast_group ovs_dp_flow_multicast_group = { + .name = OVS_FLOW_MCGROUP +}; + +static struct genl_multicast_group ovs_dp_datapath_multicast_group = { + .name = OVS_DATAPATH_MCGROUP +}; + +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_info *info, + const struct genl_multicast_group *grp) +{ + return info->nlhdr->nlmsg_flags & NLM_F_ECHO || + netlink_has_listeners(genl_info_net(info)->genl_sock, 0); +} + static void ovs_notify(struct genl_family *family, struct sk_buff *skb, struct genl_info *info) { @@ -173,6 +198,7 @@ static struct hlist_head *vport_hash_bucket(const struct datapath *dp, 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; @@ -262,16 +288,6 @@ out: 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 dp_upcall_info *upcall_info) { @@ -524,7 +540,7 @@ 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 = ovs_flow_alloc(false); + flow = ovs_flow_alloc(); err = PTR_ERR(flow); if (IS_ERR(flow)) goto err_kfree_skb; @@ -590,6 +606,18 @@ static const struct genl_ops dp_packet_genl_ops[] = { } }; +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(struct datapath *dp, struct ovs_dp_stats *stats, struct ovs_dp_megaflow_stats *mega_stats) { @@ -621,26 +649,6 @@ static void get_dp_stats(struct datapath *dp, struct ovs_dp_stats *stats, } } -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 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 struct genl_multicast_group ovs_dp_flow_multicast_group = { - .name = OVS_FLOW_MCGROUP -}; - static size_t ovs_flow_cmd_msg_size(const struct sw_flow_actions *acts) { return NLMSG_ALIGN(sizeof(struct ovs_header)) @@ -652,8 +660,8 @@ static size_t ovs_flow_cmd_msg_size(const struct sw_flow_actions *acts) + nla_total_size(acts->actions_len); /* OVS_FLOW_ATTR_ACTIONS */ } -/* Called with ovs_mutex. */ -static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp, +/* 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) { @@ -670,7 +678,7 @@ static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp, if (!ovs_header) return -EMSGSIZE; - ovs_header->dp_ifindex = get_dpifindex(dp); + ovs_header->dp_ifindex = dp_ifindex; /* Fill flow key. */ nla = nla_nest_start(skb, OVS_FLOW_ATTR_KEY); @@ -693,6 +701,7 @@ static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp, nla_nest_end(skb, nla); 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; @@ -720,9 +729,9 @@ static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp, 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 { @@ -743,113 +752,128 @@ error: return err; } -static struct sk_buff *ovs_flow_cmd_alloc_info(struct sw_flow *flow, - struct genl_info *info) +/* May not be called with RCU read lock. */ +static struct sk_buff *ovs_flow_cmd_alloc_info(const struct sw_flow_actions *acts, + struct genl_info *info, + bool always) { - size_t len; + struct sk_buff *skb; + + if (!always && !ovs_must_notify(info, &ovs_dp_flow_multicast_group)) + return NULL; - len = ovs_flow_cmd_msg_size(ovsl_dereference(flow->sf_acts)); + skb = genlmsg_new_unicast(ovs_flow_cmd_msg_size(acts), info, GFP_KERNEL); + if (!skb) + return ERR_PTR(-ENOMEM); - return genlmsg_new_unicast(len, info, GFP_KERNEL); + return skb; } -static struct sk_buff *ovs_flow_cmd_build_info(struct sw_flow *flow, - struct datapath *dp, - struct genl_info *info, - 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) { struct sk_buff *skb; int retval; - skb = ovs_flow_cmd_alloc_info(flow, info); - if (!skb) - return ERR_PTR(-ENOMEM); + skb = ovs_flow_cmd_alloc_info(ovsl_dereference(flow->sf_acts), info, + always); + if (!skb || IS_ERR(skb)) + return skb; - retval = ovs_flow_cmd_fill_info(flow, dp, skb, info->snd_portid, - info->snd_seq, 0, cmd); + retval = ovs_flow_cmd_fill_info(flow, dp_ifindex, skb, + info->snd_portid, info->snd_seq, 0, + cmd); 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 nlattr **a = info->attrs; struct ovs_header *ovs_header = info->userhdr; - struct sw_flow_key key, masked_key; - struct sw_flow *flow = NULL; + struct sw_flow *flow, *new_flow; struct sw_flow_mask mask; struct sk_buff *reply; struct datapath *dp; - struct sw_flow_actions *acts = NULL; + struct sw_flow_actions *acts; struct sw_flow_match match; - bool exact_5tuple; int error; - /* Extract key. */ + /* Must have key and actions. */ error = -EINVAL; if (!a[OVS_FLOW_ATTR_KEY]) goto error; + if (!a[OVS_FLOW_ATTR_ACTIONS]) + goto error; - ovs_match_init(&match, &key, &mask); - error = ovs_nla_get_match(&match, &exact_5tuple, + /* 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; + } + + /* Extract key. */ + ovs_match_init(&match, &new_flow->unmasked_key, &mask); + error = ovs_nla_get_match(&match, a[OVS_FLOW_ATTR_KEY], a[OVS_FLOW_ATTR_MASK]); if (error) - goto error; + goto err_kfree_flow; + + ovs_flow_mask_key(&new_flow->key, &new_flow->unmasked_key, &mask); /* Validate actions. */ - if (a[OVS_FLOW_ATTR_ACTIONS]) { - acts = ovs_nla_alloc_flow_actions(nla_len(a[OVS_FLOW_ATTR_ACTIONS])); - error = PTR_ERR(acts); - if (IS_ERR(acts)) - goto error; + acts = ovs_nla_alloc_flow_actions(nla_len(a[OVS_FLOW_ATTR_ACTIONS])); + error = PTR_ERR(acts); + if (IS_ERR(acts)) + goto err_kfree_flow; - ovs_flow_mask_key(&masked_key, &key, &mask); - error = ovs_nla_copy_actions(a[OVS_FLOW_ATTR_ACTIONS], - &masked_key, 0, &acts); - if (error) { - OVS_NLERR("Flow actions may not be safe on all matching packets.\n"); - goto err_kfree; - } - } else if (info->genlhdr->cmd == OVS_FLOW_CMD_NEW) { - error = -EINVAL; - goto error; + error = ovs_nla_copy_actions(a[OVS_FLOW_ATTR_ACTIONS], &new_flow->key, + 0, &acts); + if (error) { + OVS_NLERR("Flow actions may not be safe on all matching packets.\n"); + goto err_kfree_acts; + } + + reply = ovs_flow_cmd_alloc_info(acts, info, false); + if (IS_ERR(reply)) { + error = PTR_ERR(reply); + goto err_kfree_acts; } ovs_lock(); dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); - error = -ENODEV; - if (!dp) + if (unlikely(!dp)) { + error = -ENODEV; goto err_unlock_ovs; - + } /* Check if this is a duplicate flow */ - flow = ovs_flow_tbl_lookup(&dp->table, &key); - if (!flow) { - /* 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; - - /* Allocate flow. */ - flow = ovs_flow_alloc(!exact_5tuple); - if (IS_ERR(flow)) { - error = PTR_ERR(flow); - goto err_unlock_ovs; - } - - flow->key = masked_key; - flow->unmasked_key = key; - rcu_assign_pointer(flow->sf_acts, acts); + flow = ovs_flow_tbl_lookup(&dp->table, &new_flow->unmasked_key); + if (likely(!flow)) { + rcu_assign_pointer(new_flow->sf_acts, acts); /* Put flow in bucket. */ - error = ovs_flow_tbl_insert(&dp->table, flow, &mask); - if (error) { + error = ovs_flow_tbl_insert(&dp->table, new_flow, &mask); + if (unlikely(error)) { acts = NULL; - goto err_flow_free; + goto err_unlock_ovs; } - reply = ovs_flow_cmd_build_info(flow, dp, info, 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); + BUG_ON(error < 0); + } + ovs_unlock(); } else { - /* We found a matching flow. */ struct sw_flow_actions *old_acts; /* Bail out if we're not allowed to modify an existing flow. @@ -858,40 +882,154 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) * 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 unmasked key has to be the same for flow updates. */ - if (!ovs_flow_cmp_unmasked_key(flow, &match)) + if (unlikely(!ovs_flow_cmp_unmasked_key(flow, &match))) { + error = -EEXIST; goto err_unlock_ovs; - + } /* Update actions. */ 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); + BUG_ON(error < 0); + } + ovs_unlock(); + ovs_nla_free_flow_actions(old_acts); + ovs_flow_free(new_flow, false); + } + + if (reply) + ovs_notify(&dp_flow_genl_family, reply, info); + return 0; + +err_unlock_ovs: + ovs_unlock(); + kfree_skb(reply); +err_kfree_acts: + kfree(acts); +err_kfree_flow: + ovs_flow_free(new_flow, false); +error: + return error; +} + +static int ovs_flow_cmd_set(struct sk_buff *skb, struct genl_info *info) +{ + struct nlattr **a = info->attrs; + struct ovs_header *ovs_header = info->userhdr; + struct sw_flow_key key, masked_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; + int error; + + /* Extract key. */ + error = -EINVAL; + if (!a[OVS_FLOW_ATTR_KEY]) + goto error; + + ovs_match_init(&match, &key, &mask); + error = ovs_nla_get_match(&match, + a[OVS_FLOW_ATTR_KEY], a[OVS_FLOW_ATTR_MASK]); + if (error) + goto error; + + /* Validate actions. */ + if (a[OVS_FLOW_ATTR_ACTIONS]) { + acts = ovs_nla_alloc_flow_actions(nla_len(a[OVS_FLOW_ATTR_ACTIONS])); + error = PTR_ERR(acts); + if (IS_ERR(acts)) + goto error; + + ovs_flow_mask_key(&masked_key, &key, &mask); + error = ovs_nla_copy_actions(a[OVS_FLOW_ATTR_ACTIONS], + &masked_key, 0, &acts); + if (error) { + OVS_NLERR("Flow actions may not be safe on all matching packets.\n"); + goto err_kfree_acts; + } + } + + /* Can allocate before locking if have acts. */ + if (acts) { + reply = ovs_flow_cmd_alloc_info(acts, info, false); + if (IS_ERR(reply)) { + error = PTR_ERR(reply); + goto err_kfree_acts; + } + } - reply = ovs_flow_cmd_build_info(flow, dp, info, OVS_FLOW_CMD_NEW); + ovs_lock(); + dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex); + if (unlikely(!dp)) { + error = -ENODEV; + goto err_unlock_ovs; + } + /* Check that the flow exists. */ + flow = ovs_flow_tbl_lookup(&dp->table, &key); + if (unlikely(!flow)) { + error = -ENOENT; + goto err_unlock_ovs; + } + /* The unmasked key has to be the same for flow updates. */ + if (unlikely(!ovs_flow_cmp_unmasked_key(flow, &match))) { + error = -EEXIST; + 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); - /* Clear stats. */ - if (a[OVS_FLOW_ATTR_CLEAR]) - ovs_flow_stats_clear(flow); + 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); + 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); + if (unlikely(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 (!IS_ERR(reply)) + if (reply) ovs_notify(&dp_flow_genl_family, reply, info); - else - genl_set_err(&dp_flow_genl_family, sock_net(skb->sk), 0, - 0, PTR_ERR(reply)); + if (old_acts) + ovs_nla_free_flow_actions(old_acts); + return 0; -err_flow_free: - ovs_flow_free(flow, false); err_unlock_ovs: ovs_unlock(); -err_kfree: + kfree_skb(reply); +err_kfree_acts: kfree(acts); error: return error; @@ -914,7 +1052,7 @@ static int ovs_flow_cmd_get(struct sk_buff *skb, struct genl_info *info) } ovs_match_init(&match, &key, NULL); - err = ovs_nla_get_match(&match, NULL, a[OVS_FLOW_ATTR_KEY], NULL); + err = ovs_nla_get_match(&match, a[OVS_FLOW_ATTR_KEY], NULL); if (err) return err; @@ -931,7 +1069,8 @@ static int ovs_flow_cmd_get(struct sk_buff *skb, struct genl_info *info) goto unlock; } - reply = ovs_flow_cmd_build_info(flow, dp, info, OVS_FLOW_CMD_NEW); + reply = ovs_flow_cmd_build_info(flow, ovs_header->dp_ifindex, info, + OVS_FLOW_CMD_NEW, true); if (IS_ERR(reply)) { err = PTR_ERR(reply); goto unlock; @@ -955,45 +1094,53 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info) struct sw_flow_match match; int err; + if (likely(a[OVS_FLOW_ATTR_KEY])) { + ovs_match_init(&match, &key, NULL); + err = ovs_nla_get_match(&match, a[OVS_FLOW_ATTR_KEY], NULL); + 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]) { + if (unlikely(!a[OVS_FLOW_ATTR_KEY])) { err = ovs_flow_tbl_flush(&dp->table); goto unlock; } - ovs_match_init(&match, &key, NULL); - err = ovs_nla_get_match(&match, NULL, a[OVS_FLOW_ATTR_KEY], NULL); - if (err) - goto unlock; - flow = ovs_flow_tbl_lookup(&dp->table, &key); - if (!flow || !ovs_flow_cmp_unmasked_key(flow, &match)) { + if (unlikely(!flow || !ovs_flow_cmp_unmasked_key(flow, &match))) { err = -ENOENT; goto unlock; } - reply = ovs_flow_cmd_alloc_info(flow, info); - if (!reply) { - err = -ENOMEM; - goto unlock; - } - ovs_flow_tbl_remove(&dp->table, flow); + ovs_unlock(); - err = ovs_flow_cmd_fill_info(flow, dp, reply, info->snd_portid, - info->snd_seq, 0, OVS_FLOW_CMD_DEL); - BUG_ON(err < 0); + reply = ovs_flow_cmd_alloc_info((const struct sw_flow_actions __force *) flow->sf_acts, + info, false); + 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); + 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); - ovs_unlock(); - - ovs_notify(&dp_flow_genl_family, reply, info); return 0; unlock: ovs_unlock(); @@ -1024,7 +1171,7 @@ static int ovs_flow_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb) 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) @@ -1037,11 +1184,17 @@ static int ovs_flow_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb) return skb->len; } -static const 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_ACTIONS] = { .type = NLA_NESTED }, + [OVS_FLOW_ATTR_CLEAR] = { .type = NLA_FLAG }, +}; + +static 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. */ @@ -1057,28 +1210,22 @@ static const struct genl_ops dp_flow_genl_ops[] = { { .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 }, - [OVS_DP_ATTR_USER_FEATURES] = { .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) @@ -1093,6 +1240,7 @@ static size_t ovs_dp_cmd_msg_size(void) return msgsize; } +/* Called with ovs_mutex or RCU read lock. */ static int ovs_dp_cmd_fill_info(struct datapath *dp, struct sk_buff *skb, u32 portid, u32 seq, u32 flags, u8 cmd) { @@ -1108,9 +1256,7 @@ 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, ovs_dp_name(dp)); - rcu_read_unlock(); if (err) goto nla_put_failure; @@ -1135,25 +1281,12 @@ error: return -EMSGSIZE; } -static struct sk_buff *ovs_dp_cmd_build_info(struct datapath *dp, - struct genl_info *info, u8 cmd) +static struct sk_buff *ovs_dp_cmd_alloc_info(struct genl_info *info) { - struct sk_buff *skb; - int retval; - - skb = genlmsg_new_unicast(ovs_dp_cmd_msg_size(), info, GFP_KERNEL); - if (!skb) - return ERR_PTR(-ENOMEM); - - retval = ovs_dp_cmd_fill_info(dp, skb, info->snd_portid, info->snd_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, struct nlattr *a[OVS_DP_ATTR_MAX + 1]) @@ -1165,10 +1298,8 @@ static struct datapath *lookup_datapath(struct net *net, 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); } @@ -1205,12 +1336,14 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) 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))); @@ -1245,6 +1378,9 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) 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)) { err = PTR_ERR(vport); @@ -1263,10 +1399,9 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) goto err_destroy_ports_array; } - reply = ovs_dp_cmd_build_info(dp, info, 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_rcu(&dp->list_node, &ovs_net->dps); @@ -1276,9 +1411,8 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) 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); @@ -1287,8 +1421,8 @@ err_destroy_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; } @@ -1326,16 +1460,19 @@ static int ovs_dp_cmd_del(struct sk_buff *skb, struct genl_info *info) 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, 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(); @@ -1343,8 +1480,10 @@ static int ovs_dp_cmd_del(struct sk_buff *skb, struct genl_info *info) ovs_notify(&dp_datapath_genl_family, reply, info); return 0; -unlock: + +err_unlock_free: ovs_unlock(); + kfree_skb(reply); return err; } @@ -1354,29 +1493,30 @@ static int ovs_dp_cmd_set(struct sk_buff *skb, struct genl_info *info) 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; ovs_dp_change(dp, info->attrs); - reply = ovs_dp_cmd_build_info(dp, info, OVS_DP_CMD_NEW); - if (IS_ERR(reply)) { - err = PTR_ERR(reply); - genl_set_err(&dp_datapath_genl_family, sock_net(skb->sk), 0, - 0, err); - err = 0; - 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(); ovs_notify(&dp_datapath_genl_family, reply, info); return 0; -unlock: + +err_unlock_free: ovs_unlock(); + kfree_skb(reply); return err; } @@ -1386,24 +1526,26 @@ static int ovs_dp_cmd_get(struct sk_buff *skb, struct genl_info *info) struct datapath *dp; int err; - ovs_lock(); + reply = ovs_dp_cmd_alloc_info(info); + if (!reply) + return -ENOMEM; + + rcu_read_lock(); dp = lookup_datapath(sock_net(skb->sk), info->userhdr, info->attrs); if (IS_ERR(dp)) { err = PTR_ERR(dp); - goto unlock; - } - - reply = ovs_dp_cmd_build_info(dp, info, OVS_DP_CMD_NEW); - if (IS_ERR(reply)) { - err = PTR_ERR(reply); - goto unlock; + goto err_unlock_free; } + err = ovs_dp_cmd_fill_info(dp, reply, info->snd_portid, + info->snd_seq, 0, OVS_DP_CMD_NEW); + BUG_ON(err < 0); + rcu_read_unlock(); - ovs_unlock(); return genlmsg_reply(reply, info); -unlock: - ovs_unlock(); +err_unlock_free: + rcu_read_unlock(); + kfree_skb(reply); return err; } @@ -1430,7 +1572,13 @@ static int ovs_dp_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb) return skb->len; } -static const 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 struct genl_ops dp_datapath_genl_ops[] = { { .cmd = OVS_DP_CMD_NEW, .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */ .policy = datapath_policy, @@ -1454,27 +1602,18 @@ static const struct genl_ops dp_datapath_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 }, -}; - -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, -}; - -static 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. */ @@ -1516,7 +1655,12 @@ error: 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) { @@ -1578,33 +1722,35 @@ static int ovs_vport_cmd_new(struct sk_buff *skb, struct genl_info *info) 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(); 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) @@ -1622,22 +1768,19 @@ static int ovs_vport_cmd_new(struct sk_buff *skb, struct genl_info *info) vport = new_vport(&parms); err = PTR_ERR(vport); if (IS_ERR(vport)) - goto exit_unlock; + goto exit_unlock_free; - 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; - } + 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; } @@ -1648,28 +1791,26 @@ static int ovs_vport_cmd_set(struct sk_buff *skb, struct genl_info *info) 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 (a[OVS_VPORT_ATTR_TYPE] && nla_get_u32(a[OVS_VPORT_ATTR_TYPE]) != vport->ops->type) { err = -EINVAL; - goto exit_unlock; - } - - reply = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); - if (!reply) { - err = -ENOMEM; - goto exit_unlock; + goto exit_unlock_free; } if (a[OVS_VPORT_ATTR_OPTIONS]) { err = ovs_vport_set_options(vport, a[OVS_VPORT_ATTR_OPTIONS]); if (err) - goto exit_free; + goto exit_unlock_free; } if (a[OVS_VPORT_ATTR_UPCALL_PID]) @@ -1683,10 +1824,9 @@ static int ovs_vport_cmd_set(struct sk_buff *skb, struct genl_info *info) ovs_notify(&dp_vport_genl_family, reply, info); return 0; -exit_free: - kfree_skb(reply); -exit_unlock: +exit_unlock_free: ovs_unlock(); + kfree_skb(reply); return err; } @@ -1697,30 +1837,33 @@ static int ovs_vport_cmd_del(struct sk_buff *skb, struct genl_info *info) 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(&dp_vport_genl_family, reply, info); + return 0; -exit_unlock: +exit_unlock_free: ovs_unlock(); + kfree_skb(reply); return err; } @@ -1732,24 +1875,25 @@ static int ovs_vport_cmd_get(struct sk_buff *skb, struct genl_info *info) 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; } @@ -1792,7 +1936,16 @@ out: return skb->len; } -static const 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 struct genl_ops dp_vport_genl_ops[] = { { .cmd = OVS_VPORT_CMD_NEW, .flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */ .policy = vport_policy, @@ -1816,26 +1969,25 @@ static const struct genl_ops dp_vport_genl_ops[] = { }, }; -struct genl_family_and_ops { - struct genl_family *family; - const struct genl_ops *ops; - int n_ops; - const 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) @@ -1843,33 +1995,25 @@ static void dp_unregister_genl(int n_families) 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]; - f->family->ops = f->ops; - f->family->n_ops = f->n_ops; - f->family->mcgrps = f->group; - f->family->n_mcgrps = f->group ? 1 : 0; - err = genl_register_family(f->family); + err = genl_register_family(dp_genl_families[i]); if (err) goto error; - n_registered++; } return 0; error: - dp_unregister_genl(n_registered); + dp_unregister_genl(i); return err; } diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h index 05317380fc0..7ede507500d 100644 --- a/net/openvswitch/datapath.h +++ b/net/openvswitch/datapath.h @@ -194,7 +194,9 @@ struct sk_buff *ovs_vport_cmd_build_info(struct vport *, u32 pid, u32 seq, int ovs_execute_actions(struct datapath *dp, struct sk_buff *skb); void ovs_dp_notify_wq(struct work_struct *work); -#define OVS_NLERR(fmt, ...) \ - pr_info_once("netlink: " fmt, ##__VA_ARGS__) - +#define OVS_NLERR(fmt, ...) \ +do { \ + if (net_ratelimit()) \ + pr_info("netlink: " fmt, ##__VA_ARGS__); \ +} while (0) #endif /* datapath.h */ diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c index 2998989e76d..334751cb152 100644 --- a/net/openvswitch/flow.c +++ b/net/openvswitch/flow.c @@ -64,88 +64,110 @@ u64 ovs_flow_used_time(unsigned long flow_jiffies) void ovs_flow_stats_update(struct sw_flow *flow, struct sk_buff *skb) { struct flow_stats *stats; - __be16 tcp_flags = 0; - - if (!flow->stats.is_percpu) - stats = flow->stats.stat; - else - stats = this_cpu_ptr(flow->stats.cpu_stats); - - if ((flow->key.eth.type == htons(ETH_P_IP) || - flow->key.eth.type == htons(ETH_P_IPV6)) && - flow->key.ip.frag != OVS_FRAG_TYPE_LATER && - flow->key.ip.proto == IPPROTO_TCP && - likely(skb->len >= skb_transport_offset(skb) + sizeof(struct tcphdr))) { - tcp_flags = TCP_FLAGS_BE16(tcp_hdr(skb)); + __be16 tcp_flags = flow->key.tp.flags; + int node = numa_node_id(); + + stats = rcu_dereference(flow->stats[node]); + + /* Check if already have node-specific stats. */ + if (likely(stats)) { + spin_lock(&stats->lock); + /* Mark if we write on the pre-allocated stats. */ + if (node == 0 && unlikely(flow->stats_last_writer != node)) + flow->stats_last_writer = node; + } else { + stats = rcu_dereference(flow->stats[0]); /* Pre-allocated. */ + spin_lock(&stats->lock); + + /* If the current NUMA-node is the only writer on the + * pre-allocated stats keep using them. + */ + if (unlikely(flow->stats_last_writer != node)) { + /* A previous locker may have already allocated the + * stats, so we need to check again. If node-specific + * stats were already allocated, we update the pre- + * allocated stats as we have already locked them. + */ + if (likely(flow->stats_last_writer != NUMA_NO_NODE) + && likely(!rcu_dereference(flow->stats[node]))) { + /* Try to allocate node-specific stats. */ + struct flow_stats *new_stats; + + new_stats = + kmem_cache_alloc_node(flow_stats_cache, + GFP_THISNODE | + __GFP_NOMEMALLOC, + node); + if (likely(new_stats)) { + new_stats->used = jiffies; + new_stats->packet_count = 1; + new_stats->byte_count = skb->len; + new_stats->tcp_flags = tcp_flags; + spin_lock_init(&new_stats->lock); + + rcu_assign_pointer(flow->stats[node], + new_stats); + goto unlock; + } + } + flow->stats_last_writer = node; + } } - spin_lock(&stats->lock); stats->used = jiffies; stats->packet_count++; stats->byte_count += skb->len; stats->tcp_flags |= tcp_flags; +unlock: spin_unlock(&stats->lock); } -static void stats_read(struct flow_stats *stats, - struct ovs_flow_stats *ovs_stats, - unsigned long *used, __be16 *tcp_flags) -{ - spin_lock(&stats->lock); - if (!*used || time_after(stats->used, *used)) - *used = stats->used; - *tcp_flags |= stats->tcp_flags; - ovs_stats->n_packets += stats->packet_count; - ovs_stats->n_bytes += stats->byte_count; - spin_unlock(&stats->lock); -} - -void ovs_flow_stats_get(struct sw_flow *flow, struct ovs_flow_stats *ovs_stats, +/* Must be called with rcu_read_lock or ovs_mutex. */ +void ovs_flow_stats_get(const struct sw_flow *flow, + struct ovs_flow_stats *ovs_stats, unsigned long *used, __be16 *tcp_flags) { - int cpu; + int node; *used = 0; *tcp_flags = 0; memset(ovs_stats, 0, sizeof(*ovs_stats)); - local_bh_disable(); - if (!flow->stats.is_percpu) { - stats_read(flow->stats.stat, ovs_stats, used, tcp_flags); - } else { - for_each_possible_cpu(cpu) { - struct flow_stats *stats; + for_each_node(node) { + struct flow_stats *stats = rcu_dereference_ovsl(flow->stats[node]); - stats = per_cpu_ptr(flow->stats.cpu_stats, cpu); - stats_read(stats, ovs_stats, used, tcp_flags); + if (stats) { + /* Local CPU may write on non-local stats, so we must + * block bottom-halves here. + */ + spin_lock_bh(&stats->lock); + if (!*used || time_after(stats->used, *used)) + *used = stats->used; + *tcp_flags |= stats->tcp_flags; + ovs_stats->n_packets += stats->packet_count; + ovs_stats->n_bytes += stats->byte_count; + spin_unlock_bh(&stats->lock); } } - local_bh_enable(); -} - -static void stats_reset(struct flow_stats *stats) -{ - spin_lock(&stats->lock); - stats->used = 0; - stats->packet_count = 0; - stats->byte_count = 0; - stats->tcp_flags = 0; - spin_unlock(&stats->lock); } +/* Called with ovs_mutex. */ void ovs_flow_stats_clear(struct sw_flow *flow) { - int cpu; - - local_bh_disable(); - if (!flow->stats.is_percpu) { - stats_reset(flow->stats.stat); - } else { - for_each_possible_cpu(cpu) { - stats_reset(per_cpu_ptr(flow->stats.cpu_stats, cpu)); + int node; + + for_each_node(node) { + struct flow_stats *stats = ovsl_dereference(flow->stats[node]); + + if (stats) { + spin_lock_bh(&stats->lock); + stats->used = 0; + stats->packet_count = 0; + stats->byte_count = 0; + stats->tcp_flags = 0; + spin_unlock_bh(&stats->lock); } } - local_bh_enable(); } static int check_header(struct sk_buff *skb, int len) @@ -332,8 +354,8 @@ static int parse_icmpv6(struct sk_buff *skb, struct sw_flow_key *key, /* The ICMPv6 type and code fields use the 16-bit transport port * fields, so we need to store them in 16-bit network byte order. */ - key->ipv6.tp.src = htons(icmp->icmp6_type); - key->ipv6.tp.dst = htons(icmp->icmp6_code); + key->tp.src = htons(icmp->icmp6_type); + key->tp.dst = htons(icmp->icmp6_code); if (icmp->icmp6_code == 0 && (icmp->icmp6_type == NDISC_NEIGHBOUR_SOLICITATION || @@ -372,14 +394,14 @@ static int parse_icmpv6(struct sk_buff *skb, struct sw_flow_key *key, && opt_len == 8) { if (unlikely(!is_zero_ether_addr(key->ipv6.nd.sll))) goto invalid; - memcpy(key->ipv6.nd.sll, - &nd->opt[offset+sizeof(*nd_opt)], ETH_ALEN); + ether_addr_copy(key->ipv6.nd.sll, + &nd->opt[offset+sizeof(*nd_opt)]); } else if (nd_opt->nd_opt_type == ND_OPT_TARGET_LL_ADDR && opt_len == 8) { if (unlikely(!is_zero_ether_addr(key->ipv6.nd.tll))) goto invalid; - memcpy(key->ipv6.nd.tll, - &nd->opt[offset+sizeof(*nd_opt)], ETH_ALEN); + ether_addr_copy(key->ipv6.nd.tll, + &nd->opt[offset+sizeof(*nd_opt)]); } icmp_len -= opt_len; @@ -439,8 +461,8 @@ int ovs_flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key) * header in the linear data area. */ eth = eth_hdr(skb); - memcpy(key->eth.src, eth->h_source, ETH_ALEN); - memcpy(key->eth.dst, eth->h_dest, ETH_ALEN); + ether_addr_copy(key->eth.src, eth->h_source); + ether_addr_copy(key->eth.dst, eth->h_dest); __skb_pull(skb, 2 * ETH_ALEN); /* We are going to push all headers that we pull, so no need to @@ -495,21 +517,21 @@ int ovs_flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key) if (key->ip.proto == IPPROTO_TCP) { if (tcphdr_ok(skb)) { struct tcphdr *tcp = tcp_hdr(skb); - key->ipv4.tp.src = tcp->source; - key->ipv4.tp.dst = tcp->dest; - key->ipv4.tp.flags = TCP_FLAGS_BE16(tcp); + key->tp.src = tcp->source; + key->tp.dst = tcp->dest; + key->tp.flags = TCP_FLAGS_BE16(tcp); } } else if (key->ip.proto == IPPROTO_UDP) { if (udphdr_ok(skb)) { struct udphdr *udp = udp_hdr(skb); - key->ipv4.tp.src = udp->source; - key->ipv4.tp.dst = udp->dest; + key->tp.src = udp->source; + key->tp.dst = udp->dest; } } else if (key->ip.proto == IPPROTO_SCTP) { if (sctphdr_ok(skb)) { struct sctphdr *sctp = sctp_hdr(skb); - key->ipv4.tp.src = sctp->source; - key->ipv4.tp.dst = sctp->dest; + key->tp.src = sctp->source; + key->tp.dst = sctp->dest; } } else if (key->ip.proto == IPPROTO_ICMP) { if (icmphdr_ok(skb)) { @@ -517,8 +539,8 @@ int ovs_flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key) /* The ICMP type and code fields use the 16-bit * transport port fields, so we need to store * them in 16-bit network byte order. */ - key->ipv4.tp.src = htons(icmp->type); - key->ipv4.tp.dst = htons(icmp->code); + key->tp.src = htons(icmp->type); + key->tp.dst = htons(icmp->code); } } @@ -538,8 +560,8 @@ int ovs_flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key) key->ip.proto = ntohs(arp->ar_op); memcpy(&key->ipv4.addr.src, arp->ar_sip, sizeof(key->ipv4.addr.src)); memcpy(&key->ipv4.addr.dst, arp->ar_tip, sizeof(key->ipv4.addr.dst)); - memcpy(key->ipv4.arp.sha, arp->ar_sha, ETH_ALEN); - memcpy(key->ipv4.arp.tha, arp->ar_tha, ETH_ALEN); + ether_addr_copy(key->ipv4.arp.sha, arp->ar_sha); + ether_addr_copy(key->ipv4.arp.tha, arp->ar_tha); } } else if (key->eth.type == htons(ETH_P_IPV6)) { int nh_len; /* IPv6 Header + Extensions */ @@ -564,21 +586,21 @@ int ovs_flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key) if (key->ip.proto == NEXTHDR_TCP) { if (tcphdr_ok(skb)) { struct tcphdr *tcp = tcp_hdr(skb); - key->ipv6.tp.src = tcp->source; - key->ipv6.tp.dst = tcp->dest; - key->ipv6.tp.flags = TCP_FLAGS_BE16(tcp); + key->tp.src = tcp->source; + key->tp.dst = tcp->dest; + key->tp.flags = TCP_FLAGS_BE16(tcp); } } else if (key->ip.proto == NEXTHDR_UDP) { if (udphdr_ok(skb)) { struct udphdr *udp = udp_hdr(skb); - key->ipv6.tp.src = udp->source; - key->ipv6.tp.dst = udp->dest; + key->tp.src = udp->source; + key->tp.dst = udp->dest; } } else if (key->ip.proto == NEXTHDR_SCTP) { if (sctphdr_ok(skb)) { struct sctphdr *sctp = sctp_hdr(skb); - key->ipv6.tp.src = sctp->source; - key->ipv6.tp.dst = sctp->dest; + key->tp.src = sctp->source; + key->tp.dst = sctp->dest; } } else if (key->ip.proto == NEXTHDR_ICMP) { if (icmp6hdr_ok(skb)) { diff --git a/net/openvswitch/flow.h b/net/openvswitch/flow.h index 2d770e28a3a..ac395d2cd82 100644 --- a/net/openvswitch/flow.h +++ b/net/openvswitch/flow.h @@ -47,7 +47,7 @@ struct ovs_key_ipv4_tunnel { __be16 tun_flags; u8 ipv4_tos; u8 ipv4_ttl; -}; +} __packed __aligned(4); /* Minimize padding. */ static inline void ovs_flow_tun_key_init(struct ovs_key_ipv4_tunnel *tun_key, const struct iphdr *iph, __be64 tun_id, @@ -71,7 +71,7 @@ struct sw_flow_key { u32 priority; /* Packet QoS priority. */ u32 skb_mark; /* SKB mark. */ u16 in_port; /* Input switch port (or DP_MAX_PORTS). */ - } phy; + } __packed phy; /* Safe when right after 'tun_key'. */ struct { u8 src[ETH_ALEN]; /* Ethernet source address. */ u8 dst[ETH_ALEN]; /* Ethernet destination address. */ @@ -84,23 +84,21 @@ struct sw_flow_key { u8 ttl; /* IP TTL/hop limit. */ u8 frag; /* One of OVS_FRAG_TYPE_*. */ } ip; + struct { + __be16 src; /* TCP/UDP/SCTP source port. */ + __be16 dst; /* TCP/UDP/SCTP destination port. */ + __be16 flags; /* TCP flags. */ + } tp; union { struct { struct { __be32 src; /* IP source address. */ __be32 dst; /* IP destination address. */ } addr; - union { - struct { - __be16 src; /* TCP/UDP/SCTP source port. */ - __be16 dst; /* TCP/UDP/SCTP destination port. */ - __be16 flags; /* TCP flags. */ - } tp; - struct { - u8 sha[ETH_ALEN]; /* ARP source hardware address. */ - u8 tha[ETH_ALEN]; /* ARP target hardware address. */ - } arp; - }; + struct { + u8 sha[ETH_ALEN]; /* ARP source hardware address. */ + u8 tha[ETH_ALEN]; /* ARP target hardware address. */ + } arp; } ipv4; struct { struct { @@ -109,11 +107,6 @@ struct sw_flow_key { } addr; __be32 label; /* IPv6 flow label. */ struct { - __be16 src; /* TCP/UDP/SCTP source port. */ - __be16 dst; /* TCP/UDP/SCTP destination port. */ - __be16 flags; /* TCP flags. */ - } tp; - struct { struct in6_addr target; /* ND target address. */ u8 sll[ETH_ALEN]; /* ND source link layer address. */ u8 tll[ETH_ALEN]; /* ND target link layer address. */ @@ -155,24 +148,22 @@ struct flow_stats { __be16 tcp_flags; /* Union of seen TCP flags. */ }; -struct sw_flow_stats { - bool is_percpu; - union { - struct flow_stats *stat; - struct flow_stats __percpu *cpu_stats; - }; -}; - struct sw_flow { struct rcu_head rcu; struct hlist_node hash_node[2]; u32 hash; - + int stats_last_writer; /* NUMA-node id of the last writer on + * 'stats[0]'. + */ struct sw_flow_key key; struct sw_flow_key unmasked_key; struct sw_flow_mask *mask; struct sw_flow_actions __rcu *sf_acts; - struct sw_flow_stats stats; + struct flow_stats __rcu *stats[]; /* One for each NUMA node. First one + * is allocated at flow creation time, + * the rest are allocated on demand + * while holding the 'stats[0].lock'. + */ }; struct arp_eth_header { @@ -189,10 +180,10 @@ struct arp_eth_header { unsigned char ar_tip[4]; /* target IP address */ } __packed; -void ovs_flow_stats_update(struct sw_flow *flow, struct sk_buff *skb); -void ovs_flow_stats_get(struct sw_flow *flow, struct ovs_flow_stats *stats, +void ovs_flow_stats_update(struct sw_flow *, struct sk_buff *); +void ovs_flow_stats_get(const struct sw_flow *, struct ovs_flow_stats *, unsigned long *used, __be16 *tcp_flags); -void ovs_flow_stats_clear(struct sw_flow *flow); +void ovs_flow_stats_clear(struct sw_flow *); u64 ovs_flow_used_time(unsigned long flow_jiffies); int ovs_flow_extract(struct sk_buff *, u16 in_port, struct sw_flow_key *); diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c index 4d000acaed0..d757848da89 100644 --- a/net/openvswitch/flow_netlink.c +++ b/net/openvswitch/flow_netlink.c @@ -16,6 +16,8 @@ * 02110-1301, USA */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include "flow.h" #include "datapath.h" #include <linux/uaccess.h> @@ -202,11 +204,11 @@ static bool match_validate(const struct sw_flow_match *match, if (match->mask && (match->mask->key.ip.proto == 0xff)) mask_allowed |= 1 << OVS_KEY_ATTR_ICMPV6; - if (match->key->ipv6.tp.src == + if (match->key->tp.src == htons(NDISC_NEIGHBOUR_SOLICITATION) || - match->key->ipv6.tp.src == htons(NDISC_NEIGHBOUR_ADVERTISEMENT)) { + match->key->tp.src == htons(NDISC_NEIGHBOUR_ADVERTISEMENT)) { key_expected |= 1 << OVS_KEY_ATTR_ND; - if (match->mask && (match->mask->key.ipv6.tp.src == htons(0xffff))) + if (match->mask && (match->mask->key.tp.src == htons(0xffff))) mask_allowed |= 1 << OVS_KEY_ATTR_ND; } } @@ -216,14 +218,14 @@ static bool match_validate(const struct sw_flow_match *match, if ((key_attrs & key_expected) != key_expected) { /* Key attributes check failed. */ OVS_NLERR("Missing expected key attributes (key_attrs=%llx, expected=%llx).\n", - key_attrs, key_expected); + (unsigned long long)key_attrs, (unsigned long long)key_expected); return false; } if ((mask_attrs & mask_allowed) != mask_attrs) { /* Mask attributes check failed. */ OVS_NLERR("Contain more than allowed mask fields (mask_attrs=%llx, mask_allowed=%llx).\n", - mask_attrs, mask_allowed); + (unsigned long long)mask_attrs, (unsigned long long)mask_allowed); return false; } @@ -266,20 +268,6 @@ static bool is_all_zero(const u8 *fp, size_t size) return true; } -static bool is_all_set(const u8 *fp, size_t size) -{ - int i; - - if (!fp) - return false; - - for (i = 0; i < size; i++) - if (fp[i] != 0xff) - return false; - - return true; -} - static int __parse_flow_nlattrs(const struct nlattr *attr, const struct nlattr *a[], u64 *attrsp, bool nz) @@ -501,9 +489,8 @@ static int metadata_from_nlattrs(struct sw_flow_match *match, u64 *attrs, return 0; } -static int ovs_key_from_nlattrs(struct sw_flow_match *match, bool *exact_5tuple, - u64 attrs, const struct nlattr **a, - bool is_mask) +static int ovs_key_from_nlattrs(struct sw_flow_match *match, u64 attrs, + const struct nlattr **a, bool is_mask) { int err; u64 orig_attrs = attrs; @@ -560,11 +547,6 @@ static int ovs_key_from_nlattrs(struct sw_flow_match *match, bool *exact_5tuple SW_FLOW_KEY_PUT(match, eth.type, htons(ETH_P_802_2), is_mask); } - if (is_mask && exact_5tuple) { - if (match->mask->key.eth.type != htons(0xffff)) - *exact_5tuple = false; - } - if (attrs & (1 << OVS_KEY_ATTR_IPV4)) { const struct ovs_key_ipv4 *ipv4_key; @@ -587,13 +569,6 @@ static int ovs_key_from_nlattrs(struct sw_flow_match *match, bool *exact_5tuple SW_FLOW_KEY_PUT(match, ipv4.addr.dst, ipv4_key->ipv4_dst, is_mask); attrs &= ~(1 << OVS_KEY_ATTR_IPV4); - - if (is_mask && exact_5tuple && *exact_5tuple) { - if (ipv4_key->ipv4_proto != 0xff || - ipv4_key->ipv4_src != htonl(0xffffffff) || - ipv4_key->ipv4_dst != htonl(0xffffffff)) - *exact_5tuple = false; - } } if (attrs & (1 << OVS_KEY_ATTR_IPV6)) { @@ -625,13 +600,6 @@ static int ovs_key_from_nlattrs(struct sw_flow_match *match, bool *exact_5tuple is_mask); attrs &= ~(1 << OVS_KEY_ATTR_IPV6); - - if (is_mask && exact_5tuple && *exact_5tuple) { - if (ipv6_key->ipv6_proto != 0xff || - !is_all_set((u8 *)ipv6_key->ipv6_src, sizeof(match->key->ipv6.addr.src)) || - !is_all_set((u8 *)ipv6_key->ipv6_dst, sizeof(match->key->ipv6.addr.dst))) - *exact_5tuple = false; - } } if (attrs & (1 << OVS_KEY_ATTR_ARP)) { @@ -662,32 +630,18 @@ static int ovs_key_from_nlattrs(struct sw_flow_match *match, bool *exact_5tuple const struct ovs_key_tcp *tcp_key; tcp_key = nla_data(a[OVS_KEY_ATTR_TCP]); - if (orig_attrs & (1 << OVS_KEY_ATTR_IPV4)) { - SW_FLOW_KEY_PUT(match, ipv4.tp.src, - tcp_key->tcp_src, is_mask); - SW_FLOW_KEY_PUT(match, ipv4.tp.dst, - tcp_key->tcp_dst, is_mask); - } else { - SW_FLOW_KEY_PUT(match, ipv6.tp.src, - tcp_key->tcp_src, is_mask); - SW_FLOW_KEY_PUT(match, ipv6.tp.dst, - tcp_key->tcp_dst, is_mask); - } + SW_FLOW_KEY_PUT(match, tp.src, tcp_key->tcp_src, is_mask); + SW_FLOW_KEY_PUT(match, tp.dst, tcp_key->tcp_dst, is_mask); attrs &= ~(1 << OVS_KEY_ATTR_TCP); - - if (is_mask && exact_5tuple && *exact_5tuple && - (tcp_key->tcp_src != htons(0xffff) || - tcp_key->tcp_dst != htons(0xffff))) - *exact_5tuple = false; } if (attrs & (1 << OVS_KEY_ATTR_TCP_FLAGS)) { if (orig_attrs & (1 << OVS_KEY_ATTR_IPV4)) { - SW_FLOW_KEY_PUT(match, ipv4.tp.flags, + SW_FLOW_KEY_PUT(match, tp.flags, nla_get_be16(a[OVS_KEY_ATTR_TCP_FLAGS]), is_mask); } else { - SW_FLOW_KEY_PUT(match, ipv6.tp.flags, + SW_FLOW_KEY_PUT(match, tp.flags, nla_get_be16(a[OVS_KEY_ATTR_TCP_FLAGS]), is_mask); } @@ -698,40 +652,17 @@ static int ovs_key_from_nlattrs(struct sw_flow_match *match, bool *exact_5tuple const struct ovs_key_udp *udp_key; udp_key = nla_data(a[OVS_KEY_ATTR_UDP]); - if (orig_attrs & (1 << OVS_KEY_ATTR_IPV4)) { - SW_FLOW_KEY_PUT(match, ipv4.tp.src, - udp_key->udp_src, is_mask); - SW_FLOW_KEY_PUT(match, ipv4.tp.dst, - udp_key->udp_dst, is_mask); - } else { - SW_FLOW_KEY_PUT(match, ipv6.tp.src, - udp_key->udp_src, is_mask); - SW_FLOW_KEY_PUT(match, ipv6.tp.dst, - udp_key->udp_dst, is_mask); - } + SW_FLOW_KEY_PUT(match, tp.src, udp_key->udp_src, is_mask); + SW_FLOW_KEY_PUT(match, tp.dst, udp_key->udp_dst, is_mask); attrs &= ~(1 << OVS_KEY_ATTR_UDP); - - if (is_mask && exact_5tuple && *exact_5tuple && - (udp_key->udp_src != htons(0xffff) || - udp_key->udp_dst != htons(0xffff))) - *exact_5tuple = false; } if (attrs & (1 << OVS_KEY_ATTR_SCTP)) { const struct ovs_key_sctp *sctp_key; sctp_key = nla_data(a[OVS_KEY_ATTR_SCTP]); - if (orig_attrs & (1 << OVS_KEY_ATTR_IPV4)) { - SW_FLOW_KEY_PUT(match, ipv4.tp.src, - sctp_key->sctp_src, is_mask); - SW_FLOW_KEY_PUT(match, ipv4.tp.dst, - sctp_key->sctp_dst, is_mask); - } else { - SW_FLOW_KEY_PUT(match, ipv6.tp.src, - sctp_key->sctp_src, is_mask); - SW_FLOW_KEY_PUT(match, ipv6.tp.dst, - sctp_key->sctp_dst, is_mask); - } + SW_FLOW_KEY_PUT(match, tp.src, sctp_key->sctp_src, is_mask); + SW_FLOW_KEY_PUT(match, tp.dst, sctp_key->sctp_dst, is_mask); attrs &= ~(1 << OVS_KEY_ATTR_SCTP); } @@ -739,9 +670,9 @@ static int ovs_key_from_nlattrs(struct sw_flow_match *match, bool *exact_5tuple const struct ovs_key_icmp *icmp_key; icmp_key = nla_data(a[OVS_KEY_ATTR_ICMP]); - SW_FLOW_KEY_PUT(match, ipv4.tp.src, + SW_FLOW_KEY_PUT(match, tp.src, htons(icmp_key->icmp_type), is_mask); - SW_FLOW_KEY_PUT(match, ipv4.tp.dst, + SW_FLOW_KEY_PUT(match, tp.dst, htons(icmp_key->icmp_code), is_mask); attrs &= ~(1 << OVS_KEY_ATTR_ICMP); } @@ -750,9 +681,9 @@ static int ovs_key_from_nlattrs(struct sw_flow_match *match, bool *exact_5tuple const struct ovs_key_icmpv6 *icmpv6_key; icmpv6_key = nla_data(a[OVS_KEY_ATTR_ICMPV6]); - SW_FLOW_KEY_PUT(match, ipv6.tp.src, + SW_FLOW_KEY_PUT(match, tp.src, htons(icmpv6_key->icmpv6_type), is_mask); - SW_FLOW_KEY_PUT(match, ipv6.tp.dst, + SW_FLOW_KEY_PUT(match, tp.dst, htons(icmpv6_key->icmpv6_code), is_mask); attrs &= ~(1 << OVS_KEY_ATTR_ICMPV6); } @@ -800,7 +731,6 @@ static void sw_flow_mask_set(struct sw_flow_mask *mask, * attribute specifies the mask field of the wildcarded flow. */ int ovs_nla_get_match(struct sw_flow_match *match, - bool *exact_5tuple, const struct nlattr *key, const struct nlattr *mask) { @@ -848,13 +778,10 @@ int ovs_nla_get_match(struct sw_flow_match *match, } } - err = ovs_key_from_nlattrs(match, NULL, key_attrs, a, false); + err = ovs_key_from_nlattrs(match, key_attrs, a, false); if (err) return err; - if (exact_5tuple) - *exact_5tuple = true; - if (mask) { err = parse_flow_mask_nlattrs(mask, a, &mask_attrs); if (err) @@ -892,7 +819,7 @@ int ovs_nla_get_match(struct sw_flow_match *match, } } - err = ovs_key_from_nlattrs(match, exact_5tuple, mask_attrs, a, true); + err = ovs_key_from_nlattrs(match, mask_attrs, a, true); if (err) return err; } else { @@ -982,8 +909,8 @@ int ovs_nla_put_flow(const struct sw_flow_key *swkey, goto nla_put_failure; eth_key = nla_data(nla); - memcpy(eth_key->eth_src, output->eth.src, ETH_ALEN); - memcpy(eth_key->eth_dst, output->eth.dst, ETH_ALEN); + ether_addr_copy(eth_key->eth_src, output->eth.src); + ether_addr_copy(eth_key->eth_dst, output->eth.dst); if (swkey->eth.tci || swkey->eth.type == htons(ETH_P_8021Q)) { __be16 eth_type; @@ -1055,8 +982,8 @@ int ovs_nla_put_flow(const struct sw_flow_key *swkey, arp_key->arp_sip = output->ipv4.addr.src; arp_key->arp_tip = output->ipv4.addr.dst; arp_key->arp_op = htons(output->ip.proto); - memcpy(arp_key->arp_sha, output->ipv4.arp.sha, ETH_ALEN); - memcpy(arp_key->arp_tha, output->ipv4.arp.tha, ETH_ALEN); + ether_addr_copy(arp_key->arp_sha, output->ipv4.arp.sha); + ether_addr_copy(arp_key->arp_tha, output->ipv4.arp.tha); } if ((swkey->eth.type == htons(ETH_P_IP) || @@ -1070,19 +997,11 @@ int ovs_nla_put_flow(const struct sw_flow_key *swkey, if (!nla) goto nla_put_failure; tcp_key = nla_data(nla); - if (swkey->eth.type == htons(ETH_P_IP)) { - tcp_key->tcp_src = output->ipv4.tp.src; - tcp_key->tcp_dst = output->ipv4.tp.dst; - if (nla_put_be16(skb, OVS_KEY_ATTR_TCP_FLAGS, - output->ipv4.tp.flags)) - goto nla_put_failure; - } else if (swkey->eth.type == htons(ETH_P_IPV6)) { - tcp_key->tcp_src = output->ipv6.tp.src; - tcp_key->tcp_dst = output->ipv6.tp.dst; - if (nla_put_be16(skb, OVS_KEY_ATTR_TCP_FLAGS, - output->ipv6.tp.flags)) - goto nla_put_failure; - } + tcp_key->tcp_src = output->tp.src; + tcp_key->tcp_dst = output->tp.dst; + if (nla_put_be16(skb, OVS_KEY_ATTR_TCP_FLAGS, + output->tp.flags)) + goto nla_put_failure; } else if (swkey->ip.proto == IPPROTO_UDP) { struct ovs_key_udp *udp_key; @@ -1090,13 +1009,8 @@ int ovs_nla_put_flow(const struct sw_flow_key *swkey, if (!nla) goto nla_put_failure; udp_key = nla_data(nla); - if (swkey->eth.type == htons(ETH_P_IP)) { - udp_key->udp_src = output->ipv4.tp.src; - udp_key->udp_dst = output->ipv4.tp.dst; - } else if (swkey->eth.type == htons(ETH_P_IPV6)) { - udp_key->udp_src = output->ipv6.tp.src; - udp_key->udp_dst = output->ipv6.tp.dst; - } + udp_key->udp_src = output->tp.src; + udp_key->udp_dst = output->tp.dst; } else if (swkey->ip.proto == IPPROTO_SCTP) { struct ovs_key_sctp *sctp_key; @@ -1104,13 +1018,8 @@ int ovs_nla_put_flow(const struct sw_flow_key *swkey, if (!nla) goto nla_put_failure; sctp_key = nla_data(nla); - if (swkey->eth.type == htons(ETH_P_IP)) { - sctp_key->sctp_src = swkey->ipv4.tp.src; - sctp_key->sctp_dst = swkey->ipv4.tp.dst; - } else if (swkey->eth.type == htons(ETH_P_IPV6)) { - sctp_key->sctp_src = swkey->ipv6.tp.src; - sctp_key->sctp_dst = swkey->ipv6.tp.dst; - } + sctp_key->sctp_src = output->tp.src; + sctp_key->sctp_dst = output->tp.dst; } else if (swkey->eth.type == htons(ETH_P_IP) && swkey->ip.proto == IPPROTO_ICMP) { struct ovs_key_icmp *icmp_key; @@ -1119,8 +1028,8 @@ int ovs_nla_put_flow(const struct sw_flow_key *swkey, if (!nla) goto nla_put_failure; icmp_key = nla_data(nla); - icmp_key->icmp_type = ntohs(output->ipv4.tp.src); - icmp_key->icmp_code = ntohs(output->ipv4.tp.dst); + icmp_key->icmp_type = ntohs(output->tp.src); + icmp_key->icmp_code = ntohs(output->tp.dst); } else if (swkey->eth.type == htons(ETH_P_IPV6) && swkey->ip.proto == IPPROTO_ICMPV6) { struct ovs_key_icmpv6 *icmpv6_key; @@ -1130,8 +1039,8 @@ int ovs_nla_put_flow(const struct sw_flow_key *swkey, if (!nla) goto nla_put_failure; icmpv6_key = nla_data(nla); - icmpv6_key->icmpv6_type = ntohs(output->ipv6.tp.src); - icmpv6_key->icmpv6_code = ntohs(output->ipv6.tp.dst); + icmpv6_key->icmpv6_type = ntohs(output->tp.src); + icmpv6_key->icmpv6_code = ntohs(output->tp.dst); if (icmpv6_key->icmpv6_type == NDISC_NEIGHBOUR_SOLICITATION || icmpv6_key->icmpv6_type == NDISC_NEIGHBOUR_ADVERTISEMENT) { @@ -1143,8 +1052,8 @@ int ovs_nla_put_flow(const struct sw_flow_key *swkey, nd_key = nla_data(nla); memcpy(nd_key->nd_target, &output->ipv6.nd.target, sizeof(nd_key->nd_target)); - memcpy(nd_key->nd_sll, output->ipv6.nd.sll, ETH_ALEN); - memcpy(nd_key->nd_tll, output->ipv6.nd.tll, ETH_ALEN); + ether_addr_copy(nd_key->nd_sll, output->ipv6.nd.sll); + ether_addr_copy(nd_key->nd_tll, output->ipv6.nd.tll); } } } @@ -1309,13 +1218,10 @@ static int validate_and_copy_sample(const struct nlattr *attr, 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 ((flow_key->eth.type == htons(ETH_P_IP) || + flow_key->eth.type == htons(ETH_P_IPV6)) && + (flow_key->tp.src || flow_key->tp.dst)) + return 0; return -EINVAL; } diff --git a/net/openvswitch/flow_netlink.h b/net/openvswitch/flow_netlink.h index b31fbe28bc7..440151045d3 100644 --- a/net/openvswitch/flow_netlink.h +++ b/net/openvswitch/flow_netlink.h @@ -45,7 +45,6 @@ int ovs_nla_put_flow(const struct sw_flow_key *, int ovs_nla_get_flow_metadata(struct sw_flow *flow, const struct nlattr *attr); int ovs_nla_get_match(struct sw_flow_match *match, - bool *exact_5tuple, const struct nlattr *, const struct nlattr *); diff --git a/net/openvswitch/flow_table.c b/net/openvswitch/flow_table.c index 3c268b3d71c..574c3abc9b3 100644 --- a/net/openvswitch/flow_table.c +++ b/net/openvswitch/flow_table.c @@ -48,6 +48,7 @@ #define REHASH_INTERVAL (10 * 60 * HZ) static struct kmem_cache *flow_cache; +struct kmem_cache *flow_stats_cache __read_mostly; static u16 range_n_bytes(const struct sw_flow_key_range *range) { @@ -57,8 +58,10 @@ static u16 range_n_bytes(const struct sw_flow_key_range *range) void ovs_flow_mask_key(struct sw_flow_key *dst, const struct sw_flow_key *src, const struct sw_flow_mask *mask) { - const long *m = (long *)((u8 *)&mask->key + mask->range.start); - const long *s = (long *)((u8 *)src + mask->range.start); + const long *m = (const long *)((const u8 *)&mask->key + + mask->range.start); + const long *s = (const long *)((const u8 *)src + + mask->range.start); long *d = (long *)((u8 *)dst + mask->range.start); int i; @@ -70,10 +73,11 @@ void ovs_flow_mask_key(struct sw_flow_key *dst, const struct sw_flow_key *src, *d++ = *s++ & *m++; } -struct sw_flow *ovs_flow_alloc(bool percpu_stats) +struct sw_flow *ovs_flow_alloc(void) { struct sw_flow *flow; - int cpu; + struct flow_stats *stats; + int node; flow = kmem_cache_alloc(flow_cache, GFP_KERNEL); if (!flow) @@ -81,27 +85,22 @@ struct sw_flow *ovs_flow_alloc(bool percpu_stats) flow->sf_acts = NULL; flow->mask = NULL; + flow->stats_last_writer = NUMA_NO_NODE; - flow->stats.is_percpu = percpu_stats; + /* Initialize the default stat node. */ + stats = kmem_cache_alloc_node(flow_stats_cache, + GFP_KERNEL | __GFP_ZERO, 0); + if (!stats) + goto err; - if (!percpu_stats) { - flow->stats.stat = kzalloc(sizeof(*flow->stats.stat), GFP_KERNEL); - if (!flow->stats.stat) - goto err; + spin_lock_init(&stats->lock); - spin_lock_init(&flow->stats.stat->lock); - } else { - flow->stats.cpu_stats = alloc_percpu(struct flow_stats); - if (!flow->stats.cpu_stats) - goto err; + RCU_INIT_POINTER(flow->stats[0], stats); - for_each_possible_cpu(cpu) { - struct flow_stats *cpu_stats; + for_each_node(node) + if (node != 0) + RCU_INIT_POINTER(flow->stats[node], NULL); - cpu_stats = per_cpu_ptr(flow->stats.cpu_stats, cpu); - spin_lock_init(&cpu_stats->lock); - } - } return flow; err: kmem_cache_free(flow_cache, flow); @@ -138,11 +137,13 @@ static struct flex_array *alloc_buckets(unsigned int n_buckets) static void flow_free(struct sw_flow *flow) { - kfree((struct sf_flow_acts __force *)flow->sf_acts); - if (flow->stats.is_percpu) - free_percpu(flow->stats.cpu_stats); - else - kfree(flow->stats.stat); + int node; + + kfree((struct sw_flow_actions __force *)flow->sf_acts); + for_each_node(node) + if (flow->stats[node]) + kmem_cache_free(flow_stats_cache, + (struct flow_stats __force *)flow->stats[node]); kmem_cache_free(flow_cache, flow); } @@ -158,25 +159,6 @@ void ovs_flow_free(struct sw_flow *flow, bool deferred) if (!flow) return; - if (flow->mask) { - struct sw_flow_mask *mask = flow->mask; - - /* ovs-lock is required to protect mask-refcount and - * mask list. - */ - ASSERT_OVSL(); - BUG_ON(!mask->ref_count); - mask->ref_count--; - - if (!mask->ref_count) { - list_del_rcu(&mask->list); - if (deferred) - kfree_rcu(mask, rcu); - else - kfree(mask); - } - } - if (deferred) call_rcu(&flow->rcu, rcu_free_flow_callback); else @@ -375,7 +357,7 @@ int ovs_flow_tbl_flush(struct flow_table *flow_table) static u32 flow_hash(const struct sw_flow_key *key, int key_start, int key_end) { - u32 *hash_key = (u32 *)((u8 *)key + key_start); + const u32 *hash_key = (const u32 *)((const u8 *)key + key_start); int hash_u32s = (key_end - key_start) >> 2; /* Make sure number of hash bytes are multiple of u32. */ @@ -397,8 +379,8 @@ static bool cmp_key(const struct sw_flow_key *key1, const struct sw_flow_key *key2, int key_start, int key_end) { - const long *cp1 = (long *)((u8 *)key1 + key_start); - const long *cp2 = (long *)((u8 *)key2 + key_start); + const long *cp1 = (const long *)((const u8 *)key1 + key_start); + const long *cp2 = (const long *)((const u8 *)key2 + key_start); long diffs = 0; int i; @@ -490,6 +472,25 @@ static struct table_instance *table_instance_expand(struct table_instance *ti) return table_instance_rehash(ti, ti->n_buckets * 2); } +/* Remove 'mask' from the mask list, if it is not needed any more. */ +static void flow_mask_remove(struct flow_table *tbl, struct sw_flow_mask *mask) +{ + if (mask) { + /* ovs-lock is required to protect mask-refcount and + * mask list. + */ + ASSERT_OVSL(); + BUG_ON(!mask->ref_count); + mask->ref_count--; + + if (!mask->ref_count) { + list_del_rcu(&mask->list); + kfree_rcu(mask, rcu); + } + } +} + +/* Must be called with OVS mutex held. */ void ovs_flow_tbl_remove(struct flow_table *table, struct sw_flow *flow) { struct table_instance *ti = ovsl_dereference(table->ti); @@ -497,6 +498,11 @@ void ovs_flow_tbl_remove(struct flow_table *table, struct sw_flow *flow) BUG_ON(table->count == 0); hlist_del_rcu(&flow->hash_node[ti->node_ver]); table->count--; + + /* RCU delete the mask. 'flow->mask' is not NULLed, as it should be + * accessible as long as the RCU read lock is held. + */ + flow_mask_remove(table, flow->mask); } static struct sw_flow_mask *mask_alloc(void) @@ -513,8 +519,8 @@ static struct sw_flow_mask *mask_alloc(void) static bool mask_equal(const struct sw_flow_mask *a, const struct sw_flow_mask *b) { - u8 *a_ = (u8 *)&a->key + a->range.start; - u8 *b_ = (u8 *)&b->key + b->range.start; + const u8 *a_ = (const u8 *)&a->key + a->range.start; + const u8 *b_ = (const u8 *)&b->key + b->range.start; return (a->range.end == b->range.end) && (a->range.start == b->range.start) @@ -559,6 +565,7 @@ static int flow_mask_insert(struct flow_table *tbl, struct sw_flow *flow, return 0; } +/* Must be called with OVS mutex held. */ int ovs_flow_tbl_insert(struct flow_table *table, struct sw_flow *flow, struct sw_flow_mask *mask) { @@ -597,16 +604,28 @@ int ovs_flow_init(void) BUILD_BUG_ON(__alignof__(struct sw_flow_key) % __alignof__(long)); BUILD_BUG_ON(sizeof(struct sw_flow_key) % sizeof(long)); - flow_cache = kmem_cache_create("sw_flow", sizeof(struct sw_flow), 0, - 0, NULL); + flow_cache = kmem_cache_create("sw_flow", sizeof(struct sw_flow) + + (num_possible_nodes() + * sizeof(struct flow_stats *)), + 0, 0, NULL); if (flow_cache == NULL) return -ENOMEM; + flow_stats_cache + = kmem_cache_create("sw_flow_stats", sizeof(struct flow_stats), + 0, SLAB_HWCACHE_ALIGN, NULL); + if (flow_stats_cache == NULL) { + kmem_cache_destroy(flow_cache); + flow_cache = NULL; + return -ENOMEM; + } + return 0; } /* Uninitializes the flow module. */ void ovs_flow_exit(void) { + kmem_cache_destroy(flow_stats_cache); kmem_cache_destroy(flow_cache); } diff --git a/net/openvswitch/flow_table.h b/net/openvswitch/flow_table.h index baaeb101924..ca8a5820f61 100644 --- a/net/openvswitch/flow_table.h +++ b/net/openvswitch/flow_table.h @@ -52,10 +52,12 @@ struct flow_table { unsigned int count; }; +extern struct kmem_cache *flow_stats_cache; + int ovs_flow_init(void); void ovs_flow_exit(void); -struct sw_flow *ovs_flow_alloc(bool percpu_stats); +struct sw_flow *ovs_flow_alloc(void); void ovs_flow_free(struct sw_flow *, bool deferred); int ovs_flow_tbl_init(struct flow_table *); diff --git a/net/openvswitch/vport-gre.c b/net/openvswitch/vport-gre.c index ebb6e244255..35ec4fed09e 100644 --- a/net/openvswitch/vport-gre.c +++ b/net/openvswitch/vport-gre.c @@ -172,7 +172,7 @@ static int gre_tnl_send(struct vport *vport, struct sk_buff *skb) df = OVS_CB(skb)->tun_key->tun_flags & TUNNEL_DONT_FRAGMENT ? htons(IP_DF) : 0; - skb->local_df = 1; + skb->ignore_df = 1; return iptunnel_xmit(skb->sk, rt, skb, fl.saddr, OVS_CB(skb)->tun_key->ipv4_dst, IPPROTO_GRE, @@ -256,7 +256,7 @@ static void gre_tnl_destroy(struct vport *vport) ovs_net = net_generic(net, ovs_net_id); - rcu_assign_pointer(ovs_net->vport_net.gre_vport, NULL); + RCU_INIT_POINTER(ovs_net->vport_net.gre_vport, NULL); ovs_vport_deferred_free(vport); gre_exit(); } diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c index 729c68763fe..789af9280e7 100644 --- a/net/openvswitch/vport-internal_dev.c +++ b/net/openvswitch/vport-internal_dev.c @@ -130,7 +130,7 @@ static void do_setup(struct net_device *netdev) netdev->priv_flags &= ~IFF_TX_SKB_SHARING; netdev->priv_flags |= IFF_LIVE_ADDR_CHANGE; netdev->destructor = internal_dev_destructor; - SET_ETHTOOL_OPS(netdev, &internal_dev_ethtool_ops); + netdev->ethtool_ops = &internal_dev_ethtool_ops; netdev->tx_queue_len = 0; netdev->features = NETIF_F_LLTX | NETIF_F_SG | NETIF_F_FRAGLIST | diff --git a/net/openvswitch/vport-vxlan.c b/net/openvswitch/vport-vxlan.c index e797a50ac2b..0edbd95c60e 100644 --- a/net/openvswitch/vport-vxlan.c +++ b/net/openvswitch/vport-vxlan.c @@ -122,7 +122,7 @@ static struct vport *vxlan_tnl_create(const struct vport_parms *parms) vxlan_port = vxlan_vport(vport); strncpy(vxlan_port->name, parms->name, IFNAMSIZ); - vs = vxlan_sock_add(net, htons(dst_port), vxlan_rcv, vport, true, false); + vs = vxlan_sock_add(net, htons(dst_port), vxlan_rcv, vport, true, 0); if (IS_ERR(vs)) { ovs_vport_free(vport); return (void *)vs; @@ -170,7 +170,7 @@ static int vxlan_tnl_send(struct vport *vport, struct sk_buff *skb) df = OVS_CB(skb)->tun_key->tun_flags & TUNNEL_DONT_FRAGMENT ? htons(IP_DF) : 0; - skb->local_df = 1; + skb->ignore_df = 1; inet_get_local_port_range(net, &port_min, &port_max); src_port = vxlan_src_port(port_min, port_max, skb); @@ -180,7 +180,8 @@ static int vxlan_tnl_send(struct vport *vport, struct sk_buff *skb) OVS_CB(skb)->tun_key->ipv4_tos, OVS_CB(skb)->tun_key->ipv4_ttl, df, src_port, dst_port, - htonl(be64_to_cpu(OVS_CB(skb)->tun_key->tun_id) << 8)); + htonl(be64_to_cpu(OVS_CB(skb)->tun_key->tun_id) << 8), + false); if (err < 0) ip_rt_put(rt); error: diff --git a/net/openvswitch/vport.h b/net/openvswitch/vport.h index d7e50a17396..8d721e62f38 100644 --- a/net/openvswitch/vport.h +++ b/net/openvswitch/vport.h @@ -172,7 +172,7 @@ void ovs_vport_deferred_free(struct vport *vport); */ static inline void *vport_priv(const struct vport *vport) { - return (u8 *)vport + ALIGN(sizeof(struct vport), VPORT_ALIGN); + return (u8 *)(uintptr_t)vport + ALIGN(sizeof(struct vport), VPORT_ALIGN); } /** @@ -185,9 +185,9 @@ static inline void *vport_priv(const struct vport *vport) * the result of a hash table lookup. @priv must point to the start of the * private data area. */ -static inline struct vport *vport_from_priv(const void *priv) +static inline struct vport *vport_from_priv(void *priv) { - return (struct vport *)(priv - ALIGN(sizeof(struct vport), VPORT_ALIGN)); + return (struct vport *)((u8 *)priv - ALIGN(sizeof(struct vport), VPORT_ALIGN)); } void ovs_vport_receive(struct vport *, struct sk_buff *, diff --git a/net/packet/diag.c b/net/packet/diag.c index 533ce4ff108..92f2c7107ee 100644 --- a/net/packet/diag.c +++ b/net/packet/diag.c @@ -128,6 +128,7 @@ static int pdiag_put_fanout(struct packet_sock *po, struct sk_buff *nlskb) static int sk_diag_fill(struct sock *sk, struct sk_buff *skb, struct packet_diag_req *req, + bool may_report_filterinfo, struct user_namespace *user_ns, u32 portid, u32 seq, u32 flags, int sk_ino) { @@ -172,7 +173,8 @@ static int sk_diag_fill(struct sock *sk, struct sk_buff *skb, goto out_nlmsg_trim; if ((req->pdiag_show & PACKET_SHOW_FILTER) && - sock_diag_put_filterinfo(user_ns, sk, skb, PACKET_DIAG_FILTER)) + sock_diag_put_filterinfo(may_report_filterinfo, sk, skb, + PACKET_DIAG_FILTER)) goto out_nlmsg_trim; return nlmsg_end(skb, nlh); @@ -188,9 +190,11 @@ static int packet_diag_dump(struct sk_buff *skb, struct netlink_callback *cb) struct packet_diag_req *req; struct net *net; struct sock *sk; + bool may_report_filterinfo; net = sock_net(skb->sk); req = nlmsg_data(cb->nlh); + may_report_filterinfo = netlink_net_capable(cb->skb, CAP_NET_ADMIN); mutex_lock(&net->packet.sklist_lock); sk_for_each(sk, &net->packet.sklist) { @@ -200,6 +204,7 @@ static int packet_diag_dump(struct sk_buff *skb, struct netlink_callback *cb) goto next; if (sk_diag_fill(sk, skb, req, + may_report_filterinfo, sk_user_ns(NETLINK_CB(cb->skb).sk), NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, NLM_F_MULTI, diff --git a/net/phonet/pn_netlink.c b/net/phonet/pn_netlink.c index dc15f430080..b64151ade6b 100644 --- a/net/phonet/pn_netlink.c +++ b/net/phonet/pn_netlink.c @@ -70,10 +70,10 @@ static int addr_doit(struct sk_buff *skb, struct nlmsghdr *nlh) int err; u8 pnaddr; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; - if (!capable(CAP_SYS_ADMIN)) + if (!netlink_capable(skb, CAP_SYS_ADMIN)) return -EPERM; ASSERT_RTNL(); @@ -233,10 +233,10 @@ static int route_doit(struct sk_buff *skb, struct nlmsghdr *nlh) int err; u8 dst; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; - if (!capable(CAP_SYS_ADMIN)) + if (!netlink_capable(skb, CAP_SYS_ADMIN)) return -EPERM; ASSERT_RTNL(); diff --git a/net/rds/ib_send.c b/net/rds/ib_send.c index 37be6e226d1..1dde91e3dc7 100644 --- a/net/rds/ib_send.c +++ b/net/rds/ib_send.c @@ -298,7 +298,7 @@ void rds_ib_send_cq_comp_handler(struct ib_cq *cq, void *context) rds_ib_stats_inc(s_ib_tx_cq_event); if (wc.wr_id == RDS_IB_ACK_WR_ID) { - if (ic->i_ack_queued + HZ/2 < jiffies) + if (time_after(jiffies, ic->i_ack_queued + HZ/2)) rds_ib_stats_inc(s_ib_tx_stalled); rds_ib_ack_send_complete(ic); continue; @@ -315,7 +315,7 @@ void rds_ib_send_cq_comp_handler(struct ib_cq *cq, void *context) rm = rds_ib_send_unmap_op(ic, send, wc.status); - if (send->s_queued + HZ/2 < jiffies) + if (time_after(jiffies, send->s_queued + HZ/2)) rds_ib_stats_inc(s_ib_tx_stalled); if (send->s_op) { diff --git a/net/rds/iw_send.c b/net/rds/iw_send.c index e40c3c5db2c..9105ea03aec 100644 --- a/net/rds/iw_send.c +++ b/net/rds/iw_send.c @@ -232,7 +232,7 @@ void rds_iw_send_cq_comp_handler(struct ib_cq *cq, void *context) } if (wc.wr_id == RDS_IW_ACK_WR_ID) { - if (ic->i_ack_queued + HZ/2 < jiffies) + if (time_after(jiffies, ic->i_ack_queued + HZ/2)) rds_iw_stats_inc(s_iw_tx_stalled); rds_iw_ack_send_complete(ic); continue; @@ -267,7 +267,7 @@ void rds_iw_send_cq_comp_handler(struct ib_cq *cq, void *context) send->s_wr.opcode = 0xdead; send->s_wr.num_sge = 1; - if (send->s_queued + HZ/2 < jiffies) + if (time_after(jiffies, send->s_queued + HZ/2)) rds_iw_stats_inc(s_iw_tx_stalled); /* If a RDMA operation produced an error, signal this right diff --git a/net/rds/iw_sysctl.c b/net/rds/iw_sysctl.c index 89c91515ed0..139239d2cb2 100644 --- a/net/rds/iw_sysctl.c +++ b/net/rds/iw_sysctl.c @@ -111,8 +111,7 @@ static struct ctl_table rds_iw_sysctl_table[] = { void rds_iw_sysctl_exit(void) { - if (rds_iw_sysctl_hdr) - unregister_net_sysctl_table(rds_iw_sysctl_hdr); + unregister_net_sysctl_table(rds_iw_sysctl_hdr); } int rds_iw_sysctl_init(void) diff --git a/net/rds/rdma_transport.c b/net/rds/rdma_transport.c index c2be901d19e..6cd9d1deafc 100644 --- a/net/rds/rdma_transport.c +++ b/net/rds/rdma_transport.c @@ -168,7 +168,7 @@ static int rds_rdma_listen_init(void) return ret; } - sin.sin_family = AF_INET, + sin.sin_family = AF_INET; sin.sin_addr.s_addr = (__force u32)htonl(INADDR_ANY); sin.sin_port = (__force u16)htons(RDS_PORT); diff --git a/net/rds/sysctl.c b/net/rds/sysctl.c index b5cb2aa08f3..c3b0cd43eb5 100644 --- a/net/rds/sysctl.c +++ b/net/rds/sysctl.c @@ -94,8 +94,7 @@ static struct ctl_table rds_sysctl_rds_table[] = { void rds_sysctl_exit(void) { - if (rds_sysctl_reg_table) - unregister_net_sysctl_table(rds_sysctl_reg_table); + unregister_net_sysctl_table(rds_sysctl_reg_table); } int rds_sysctl_init(void) diff --git a/net/rds/tcp_listen.c b/net/rds/tcp_listen.c index 4e638f85118..23ab4dcd1d9 100644 --- a/net/rds/tcp_listen.c +++ b/net/rds/tcp_listen.c @@ -153,7 +153,7 @@ int rds_tcp_listen_init(void) sock->sk->sk_data_ready = rds_tcp_listen_data_ready; write_unlock_bh(&sock->sk->sk_callback_lock); - sin.sin_family = PF_INET, + sin.sin_family = PF_INET; sin.sin_addr.s_addr = (__force u32)htonl(INADDR_ANY); sin.sin_port = (__force u16)htons(RDS_TCP_PORT); diff --git a/net/rxrpc/ar-key.c b/net/rxrpc/ar-key.c index 7633a752c65..0ad080790a3 100644 --- a/net/rxrpc/ar-key.c +++ b/net/rxrpc/ar-key.c @@ -99,7 +99,7 @@ static int rxrpc_instantiate_xdr_rxkad(struct key *key, const __be32 *xdr, _debug("tktlen: %x", tktlen); if (tktlen > AFSTOKEN_RK_TIX_MAX) return -EKEYREJECTED; - if (8 * 4 + tktlen != toklen) + if (toklen < 8 * 4 + tktlen) return -EKEYREJECTED; plen = sizeof(*token) + sizeof(*token->kad) + tktlen; diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 8a5ba5add4b..648778aef1a 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -948,7 +948,7 @@ static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n) u32 portid = skb ? NETLINK_CB(skb).portid : 0; int ret = 0, ovr = 0; - if ((n->nlmsg_type != RTM_GETACTION) && !capable(CAP_NET_ADMIN)) + if ((n->nlmsg_type != RTM_GETACTION) && !netlink_capable(skb, CAP_NET_ADMIN)) return -EPERM; ret = nlmsg_parse(n, sizeof(struct tcamsg), tca, TCA_ACT_MAX, NULL); diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 29a30a14c31..45527e6b52d 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -134,7 +134,8 @@ static int tc_ctl_tfilter(struct sk_buff *skb, struct nlmsghdr *n) int err; int tp_created = 0; - if ((n->nlmsg_type != RTM_GETTFILTER) && !capable(CAP_NET_ADMIN)) + if ((n->nlmsg_type != RTM_GETTFILTER) && + !netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) return -EPERM; replay: @@ -317,7 +318,8 @@ replay: } } - err = tp->ops->change(net, skb, tp, cl, t->tcm_handle, tca, &fh); + err = tp->ops->change(net, skb, tp, cl, t->tcm_handle, tca, &fh, + n->nlmsg_flags & NLM_F_CREATE ? TCA_ACT_NOREPLACE : TCA_ACT_REPLACE); if (err == 0) { if (tp_created) { spin_lock_bh(root_lock); @@ -504,7 +506,7 @@ void tcf_exts_destroy(struct tcf_proto *tp, struct tcf_exts *exts) EXPORT_SYMBOL(tcf_exts_destroy); int tcf_exts_validate(struct net *net, struct tcf_proto *tp, struct nlattr **tb, - struct nlattr *rate_tlv, struct tcf_exts *exts) + struct nlattr *rate_tlv, struct tcf_exts *exts, bool ovr) { #ifdef CONFIG_NET_CLS_ACT { @@ -513,7 +515,7 @@ int tcf_exts_validate(struct net *net, struct tcf_proto *tp, struct nlattr **tb, INIT_LIST_HEAD(&exts->actions); if (exts->police && tb[exts->police]) { act = tcf_action_init_1(net, tb[exts->police], rate_tlv, - "police", TCA_ACT_NOREPLACE, + "police", ovr, TCA_ACT_BIND); if (IS_ERR(act)) return PTR_ERR(act); @@ -523,7 +525,7 @@ int tcf_exts_validate(struct net *net, struct tcf_proto *tp, struct nlattr **tb, } else if (exts->action && tb[exts->action]) { int err; err = tcf_action_init(net, tb[exts->action], rate_tlv, - NULL, TCA_ACT_NOREPLACE, + NULL, ovr, TCA_ACT_BIND, &exts->actions); if (err) return err; @@ -543,14 +545,12 @@ void tcf_exts_change(struct tcf_proto *tp, struct tcf_exts *dst, struct tcf_exts *src) { #ifdef CONFIG_NET_CLS_ACT - if (!list_empty(&src->actions)) { - LIST_HEAD(tmp); - tcf_tree_lock(tp); - list_splice_init(&dst->actions, &tmp); - list_splice(&src->actions, &dst->actions); - tcf_tree_unlock(tp); - tcf_action_destroy(&tmp, TCA_ACT_UNBIND); - } + LIST_HEAD(tmp); + tcf_tree_lock(tp); + list_splice_init(&dst->actions, &tmp); + list_splice(&src->actions, &dst->actions); + tcf_tree_unlock(tp); + tcf_action_destroy(&tmp, TCA_ACT_UNBIND); #endif } EXPORT_SYMBOL(tcf_exts_change); diff --git a/net/sched/cls_basic.c b/net/sched/cls_basic.c index e98ca99c202..0ae1813e3e9 100644 --- a/net/sched/cls_basic.c +++ b/net/sched/cls_basic.c @@ -130,14 +130,14 @@ static const struct nla_policy basic_policy[TCA_BASIC_MAX + 1] = { static int basic_set_parms(struct net *net, struct tcf_proto *tp, struct basic_filter *f, unsigned long base, struct nlattr **tb, - struct nlattr *est) + struct nlattr *est, bool ovr) { int err; struct tcf_exts e; struct tcf_ematch_tree t; tcf_exts_init(&e, TCA_BASIC_ACT, TCA_BASIC_POLICE); - err = tcf_exts_validate(net, tp, tb, est, &e); + err = tcf_exts_validate(net, tp, tb, est, &e, ovr); if (err < 0) return err; @@ -161,7 +161,7 @@ errout: static int basic_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, - struct nlattr **tca, unsigned long *arg) + struct nlattr **tca, unsigned long *arg, bool ovr) { int err; struct basic_head *head = tp->root; @@ -179,7 +179,7 @@ static int basic_change(struct net *net, struct sk_buff *in_skb, if (f != NULL) { if (handle && f->handle != handle) return -EINVAL; - return basic_set_parms(net, tp, f, base, tb, tca[TCA_RATE]); + return basic_set_parms(net, tp, f, base, tb, tca[TCA_RATE], ovr); } err = -ENOBUFS; @@ -206,7 +206,7 @@ static int basic_change(struct net *net, struct sk_buff *in_skb, f->handle = head->hgenerator; } - err = basic_set_parms(net, tp, f, base, tb, tca[TCA_RATE]); + err = basic_set_parms(net, tp, f, base, tb, tca[TCA_RATE], ovr); if (err < 0) goto errout; diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c index 8e3cf49118e..13f64df2c71 100644 --- a/net/sched/cls_bpf.c +++ b/net/sched/cls_bpf.c @@ -156,11 +156,11 @@ static void cls_bpf_put(struct tcf_proto *tp, unsigned long f) static int cls_bpf_modify_existing(struct net *net, struct tcf_proto *tp, struct cls_bpf_prog *prog, unsigned long base, struct nlattr **tb, - struct nlattr *est) + struct nlattr *est, bool ovr) { struct sock_filter *bpf_ops, *bpf_old; struct tcf_exts exts; - struct sock_fprog tmp; + struct sock_fprog_kern tmp; struct sk_filter *fp, *fp_old; u16 bpf_size, bpf_len; u32 classid; @@ -170,7 +170,7 @@ static int cls_bpf_modify_existing(struct net *net, struct tcf_proto *tp, return -EINVAL; tcf_exts_init(&exts, TCA_BPF_ACT, TCA_BPF_POLICE); - ret = tcf_exts_validate(net, tp, tb, est, &exts); + ret = tcf_exts_validate(net, tp, tb, est, &exts, ovr); if (ret < 0) return ret; @@ -191,7 +191,7 @@ static int cls_bpf_modify_existing(struct net *net, struct tcf_proto *tp, memcpy(bpf_ops, nla_data(tb[TCA_BPF_OPS]), bpf_size); tmp.len = bpf_len; - tmp.filter = (struct sock_filter __user *) bpf_ops; + tmp.filter = bpf_ops; ret = sk_unattached_filter_create(&fp, &tmp); if (ret) @@ -242,7 +242,7 @@ static u32 cls_bpf_grab_new_handle(struct tcf_proto *tp, static int cls_bpf_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, struct nlattr **tca, - unsigned long *arg) + unsigned long *arg, bool ovr) { struct cls_bpf_head *head = tp->root; struct cls_bpf_prog *prog = (struct cls_bpf_prog *) *arg; @@ -260,7 +260,7 @@ static int cls_bpf_change(struct net *net, struct sk_buff *in_skb, if (handle && prog->handle != handle) return -EINVAL; return cls_bpf_modify_existing(net, tp, prog, base, tb, - tca[TCA_RATE]); + tca[TCA_RATE], ovr); } prog = kzalloc(sizeof(*prog), GFP_KERNEL); @@ -277,7 +277,7 @@ static int cls_bpf_change(struct net *net, struct sk_buff *in_skb, goto errout; } - ret = cls_bpf_modify_existing(net, tp, prog, base, tb, tca[TCA_RATE]); + ret = cls_bpf_modify_existing(net, tp, prog, base, tb, tca[TCA_RATE], ovr); if (ret < 0) goto errout; diff --git a/net/sched/cls_cgroup.c b/net/sched/cls_cgroup.c index 8e2158ab551..cacf01bd04f 100644 --- a/net/sched/cls_cgroup.c +++ b/net/sched/cls_cgroup.c @@ -83,7 +83,7 @@ static const struct nla_policy cgroup_policy[TCA_CGROUP_MAX + 1] = { static int cls_cgroup_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, struct nlattr **tca, - unsigned long *arg) + unsigned long *arg, bool ovr) { struct nlattr *tb[TCA_CGROUP_MAX + 1]; struct cls_cgroup_head *head = tp->root; @@ -119,7 +119,7 @@ static int cls_cgroup_change(struct net *net, struct sk_buff *in_skb, return err; tcf_exts_init(&e, TCA_CGROUP_ACT, TCA_CGROUP_POLICE); - err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &e); + err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &e, ovr); if (err < 0) return err; diff --git a/net/sched/cls_flow.c b/net/sched/cls_flow.c index 257029c5433..35be16f7c19 100644 --- a/net/sched/cls_flow.c +++ b/net/sched/cls_flow.c @@ -349,7 +349,7 @@ static const struct nla_policy flow_policy[TCA_FLOW_MAX + 1] = { static int flow_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, struct nlattr **tca, - unsigned long *arg) + unsigned long *arg, bool ovr) { struct flow_head *head = tp->root; struct flow_filter *f; @@ -393,7 +393,7 @@ static int flow_change(struct net *net, struct sk_buff *in_skb, } tcf_exts_init(&e, TCA_FLOW_ACT, TCA_FLOW_POLICE); - err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &e); + err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &e, ovr); if (err < 0) return err; diff --git a/net/sched/cls_fw.c b/net/sched/cls_fw.c index 63a3ce75c02..861b03ccfed 100644 --- a/net/sched/cls_fw.c +++ b/net/sched/cls_fw.c @@ -169,7 +169,7 @@ static const struct nla_policy fw_policy[TCA_FW_MAX + 1] = { static int fw_change_attrs(struct net *net, struct tcf_proto *tp, struct fw_filter *f, - struct nlattr **tb, struct nlattr **tca, unsigned long base) + struct nlattr **tb, struct nlattr **tca, unsigned long base, bool ovr) { struct fw_head *head = tp->root; struct tcf_exts e; @@ -177,7 +177,7 @@ fw_change_attrs(struct net *net, struct tcf_proto *tp, struct fw_filter *f, int err; tcf_exts_init(&e, TCA_FW_ACT, TCA_FW_POLICE); - err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &e); + err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &e, ovr); if (err < 0) return err; @@ -218,7 +218,7 @@ static int fw_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, struct nlattr **tca, - unsigned long *arg) + unsigned long *arg, bool ovr) { struct fw_head *head = tp->root; struct fw_filter *f = (struct fw_filter *) *arg; @@ -236,7 +236,7 @@ static int fw_change(struct net *net, struct sk_buff *in_skb, if (f != NULL) { if (f->id != handle && handle) return -EINVAL; - return fw_change_attrs(net, tp, f, tb, tca, base); + return fw_change_attrs(net, tp, f, tb, tca, base, ovr); } if (!handle) @@ -264,7 +264,7 @@ static int fw_change(struct net *net, struct sk_buff *in_skb, tcf_exts_init(&f->exts, TCA_FW_ACT, TCA_FW_POLICE); f->id = handle; - err = fw_change_attrs(net, tp, f, tb, tca, base); + err = fw_change_attrs(net, tp, f, tb, tca, base, ovr); if (err < 0) goto errout; diff --git a/net/sched/cls_route.c b/net/sched/cls_route.c index 1ad3068f2ce..dd9fc2523c7 100644 --- a/net/sched/cls_route.c +++ b/net/sched/cls_route.c @@ -333,7 +333,8 @@ static const struct nla_policy route4_policy[TCA_ROUTE4_MAX + 1] = { static int route4_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, struct route4_filter *f, u32 handle, struct route4_head *head, - struct nlattr **tb, struct nlattr *est, int new) + struct nlattr **tb, struct nlattr *est, int new, + bool ovr) { int err; u32 id = 0, to = 0, nhandle = 0x8000; @@ -343,7 +344,7 @@ static int route4_set_parms(struct net *net, struct tcf_proto *tp, struct tcf_exts e; tcf_exts_init(&e, TCA_ROUTE4_ACT, TCA_ROUTE4_POLICE); - err = tcf_exts_validate(net, tp, tb, est, &e); + err = tcf_exts_validate(net, tp, tb, est, &e, ovr); if (err < 0) return err; @@ -428,7 +429,7 @@ static int route4_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, struct nlattr **tca, - unsigned long *arg) + unsigned long *arg, bool ovr) { struct route4_head *head = tp->root; struct route4_filter *f, *f1, **fp; @@ -455,7 +456,7 @@ static int route4_change(struct net *net, struct sk_buff *in_skb, old_handle = f->handle; err = route4_set_parms(net, tp, base, f, handle, head, tb, - tca[TCA_RATE], 0); + tca[TCA_RATE], 0, ovr); if (err < 0) return err; @@ -479,7 +480,7 @@ static int route4_change(struct net *net, struct sk_buff *in_skb, tcf_exts_init(&f->exts, TCA_ROUTE4_ACT, TCA_ROUTE4_POLICE); err = route4_set_parms(net, tp, base, f, handle, head, tb, - tca[TCA_RATE], 1); + tca[TCA_RATE], 1, ovr); if (err < 0) goto errout; diff --git a/net/sched/cls_rsvp.h b/net/sched/cls_rsvp.h index 19f8e5dfa8b..1020e233a5d 100644 --- a/net/sched/cls_rsvp.h +++ b/net/sched/cls_rsvp.h @@ -415,7 +415,7 @@ static int rsvp_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, struct nlattr **tca, - unsigned long *arg) + unsigned long *arg, bool ovr) { struct rsvp_head *data = tp->root; struct rsvp_filter *f, **fp; @@ -436,7 +436,7 @@ static int rsvp_change(struct net *net, struct sk_buff *in_skb, return err; tcf_exts_init(&e, TCA_RSVP_ACT, TCA_RSVP_POLICE); - err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &e); + err = tcf_exts_validate(net, tp, tb, tca[TCA_RATE], &e, ovr); if (err < 0) return err; diff --git a/net/sched/cls_tcindex.c b/net/sched/cls_tcindex.c index eed8404443d..c721cd4a469 100644 --- a/net/sched/cls_tcindex.c +++ b/net/sched/cls_tcindex.c @@ -188,11 +188,17 @@ static const struct nla_policy tcindex_policy[TCA_TCINDEX_MAX + 1] = { [TCA_TCINDEX_CLASSID] = { .type = NLA_U32 }, }; +static void tcindex_filter_result_init(struct tcindex_filter_result *r) +{ + memset(r, 0, sizeof(*r)); + tcf_exts_init(&r->exts, TCA_TCINDEX_ACT, TCA_TCINDEX_POLICE); +} + static int tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, u32 handle, struct tcindex_data *p, struct tcindex_filter_result *r, struct nlattr **tb, - struct nlattr *est) + struct nlattr *est, bool ovr) { int err, balloc = 0; struct tcindex_filter_result new_filter_result, *old_r = r; @@ -202,20 +208,16 @@ tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, struct tcf_exts e; tcf_exts_init(&e, TCA_TCINDEX_ACT, TCA_TCINDEX_POLICE); - err = tcf_exts_validate(net, tp, tb, est, &e); + err = tcf_exts_validate(net, tp, tb, est, &e, ovr); if (err < 0) return err; memcpy(&cp, p, sizeof(cp)); - memset(&new_filter_result, 0, sizeof(new_filter_result)); - tcf_exts_init(&new_filter_result.exts, TCA_TCINDEX_ACT, TCA_TCINDEX_POLICE); + tcindex_filter_result_init(&new_filter_result); + tcindex_filter_result_init(&cr); if (old_r) - memcpy(&cr, r, sizeof(cr)); - else { - memset(&cr, 0, sizeof(cr)); - tcf_exts_init(&cr.exts, TCA_TCINDEX_ACT, TCA_TCINDEX_POLICE); - } + cr.res = r->res; if (tb[TCA_TCINDEX_HASH]) cp.hash = nla_get_u32(tb[TCA_TCINDEX_HASH]); @@ -267,9 +269,14 @@ tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, err = -ENOMEM; if (!cp.perfect && !cp.h) { if (valid_perfect_hash(&cp)) { + int i; + cp.perfect = kcalloc(cp.hash, sizeof(*r), GFP_KERNEL); if (!cp.perfect) goto errout; + for (i = 0; i < cp.hash; i++) + tcf_exts_init(&cp.perfect[i].exts, TCA_TCINDEX_ACT, + TCA_TCINDEX_POLICE); balloc = 1; } else { cp.h = kcalloc(cp.hash, sizeof(f), GFP_KERNEL); @@ -295,14 +302,17 @@ tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, tcf_bind_filter(tp, &cr.res, base); } - tcf_exts_change(tp, &cr.exts, &e); + if (old_r) + tcf_exts_change(tp, &r->exts, &e); + else + tcf_exts_change(tp, &cr.exts, &e); tcf_tree_lock(tp); if (old_r && old_r != r) - memset(old_r, 0, sizeof(*old_r)); + tcindex_filter_result_init(old_r); memcpy(p, &cp, sizeof(cp)); - memcpy(r, &cr, sizeof(cr)); + r->res = cr.res; if (r == &new_filter_result) { struct tcindex_filter **fp; @@ -331,7 +341,7 @@ errout: static int tcindex_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, - struct nlattr **tca, unsigned long *arg) + struct nlattr **tca, unsigned long *arg, bool ovr) { struct nlattr *opt = tca[TCA_OPTIONS]; struct nlattr *tb[TCA_TCINDEX_MAX + 1]; @@ -351,7 +361,7 @@ tcindex_change(struct net *net, struct sk_buff *in_skb, return err; return tcindex_set_parms(net, tp, base, handle, p, r, tb, - tca[TCA_RATE]); + tca[TCA_RATE], ovr); } diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c index 84c28daff84..c39b583ace3 100644 --- a/net/sched/cls_u32.c +++ b/net/sched/cls_u32.c @@ -486,13 +486,13 @@ static const struct nla_policy u32_policy[TCA_U32_MAX + 1] = { static int u32_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, struct tc_u_hnode *ht, struct tc_u_knode *n, struct nlattr **tb, - struct nlattr *est) + struct nlattr *est, bool ovr) { int err; struct tcf_exts e; tcf_exts_init(&e, TCA_U32_ACT, TCA_U32_POLICE); - err = tcf_exts_validate(net, tp, tb, est, &e); + err = tcf_exts_validate(net, tp, tb, est, &e, ovr); if (err < 0) return err; @@ -545,7 +545,7 @@ errout: static int u32_change(struct net *net, struct sk_buff *in_skb, struct tcf_proto *tp, unsigned long base, u32 handle, struct nlattr **tca, - unsigned long *arg) + unsigned long *arg, bool ovr) { struct tc_u_common *tp_c = tp->data; struct tc_u_hnode *ht; @@ -569,7 +569,7 @@ static int u32_change(struct net *net, struct sk_buff *in_skb, return -EINVAL; return u32_set_parms(net, tp, base, n->ht_up, n, tb, - tca[TCA_RATE]); + tca[TCA_RATE], ovr); } if (tb[TCA_U32_DIVISOR]) { @@ -656,7 +656,7 @@ static int u32_change(struct net *net, struct sk_buff *in_skb, } #endif - err = u32_set_parms(net, tp, base, ht, n, tb, tca[TCA_RATE]); + err = u32_set_parms(net, tp, base, ht, n, tb, tca[TCA_RATE], ovr); if (err == 0) { struct tc_u_knode **ins; for (ins = &ht->ht[TC_U32_HASH(handle)]; *ins; ins = &(*ins)->next) diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index a0b84e0e22d..fd14df56e5f 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1084,7 +1084,8 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n) struct Qdisc *p = NULL; int err; - if ((n->nlmsg_type != RTM_GETQDISC) && !capable(CAP_NET_ADMIN)) + if ((n->nlmsg_type != RTM_GETQDISC) && + !netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) return -EPERM; err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, NULL); @@ -1151,7 +1152,7 @@ static int tc_modify_qdisc(struct sk_buff *skb, struct nlmsghdr *n) struct Qdisc *q, *p; int err; - if (!capable(CAP_NET_ADMIN)) + if (!netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) return -EPERM; replay: @@ -1490,7 +1491,8 @@ static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n) u32 qid; int err; - if ((n->nlmsg_type != RTM_GETTCLASS) && !capable(CAP_NET_ADMIN)) + if ((n->nlmsg_type != RTM_GETTCLASS) && + !netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN)) return -EPERM; err = nlmsg_parse(n, sizeof(*tcm), tca, TCA_MAX, NULL); diff --git a/net/sched/sch_choke.c b/net/sched/sch_choke.c index 2aee02802c2..ed30e436128 100644 --- a/net/sched/sch_choke.c +++ b/net/sched/sch_choke.c @@ -391,12 +391,7 @@ static const struct nla_policy choke_policy[TCA_CHOKE_MAX + 1] = { static void choke_free(void *addr) { - if (addr) { - if (is_vmalloc_addr(addr)) - vfree(addr); - else - kfree(addr); - } + kvfree(addr); } static int choke_change(struct Qdisc *sch, struct nlattr *opt) diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c index 23c682b42f9..ba32c2b005d 100644 --- a/net/sched/sch_fq.c +++ b/net/sched/sch_fq.c @@ -591,10 +591,7 @@ static void *fq_alloc_node(size_t sz, int node) static void fq_free(void *addr) { - if (addr && is_vmalloc_addr(addr)) - vfree(addr); - else - kfree(addr); + kvfree(addr); } static int fq_resize(struct Qdisc *sch, u32 log) diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index 0bf432c782c..063b726bf1f 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -365,12 +365,7 @@ static void *fq_codel_zalloc(size_t sz) static void fq_codel_free(void *addr) { - if (addr) { - if (is_vmalloc_addr(addr)) - vfree(addr); - else - kfree(addr); - } + kvfree(addr); } static void fq_codel_destroy(struct Qdisc *sch) diff --git a/net/sched/sch_hhf.c b/net/sched/sch_hhf.c index edee03d922e..d85b6812a7d 100644 --- a/net/sched/sch_hhf.c +++ b/net/sched/sch_hhf.c @@ -414,7 +414,7 @@ static int hhf_enqueue(struct sk_buff *skb, struct Qdisc *sch) } bucket->deficit = weight * q->quantum; } - if (++sch->q.qlen < sch->limit) + if (++sch->q.qlen <= sch->limit) return NET_XMIT_SUCCESS; q->drop_overlimit++; @@ -494,12 +494,7 @@ static void *hhf_zalloc(size_t sz) static void hhf_free(void *addr) { - if (addr) { - if (is_vmalloc_addr(addr)) - vfree(addr); - else - kfree(addr); - } + kvfree(addr); } static void hhf_destroy(struct Qdisc *sch) @@ -553,11 +548,6 @@ static int hhf_change(struct Qdisc *sch, struct nlattr *opt) if (err < 0) return err; - sch_tree_lock(sch); - - if (tb[TCA_HHF_BACKLOG_LIMIT]) - sch->limit = nla_get_u32(tb[TCA_HHF_BACKLOG_LIMIT]); - if (tb[TCA_HHF_QUANTUM]) new_quantum = nla_get_u32(tb[TCA_HHF_QUANTUM]); @@ -567,6 +557,12 @@ static int hhf_change(struct Qdisc *sch, struct nlattr *opt) non_hh_quantum = (u64)new_quantum * new_hhf_non_hh_weight; if (non_hh_quantum > INT_MAX) return -EINVAL; + + sch_tree_lock(sch); + + if (tb[TCA_HHF_BACKLOG_LIMIT]) + sch->limit = nla_get_u32(tb[TCA_HHF_BACKLOG_LIMIT]); + q->quantum = new_quantum; q->hhf_non_hh_weight = new_hhf_non_hh_weight; diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index f1669a00f57..111d70fddae 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -648,12 +648,7 @@ static void netem_reset(struct Qdisc *sch) static void dist_free(struct disttable *d) { - if (d) { - if (is_vmalloc_addr(d)) - vfree(d); - else - kfree(d); - } + kvfree(d); } /* diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index 87317ff0b4e..1af2f73906d 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -716,12 +716,7 @@ static void *sfq_alloc(size_t sz) static void sfq_free(void *addr) { - if (addr) { - if (is_vmalloc_addr(addr)) - vfree(addr); - else - kfree(addr); - } + kvfree(addr); } static void sfq_destroy(struct Qdisc *sch) diff --git a/net/sctp/auth.c b/net/sctp/auth.c index 683c7d1b130..0e8529113dc 100644 --- a/net/sctp/auth.c +++ b/net/sctp/auth.c @@ -386,14 +386,13 @@ nomem: */ int sctp_auth_asoc_init_active_key(struct sctp_association *asoc, gfp_t gfp) { - struct net *net = sock_net(asoc->base.sk); struct sctp_auth_bytes *secret; struct sctp_shared_key *ep_key; /* If we don't support AUTH, or peer is not capable * we don't need to do anything. */ - if (!net->sctp.auth_enable || !asoc->peer.auth_capable) + if (!asoc->ep->auth_enable || !asoc->peer.auth_capable) return 0; /* If the key_id is non-zero and we couldn't find an @@ -440,16 +439,16 @@ struct sctp_shared_key *sctp_auth_get_shkey( */ int sctp_auth_init_hmacs(struct sctp_endpoint *ep, gfp_t gfp) { - struct net *net = sock_net(ep->base.sk); struct crypto_hash *tfm = NULL; __u16 id; - /* if the transforms are already allocted, we are done */ - if (!net->sctp.auth_enable) { + /* If AUTH extension is disabled, we are done */ + if (!ep->auth_enable) { ep->auth_hmacs = NULL; return 0; } + /* If the transforms are already allocated, we are done */ if (ep->auth_hmacs) return 0; @@ -665,12 +664,10 @@ static int __sctp_auth_cid(sctp_cid_t chunk, struct sctp_chunks_param *param) /* Check if peer requested that this chunk is authenticated */ int sctp_auth_send_cid(sctp_cid_t chunk, const struct sctp_association *asoc) { - struct net *net; if (!asoc) return 0; - net = sock_net(asoc->base.sk); - if (!net->sctp.auth_enable || !asoc->peer.auth_capable) + if (!asoc->ep->auth_enable || !asoc->peer.auth_capable) return 0; return __sctp_auth_cid(chunk, asoc->peer.peer_chunks); @@ -679,12 +676,10 @@ int sctp_auth_send_cid(sctp_cid_t chunk, const struct sctp_association *asoc) /* Check if we requested that peer authenticate this chunk. */ int sctp_auth_recv_cid(sctp_cid_t chunk, const struct sctp_association *asoc) { - struct net *net; if (!asoc) return 0; - net = sock_net(asoc->base.sk); - if (!net->sctp.auth_enable) + if (!asoc->ep->auth_enable) return 0; return __sctp_auth_cid(chunk, diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c index 8e5fdea0521..3d9f429858d 100644 --- a/net/sctp/endpointola.c +++ b/net/sctp/endpointola.c @@ -68,7 +68,8 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, if (!ep->digest) return NULL; - if (net->sctp.auth_enable) { + ep->auth_enable = net->sctp.auth_enable; + if (ep->auth_enable) { /* Allocate space for HMACS and CHUNKS authentication * variables. There are arrays that we encode directly * into parameters to make the rest of the operations easier. diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 2b1738ef939..1999592ba88 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -216,7 +216,7 @@ static int sctp_v6_xmit(struct sk_buff *skb, struct sctp_transport *transport) IP6_ECN_flow_xmit(sk, fl6->flowlabel); if (!(transport->param_flags & SPP_PMTUD_ENABLE)) - skb->local_df = 1; + skb->ignore_df = 1; SCTP_INC_STATS(sock_net(sk), SCTP_MIB_OUTSCTPPACKS); @@ -943,7 +943,6 @@ static struct inet_protosw sctpv6_seqpacket_protosw = { .protocol = IPPROTO_SCTP, .prot = &sctpv6_prot, .ops = &inet6_seqpacket_ops, - .no_check = 0, .flags = SCTP_PROTOSW_FLAG }; static struct inet_protosw sctpv6_stream_protosw = { @@ -951,7 +950,6 @@ static struct inet_protosw sctpv6_stream_protosw = { .protocol = IPPROTO_SCTP, .prot = &sctpv6_prot, .ops = &inet6_seqpacket_ops, - .no_check = 0, .flags = SCTP_PROTOSW_FLAG, }; diff --git a/net/sctp/output.c b/net/sctp/output.c index 0f4d15fc262..01ab8e0723f 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -591,7 +591,7 @@ int sctp_packet_transmit(struct sctp_packet *packet) pr_debug("***sctp_transmit_packet*** skb->len:%d\n", nskb->len); - nskb->local_df = packet->ipfragok; + nskb->ignore_df = packet->ipfragok; tp->af_specific->sctp_xmit(nskb, tp); out: diff --git a/net/sctp/proc.c b/net/sctp/proc.c index 0947f1e15eb..34229ee7f37 100644 --- a/net/sctp/proc.c +++ b/net/sctp/proc.c @@ -78,7 +78,7 @@ static int sctp_snmp_seq_show(struct seq_file *seq, void *v) for (i = 0; sctp_snmp_list[i].name != NULL; i++) seq_printf(seq, "%-32s\t%ld\n", sctp_snmp_list[i].name, - snmp_fold_field((void __percpu **)net->sctp.sctp_statistics, + snmp_fold_field(net->sctp.sctp_statistics, sctp_snmp_list[i].entry)); return 0; diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index c09757fbf80..6789d785e69 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -491,8 +491,13 @@ static void sctp_v4_get_dst(struct sctp_transport *t, union sctp_addr *saddr, continue; if ((laddr->state == SCTP_ADDR_SRC) && (AF_INET == laddr->a.sa.sa_family)) { - fl4->saddr = laddr->a.v4.sin_addr.s_addr; fl4->fl4_sport = laddr->a.v4.sin_port; + flowi4_update_output(fl4, + asoc->base.sk->sk_bound_dev_if, + RT_CONN_FLAGS(asoc->base.sk), + daddr->v4.sin_addr.s_addr, + laddr->a.v4.sin_addr.s_addr); + rt = ip_route_output_key(sock_net(sk), fl4); if (!IS_ERR(rt)) { dst = &rt->dst; @@ -1012,7 +1017,6 @@ static struct inet_protosw sctp_seqpacket_protosw = { .protocol = IPPROTO_SCTP, .prot = &sctp_prot, .ops = &inet_seqpacket_ops, - .no_check = 0, .flags = SCTP_PROTOSW_FLAG }; static struct inet_protosw sctp_stream_protosw = { @@ -1020,7 +1024,6 @@ static struct inet_protosw sctp_stream_protosw = { .protocol = IPPROTO_SCTP, .prot = &sctp_prot, .ops = &inet_seqpacket_ops, - .no_check = 0, .flags = SCTP_PROTOSW_FLAG }; @@ -1100,14 +1103,15 @@ int sctp_register_pf(struct sctp_pf *pf, sa_family_t family) static inline int init_sctp_mibs(struct net *net) { - return snmp_mib_init((void __percpu **)net->sctp.sctp_statistics, - sizeof(struct sctp_mib), - __alignof__(struct sctp_mib)); + net->sctp.sctp_statistics = alloc_percpu(struct sctp_mib); + if (!net->sctp.sctp_statistics) + return -ENOMEM; + return 0; } static inline void cleanup_sctp_mibs(struct net *net) { - snmp_mib_free((void __percpu **)net->sctp.sctp_statistics); + free_percpu(net->sctp.sctp_statistics); } static void sctp_v4_pf_init(void) diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 3a1767ef320..fee5552ddf9 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -219,6 +219,7 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, gfp_t gfp, int vparam_len) { struct net *net = sock_net(asoc->base.sk); + struct sctp_endpoint *ep = asoc->ep; sctp_inithdr_t init; union sctp_params addrs; size_t chunksize; @@ -278,7 +279,7 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, chunksize += vparam_len; /* Account for AUTH related parameters */ - if (net->sctp.auth_enable) { + if (ep->auth_enable) { /* Add random parameter length*/ chunksize += sizeof(asoc->c.auth_random); @@ -363,7 +364,7 @@ struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, } /* Add SCTP-AUTH chunks to the parameter list */ - if (net->sctp.auth_enable) { + if (ep->auth_enable) { sctp_addto_chunk(retval, sizeof(asoc->c.auth_random), asoc->c.auth_random); if (auth_hmacs) @@ -2010,7 +2011,7 @@ static void sctp_process_ext_param(struct sctp_association *asoc, /* if the peer reports AUTH, assume that he * supports AUTH. */ - if (net->sctp.auth_enable) + if (asoc->ep->auth_enable) asoc->peer.auth_capable = 1; break; case SCTP_CID_ASCONF: @@ -2102,6 +2103,7 @@ static sctp_ierror_t sctp_process_unk_param(const struct sctp_association *asoc, * SCTP_IERROR_NO_ERROR - continue with the chunk */ static sctp_ierror_t sctp_verify_param(struct net *net, + const struct sctp_endpoint *ep, const struct sctp_association *asoc, union sctp_params param, sctp_cid_t cid, @@ -2152,7 +2154,7 @@ static sctp_ierror_t sctp_verify_param(struct net *net, goto fallthrough; case SCTP_PARAM_RANDOM: - if (!net->sctp.auth_enable) + if (!ep->auth_enable) goto fallthrough; /* SCTP-AUTH: Secion 6.1 @@ -2169,7 +2171,7 @@ static sctp_ierror_t sctp_verify_param(struct net *net, break; case SCTP_PARAM_CHUNKS: - if (!net->sctp.auth_enable) + if (!ep->auth_enable) goto fallthrough; /* SCTP-AUTH: Section 3.2 @@ -2185,7 +2187,7 @@ static sctp_ierror_t sctp_verify_param(struct net *net, break; case SCTP_PARAM_HMAC_ALGO: - if (!net->sctp.auth_enable) + if (!ep->auth_enable) goto fallthrough; hmacs = (struct sctp_hmac_algo_param *)param.p; @@ -2220,10 +2222,9 @@ fallthrough: } /* Verify the INIT packet before we process it. */ -int sctp_verify_init(struct net *net, const struct sctp_association *asoc, - sctp_cid_t cid, - sctp_init_chunk_t *peer_init, - struct sctp_chunk *chunk, +int sctp_verify_init(struct net *net, const struct sctp_endpoint *ep, + const struct sctp_association *asoc, sctp_cid_t cid, + sctp_init_chunk_t *peer_init, struct sctp_chunk *chunk, struct sctp_chunk **errp) { union sctp_params param; @@ -2264,8 +2265,8 @@ int sctp_verify_init(struct net *net, const struct sctp_association *asoc, /* Verify all the variable length parameters */ sctp_walk_params(param, peer_init, init_hdr.params) { - - result = sctp_verify_param(net, asoc, param, cid, chunk, errp); + result = sctp_verify_param(net, ep, asoc, param, cid, + chunk, errp); switch (result) { case SCTP_IERROR_ABORT: case SCTP_IERROR_NOMEM: @@ -2497,6 +2498,7 @@ static int sctp_process_param(struct sctp_association *asoc, struct sctp_af *af; union sctp_addr_param *addr_param; struct sctp_transport *t; + struct sctp_endpoint *ep = asoc->ep; /* We maintain all INIT parameters in network byte order all the * time. This allows us to not worry about whether the parameters @@ -2636,7 +2638,7 @@ do_addr_param: goto fall_through; case SCTP_PARAM_RANDOM: - if (!net->sctp.auth_enable) + if (!ep->auth_enable) goto fall_through; /* Save peer's random parameter */ @@ -2649,7 +2651,7 @@ do_addr_param: break; case SCTP_PARAM_HMAC_ALGO: - if (!net->sctp.auth_enable) + if (!ep->auth_enable) goto fall_through; /* Save peer's HMAC list */ @@ -2665,7 +2667,7 @@ do_addr_param: break; case SCTP_PARAM_CHUNKS: - if (!net->sctp.auth_enable) + if (!ep->auth_enable) goto fall_through; asoc->peer.peer_chunks = kmemdup(param.p, diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 5d6883ff00c..fef2acdf4a2 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -496,11 +496,10 @@ static void sctp_do_8_2_transport_strike(sctp_cmd_seq_t *commands, /* If the transport error count is greater than the pf_retrans * threshold, and less than pathmaxrtx, and if the current state - * is not SCTP_UNCONFIRMED, then mark this transport as Partially - * Failed, see SCTP Quick Failover Draft, section 5.1 + * is SCTP_ACTIVE, then mark this transport as Partially Failed, + * see SCTP Quick Failover Draft, section 5.1 */ - if ((transport->state != SCTP_PF) && - (transport->state != SCTP_UNCONFIRMED) && + if ((transport->state == SCTP_ACTIVE) && (asoc->pf_retrans < transport->pathmaxrxt) && (transport->error_count > asoc->pf_retrans)) { diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index ae9fbeba40b..5170a1ff95a 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -357,7 +357,7 @@ sctp_disposition_t sctp_sf_do_5_1B_init(struct net *net, /* Verify the INIT chunk before processing it. */ err_chunk = NULL; - if (!sctp_verify_init(net, asoc, chunk->chunk_hdr->type, + if (!sctp_verify_init(net, ep, asoc, chunk->chunk_hdr->type, (sctp_init_chunk_t *)chunk->chunk_hdr, chunk, &err_chunk)) { /* This chunk contains fatal error. It is to be discarded. @@ -524,7 +524,7 @@ sctp_disposition_t sctp_sf_do_5_1C_ack(struct net *net, /* Verify the INIT chunk before processing it. */ err_chunk = NULL; - if (!sctp_verify_init(net, asoc, chunk->chunk_hdr->type, + if (!sctp_verify_init(net, ep, asoc, chunk->chunk_hdr->type, (sctp_init_chunk_t *)chunk->chunk_hdr, chunk, &err_chunk)) { @@ -1430,7 +1430,7 @@ static sctp_disposition_t sctp_sf_do_unexpected_init( /* Verify the INIT chunk before processing it. */ err_chunk = NULL; - if (!sctp_verify_init(net, asoc, chunk->chunk_hdr->type, + if (!sctp_verify_init(net, ep, asoc, chunk->chunk_hdr->type, (sctp_init_chunk_t *)chunk->chunk_hdr, chunk, &err_chunk)) { /* This chunk contains fatal error. It is to be discarded. diff --git a/net/sctp/socket.c b/net/sctp/socket.c index ff20e2dbbbc..42989968940 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -71,6 +71,7 @@ #include <net/route.h> #include <net/ipv6.h> #include <net/inet_common.h> +#include <net/busy_poll.h> #include <linux/socket.h> /* for sa_family_t */ #include <linux/export.h> @@ -3321,10 +3322,10 @@ static int sctp_setsockopt_auth_chunk(struct sock *sk, char __user *optval, unsigned int optlen) { - struct net *net = sock_net(sk); + struct sctp_endpoint *ep = sctp_sk(sk)->ep; struct sctp_authchunk val; - if (!net->sctp.auth_enable) + if (!ep->auth_enable) return -EACCES; if (optlen != sizeof(struct sctp_authchunk)) @@ -3341,7 +3342,7 @@ static int sctp_setsockopt_auth_chunk(struct sock *sk, } /* add this chunk id to the endpoint */ - return sctp_auth_ep_add_chunkid(sctp_sk(sk)->ep, val.sauth_chunk); + return sctp_auth_ep_add_chunkid(ep, val.sauth_chunk); } /* @@ -3354,12 +3355,12 @@ static int sctp_setsockopt_hmac_ident(struct sock *sk, char __user *optval, unsigned int optlen) { - struct net *net = sock_net(sk); + struct sctp_endpoint *ep = sctp_sk(sk)->ep; struct sctp_hmacalgo *hmacs; u32 idents; int err; - if (!net->sctp.auth_enable) + if (!ep->auth_enable) return -EACCES; if (optlen < sizeof(struct sctp_hmacalgo)) @@ -3376,7 +3377,7 @@ static int sctp_setsockopt_hmac_ident(struct sock *sk, goto out; } - err = sctp_auth_ep_set_hmacs(sctp_sk(sk)->ep, hmacs); + err = sctp_auth_ep_set_hmacs(ep, hmacs); out: kfree(hmacs); return err; @@ -3392,12 +3393,12 @@ static int sctp_setsockopt_auth_key(struct sock *sk, char __user *optval, unsigned int optlen) { - struct net *net = sock_net(sk); + struct sctp_endpoint *ep = sctp_sk(sk)->ep; struct sctp_authkey *authkey; struct sctp_association *asoc; int ret; - if (!net->sctp.auth_enable) + if (!ep->auth_enable) return -EACCES; if (optlen <= sizeof(struct sctp_authkey)) @@ -3418,7 +3419,7 @@ static int sctp_setsockopt_auth_key(struct sock *sk, goto out; } - ret = sctp_auth_set_key(sctp_sk(sk)->ep, asoc, authkey); + ret = sctp_auth_set_key(ep, asoc, authkey); out: kzfree(authkey); return ret; @@ -3434,11 +3435,11 @@ static int sctp_setsockopt_active_key(struct sock *sk, char __user *optval, unsigned int optlen) { - struct net *net = sock_net(sk); + struct sctp_endpoint *ep = sctp_sk(sk)->ep; struct sctp_authkeyid val; struct sctp_association *asoc; - if (!net->sctp.auth_enable) + if (!ep->auth_enable) return -EACCES; if (optlen != sizeof(struct sctp_authkeyid)) @@ -3450,8 +3451,7 @@ static int sctp_setsockopt_active_key(struct sock *sk, if (!asoc && val.scact_assoc_id && sctp_style(sk, UDP)) return -EINVAL; - return sctp_auth_set_active_key(sctp_sk(sk)->ep, asoc, - val.scact_keynumber); + return sctp_auth_set_active_key(ep, asoc, val.scact_keynumber); } /* @@ -3463,11 +3463,11 @@ static int sctp_setsockopt_del_key(struct sock *sk, char __user *optval, unsigned int optlen) { - struct net *net = sock_net(sk); + struct sctp_endpoint *ep = sctp_sk(sk)->ep; struct sctp_authkeyid val; struct sctp_association *asoc; - if (!net->sctp.auth_enable) + if (!ep->auth_enable) return -EACCES; if (optlen != sizeof(struct sctp_authkeyid)) @@ -3479,8 +3479,7 @@ static int sctp_setsockopt_del_key(struct sock *sk, if (!asoc && val.scact_assoc_id && sctp_style(sk, UDP)) return -EINVAL; - return sctp_auth_del_key_id(sctp_sk(sk)->ep, asoc, - val.scact_keynumber); + return sctp_auth_del_key_id(ep, asoc, val.scact_keynumber); } @@ -5387,16 +5386,16 @@ static int sctp_getsockopt_maxburst(struct sock *sk, int len, static int sctp_getsockopt_hmac_ident(struct sock *sk, int len, char __user *optval, int __user *optlen) { - struct net *net = sock_net(sk); + struct sctp_endpoint *ep = sctp_sk(sk)->ep; struct sctp_hmacalgo __user *p = (void __user *)optval; struct sctp_hmac_algo_param *hmacs; __u16 data_len = 0; u32 num_idents; - if (!net->sctp.auth_enable) + if (!ep->auth_enable) return -EACCES; - hmacs = sctp_sk(sk)->ep->auth_hmacs_list; + hmacs = ep->auth_hmacs_list; data_len = ntohs(hmacs->param_hdr.length) - sizeof(sctp_paramhdr_t); if (len < sizeof(struct sctp_hmacalgo) + data_len) @@ -5417,11 +5416,11 @@ static int sctp_getsockopt_hmac_ident(struct sock *sk, int len, static int sctp_getsockopt_active_key(struct sock *sk, int len, char __user *optval, int __user *optlen) { - struct net *net = sock_net(sk); + struct sctp_endpoint *ep = sctp_sk(sk)->ep; struct sctp_authkeyid val; struct sctp_association *asoc; - if (!net->sctp.auth_enable) + if (!ep->auth_enable) return -EACCES; if (len < sizeof(struct sctp_authkeyid)) @@ -5436,7 +5435,7 @@ static int sctp_getsockopt_active_key(struct sock *sk, int len, if (asoc) val.scact_keynumber = asoc->active_key_id; else - val.scact_keynumber = sctp_sk(sk)->ep->active_key_id; + val.scact_keynumber = ep->active_key_id; len = sizeof(struct sctp_authkeyid); if (put_user(len, optlen)) @@ -5450,7 +5449,7 @@ static int sctp_getsockopt_active_key(struct sock *sk, int len, static int sctp_getsockopt_peer_auth_chunks(struct sock *sk, int len, char __user *optval, int __user *optlen) { - struct net *net = sock_net(sk); + struct sctp_endpoint *ep = sctp_sk(sk)->ep; struct sctp_authchunks __user *p = (void __user *)optval; struct sctp_authchunks val; struct sctp_association *asoc; @@ -5458,7 +5457,7 @@ static int sctp_getsockopt_peer_auth_chunks(struct sock *sk, int len, u32 num_chunks = 0; char __user *to; - if (!net->sctp.auth_enable) + if (!ep->auth_enable) return -EACCES; if (len < sizeof(struct sctp_authchunks)) @@ -5495,7 +5494,7 @@ num: static int sctp_getsockopt_local_auth_chunks(struct sock *sk, int len, char __user *optval, int __user *optlen) { - struct net *net = sock_net(sk); + struct sctp_endpoint *ep = sctp_sk(sk)->ep; struct sctp_authchunks __user *p = (void __user *)optval; struct sctp_authchunks val; struct sctp_association *asoc; @@ -5503,7 +5502,7 @@ static int sctp_getsockopt_local_auth_chunks(struct sock *sk, int len, u32 num_chunks = 0; char __user *to; - if (!net->sctp.auth_enable) + if (!ep->auth_enable) return -EACCES; if (len < sizeof(struct sctp_authchunks)) @@ -5520,7 +5519,7 @@ static int sctp_getsockopt_local_auth_chunks(struct sock *sk, int len, if (asoc) ch = (struct sctp_chunks_param *)asoc->c.auth_chunks; else - ch = sctp_sk(sk)->ep->auth_chunk_list; + ch = ep->auth_chunk_list; if (!ch) goto num; @@ -5947,8 +5946,9 @@ static long sctp_get_port_local(struct sock *sk, union sctp_addr *addr) /* Search for an available port. */ int low, high, remaining, index; unsigned int rover; + struct net *net = sock_net(sk); - inet_get_local_port_range(sock_net(sk), &low, &high); + inet_get_local_port_range(net, &low, &high); remaining = (high - low) + 1; rover = prandom_u32() % remaining + low; @@ -5956,7 +5956,7 @@ static long sctp_get_port_local(struct sock *sk, union sctp_addr *addr) rover++; if ((rover < low) || (rover > high)) rover = low; - if (inet_is_reserved_local_port(rover)) + if (inet_is_local_reserved_port(net, rover)) continue; index = sctp_phashfn(sock_net(sk), rover); head = &sctp_port_hashtable[index]; @@ -6559,6 +6559,10 @@ static struct sk_buff *sctp_skb_recv_datagram(struct sock *sk, int flags, if (sk->sk_shutdown & RCV_SHUTDOWN) break; + if (sk_can_busy_loop(sk) && + sk_busy_loop(sk, noblock)) + continue; + /* User doesn't want to wait. */ error = -EAGAIN; if (!timeo) @@ -6942,7 +6946,8 @@ void sctp_copy_sock(struct sock *newsk, struct sock *sk, newsk->sk_type = sk->sk_type; newsk->sk_bound_dev_if = sk->sk_bound_dev_if; newsk->sk_flags = sk->sk_flags; - newsk->sk_no_check = sk->sk_no_check; + newsk->sk_no_check_tx = sk->sk_no_check_tx; + newsk->sk_no_check_rx = sk->sk_no_check_rx; newsk->sk_reuse = sk->sk_reuse; newsk->sk_shutdown = sk->sk_shutdown; diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c index 35c8923b555..7e5eb755499 100644 --- a/net/sctp/sysctl.c +++ b/net/sctp/sysctl.c @@ -64,6 +64,9 @@ static int proc_sctp_do_rto_min(struct ctl_table *ctl, int write, static int proc_sctp_do_rto_max(struct ctl_table *ctl, int write, void __user *buffer, size_t *lenp, loff_t *ppos); +static int proc_sctp_do_auth(struct ctl_table *ctl, int write, + void __user *buffer, size_t *lenp, + loff_t *ppos); static struct ctl_table sctp_table[] = { { @@ -266,7 +269,7 @@ static struct ctl_table sctp_net_table[] = { .data = &init_net.sctp.auth_enable, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_sctp_do_auth, }, { .procname = "addr_scope_policy", @@ -400,22 +403,54 @@ static int proc_sctp_do_rto_max(struct ctl_table *ctl, int write, return ret; } -int sctp_sysctl_net_register(struct net *net) +static int proc_sctp_do_auth(struct ctl_table *ctl, int write, + void __user *buffer, size_t *lenp, + loff_t *ppos) { - struct ctl_table *table = sctp_net_table; + struct net *net = current->nsproxy->net_ns; + struct ctl_table tbl; + int new_value, ret; + + memset(&tbl, 0, sizeof(struct ctl_table)); + tbl.maxlen = sizeof(unsigned int); - if (!net_eq(net, &init_net)) { - int i; + if (write) + tbl.data = &new_value; + else + tbl.data = &net->sctp.auth_enable; - table = kmemdup(sctp_net_table, sizeof(sctp_net_table), GFP_KERNEL); - if (!table) - return -ENOMEM; + ret = proc_dointvec(&tbl, write, buffer, lenp, ppos); - for (i = 0; table[i].data; i++) - table[i].data += (char *)(&net->sctp) - (char *)&init_net.sctp; + if (write) { + struct sock *sk = net->sctp.ctl_sock; + + net->sctp.auth_enable = new_value; + /* Update the value in the control socket */ + lock_sock(sk); + sctp_sk(sk)->ep->auth_enable = new_value; + release_sock(sk); } + return ret; +} + +int sctp_sysctl_net_register(struct net *net) +{ + struct ctl_table *table; + int i; + + table = kmemdup(sctp_net_table, sizeof(sctp_net_table), GFP_KERNEL); + if (!table) + return -ENOMEM; + + for (i = 0; table[i].data; i++) + table[i].data += (char *)(&net->sctp) - (char *)&init_net.sctp; + net->sctp.sysctl_header = register_net_sysctl(net, "net/sctp", table); + if (net->sctp.sysctl_header == NULL) { + kfree(table); + return -ENOMEM; + } return 0; } diff --git a/net/sctp/ulpqueue.c b/net/sctp/ulpqueue.c index 7144eb6a1b9..d49dc2ed30a 100644 --- a/net/sctp/ulpqueue.c +++ b/net/sctp/ulpqueue.c @@ -38,6 +38,7 @@ #include <linux/types.h> #include <linux/skbuff.h> #include <net/sock.h> +#include <net/busy_poll.h> #include <net/sctp/structs.h> #include <net/sctp/sctp.h> #include <net/sctp/sm.h> @@ -204,6 +205,9 @@ int sctp_ulpq_tail_event(struct sctp_ulpq *ulpq, struct sctp_ulpevent *event) if (sock_flag(sk, SOCK_DEAD) || (sk->sk_shutdown & RCV_SHUTDOWN)) goto out_free; + if (!sctp_ulpevent_is_notification(event)) + sk_mark_napi_id(sk, skb); + /* Check if the user wishes to receive this event. */ if (!sctp_ulpevent_is_enabled(event, &sctp_sk(sk)->subscribe)) goto out_free; diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 25a3dcf15ca..1dec6043e4d 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -866,8 +866,6 @@ static void xs_reset_transport(struct sock_xprt *transport) xs_restore_old_callbacks(transport, sk); write_unlock_bh(&sk->sk_callback_lock); - sk->sk_no_check = 0; - trace_rpc_socket_close(&transport->xprt, sock); sock_release(sock); } @@ -2046,7 +2044,6 @@ static void xs_udp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock) sk->sk_user_data = xprt; sk->sk_data_ready = xs_udp_data_ready; sk->sk_write_space = xs_udp_write_space; - sk->sk_no_check = UDP_CSUM_NORCV; sk->sk_allocation = GFP_ATOMIC; xprt_set_connected(xprt); diff --git a/net/tipc/Makefile b/net/tipc/Makefile index b282f7130d2..a080c66d819 100644 --- a/net/tipc/Makefile +++ b/net/tipc/Makefile @@ -5,7 +5,7 @@ obj-$(CONFIG_TIPC) := tipc.o tipc-y += addr.o bcast.o bearer.o config.o \ - core.o handler.o link.o discover.o msg.o \ + core.o link.o discover.o msg.o \ name_distr.o subscr.o name_table.o net.o \ netlink.o node.o node_subscr.o port.o ref.o \ socket.o log.o eth_media.o server.o diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c index 95ab5ef9292..671f9817b4f 100644 --- a/net/tipc/bcast.c +++ b/net/tipc/bcast.c @@ -71,7 +71,7 @@ struct tipc_bcbearer_pair { * Note: The fields labelled "temporary" are incorporated into the bearer * to avoid consuming potentially limited stack space through the use of * large local variables within multicast routines. Concurrent access is - * prevented through use of the spinlock "bc_lock". + * prevented through use of the spinlock "bclink_lock". */ struct tipc_bcbearer { struct tipc_bearer bearer; @@ -84,34 +84,64 @@ struct tipc_bcbearer { /** * struct tipc_bclink - link used for broadcast messages + * @lock: spinlock governing access to structure * @link: (non-standard) broadcast link structure * @node: (non-standard) node structure representing b'cast link's peer node + * @flags: represent bclink states * @bcast_nodes: map of broadcast-capable nodes * @retransmit_to: node that most recently requested a retransmit * * Handles sequence numbering, fragmentation, bundling, etc. */ struct tipc_bclink { + spinlock_t lock; struct tipc_link link; struct tipc_node node; + unsigned int flags; struct tipc_node_map bcast_nodes; struct tipc_node *retransmit_to; }; -static struct tipc_bcbearer bcast_bearer; -static struct tipc_bclink bcast_link; - -static struct tipc_bcbearer *bcbearer = &bcast_bearer; -static struct tipc_bclink *bclink = &bcast_link; -static struct tipc_link *bcl = &bcast_link.link; - -static DEFINE_SPINLOCK(bc_lock); +static struct tipc_bcbearer *bcbearer; +static struct tipc_bclink *bclink; +static struct tipc_link *bcl; const char tipc_bclink_name[] = "broadcast-link"; static void tipc_nmap_diff(struct tipc_node_map *nm_a, struct tipc_node_map *nm_b, struct tipc_node_map *nm_diff); +static void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node); +static void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node); + +static void tipc_bclink_lock(void) +{ + spin_lock_bh(&bclink->lock); +} + +static void tipc_bclink_unlock(void) +{ + struct tipc_node *node = NULL; + + if (likely(!bclink->flags)) { + spin_unlock_bh(&bclink->lock); + return; + } + + if (bclink->flags & TIPC_BCLINK_RESET) { + bclink->flags &= ~TIPC_BCLINK_RESET; + node = tipc_bclink_retransmit_to(); + } + spin_unlock_bh(&bclink->lock); + + if (node) + tipc_link_reset_all(node); +} + +void tipc_bclink_set_flags(unsigned int flags) +{ + bclink->flags |= flags; +} static u32 bcbuf_acks(struct sk_buff *buf) { @@ -130,16 +160,16 @@ static void bcbuf_decr_acks(struct sk_buff *buf) void tipc_bclink_add_node(u32 addr) { - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); tipc_nmap_add(&bclink->bcast_nodes, addr); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); } void tipc_bclink_remove_node(u32 addr) { - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); tipc_nmap_remove(&bclink->bcast_nodes, addr); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); } static void bclink_set_last_sent(void) @@ -165,7 +195,7 @@ static void bclink_update_last_sent(struct tipc_node *node, u32 seqno) /** * tipc_bclink_retransmit_to - get most recent node to request retransmission * - * Called with bc_lock locked + * Called with bclink_lock locked */ struct tipc_node *tipc_bclink_retransmit_to(void) { @@ -177,7 +207,7 @@ struct tipc_node *tipc_bclink_retransmit_to(void) * @after: sequence number of last packet to *not* retransmit * @to: sequence number of last packet to retransmit * - * Called with bc_lock locked + * Called with bclink_lock locked */ static void bclink_retransmit_pkt(u32 after, u32 to) { @@ -194,7 +224,7 @@ static void bclink_retransmit_pkt(u32 after, u32 to) * @n_ptr: node that sent acknowledgement info * @acked: broadcast sequence # that has been acknowledged * - * Node is locked, bc_lock unlocked. + * Node is locked, bclink_lock unlocked. */ void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked) { @@ -202,8 +232,7 @@ void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked) struct sk_buff *next; unsigned int released = 0; - spin_lock_bh(&bc_lock); - + tipc_bclink_lock(); /* Bail out if tx queue is empty (no clean up is required) */ crs = bcl->first_out; if (!crs) @@ -267,13 +296,13 @@ void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked) if (unlikely(released && !list_empty(&bcl->waiting_ports))) tipc_link_wakeup_ports(bcl, 0); exit: - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); } /** * tipc_bclink_update_link_state - update broadcast link state * - * tipc_net_lock and node lock set + * RCU and node lock set */ void tipc_bclink_update_link_state(struct tipc_node *n_ptr, u32 last_sent) { @@ -320,10 +349,10 @@ void tipc_bclink_update_link_state(struct tipc_node *n_ptr, u32 last_sent) ? buf_seqno(n_ptr->bclink.deferred_head) - 1 : n_ptr->bclink.last_sent); - spin_lock_bh(&bc_lock); - tipc_bearer_send(&bcbearer->bearer, buf, NULL); + tipc_bclink_lock(); + tipc_bearer_send(MAX_BEARERS, buf, NULL); bcl->stats.sent_nacks++; - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); kfree_skb(buf); n_ptr->bclink.oos_state++; @@ -335,8 +364,6 @@ void tipc_bclink_update_link_state(struct tipc_node *n_ptr, u32 last_sent) * * Delay any upcoming NACK by this node if another node has already * requested the first message this node is going to ask for. - * - * Only tipc_net_lock set. */ static void bclink_peek_nack(struct tipc_msg *msg) { @@ -362,7 +389,7 @@ int tipc_bclink_xmit(struct sk_buff *buf) { int res; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); if (!bclink->bcast_nodes.count) { res = msg_data_sz(buf_msg(buf)); @@ -377,14 +404,14 @@ int tipc_bclink_xmit(struct sk_buff *buf) bcl->stats.accu_queue_sz += bcl->out_queue_size; } exit: - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); return res; } /** * bclink_accept_pkt - accept an incoming, in-sequence broadcast packet * - * Called with both sending node's lock and bc_lock taken. + * Called with both sending node's lock and bclink_lock taken. */ static void bclink_accept_pkt(struct tipc_node *node, u32 seqno) { @@ -408,7 +435,7 @@ static void bclink_accept_pkt(struct tipc_node *node, u32 seqno) /** * tipc_bclink_rcv - receive a broadcast packet, and deliver upwards * - * tipc_net_lock is read_locked, no other locks set + * RCU is locked, no other locks set */ void tipc_bclink_rcv(struct sk_buff *buf) { @@ -439,12 +466,12 @@ void tipc_bclink_rcv(struct sk_buff *buf) if (msg_destnode(msg) == tipc_own_addr) { tipc_bclink_acknowledge(node, msg_bcast_ack(msg)); tipc_node_unlock(node); - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); bcl->stats.recv_nacks++; bclink->retransmit_to = node; bclink_retransmit_pkt(msg_bcgap_after(msg), msg_bcgap_to(msg)); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); } else { tipc_node_unlock(node); bclink_peek_nack(msg); @@ -462,51 +489,47 @@ receive: /* Deliver message to destination */ if (likely(msg_isdata(msg))) { - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); bclink_accept_pkt(node, seqno); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); tipc_node_unlock(node); if (likely(msg_mcast(msg))) tipc_port_mcast_rcv(buf, NULL); else kfree_skb(buf); } else if (msg_user(msg) == MSG_BUNDLER) { - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); bclink_accept_pkt(node, seqno); bcl->stats.recv_bundles++; bcl->stats.recv_bundled += msg_msgcnt(msg); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); tipc_node_unlock(node); tipc_link_bundle_rcv(buf); } else if (msg_user(msg) == MSG_FRAGMENTER) { - int ret; - ret = tipc_link_frag_rcv(&node->bclink.reasm_head, - &node->bclink.reasm_tail, - &buf); - if (ret == LINK_REASM_ERROR) + tipc_buf_append(&node->bclink.reasm_buf, &buf); + if (unlikely(!buf && !node->bclink.reasm_buf)) goto unlock; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); bclink_accept_pkt(node, seqno); bcl->stats.recv_fragments++; - if (ret == LINK_REASM_COMPLETE) { + if (buf) { bcl->stats.recv_fragmented++; - /* Point msg to inner header */ msg = buf_msg(buf); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); goto receive; } - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); tipc_node_unlock(node); } else if (msg_user(msg) == NAME_DISTRIBUTOR) { - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); bclink_accept_pkt(node, seqno); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); tipc_node_unlock(node); tipc_named_rcv(buf); } else { - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); bclink_accept_pkt(node, seqno); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); tipc_node_unlock(node); kfree_skb(buf); } @@ -552,14 +575,14 @@ receive: } else deferred = 0; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); if (deferred) bcl->stats.deferred_recv++; else bcl->stats.duplicates++; - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); unlock: tipc_node_unlock(node); @@ -627,13 +650,13 @@ static int tipc_bcbearer_send(struct sk_buff *buf, struct tipc_bearer *unused1, if (bp_index == 0) { /* Use original buffer for first bearer */ - tipc_bearer_send(b, buf, &b->bcast_addr); + tipc_bearer_send(b->identity, buf, &b->bcast_addr); } else { /* Avoid concurrent buffer access */ tbuf = pskb_copy(buf, GFP_ATOMIC); if (!tbuf) break; - tipc_bearer_send(b, tbuf, &b->bcast_addr); + tipc_bearer_send(b->identity, tbuf, &b->bcast_addr); kfree_skb(tbuf); /* Bearer keeps a clone */ } @@ -655,20 +678,27 @@ static int tipc_bcbearer_send(struct sk_buff *buf, struct tipc_bearer *unused1, /** * tipc_bcbearer_sort - create sets of bearer pairs used by broadcast bearer */ -void tipc_bcbearer_sort(void) +void tipc_bcbearer_sort(struct tipc_node_map *nm_ptr, u32 node, bool action) { struct tipc_bcbearer_pair *bp_temp = bcbearer->bpairs_temp; struct tipc_bcbearer_pair *bp_curr; + struct tipc_bearer *b; int b_index; int pri; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); + + if (action) + tipc_nmap_add(nm_ptr, node); + else + tipc_nmap_remove(nm_ptr, node); /* Group bearers by priority (can assume max of two per priority) */ memset(bp_temp, 0, sizeof(bcbearer->bpairs_temp)); + rcu_read_lock(); for (b_index = 0; b_index < MAX_BEARERS; b_index++) { - struct tipc_bearer *b = bearer_list[b_index]; + b = rcu_dereference_rtnl(bearer_list[b_index]); if (!b || !b->nodes.count) continue; @@ -677,6 +707,7 @@ void tipc_bcbearer_sort(void) else bp_temp[b->priority].secondary = b; } + rcu_read_unlock(); /* Create array of bearer pairs for broadcasting */ bp_curr = bcbearer->bpairs; @@ -702,7 +733,7 @@ void tipc_bcbearer_sort(void) bp_curr++; } - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); } @@ -714,7 +745,7 @@ int tipc_bclink_stats(char *buf, const u32 buf_size) if (!bcl) return 0; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); s = &bcl->stats; @@ -743,7 +774,7 @@ int tipc_bclink_stats(char *buf, const u32 buf_size) s->queue_sz_counts ? (s->accu_queue_sz / s->queue_sz_counts) : 0); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); return ret; } @@ -752,9 +783,9 @@ int tipc_bclink_reset_stats(void) if (!bcl) return -ENOPROTOOPT; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); memset(&bcl->stats, 0, sizeof(bcl->stats)); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); return 0; } @@ -765,46 +796,59 @@ int tipc_bclink_set_queue_limits(u32 limit) if ((limit < TIPC_MIN_LINK_WIN) || (limit > TIPC_MAX_LINK_WIN)) return -EINVAL; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); tipc_link_set_queue_limits(bcl, limit); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); return 0; } -void tipc_bclink_init(void) +int tipc_bclink_init(void) { + bcbearer = kzalloc(sizeof(*bcbearer), GFP_ATOMIC); + if (!bcbearer) + return -ENOMEM; + + bclink = kzalloc(sizeof(*bclink), GFP_ATOMIC); + if (!bclink) { + kfree(bcbearer); + return -ENOMEM; + } + + bcl = &bclink->link; bcbearer->bearer.media = &bcbearer->media; bcbearer->media.send_msg = tipc_bcbearer_send; sprintf(bcbearer->media.name, "tipc-broadcast"); + spin_lock_init(&bclink->lock); INIT_LIST_HEAD(&bcl->waiting_ports); bcl->next_out_no = 1; spin_lock_init(&bclink->node.lock); bcl->owner = &bclink->node; bcl->max_pkt = MAX_PKT_DEFAULT_MCAST; tipc_link_set_queue_limits(bcl, BCLINK_WIN_DEFAULT); - bcl->b_ptr = &bcbearer->bearer; - bearer_list[BCBEARER] = &bcbearer->bearer; + bcl->bearer_id = MAX_BEARERS; + rcu_assign_pointer(bearer_list[MAX_BEARERS], &bcbearer->bearer); bcl->state = WORKING_WORKING; strlcpy(bcl->name, tipc_bclink_name, TIPC_MAX_LINK_NAME); + return 0; } void tipc_bclink_stop(void) { - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); tipc_link_purge_queues(bcl); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); - bearer_list[BCBEARER] = NULL; - memset(bclink, 0, sizeof(*bclink)); - memset(bcbearer, 0, sizeof(*bcbearer)); + RCU_INIT_POINTER(bearer_list[BCBEARER], NULL); + synchronize_net(); + kfree(bcbearer); + kfree(bclink); } - /** * tipc_nmap_add - add a node to a node map */ -void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node) +static void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node) { int n = tipc_node(node); int w = n / WSIZE; @@ -819,7 +863,7 @@ void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node) /** * tipc_nmap_remove - remove a node from a node map */ -void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node) +static void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node) { int n = tipc_node(node); int w = n / WSIZE; diff --git a/net/tipc/bcast.h b/net/tipc/bcast.h index a80ef54b818..00330c45df3 100644 --- a/net/tipc/bcast.h +++ b/net/tipc/bcast.h @@ -39,6 +39,7 @@ #define MAX_NODES 4096 #define WSIZE 32 +#define TIPC_BCLINK_RESET 1 /** * struct tipc_node_map - set of node identifiers @@ -69,9 +70,6 @@ struct tipc_node; extern const char tipc_bclink_name[]; -void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node); -void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node); - /** * tipc_nmap_equal - test for equality of node maps */ @@ -84,8 +82,9 @@ static inline int tipc_nmap_equal(struct tipc_node_map *nm_a, void tipc_port_list_add(struct tipc_port_list *pl_ptr, u32 port); void tipc_port_list_free(struct tipc_port_list *pl_ptr); -void tipc_bclink_init(void); +int tipc_bclink_init(void); void tipc_bclink_stop(void); +void tipc_bclink_set_flags(unsigned int flags); void tipc_bclink_add_node(u32 addr); void tipc_bclink_remove_node(u32 addr); struct tipc_node *tipc_bclink_retransmit_to(void); @@ -98,6 +97,6 @@ void tipc_bclink_update_link_state(struct tipc_node *n_ptr, u32 last_sent); int tipc_bclink_stats(char *stats_buf, const u32 buf_size); int tipc_bclink_reset_stats(void); int tipc_bclink_set_queue_limits(u32 limit); -void tipc_bcbearer_sort(void); +void tipc_bcbearer_sort(struct tipc_node_map *nm_ptr, u32 node, bool action); #endif diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 3fef7eb776d..264474394f9 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -49,7 +49,7 @@ static struct tipc_media * const media_info_array[] = { NULL }; -struct tipc_bearer *bearer_list[MAX_BEARERS + 1]; +struct tipc_bearer __rcu *bearer_list[MAX_BEARERS + 1]; static void bearer_disable(struct tipc_bearer *b_ptr, bool shutting_down); @@ -178,7 +178,7 @@ struct tipc_bearer *tipc_bearer_find(const char *name) u32 i; for (i = 0; i < MAX_BEARERS; i++) { - b_ptr = bearer_list[i]; + b_ptr = rtnl_dereference(bearer_list[i]); if (b_ptr && (!strcmp(b_ptr->name, name))) return b_ptr; } @@ -198,10 +198,9 @@ struct sk_buff *tipc_bearer_get_names(void) if (!buf) return NULL; - read_lock_bh(&tipc_net_lock); for (i = 0; media_info_array[i] != NULL; i++) { for (j = 0; j < MAX_BEARERS; j++) { - b = bearer_list[j]; + b = rtnl_dereference(bearer_list[j]); if (!b) continue; if (b->media == media_info_array[i]) { @@ -211,22 +210,33 @@ struct sk_buff *tipc_bearer_get_names(void) } } } - read_unlock_bh(&tipc_net_lock); return buf; } -void tipc_bearer_add_dest(struct tipc_bearer *b_ptr, u32 dest) +void tipc_bearer_add_dest(u32 bearer_id, u32 dest) { - tipc_nmap_add(&b_ptr->nodes, dest); - tipc_bcbearer_sort(); - tipc_disc_add_dest(b_ptr->link_req); + struct tipc_bearer *b_ptr; + + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[bearer_id]); + if (b_ptr) { + tipc_bcbearer_sort(&b_ptr->nodes, dest, true); + tipc_disc_add_dest(b_ptr->link_req); + } + rcu_read_unlock(); } -void tipc_bearer_remove_dest(struct tipc_bearer *b_ptr, u32 dest) +void tipc_bearer_remove_dest(u32 bearer_id, u32 dest) { - tipc_nmap_remove(&b_ptr->nodes, dest); - tipc_bcbearer_sort(); - tipc_disc_remove_dest(b_ptr->link_req); + struct tipc_bearer *b_ptr; + + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[bearer_id]); + if (b_ptr) { + tipc_bcbearer_sort(&b_ptr->nodes, dest, false); + tipc_disc_remove_dest(b_ptr->link_req); + } + rcu_read_unlock(); } /** @@ -271,13 +281,11 @@ int tipc_enable_bearer(const char *name, u32 disc_domain, u32 priority) return -EINVAL; } - write_lock_bh(&tipc_net_lock); - m_ptr = tipc_media_find(b_names.media_name); if (!m_ptr) { pr_warn("Bearer <%s> rejected, media <%s> not registered\n", name, b_names.media_name); - goto exit; + return -EINVAL; } if (priority == TIPC_MEDIA_LINK_PRI) @@ -287,7 +295,7 @@ restart: bearer_id = MAX_BEARERS; with_this_prio = 1; for (i = MAX_BEARERS; i-- != 0; ) { - b_ptr = bearer_list[i]; + b_ptr = rtnl_dereference(bearer_list[i]); if (!b_ptr) { bearer_id = i; continue; @@ -295,14 +303,14 @@ restart: if (!strcmp(name, b_ptr->name)) { pr_warn("Bearer <%s> rejected, already enabled\n", name); - goto exit; + return -EINVAL; } if ((b_ptr->priority == priority) && (++with_this_prio > 2)) { if (priority-- == 0) { pr_warn("Bearer <%s> rejected, duplicate priority\n", name); - goto exit; + return -EINVAL; } pr_warn("Bearer <%s> priority adjustment required %u->%u\n", name, priority + 1, priority); @@ -312,21 +320,20 @@ restart: if (bearer_id >= MAX_BEARERS) { pr_warn("Bearer <%s> rejected, bearer limit reached (%u)\n", name, MAX_BEARERS); - goto exit; + return -EINVAL; } b_ptr = kzalloc(sizeof(*b_ptr), GFP_ATOMIC); - if (!b_ptr) { - res = -ENOMEM; - goto exit; - } + if (!b_ptr) + return -ENOMEM; + strcpy(b_ptr->name, name); b_ptr->media = m_ptr; res = m_ptr->enable_media(b_ptr); if (res) { pr_warn("Bearer <%s> rejected, enable failure (%d)\n", name, -res); - goto exit; + return -EINVAL; } b_ptr->identity = bearer_id; @@ -341,16 +348,14 @@ restart: bearer_disable(b_ptr, false); pr_warn("Bearer <%s> rejected, discovery object creation failed\n", name); - goto exit; + return -EINVAL; } - bearer_list[bearer_id] = b_ptr; + rcu_assign_pointer(bearer_list[bearer_id], b_ptr); pr_info("Enabled bearer <%s>, discovery domain %s, priority %u\n", name, tipc_addr_string_fill(addr_string, disc_domain), priority); -exit: - write_unlock_bh(&tipc_net_lock); return res; } @@ -359,19 +364,16 @@ exit: */ static int tipc_reset_bearer(struct tipc_bearer *b_ptr) { - read_lock_bh(&tipc_net_lock); pr_info("Resetting bearer <%s>\n", b_ptr->name); - tipc_disc_delete(b_ptr->link_req); tipc_link_reset_list(b_ptr->identity); - tipc_disc_create(b_ptr, &b_ptr->bcast_addr); - read_unlock_bh(&tipc_net_lock); + tipc_disc_reset(b_ptr); return 0; } /** * bearer_disable * - * Note: This routine assumes caller holds tipc_net_lock. + * Note: This routine assumes caller holds RTNL lock. */ static void bearer_disable(struct tipc_bearer *b_ptr, bool shutting_down) { @@ -385,12 +387,12 @@ static void bearer_disable(struct tipc_bearer *b_ptr, bool shutting_down) tipc_disc_delete(b_ptr->link_req); for (i = 0; i < MAX_BEARERS; i++) { - if (b_ptr == bearer_list[i]) { - bearer_list[i] = NULL; + if (b_ptr == rtnl_dereference(bearer_list[i])) { + RCU_INIT_POINTER(bearer_list[i], NULL); break; } } - kfree(b_ptr); + kfree_rcu(b_ptr, rcu); } int tipc_disable_bearer(const char *name) @@ -398,7 +400,6 @@ int tipc_disable_bearer(const char *name) struct tipc_bearer *b_ptr; int res; - write_lock_bh(&tipc_net_lock); b_ptr = tipc_bearer_find(name); if (b_ptr == NULL) { pr_warn("Attempt to disable unknown bearer <%s>\n", name); @@ -407,32 +408,9 @@ int tipc_disable_bearer(const char *name) bearer_disable(b_ptr, false); res = 0; } - write_unlock_bh(&tipc_net_lock); return res; } - -/* tipc_l2_media_addr_set - initialize Ethernet media address structure - * - * Media-dependent "value" field stores MAC address in first 6 bytes - * and zeroes out the remaining bytes. - */ -void tipc_l2_media_addr_set(const struct tipc_bearer *b, - struct tipc_media_addr *a, char *mac) -{ - int len = b->media->hwaddr_len; - - if (unlikely(sizeof(a->value) < len)) { - WARN_ONCE(1, "Media length invalid\n"); - return; - } - - memcpy(a->value, mac, len); - memset(a->value + len, 0, sizeof(a->value) - len); - a->media_id = b->media->type_id; - a->broadcast = !memcmp(mac, b->bcast_addr.value, len); -} - int tipc_enable_l2_media(struct tipc_bearer *b) { struct net_device *dev; @@ -443,33 +421,37 @@ int tipc_enable_l2_media(struct tipc_bearer *b) if (!dev) return -ENODEV; - /* Associate TIPC bearer with Ethernet bearer */ - b->media_ptr = dev; - memset(b->bcast_addr.value, 0, sizeof(b->bcast_addr.value)); + /* Associate TIPC bearer with L2 bearer */ + rcu_assign_pointer(b->media_ptr, dev); + memset(&b->bcast_addr, 0, sizeof(b->bcast_addr)); memcpy(b->bcast_addr.value, dev->broadcast, b->media->hwaddr_len); b->bcast_addr.media_id = b->media->type_id; b->bcast_addr.broadcast = 1; b->mtu = dev->mtu; - tipc_l2_media_addr_set(b, &b->addr, (char *)dev->dev_addr); + b->media->raw2addr(b, &b->addr, (char *)dev->dev_addr); rcu_assign_pointer(dev->tipc_ptr, b); return 0; } -/* tipc_disable_l2_media - detach TIPC bearer from an Ethernet interface +/* tipc_disable_l2_media - detach TIPC bearer from an L2 interface * - * Mark Ethernet bearer as inactive so that incoming buffers are thrown away, + * Mark L2 bearer as inactive so that incoming buffers are thrown away, * then get worker thread to complete bearer cleanup. (Can't do cleanup * here because cleanup code needs to sleep and caller holds spinlocks.) */ void tipc_disable_l2_media(struct tipc_bearer *b) { - struct net_device *dev = (struct net_device *)b->media_ptr; + struct net_device *dev; + + dev = (struct net_device *)rtnl_dereference(b->media_ptr); + RCU_INIT_POINTER(b->media_ptr, NULL); RCU_INIT_POINTER(dev->tipc_ptr, NULL); + synchronize_net(); dev_put(dev); } /** - * tipc_l2_send_msg - send a TIPC packet out over an Ethernet interface + * tipc_l2_send_msg - send a TIPC packet out over an L2 interface * @buf: the packet to be sent * @b_ptr: the bearer through which the packet is to be sent * @dest: peer destination address @@ -478,8 +460,12 @@ int tipc_l2_send_msg(struct sk_buff *buf, struct tipc_bearer *b, struct tipc_media_addr *dest) { struct sk_buff *clone; + struct net_device *dev; int delta; - struct net_device *dev = (struct net_device *)b->media_ptr; + + dev = (struct net_device *)rcu_dereference_rtnl(b->media_ptr); + if (!dev) + return 0; clone = skb_clone(buf, GFP_ATOMIC); if (!clone) @@ -507,10 +493,16 @@ int tipc_l2_send_msg(struct sk_buff *buf, struct tipc_bearer *b, * The media send routine must not alter the buffer being passed in * as it may be needed for later retransmission! */ -void tipc_bearer_send(struct tipc_bearer *b, struct sk_buff *buf, +void tipc_bearer_send(u32 bearer_id, struct sk_buff *buf, struct tipc_media_addr *dest) { - b->media->send_msg(buf, b, dest); + struct tipc_bearer *b_ptr; + + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[bearer_id]); + if (likely(b_ptr)) + b_ptr->media->send_msg(buf, b_ptr, dest); + rcu_read_unlock(); } /** @@ -535,7 +527,7 @@ static int tipc_l2_rcv_msg(struct sk_buff *buf, struct net_device *dev, } rcu_read_lock(); - b_ptr = rcu_dereference(dev->tipc_ptr); + b_ptr = rcu_dereference_rtnl(dev->tipc_ptr); if (likely(b_ptr)) { if (likely(buf->pkt_type <= PACKET_BROADCAST)) { buf->next = NULL; @@ -568,12 +560,9 @@ static int tipc_l2_device_event(struct notifier_block *nb, unsigned long evt, if (!net_eq(dev_net(dev), &init_net)) return NOTIFY_DONE; - rcu_read_lock(); - b_ptr = rcu_dereference(dev->tipc_ptr); - if (!b_ptr) { - rcu_read_unlock(); + b_ptr = rtnl_dereference(dev->tipc_ptr); + if (!b_ptr) return NOTIFY_DONE; - } b_ptr->mtu = dev->mtu; @@ -586,17 +575,15 @@ static int tipc_l2_device_event(struct notifier_block *nb, unsigned long evt, tipc_reset_bearer(b_ptr); break; case NETDEV_CHANGEADDR: - tipc_l2_media_addr_set(b_ptr, &b_ptr->addr, + b_ptr->media->raw2addr(b_ptr, &b_ptr->addr, (char *)dev->dev_addr); tipc_reset_bearer(b_ptr); break; case NETDEV_UNREGISTER: case NETDEV_CHANGENAME: - tipc_disable_bearer(b_ptr->name); + bearer_disable(b_ptr, false); break; } - rcu_read_unlock(); - return NOTIFY_OK; } @@ -633,7 +620,7 @@ void tipc_bearer_stop(void) u32 i; for (i = 0; i < MAX_BEARERS; i++) { - b_ptr = bearer_list[i]; + b_ptr = rtnl_dereference(bearer_list[i]); if (b_ptr) { bearer_disable(b_ptr, true); bearer_list[i] = NULL; diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h index ba48145e871..78fccc49de2 100644 --- a/net/tipc/bearer.h +++ b/net/tipc/bearer.h @@ -42,14 +42,12 @@ #define MAX_BEARERS 2 #define MAX_MEDIA 2 -/* - * Identifiers associated with TIPC message header media address info - * - * - address info field is 20 bytes long - * - media type identifier located at offset 3 - * - remaining bytes vary according to media type +/* Identifiers associated with TIPC message header media address info + * - address info field is 32 bytes long + * - the field's actual content and length is defined per media + * - remaining unused bytes in the field are set to zero */ -#define TIPC_MEDIA_ADDR_SIZE 20 +#define TIPC_MEDIA_ADDR_SIZE 32 #define TIPC_MEDIA_TYPE_OFFSET 3 /* @@ -77,9 +75,10 @@ struct tipc_bearer; * @send_msg: routine which handles buffer transmission * @enable_media: routine which enables a media * @disable_media: routine which disables a media - * @addr2str: routine which converts media address to string - * @addr2msg: routine which converts media address to protocol message area - * @msg2addr: routine which converts media address from protocol message area + * @addr2str: convert media address format to string + * @addr2msg: convert from media addr format to discovery msg addr format + * @msg2addr: convert from discovery msg addr format to media addr format + * @raw2addr: convert from raw addr format to media addr format * @priority: default link (and bearer) priority * @tolerance: default time (in ms) before declaring link failure * @window: default window (in packets) before declaring link congestion @@ -93,10 +92,16 @@ struct tipc_media { struct tipc_media_addr *dest); int (*enable_media)(struct tipc_bearer *b_ptr); void (*disable_media)(struct tipc_bearer *b_ptr); - int (*addr2str)(struct tipc_media_addr *a, char *str_buf, int str_size); - int (*addr2msg)(struct tipc_media_addr *a, char *msg_area); - int (*msg2addr)(const struct tipc_bearer *b_ptr, - struct tipc_media_addr *a, char *msg_area); + int (*addr2str)(struct tipc_media_addr *addr, + char *strbuf, + int bufsz); + int (*addr2msg)(char *msg, struct tipc_media_addr *addr); + int (*msg2addr)(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg); + int (*raw2addr)(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *raw); u32 priority; u32 tolerance; u32 window; @@ -113,6 +118,7 @@ struct tipc_media { * @name: bearer name (format = media:interface) * @media: ptr to media structure associated with bearer * @bcast_addr: media address used in broadcasting + * @rcu: rcu struct for tipc_bearer * @priority: default link priority for bearer * @window: default window size for bearer * @tolerance: default link tolerance for bearer @@ -127,12 +133,13 @@ struct tipc_media { * care of initializing all other fields. */ struct tipc_bearer { - void *media_ptr; /* initalized by media */ + void __rcu *media_ptr; /* initalized by media */ u32 mtu; /* initalized by media */ struct tipc_media_addr addr; /* initalized by media */ char name[TIPC_MAX_BEARER_NAME]; struct tipc_media *media; struct tipc_media_addr bcast_addr; + struct rcu_head rcu; u32 priority; u32 window; u32 tolerance; @@ -150,7 +157,7 @@ struct tipc_bearer_names { struct tipc_link; -extern struct tipc_bearer *bearer_list[]; +extern struct tipc_bearer __rcu *bearer_list[]; /* * TIPC routines available to supported media types @@ -173,22 +180,20 @@ int tipc_media_set_priority(const char *name, u32 new_value); int tipc_media_set_window(const char *name, u32 new_value); void tipc_media_addr_printf(char *buf, int len, struct tipc_media_addr *a); struct sk_buff *tipc_media_get_names(void); -void tipc_l2_media_addr_set(const struct tipc_bearer *b, - struct tipc_media_addr *a, char *mac); int tipc_enable_l2_media(struct tipc_bearer *b); void tipc_disable_l2_media(struct tipc_bearer *b); int tipc_l2_send_msg(struct sk_buff *buf, struct tipc_bearer *b, struct tipc_media_addr *dest); struct sk_buff *tipc_bearer_get_names(void); -void tipc_bearer_add_dest(struct tipc_bearer *b_ptr, u32 dest); -void tipc_bearer_remove_dest(struct tipc_bearer *b_ptr, u32 dest); +void tipc_bearer_add_dest(u32 bearer_id, u32 dest); +void tipc_bearer_remove_dest(u32 bearer_id, u32 dest); struct tipc_bearer *tipc_bearer_find(const char *name); struct tipc_media *tipc_media_find(const char *name); int tipc_bearer_setup(void); void tipc_bearer_cleanup(void); void tipc_bearer_stop(void); -void tipc_bearer_send(struct tipc_bearer *b, struct sk_buff *buf, +void tipc_bearer_send(u32 bearer_id, struct sk_buff *buf, struct tipc_media_addr *dest); #endif /* _TIPC_BEARER_H */ diff --git a/net/tipc/config.c b/net/tipc/config.c index 4b981c05382..2b42403ad33 100644 --- a/net/tipc/config.c +++ b/net/tipc/config.c @@ -42,8 +42,6 @@ #define REPLY_TRUNCATED "<truncated>\n" -static DEFINE_MUTEX(config_mutex); - static const void *req_tlv_area; /* request message TLV area */ static int req_tlv_space; /* request message TLV area size */ static int rep_headroom; /* reply message headroom to use */ @@ -179,8 +177,10 @@ static struct sk_buff *cfg_set_own_addr(void) if (tipc_own_addr) return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED " (cannot change node address once assigned)"); - tipc_net_start(addr); - return tipc_cfg_reply_none(); + if (!tipc_net_start(addr)) + return tipc_cfg_reply_none(); + + return tipc_cfg_reply_error_string("cannot change to network mode"); } static struct sk_buff *cfg_set_max_ports(void) @@ -223,7 +223,7 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area { struct sk_buff *rep_tlv_buf; - mutex_lock(&config_mutex); + rtnl_lock(); /* Save request and reply details in a well-known location */ req_tlv_area = request_area; @@ -337,6 +337,6 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area /* Return reply buffer */ exit: - mutex_unlock(&config_mutex); + rtnl_unlock(); return rep_tlv_buf; } diff --git a/net/tipc/core.c b/net/tipc/core.c index 50d57429ebc..676d18015dd 100644 --- a/net/tipc/core.c +++ b/net/tipc/core.c @@ -80,7 +80,6 @@ struct sk_buff *tipc_buf_acquire(u32 size) */ static void tipc_core_stop(void) { - tipc_handler_stop(); tipc_net_stop(); tipc_bearer_cleanup(); tipc_netlink_stop(); @@ -100,10 +99,6 @@ static int tipc_core_start(void) get_random_bytes(&tipc_random, sizeof(tipc_random)); - err = tipc_handler_start(); - if (err) - goto out_handler; - err = tipc_ref_table_init(tipc_max_ports, tipc_random); if (err) goto out_reftbl; @@ -146,8 +141,6 @@ out_netlink: out_nametbl: tipc_ref_table_stop(); out_reftbl: - tipc_handler_stop(); -out_handler: return err; } @@ -161,10 +154,11 @@ static int __init tipc_init(void) tipc_max_ports = CONFIG_TIPC_PORTS; tipc_net_id = 4711; - sysctl_tipc_rmem[0] = CONN_OVERLOAD_LIMIT >> 4 << TIPC_LOW_IMPORTANCE; - sysctl_tipc_rmem[1] = CONN_OVERLOAD_LIMIT >> 4 << + sysctl_tipc_rmem[0] = TIPC_CONN_OVERLOAD_LIMIT >> 4 << + TIPC_LOW_IMPORTANCE; + sysctl_tipc_rmem[1] = TIPC_CONN_OVERLOAD_LIMIT >> 4 << TIPC_CRITICAL_IMPORTANCE; - sysctl_tipc_rmem[2] = CONN_OVERLOAD_LIMIT; + sysctl_tipc_rmem[2] = TIPC_CONN_OVERLOAD_LIMIT; res = tipc_core_start(); if (res) diff --git a/net/tipc/core.h b/net/tipc/core.h index 8985bbcb942..bb26ed1ee96 100644 --- a/net/tipc/core.h +++ b/net/tipc/core.h @@ -56,7 +56,8 @@ #include <linux/list.h> #include <linux/slab.h> #include <linux/vmalloc.h> - +#include <linux/rtnetlink.h> +#include <linux/etherdevice.h> #define TIPC_MOD_VER "2.0.0" @@ -89,8 +90,6 @@ extern int tipc_random __read_mostly; /* * Routines available to privileged subsystems */ -int tipc_handler_start(void); -void tipc_handler_stop(void); int tipc_netlink_start(void); void tipc_netlink_stop(void); int tipc_socket_init(void); @@ -109,12 +108,10 @@ void tipc_unregister_sysctl(void); #endif /* - * TIPC timer and signal code + * TIPC timer code */ typedef void (*Handler) (unsigned long); -u32 tipc_k_signal(Handler routine, unsigned long argument); - /** * k_init_timer - initialize a timer * @timer: pointer to timer structure @@ -191,6 +188,7 @@ static inline void k_term_timer(struct timer_list *timer) struct tipc_skb_cb { void *handle; bool deferred; + struct sk_buff *tail; }; #define TIPC_SKB_CB(__skb) ((struct tipc_skb_cb *)&((__skb)->cb[0])) diff --git a/net/tipc/discover.c b/net/tipc/discover.c index 542fe3413dc..aa722a42ef8 100644 --- a/net/tipc/discover.c +++ b/net/tipc/discover.c @@ -1,7 +1,7 @@ /* * net/tipc/discover.c * - * Copyright (c) 2003-2006, Ericsson AB + * Copyright (c) 2003-2006, 2014, Ericsson AB * Copyright (c) 2005-2006, 2010-2011, Wind River Systems * All rights reserved. * @@ -46,8 +46,9 @@ /** * struct tipc_link_req - information about an ongoing link setup request - * @bearer: bearer issuing requests + * @bearer_id: identity of bearer issuing requests * @dest: destination address for request messages + * @domain: network domain to which links can be established * @num_nodes: number of nodes currently discovered (i.e. with an active link) * @lock: spinlock for controlling access to requests * @buf: request message to be (repeatedly) sent @@ -55,8 +56,9 @@ * @timer_intv: current interval between requests (in ms) */ struct tipc_link_req { - struct tipc_bearer *bearer; + u32 bearer_id; struct tipc_media_addr dest; + u32 domain; int num_nodes; spinlock_t lock; struct sk_buff *buf; @@ -69,22 +71,19 @@ struct tipc_link_req { * @type: message type (request or response) * @b_ptr: ptr to bearer issuing message */ -static struct sk_buff *tipc_disc_init_msg(u32 type, struct tipc_bearer *b_ptr) +static void tipc_disc_init_msg(struct sk_buff *buf, u32 type, + struct tipc_bearer *b_ptr) { - struct sk_buff *buf = tipc_buf_acquire(INT_H_SIZE); struct tipc_msg *msg; u32 dest_domain = b_ptr->domain; - if (buf) { - msg = buf_msg(buf); - tipc_msg_init(msg, LINK_CONFIG, type, INT_H_SIZE, dest_domain); - msg_set_non_seq(msg, 1); - msg_set_node_sig(msg, tipc_random); - msg_set_dest_domain(msg, dest_domain); - msg_set_bc_netid(msg, tipc_net_id); - b_ptr->media->addr2msg(&b_ptr->addr, msg_media_addr(msg)); - } - return buf; + msg = buf_msg(buf); + tipc_msg_init(msg, LINK_CONFIG, type, INT_H_SIZE, dest_domain); + msg_set_non_seq(msg, 1); + msg_set_node_sig(msg, tipc_random); + msg_set_dest_domain(msg, dest_domain); + msg_set_bc_netid(msg, tipc_net_id); + b_ptr->media->addr2msg(msg_media_addr(msg), &b_ptr->addr); } /** @@ -107,146 +106,150 @@ static void disc_dupl_alert(struct tipc_bearer *b_ptr, u32 node_addr, } /** - * tipc_disc_rcv - handle incoming link setup message (request or response) + * tipc_disc_rcv - handle incoming discovery message (request or response) * @buf: buffer containing message - * @b_ptr: bearer that message arrived on + * @bearer: bearer that message arrived on */ -void tipc_disc_rcv(struct sk_buff *buf, struct tipc_bearer *b_ptr) +void tipc_disc_rcv(struct sk_buff *buf, struct tipc_bearer *bearer) { - struct tipc_node *n_ptr; + struct tipc_node *node; struct tipc_link *link; - struct tipc_media_addr media_addr; + struct tipc_media_addr maddr; struct sk_buff *rbuf; struct tipc_msg *msg = buf_msg(buf); - u32 dest = msg_dest_domain(msg); - u32 orig = msg_prevnode(msg); + u32 ddom = msg_dest_domain(msg); + u32 onode = msg_prevnode(msg); u32 net_id = msg_bc_netid(msg); - u32 type = msg_type(msg); + u32 mtyp = msg_type(msg); u32 signature = msg_node_sig(msg); - int addr_mismatch; - int link_fully_up; - - media_addr.broadcast = 1; - b_ptr->media->msg2addr(b_ptr, &media_addr, msg_media_addr(msg)); + bool addr_match = false; + bool sign_match = false; + bool link_up = false; + bool accept_addr = false; + bool accept_sign = false; + bool respond = false; + + bearer->media->msg2addr(bearer, &maddr, msg_media_addr(msg)); kfree_skb(buf); /* Ensure message from node is valid and communication is permitted */ if (net_id != tipc_net_id) return; - if (media_addr.broadcast) + if (maddr.broadcast) return; - if (!tipc_addr_domain_valid(dest)) + if (!tipc_addr_domain_valid(ddom)) return; - if (!tipc_addr_node_valid(orig)) + if (!tipc_addr_node_valid(onode)) return; - if (orig == tipc_own_addr) { - if (memcmp(&media_addr, &b_ptr->addr, sizeof(media_addr))) - disc_dupl_alert(b_ptr, tipc_own_addr, &media_addr); + + if (in_own_node(onode)) { + if (memcmp(&maddr, &bearer->addr, sizeof(maddr))) + disc_dupl_alert(bearer, tipc_own_addr, &maddr); return; } - if (!tipc_in_scope(dest, tipc_own_addr)) + if (!tipc_in_scope(ddom, tipc_own_addr)) return; - if (!tipc_in_scope(b_ptr->domain, orig)) + if (!tipc_in_scope(bearer->domain, onode)) return; - /* Locate structure corresponding to requesting node */ - n_ptr = tipc_node_find(orig); - if (!n_ptr) { - n_ptr = tipc_node_create(orig); - if (!n_ptr) - return; - } - tipc_node_lock(n_ptr); + /* Locate, or if necessary, create, node: */ + node = tipc_node_find(onode); + if (!node) + node = tipc_node_create(onode); + if (!node) + return; - /* Prepare to validate requesting node's signature and media address */ - link = n_ptr->links[b_ptr->identity]; - addr_mismatch = (link != NULL) && - memcmp(&link->media_addr, &media_addr, sizeof(media_addr)); + tipc_node_lock(node); + link = node->links[bearer->identity]; - /* - * Ensure discovery message's signature is correct - * - * If signature is incorrect and there is no working link to the node, - * accept the new signature but invalidate all existing links to the - * node so they won't re-activate without a new discovery message. - * - * If signature is incorrect and the requested link to the node is - * working, accept the new signature. (This is an instance of delayed - * rediscovery, where a link endpoint was able to re-establish contact - * with its peer endpoint on a node that rebooted before receiving a - * discovery message from that node.) - * - * If signature is incorrect and there is a working link to the node - * that is not the requested link, reject the request (must be from - * a duplicate node). - */ - if (signature != n_ptr->signature) { - if (n_ptr->working_links == 0) { - struct tipc_link *curr_link; - int i; - - for (i = 0; i < MAX_BEARERS; i++) { - curr_link = n_ptr->links[i]; - if (curr_link) { - memset(&curr_link->media_addr, 0, - sizeof(media_addr)); - tipc_link_reset(curr_link); - } - } - addr_mismatch = (link != NULL); - } else if (tipc_link_is_up(link) && !addr_mismatch) { - /* delayed rediscovery */ - } else { - disc_dupl_alert(b_ptr, orig, &media_addr); - tipc_node_unlock(n_ptr); - return; - } - n_ptr->signature = signature; + /* Prepare to validate requesting node's signature and media address */ + sign_match = (signature == node->signature); + addr_match = link && !memcmp(&link->media_addr, &maddr, sizeof(maddr)); + link_up = link && tipc_link_is_up(link); + + + /* These three flags give us eight permutations: */ + + if (sign_match && addr_match && link_up) { + /* All is fine. Do nothing. */ + } else if (sign_match && addr_match && !link_up) { + /* Respond. The link will come up in due time */ + respond = true; + } else if (sign_match && !addr_match && link_up) { + /* Peer has changed i/f address without rebooting. + * If so, the link will reset soon, and the next + * discovery will be accepted. So we can ignore it. + * It may also be an cloned or malicious peer having + * chosen the same node address and signature as an + * existing one. + * Ignore requests until the link goes down, if ever. + */ + disc_dupl_alert(bearer, onode, &maddr); + } else if (sign_match && !addr_match && !link_up) { + /* Peer link has changed i/f address without rebooting. + * It may also be a cloned or malicious peer; we can't + * distinguish between the two. + * The signature is correct, so we must accept. + */ + accept_addr = true; + respond = true; + } else if (!sign_match && addr_match && link_up) { + /* Peer node rebooted. Two possibilities: + * - Delayed re-discovery; this link endpoint has already + * reset and re-established contact with the peer, before + * receiving a discovery message from that node. + * (The peer happened to receive one from this node first). + * - The peer came back so fast that our side has not + * discovered it yet. Probing from this side will soon + * reset the link, since there can be no working link + * endpoint at the peer end, and the link will re-establish. + * Accept the signature, since it comes from a known peer. + */ + accept_sign = true; + } else if (!sign_match && addr_match && !link_up) { + /* The peer node has rebooted. + * Accept signature, since it is a known peer. + */ + accept_sign = true; + respond = true; + } else if (!sign_match && !addr_match && link_up) { + /* Peer rebooted with new address, or a new/duplicate peer. + * Ignore until the link goes down, if ever. + */ + disc_dupl_alert(bearer, onode, &maddr); + } else if (!sign_match && !addr_match && !link_up) { + /* Peer rebooted with new address, or it is a new peer. + * Accept signature and address. + */ + accept_sign = true; + accept_addr = true; + respond = true; } - /* - * Ensure requesting node's media address is correct - * - * If media address doesn't match and the link is working, reject the - * request (must be from a duplicate node). - * - * If media address doesn't match and the link is not working, accept - * the new media address and reset the link to ensure it starts up - * cleanly. - */ - if (addr_mismatch) { - if (tipc_link_is_up(link)) { - disc_dupl_alert(b_ptr, orig, &media_addr); - tipc_node_unlock(n_ptr); - return; - } else { - memcpy(&link->media_addr, &media_addr, - sizeof(media_addr)); - tipc_link_reset(link); - } - } + if (accept_sign) + node->signature = signature; - /* Create a link endpoint for this bearer, if necessary */ - if (!link) { - link = tipc_link_create(n_ptr, b_ptr, &media_addr); - if (!link) { - tipc_node_unlock(n_ptr); - return; + if (accept_addr) { + if (!link) + link = tipc_link_create(node, bearer, &maddr); + if (link) { + memcpy(&link->media_addr, &maddr, sizeof(maddr)); + tipc_link_reset(link); + } else { + respond = false; } } - /* Accept discovery message & send response, if necessary */ - link_fully_up = link_working_working(link); - - if ((type == DSC_REQ_MSG) && !link_fully_up) { - rbuf = tipc_disc_init_msg(DSC_RESP_MSG, b_ptr); + /* Send response, if necessary */ + if (respond && (mtyp == DSC_REQ_MSG)) { + rbuf = tipc_buf_acquire(INT_H_SIZE); if (rbuf) { - tipc_bearer_send(b_ptr, rbuf, &media_addr); + tipc_disc_init_msg(rbuf, DSC_RESP_MSG, bearer); + tipc_bearer_send(bearer->identity, rbuf, &maddr); kfree_skb(rbuf); } } - - tipc_node_unlock(n_ptr); + tipc_node_unlock(node); } /** @@ -303,7 +306,7 @@ static void disc_timeout(struct tipc_link_req *req) spin_lock_bh(&req->lock); /* Stop searching if only desired node has been found */ - if (tipc_node(req->bearer->domain) && req->num_nodes) { + if (tipc_node(req->domain) && req->num_nodes) { req->timer_intv = TIPC_LINK_REQ_INACTIVE; goto exit; } @@ -315,7 +318,7 @@ static void disc_timeout(struct tipc_link_req *req) * hold at fast polling rate if don't have any associated nodes, * otherwise hold at slow polling rate */ - tipc_bearer_send(req->bearer, req->buf, &req->dest); + tipc_bearer_send(req->bearer_id, req->buf, &req->dest); req->timer_intv *= 2; @@ -347,21 +350,23 @@ int tipc_disc_create(struct tipc_bearer *b_ptr, struct tipc_media_addr *dest) if (!req) return -ENOMEM; - req->buf = tipc_disc_init_msg(DSC_REQ_MSG, b_ptr); + req->buf = tipc_buf_acquire(INT_H_SIZE); if (!req->buf) { kfree(req); - return -ENOMSG; + return -ENOMEM; } + tipc_disc_init_msg(req->buf, DSC_REQ_MSG, b_ptr); memcpy(&req->dest, dest, sizeof(*dest)); - req->bearer = b_ptr; + req->bearer_id = b_ptr->identity; + req->domain = b_ptr->domain; req->num_nodes = 0; req->timer_intv = TIPC_LINK_REQ_INIT; spin_lock_init(&req->lock); k_init_timer(&req->timer, (Handler)disc_timeout, (unsigned long)req); k_start_timer(&req->timer, req->timer_intv); b_ptr->link_req = req; - tipc_bearer_send(req->bearer, req->buf, &req->dest); + tipc_bearer_send(req->bearer_id, req->buf, &req->dest); return 0; } @@ -376,3 +381,23 @@ void tipc_disc_delete(struct tipc_link_req *req) kfree_skb(req->buf); kfree(req); } + +/** + * tipc_disc_reset - reset object to send periodic link setup requests + * @b_ptr: ptr to bearer issuing requests + * @dest_domain: network domain to which links can be established + */ +void tipc_disc_reset(struct tipc_bearer *b_ptr) +{ + struct tipc_link_req *req = b_ptr->link_req; + + spin_lock_bh(&req->lock); + tipc_disc_init_msg(req->buf, DSC_REQ_MSG, b_ptr); + req->bearer_id = b_ptr->identity; + req->domain = b_ptr->domain; + req->num_nodes = 0; + req->timer_intv = TIPC_LINK_REQ_INIT; + k_start_timer(&req->timer, req->timer_intv); + tipc_bearer_send(req->bearer_id, req->buf, &req->dest); + spin_unlock_bh(&req->lock); +} diff --git a/net/tipc/discover.h b/net/tipc/discover.h index 07f34729459..515b57392f4 100644 --- a/net/tipc/discover.h +++ b/net/tipc/discover.h @@ -41,6 +41,7 @@ struct tipc_link_req; int tipc_disc_create(struct tipc_bearer *b_ptr, struct tipc_media_addr *dest); void tipc_disc_delete(struct tipc_link_req *req); +void tipc_disc_reset(struct tipc_bearer *b_ptr); void tipc_disc_add_dest(struct tipc_link_req *req); void tipc_disc_remove_dest(struct tipc_link_req *req); void tipc_disc_rcv(struct sk_buff *buf, struct tipc_bearer *b_ptr); diff --git a/net/tipc/eth_media.c b/net/tipc/eth_media.c index 67cf3f935db..5e1426f1751 100644 --- a/net/tipc/eth_media.c +++ b/net/tipc/eth_media.c @@ -1,7 +1,7 @@ /* * net/tipc/eth_media.c: Ethernet bearer support for TIPC * - * Copyright (c) 2001-2007, 2013, Ericsson AB + * Copyright (c) 2001-2007, 2013-2014, Ericsson AB * Copyright (c) 2005-2008, 2011-2013, Wind River Systems * All rights reserved. * @@ -37,39 +37,52 @@ #include "core.h" #include "bearer.h" -#define ETH_ADDR_OFFSET 4 /* message header offset of MAC address */ +#define ETH_ADDR_OFFSET 4 /* MAC addr position inside address field */ -/* convert Ethernet address to string */ -static int tipc_eth_addr2str(struct tipc_media_addr *a, char *str_buf, - int str_size) +/* Convert Ethernet address (media address format) to string */ +static int tipc_eth_addr2str(struct tipc_media_addr *addr, + char *strbuf, int bufsz) { - if (str_size < 18) /* 18 = strlen("aa:bb:cc:dd:ee:ff\0") */ + if (bufsz < 18) /* 18 = strlen("aa:bb:cc:dd:ee:ff\0") */ return 1; - sprintf(str_buf, "%pM", a->value); + sprintf(strbuf, "%pM", addr->value); return 0; } -/* convert Ethernet address format to message header format */ -static int tipc_eth_addr2msg(struct tipc_media_addr *a, char *msg_area) +/* Convert from media address format to discovery message addr format */ +static int tipc_eth_addr2msg(char *msg, struct tipc_media_addr *addr) { - memset(msg_area, 0, TIPC_MEDIA_ADDR_SIZE); - msg_area[TIPC_MEDIA_TYPE_OFFSET] = TIPC_MEDIA_TYPE_ETH; - memcpy(msg_area + ETH_ADDR_OFFSET, a->value, ETH_ALEN); + memset(msg, 0, TIPC_MEDIA_ADDR_SIZE); + msg[TIPC_MEDIA_TYPE_OFFSET] = TIPC_MEDIA_TYPE_ETH; + memcpy(msg + ETH_ADDR_OFFSET, addr->value, ETH_ALEN); return 0; } -/* convert message header address format to Ethernet format */ -static int tipc_eth_msg2addr(const struct tipc_bearer *tb_ptr, - struct tipc_media_addr *a, char *msg_area) +/* Convert raw mac address format to media addr format */ +static int tipc_eth_raw2addr(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg) { - if (msg_area[TIPC_MEDIA_TYPE_OFFSET] != TIPC_MEDIA_TYPE_ETH) - return 1; + char bcast_mac[ETH_ALEN] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff}; - tipc_l2_media_addr_set(tb_ptr, a, msg_area + ETH_ADDR_OFFSET); + memset(addr, 0, sizeof(*addr)); + ether_addr_copy(addr->value, msg); + addr->media_id = TIPC_MEDIA_TYPE_ETH; + addr->broadcast = !memcmp(addr->value, bcast_mac, ETH_ALEN); return 0; } +/* Convert discovery msg addr format to Ethernet media addr format */ +static int tipc_eth_msg2addr(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg) +{ + /* Skip past preamble: */ + msg += ETH_ADDR_OFFSET; + return tipc_eth_raw2addr(b, addr, msg); +} + /* Ethernet media registration info */ struct tipc_media eth_media_info = { .send_msg = tipc_l2_send_msg, @@ -78,6 +91,7 @@ struct tipc_media eth_media_info = { .addr2str = tipc_eth_addr2str, .addr2msg = tipc_eth_addr2msg, .msg2addr = tipc_eth_msg2addr, + .raw2addr = tipc_eth_raw2addr, .priority = TIPC_DEF_LINK_PRI, .tolerance = TIPC_DEF_LINK_TOL, .window = TIPC_DEF_LINK_WIN, @@ -85,4 +99,3 @@ struct tipc_media eth_media_info = { .hwaddr_len = ETH_ALEN, .name = "eth" }; - diff --git a/net/tipc/handler.c b/net/tipc/handler.c deleted file mode 100644 index 1fabf160501..00000000000 --- a/net/tipc/handler.c +++ /dev/null @@ -1,134 +0,0 @@ -/* - * net/tipc/handler.c: TIPC signal handling - * - * Copyright (c) 2000-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems - * All rights reserved. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions are met: - * - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * 3. Neither the names of the copyright holders nor the names of its - * contributors may be used to endorse or promote products derived from - * this software without specific prior written permission. - * - * Alternatively, this software may be distributed under the terms of the - * GNU General Public License ("GPL") version 2 as published by the Free - * Software Foundation. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" - * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE - * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR - * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF - * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS - * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN - * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) - * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE - * POSSIBILITY OF SUCH DAMAGE. - */ - -#include "core.h" - -struct queue_item { - struct list_head next_signal; - void (*handler) (unsigned long); - unsigned long data; -}; - -static struct kmem_cache *tipc_queue_item_cache; -static struct list_head signal_queue_head; -static DEFINE_SPINLOCK(qitem_lock); -static int handler_enabled __read_mostly; - -static void process_signal_queue(unsigned long dummy); - -static DECLARE_TASKLET_DISABLED(tipc_tasklet, process_signal_queue, 0); - - -unsigned int tipc_k_signal(Handler routine, unsigned long argument) -{ - struct queue_item *item; - - spin_lock_bh(&qitem_lock); - if (!handler_enabled) { - spin_unlock_bh(&qitem_lock); - return -ENOPROTOOPT; - } - - item = kmem_cache_alloc(tipc_queue_item_cache, GFP_ATOMIC); - if (!item) { - pr_err("Signal queue out of memory\n"); - spin_unlock_bh(&qitem_lock); - return -ENOMEM; - } - item->handler = routine; - item->data = argument; - list_add_tail(&item->next_signal, &signal_queue_head); - spin_unlock_bh(&qitem_lock); - tasklet_schedule(&tipc_tasklet); - return 0; -} - -static void process_signal_queue(unsigned long dummy) -{ - struct queue_item *__volatile__ item; - struct list_head *l, *n; - - spin_lock_bh(&qitem_lock); - list_for_each_safe(l, n, &signal_queue_head) { - item = list_entry(l, struct queue_item, next_signal); - list_del(&item->next_signal); - spin_unlock_bh(&qitem_lock); - item->handler(item->data); - spin_lock_bh(&qitem_lock); - kmem_cache_free(tipc_queue_item_cache, item); - } - spin_unlock_bh(&qitem_lock); -} - -int tipc_handler_start(void) -{ - tipc_queue_item_cache = - kmem_cache_create("tipc_queue_items", sizeof(struct queue_item), - 0, SLAB_HWCACHE_ALIGN, NULL); - if (!tipc_queue_item_cache) - return -ENOMEM; - - INIT_LIST_HEAD(&signal_queue_head); - tasklet_enable(&tipc_tasklet); - handler_enabled = 1; - return 0; -} - -void tipc_handler_stop(void) -{ - struct list_head *l, *n; - struct queue_item *item; - - spin_lock_bh(&qitem_lock); - if (!handler_enabled) { - spin_unlock_bh(&qitem_lock); - return; - } - handler_enabled = 0; - spin_unlock_bh(&qitem_lock); - - tasklet_kill(&tipc_tasklet); - - spin_lock_bh(&qitem_lock); - list_for_each_safe(l, n, &signal_queue_head) { - item = list_entry(l, struct queue_item, next_signal); - list_del(&item->next_signal); - kmem_cache_free(tipc_queue_item_cache, item); - } - spin_unlock_bh(&qitem_lock); - - kmem_cache_destroy(tipc_queue_item_cache); -} diff --git a/net/tipc/ib_media.c b/net/tipc/ib_media.c index 844a77e2582..8522eef9c13 100644 --- a/net/tipc/ib_media.c +++ b/net/tipc/ib_media.c @@ -42,7 +42,7 @@ #include "core.h" #include "bearer.h" -/* convert InfiniBand address to string */ +/* convert InfiniBand address (media address format) media address to string */ static int tipc_ib_addr2str(struct tipc_media_addr *a, char *str_buf, int str_size) { @@ -54,23 +54,35 @@ static int tipc_ib_addr2str(struct tipc_media_addr *a, char *str_buf, return 0; } -/* convert InfiniBand address format to message header format */ -static int tipc_ib_addr2msg(struct tipc_media_addr *a, char *msg_area) +/* Convert from media address format to discovery message addr format */ +static int tipc_ib_addr2msg(char *msg, struct tipc_media_addr *addr) { - memset(msg_area, 0, TIPC_MEDIA_ADDR_SIZE); - msg_area[TIPC_MEDIA_TYPE_OFFSET] = TIPC_MEDIA_TYPE_IB; - memcpy(msg_area, a->value, INFINIBAND_ALEN); + memset(msg, 0, TIPC_MEDIA_ADDR_SIZE); + memcpy(msg, addr->value, INFINIBAND_ALEN); return 0; } -/* convert message header address format to InfiniBand format */ -static int tipc_ib_msg2addr(const struct tipc_bearer *tb_ptr, - struct tipc_media_addr *a, char *msg_area) +/* Convert raw InfiniBand address format to media addr format */ +static int tipc_ib_raw2addr(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg) { - tipc_l2_media_addr_set(tb_ptr, a, msg_area); + memset(addr, 0, sizeof(*addr)); + memcpy(addr->value, msg, INFINIBAND_ALEN); + addr->media_id = TIPC_MEDIA_TYPE_IB; + addr->broadcast = !memcmp(msg, b->bcast_addr.value, + INFINIBAND_ALEN); return 0; } +/* Convert discovery msg addr format to InfiniBand media addr format */ +static int tipc_ib_msg2addr(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg) +{ + return tipc_ib_raw2addr(b, addr, msg); +} + /* InfiniBand media registration info */ struct tipc_media ib_media_info = { .send_msg = tipc_l2_send_msg, @@ -79,6 +91,7 @@ struct tipc_media ib_media_info = { .addr2str = tipc_ib_addr2str, .addr2msg = tipc_ib_addr2msg, .msg2addr = tipc_ib_msg2addr, + .raw2addr = tipc_ib_raw2addr, .priority = TIPC_DEF_LINK_PRI, .tolerance = TIPC_DEF_LINK_TOL, .window = TIPC_DEF_LINK_WIN, @@ -86,4 +99,3 @@ struct tipc_media ib_media_info = { .hwaddr_len = INFINIBAND_ALEN, .name = "ib" }; - diff --git a/net/tipc/link.c b/net/tipc/link.c index c5190ab7529..ad2c57f5868 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -37,6 +37,7 @@ #include "core.h" #include "link.h" #include "port.h" +#include "socket.h" #include "name_distr.h" #include "discover.h" #include "config.h" @@ -101,9 +102,18 @@ static unsigned int align(unsigned int i) static void link_init_max_pkt(struct tipc_link *l_ptr) { + struct tipc_bearer *b_ptr; u32 max_pkt; - max_pkt = (l_ptr->b_ptr->mtu & ~3); + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[l_ptr->bearer_id]); + if (!b_ptr) { + rcu_read_unlock(); + return; + } + max_pkt = (b_ptr->mtu & ~3); + rcu_read_unlock(); + if (max_pkt > MAX_MSG_SIZE) max_pkt = MAX_MSG_SIZE; @@ -248,7 +258,7 @@ struct tipc_link *tipc_link_create(struct tipc_node *n_ptr, l_ptr->owner = n_ptr; l_ptr->checkpoint = 1; l_ptr->peer_session = INVALID_SESSION; - l_ptr->b_ptr = b_ptr; + l_ptr->bearer_id = b_ptr->identity; link_set_supervision_props(l_ptr, b_ptr->tolerance); l_ptr->state = RESET_UNKNOWN; @@ -263,6 +273,7 @@ struct tipc_link *tipc_link_create(struct tipc_node *n_ptr, l_ptr->priority = b_ptr->priority; tipc_link_set_queue_limits(l_ptr, b_ptr->window); + l_ptr->net_plane = b_ptr->net_plane; link_init_max_pkt(l_ptr); l_ptr->next_out_no = 1; @@ -287,14 +298,14 @@ void tipc_link_delete_list(unsigned int bearer_id, bool shutting_down) rcu_read_lock(); list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { - spin_lock_bh(&n_ptr->lock); + tipc_node_lock(n_ptr); l_ptr = n_ptr->links[bearer_id]; if (l_ptr) { tipc_link_reset(l_ptr); if (shutting_down || !tipc_node_is_up(n_ptr)) { tipc_node_detach_link(l_ptr->owner, l_ptr); tipc_link_reset_fragments(l_ptr); - spin_unlock_bh(&n_ptr->lock); + tipc_node_unlock(n_ptr); /* Nobody else can access this link now: */ del_timer_sync(&l_ptr->timer); @@ -302,12 +313,12 @@ void tipc_link_delete_list(unsigned int bearer_id, bool shutting_down) } else { /* Detach/delete when failover is finished: */ l_ptr->flags |= LINK_STOPPED; - spin_unlock_bh(&n_ptr->lock); + tipc_node_unlock(n_ptr); del_timer_sync(&l_ptr->timer); } continue; } - spin_unlock_bh(&n_ptr->lock); + tipc_node_unlock(n_ptr); } rcu_read_unlock(); } @@ -388,9 +399,8 @@ static void link_release_outqueue(struct tipc_link *l_ptr) */ void tipc_link_reset_fragments(struct tipc_link *l_ptr) { - kfree_skb(l_ptr->reasm_head); - l_ptr->reasm_head = NULL; - l_ptr->reasm_tail = NULL; + kfree_skb(l_ptr->reasm_buf); + l_ptr->reasm_buf = NULL; } /** @@ -426,7 +436,7 @@ void tipc_link_reset(struct tipc_link *l_ptr) return; tipc_node_link_down(l_ptr->owner, l_ptr); - tipc_bearer_remove_dest(l_ptr->b_ptr, l_ptr->addr); + tipc_bearer_remove_dest(l_ptr->bearer_id, l_ptr->addr); if (was_active_link && tipc_node_active_links(l_ptr->owner)) { l_ptr->reset_checkpoint = checkpoint; @@ -464,11 +474,11 @@ void tipc_link_reset_list(unsigned int bearer_id) rcu_read_lock(); list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { - spin_lock_bh(&n_ptr->lock); + tipc_node_lock(n_ptr); l_ptr = n_ptr->links[bearer_id]; if (l_ptr) tipc_link_reset(l_ptr); - spin_unlock_bh(&n_ptr->lock); + tipc_node_unlock(n_ptr); } rcu_read_unlock(); } @@ -477,7 +487,7 @@ static void link_activate(struct tipc_link *l_ptr) { l_ptr->next_in_no = l_ptr->stats.recv_info = 1; tipc_node_link_up(l_ptr->owner, l_ptr); - tipc_bearer_add_dest(l_ptr->b_ptr, l_ptr->addr); + tipc_bearer_add_dest(l_ptr->bearer_id, l_ptr->addr); } /** @@ -777,7 +787,7 @@ int __tipc_link_xmit(struct tipc_link *l_ptr, struct sk_buff *buf) if (likely(!link_congested(l_ptr))) { link_add_to_outqueue(l_ptr, buf, msg); - tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr); + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); l_ptr->unacked_window = 0; return dsz; } @@ -825,7 +835,6 @@ int tipc_link_xmit(struct sk_buff *buf, u32 dest, u32 selector) struct tipc_node *n_ptr; int res = -ELINKCONG; - read_lock_bh(&tipc_net_lock); n_ptr = tipc_node_find(dest); if (n_ptr) { tipc_node_lock(n_ptr); @@ -838,7 +847,6 @@ int tipc_link_xmit(struct sk_buff *buf, u32 dest, u32 selector) } else { kfree_skb(buf); } - read_unlock_bh(&tipc_net_lock); return res; } @@ -902,7 +910,6 @@ void tipc_link_names_xmit(struct list_head *message_list, u32 dest) if (list_empty(message_list)) return; - read_lock_bh(&tipc_net_lock); n_ptr = tipc_node_find(dest); if (n_ptr) { tipc_node_lock(n_ptr); @@ -917,7 +924,6 @@ void tipc_link_names_xmit(struct list_head *message_list, u32 dest) } tipc_node_unlock(n_ptr); } - read_unlock_bh(&tipc_net_lock); /* discard the messages if they couldn't be sent */ list_for_each_safe(buf, temp_buf, ((struct sk_buff *)message_list)) { @@ -941,7 +947,7 @@ static int tipc_link_xmit_fast(struct tipc_link *l_ptr, struct sk_buff *buf, if (likely(!link_congested(l_ptr))) { if (likely(msg_size(msg) <= l_ptr->max_pkt)) { link_add_to_outqueue(l_ptr, buf, msg); - tipc_bearer_send(l_ptr->b_ptr, buf, + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); l_ptr->unacked_window = 0; return res; @@ -979,7 +985,6 @@ again: if (unlikely(res < 0)) return res; - read_lock_bh(&tipc_net_lock); node = tipc_node_find(destaddr); if (likely(node)) { tipc_node_lock(node); @@ -990,7 +995,6 @@ again: &sender->max_pkt); exit: tipc_node_unlock(node); - read_unlock_bh(&tipc_net_lock); return res; } @@ -1007,7 +1011,6 @@ exit: */ sender->max_pkt = l_ptr->max_pkt; tipc_node_unlock(node); - read_unlock_bh(&tipc_net_lock); if ((msg_hdr_sz(hdr) + res) <= sender->max_pkt) @@ -1018,7 +1021,6 @@ exit: } tipc_node_unlock(node); } - read_unlock_bh(&tipc_net_lock); /* Couldn't find a link to the destination node */ kfree_skb(buf); @@ -1204,7 +1206,7 @@ static u32 tipc_link_push_packet(struct tipc_link *l_ptr) if (r_q_size && buf) { msg_set_ack(buf_msg(buf), mod(l_ptr->next_in_no - 1)); msg_set_bcast_ack(buf_msg(buf), l_ptr->owner->bclink.last_in); - tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr); + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); l_ptr->retransm_queue_head = mod(++r_q_head); l_ptr->retransm_queue_size = --r_q_size; l_ptr->stats.retransmitted++; @@ -1216,7 +1218,7 @@ static u32 tipc_link_push_packet(struct tipc_link *l_ptr) if (buf) { msg_set_ack(buf_msg(buf), mod(l_ptr->next_in_no - 1)); msg_set_bcast_ack(buf_msg(buf), l_ptr->owner->bclink.last_in); - tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr); + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); l_ptr->unacked_window = 0; kfree_skb(buf); l_ptr->proto_msg_queue = NULL; @@ -1233,7 +1235,8 @@ static u32 tipc_link_push_packet(struct tipc_link *l_ptr) if (mod(next - first) < l_ptr->queue_limit[0]) { msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); - tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr); + tipc_bearer_send(l_ptr->bearer_id, buf, + &l_ptr->media_addr); if (msg_user(msg) == MSG_BUNDLER) msg_set_type(msg, CLOSED_MSG); l_ptr->next_out = buf->next; @@ -1256,33 +1259,24 @@ void tipc_link_push_queue(struct tipc_link *l_ptr) } while (!res); } -static void link_reset_all(unsigned long addr) +void tipc_link_reset_all(struct tipc_node *node) { - struct tipc_node *n_ptr; char addr_string[16]; u32 i; - read_lock_bh(&tipc_net_lock); - n_ptr = tipc_node_find((u32)addr); - if (!n_ptr) { - read_unlock_bh(&tipc_net_lock); - return; /* node no longer exists */ - } - - tipc_node_lock(n_ptr); + tipc_node_lock(node); pr_warn("Resetting all links to %s\n", - tipc_addr_string_fill(addr_string, n_ptr->addr)); + tipc_addr_string_fill(addr_string, node->addr)); for (i = 0; i < MAX_BEARERS; i++) { - if (n_ptr->links[i]) { - link_print(n_ptr->links[i], "Resetting link\n"); - tipc_link_reset(n_ptr->links[i]); + if (node->links[i]) { + link_print(node->links[i], "Resetting link\n"); + tipc_link_reset(node->links[i]); } } - tipc_node_unlock(n_ptr); - read_unlock_bh(&tipc_net_lock); + tipc_node_unlock(node); } static void link_retransmit_failure(struct tipc_link *l_ptr, @@ -1319,10 +1313,9 @@ static void link_retransmit_failure(struct tipc_link *l_ptr, n_ptr->bclink.oos_state, n_ptr->bclink.last_sent); - tipc_k_signal((Handler)link_reset_all, (unsigned long)n_ptr->addr); - tipc_node_unlock(n_ptr); + tipc_bclink_set_flags(TIPC_BCLINK_RESET); l_ptr->stale_count = 0; } } @@ -1352,7 +1345,7 @@ void tipc_link_retransmit(struct tipc_link *l_ptr, struct sk_buff *buf, msg = buf_msg(buf); msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); - tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr); + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); buf = buf->next; retransmits--; l_ptr->stats.retransmitted++; @@ -1440,14 +1433,13 @@ static int link_recv_buf_validate(struct sk_buff *buf) /** * tipc_rcv - process TIPC packets/messages arriving from off-node * @head: pointer to message buffer chain - * @tb_ptr: pointer to bearer message arrived on + * @b_ptr: pointer to bearer message arrived on * * Invoked with no locks held. Bearer pointer must point to a valid bearer * structure (i.e. cannot be NULL), but bearer can be inactive. */ void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr) { - read_lock_bh(&tipc_net_lock); while (head) { struct tipc_node *n_ptr; struct tipc_link *l_ptr; @@ -1497,14 +1489,14 @@ void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr) goto unlock_discard; /* Verify that communication with node is currently allowed */ - if ((n_ptr->block_setup & WAIT_PEER_DOWN) && - msg_user(msg) == LINK_PROTOCOL && - (msg_type(msg) == RESET_MSG || - msg_type(msg) == ACTIVATE_MSG) && - !msg_redundant_link(msg)) - n_ptr->block_setup &= ~WAIT_PEER_DOWN; - - if (n_ptr->block_setup) + if ((n_ptr->action_flags & TIPC_WAIT_PEER_LINKS_DOWN) && + msg_user(msg) == LINK_PROTOCOL && + (msg_type(msg) == RESET_MSG || + msg_type(msg) == ACTIVATE_MSG) && + !msg_redundant_link(msg)) + n_ptr->action_flags &= ~TIPC_WAIT_PEER_LINKS_DOWN; + + if (tipc_node_blocked(n_ptr)) goto unlock_discard; /* Validate message sequence number info */ @@ -1581,17 +1573,12 @@ void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr) } msg = buf_msg(buf); } else if (msg_user(msg) == MSG_FRAGMENTER) { - int rc; - l_ptr->stats.recv_fragments++; - rc = tipc_link_frag_rcv(&l_ptr->reasm_head, - &l_ptr->reasm_tail, - &buf); - if (rc == LINK_REASM_COMPLETE) { + if (tipc_buf_append(&l_ptr->reasm_buf, &buf)) { l_ptr->stats.recv_fragmented++; msg = buf_msg(buf); } else { - if (rc == LINK_REASM_ERROR) + if (!l_ptr->reasm_buf) tipc_link_reset(l_ptr); tipc_node_unlock(n_ptr); continue; @@ -1604,7 +1591,7 @@ void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr) case TIPC_HIGH_IMPORTANCE: case TIPC_CRITICAL_IMPORTANCE: tipc_node_unlock(n_ptr); - tipc_port_rcv(buf); + tipc_sk_rcv(buf); continue; case MSG_BUNDLER: l_ptr->stats.recv_bundles++; @@ -1635,7 +1622,6 @@ unlock_discard: discard: kfree_skb(buf); } - read_unlock_bh(&tipc_net_lock); } /** @@ -1747,12 +1733,12 @@ void tipc_link_proto_xmit(struct tipc_link *l_ptr, u32 msg_typ, int probe_msg, return; /* Abort non-RESET send if communication with node is prohibited */ - if ((l_ptr->owner->block_setup) && (msg_typ != RESET_MSG)) + if ((tipc_node_blocked(l_ptr->owner)) && (msg_typ != RESET_MSG)) return; /* Create protocol message with "out-of-sequence" sequence number */ msg_set_type(msg, msg_typ); - msg_set_net_plane(msg, l_ptr->b_ptr->net_plane); + msg_set_net_plane(msg, l_ptr->net_plane); msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); msg_set_last_bcast(msg, tipc_bclink_get_last_sent()); @@ -1818,7 +1804,7 @@ void tipc_link_proto_xmit(struct tipc_link *l_ptr, u32 msg_typ, int probe_msg, skb_copy_to_linear_data(buf, msg, sizeof(l_ptr->proto_msg)); buf->priority = TC_PRIO_CONTROL; - tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr); + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); l_ptr->unacked_window = 0; kfree_skb(buf); } @@ -1840,12 +1826,9 @@ static void tipc_link_proto_rcv(struct tipc_link *l_ptr, struct sk_buff *buf) if (l_ptr->exp_msg_count) goto exit; - /* record unnumbered packet arrival (force mismatch on next timeout) */ - l_ptr->checkpoint--; - - if (l_ptr->b_ptr->net_plane != msg_net_plane(msg)) + if (l_ptr->net_plane != msg_net_plane(msg)) if (tipc_own_addr > msg_prevnode(msg)) - l_ptr->b_ptr->net_plane = msg_net_plane(msg); + l_ptr->net_plane = msg_net_plane(msg); switch (msg_type(msg)) { @@ -1862,7 +1845,7 @@ static void tipc_link_proto_rcv(struct tipc_link *l_ptr, struct sk_buff *buf) * peer has lost contact -- don't allow peer's links * to reactivate before we recognize loss & clean up */ - l_ptr->owner->block_setup = WAIT_NODE_DOWN; + l_ptr->owner->action_flags |= TIPC_WAIT_OWN_LINKS_DOWN; } link_state_event(l_ptr, RESET_MSG); @@ -1918,6 +1901,10 @@ static void tipc_link_proto_rcv(struct tipc_link *l_ptr, struct sk_buff *buf) tipc_link_reset(l_ptr); /* Enforce change to take effect */ break; } + + /* Record reception; force mismatch at next timeout: */ + l_ptr->checkpoint--; + link_state_event(l_ptr, TRAFFIC_MSG_EVT); l_ptr->stats.recv_states++; if (link_reset_unknown(l_ptr)) @@ -2177,9 +2164,7 @@ static struct sk_buff *tipc_link_failover_rcv(struct tipc_link *l_ptr, } if (msg_user(msg) == MSG_FRAGMENTER) { l_ptr->stats.recv_fragments++; - tipc_link_frag_rcv(&l_ptr->reasm_head, - &l_ptr->reasm_tail, - &buf); + tipc_buf_append(&l_ptr->reasm_buf, &buf); } } exit: @@ -2317,53 +2302,6 @@ static int tipc_link_frag_xmit(struct tipc_link *l_ptr, struct sk_buff *buf) return dsz; } -/* tipc_link_frag_rcv(): Called with node lock on. Returns - * the reassembled buffer if message is complete. - */ -int tipc_link_frag_rcv(struct sk_buff **head, struct sk_buff **tail, - struct sk_buff **fbuf) -{ - struct sk_buff *frag = *fbuf; - struct tipc_msg *msg = buf_msg(frag); - u32 fragid = msg_type(msg); - bool headstolen; - int delta; - - skb_pull(frag, msg_hdr_sz(msg)); - if (fragid == FIRST_FRAGMENT) { - if (*head || skb_unclone(frag, GFP_ATOMIC)) - goto out_free; - *head = frag; - skb_frag_list_init(*head); - *fbuf = NULL; - return 0; - } else if (*head && - skb_try_coalesce(*head, frag, &headstolen, &delta)) { - kfree_skb_partial(frag, headstolen); - } else { - if (!*head) - goto out_free; - if (!skb_has_frag_list(*head)) - skb_shinfo(*head)->frag_list = frag; - else - (*tail)->next = frag; - *tail = frag; - (*head)->truesize += frag->truesize; - } - if (fragid == LAST_FRAGMENT) { - *fbuf = *head; - *tail = *head = NULL; - return LINK_REASM_COMPLETE; - } - *fbuf = NULL; - return 0; -out_free: - pr_warn_ratelimited("Link unable to reassemble fragmented message\n"); - kfree_skb(*fbuf); - *fbuf = NULL; - return LINK_REASM_ERROR; -} - static void link_set_supervision_props(struct tipc_link *l_ptr, u32 tolerance) { if ((tolerance < TIPC_MIN_LINK_TOL) || (tolerance > TIPC_MAX_LINK_TOL)) @@ -2397,8 +2335,6 @@ void tipc_link_set_queue_limits(struct tipc_link *l_ptr, u32 window) /* tipc_link_find_owner - locate owner node of link by link's name * @name: pointer to link name string * @bearer_id: pointer to index in 'node->links' array where the link was found. - * Caller must hold 'tipc_net_lock' to ensure node and bearer are not deleted; - * this also prevents link deletion. * * Returns pointer to node owning the link, or 0 if no matching link is found. */ @@ -2460,7 +2396,7 @@ static int link_value_is_valid(u16 cmd, u32 new_value) * @new_value: new value of link, bearer, or media setting * @cmd: which link, bearer, or media attribute to set (TIPC_CMD_SET_LINK_*) * - * Caller must hold 'tipc_net_lock' to ensure link/bearer/media is not deleted. + * Caller must hold RTNL lock to ensure link/bearer/media is not deleted. * * Returns 0 if value updated and negative value on error. */ @@ -2566,9 +2502,7 @@ struct sk_buff *tipc_link_cmd_config(const void *req_tlv_area, int req_tlv_space " (cannot change setting on broadcast link)"); } - read_lock_bh(&tipc_net_lock); res = link_cmd_set_value(args->name, new_value, cmd); - read_unlock_bh(&tipc_net_lock); if (res) return tipc_cfg_reply_error_string("cannot change link setting"); @@ -2602,22 +2536,18 @@ struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_ return tipc_cfg_reply_error_string("link not found"); return tipc_cfg_reply_none(); } - read_lock_bh(&tipc_net_lock); node = tipc_link_find_owner(link_name, &bearer_id); - if (!node) { - read_unlock_bh(&tipc_net_lock); + if (!node) return tipc_cfg_reply_error_string("link not found"); - } + tipc_node_lock(node); l_ptr = node->links[bearer_id]; if (!l_ptr) { tipc_node_unlock(node); - read_unlock_bh(&tipc_net_lock); return tipc_cfg_reply_error_string("link not found"); } link_reset_statistics(l_ptr); tipc_node_unlock(node); - read_unlock_bh(&tipc_net_lock); return tipc_cfg_reply_none(); } @@ -2650,18 +2580,15 @@ static int tipc_link_stats(const char *name, char *buf, const u32 buf_size) if (!strcmp(name, tipc_bclink_name)) return tipc_bclink_stats(buf, buf_size); - read_lock_bh(&tipc_net_lock); node = tipc_link_find_owner(name, &bearer_id); - if (!node) { - read_unlock_bh(&tipc_net_lock); + if (!node) return 0; - } + tipc_node_lock(node); l = node->links[bearer_id]; if (!l) { tipc_node_unlock(node); - read_unlock_bh(&tipc_net_lock); return 0; } @@ -2727,7 +2654,6 @@ static int tipc_link_stats(const char *name, char *buf, const u32 buf_size) (s->accu_queue_sz / s->queue_sz_counts) : 0); tipc_node_unlock(node); - read_unlock_bh(&tipc_net_lock); return ret; } @@ -2778,7 +2704,6 @@ u32 tipc_link_get_max_pkt(u32 dest, u32 selector) if (dest == tipc_own_addr) return MAX_MSG_SIZE; - read_lock_bh(&tipc_net_lock); n_ptr = tipc_node_find(dest); if (n_ptr) { tipc_node_lock(n_ptr); @@ -2787,13 +2712,18 @@ u32 tipc_link_get_max_pkt(u32 dest, u32 selector) res = l_ptr->max_pkt; tipc_node_unlock(n_ptr); } - read_unlock_bh(&tipc_net_lock); return res; } static void link_print(struct tipc_link *l_ptr, const char *str) { - pr_info("%s Link %x<%s>:", str, l_ptr->addr, l_ptr->b_ptr->name); + struct tipc_bearer *b_ptr; + + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[l_ptr->bearer_id]); + if (b_ptr) + pr_info("%s Link %x<%s>:", str, l_ptr->addr, b_ptr->name); + rcu_read_unlock(); if (link_working_unknown(l_ptr)) pr_cont(":WU\n"); diff --git a/net/tipc/link.h b/net/tipc/link.h index 8c0b49b5b2e..200d518b218 100644 --- a/net/tipc/link.h +++ b/net/tipc/link.h @@ -40,11 +40,6 @@ #include "msg.h" #include "node.h" -/* Link reassembly status codes - */ -#define LINK_REASM_ERROR -1 -#define LINK_REASM_COMPLETE 1 - /* Out-of-range value for link sequence numbers */ #define INVALID_LINK_SEQ 0x10000 @@ -107,7 +102,7 @@ struct tipc_stats { * @checkpoint: reference point for triggering link continuity checking * @peer_session: link session # being used by peer end of link * @peer_bearer_id: bearer id used by link's peer endpoint - * @b_ptr: pointer to bearer used by link + * @bearer_id: local bearer id used by link * @tolerance: minimum link continuity loss needed to reset link [in ms] * @continuity_interval: link continuity testing interval [in ms] * @abort_limit: # of unacknowledged continuity probes needed to reset link @@ -116,6 +111,7 @@ struct tipc_stats { * @proto_msg: template for control messages generated by link * @pmsg: convenience pointer to "proto_msg" field * @priority: current link priority + * @net_plane: current link network plane ('A' through 'H') * @queue_limit: outbound message queue congestion thresholds (indexed by user) * @exp_msg_count: # of tunnelled messages expected during link changeover * @reset_checkpoint: seq # of last acknowledged message at time of link reset @@ -139,8 +135,7 @@ struct tipc_stats { * @next_out: ptr to first unsent outbound message in queue * @waiting_ports: linked list of ports waiting for link congestion to abate * @long_msg_seq_no: next identifier to use for outbound fragmented messages - * @reasm_head: list head of partially reassembled inbound message fragments - * @reasm_tail: last fragment received + * @reasm_buf: head of partially reassembled inbound message fragments * @stats: collects statistics regarding link activity */ struct tipc_link { @@ -155,7 +150,7 @@ struct tipc_link { u32 checkpoint; u32 peer_session; u32 peer_bearer_id; - struct tipc_bearer *b_ptr; + u32 bearer_id; u32 tolerance; u32 continuity_interval; u32 abort_limit; @@ -167,6 +162,7 @@ struct tipc_link { } proto_msg; struct tipc_msg *pmsg; u32 priority; + char net_plane; u32 queue_limit[15]; /* queue_limit[0]==window limit */ /* Changeover */ @@ -202,8 +198,7 @@ struct tipc_link { /* Fragmentation/reassembly */ u32 long_msg_seq_no; - struct sk_buff *reasm_head; - struct sk_buff *reasm_tail; + struct sk_buff *reasm_buf; /* Statistics */ struct tipc_stats stats; @@ -228,6 +223,7 @@ struct sk_buff *tipc_link_cmd_show_stats(const void *req_tlv_area, int req_tlv_space); struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_space); +void tipc_link_reset_all(struct tipc_node *node); void tipc_link_reset(struct tipc_link *l_ptr); void tipc_link_reset_list(unsigned int bearer_id); int tipc_link_xmit(struct sk_buff *buf, u32 dest, u32 selector); @@ -239,9 +235,6 @@ int tipc_link_iovec_xmit_fast(struct tipc_port *sender, struct iovec const *msg_sect, unsigned int len, u32 destnode); void tipc_link_bundle_rcv(struct sk_buff *buf); -int tipc_link_frag_rcv(struct sk_buff **reasm_head, - struct sk_buff **reasm_tail, - struct sk_buff **fbuf); void tipc_link_proto_xmit(struct tipc_link *l_ptr, u32 msg_typ, int prob, u32 gap, u32 tolerance, u32 priority, u32 acked_mtu); void tipc_link_push_queue(struct tipc_link *l_ptr); diff --git a/net/tipc/msg.c b/net/tipc/msg.c index e525f8ce1de..8be6e94a1ca 100644 --- a/net/tipc/msg.c +++ b/net/tipc/msg.c @@ -1,7 +1,7 @@ /* * net/tipc/msg.c: TIPC message header routines * - * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2000-2006, 2014, Ericsson AB * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * @@ -99,3 +99,56 @@ int tipc_msg_build(struct tipc_msg *hdr, struct iovec const *msg_sect, } return dsz; } + +/* tipc_buf_append(): Append a buffer to the fragment list of another buffer + * Let first buffer become head buffer + * Returns 1 and sets *buf to headbuf if chain is complete, otherwise 0 + * Leaves headbuf pointer at NULL if failure + */ +int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf) +{ + struct sk_buff *head = *headbuf; + struct sk_buff *frag = *buf; + struct sk_buff *tail; + struct tipc_msg *msg = buf_msg(frag); + u32 fragid = msg_type(msg); + bool headstolen; + int delta; + + skb_pull(frag, msg_hdr_sz(msg)); + + if (fragid == FIRST_FRAGMENT) { + if (head || skb_unclone(frag, GFP_ATOMIC)) + goto out_free; + head = *headbuf = frag; + skb_frag_list_init(head); + return 0; + } + if (!head) + goto out_free; + tail = TIPC_SKB_CB(head)->tail; + if (skb_try_coalesce(head, frag, &headstolen, &delta)) { + kfree_skb_partial(frag, headstolen); + } else { + if (!skb_has_frag_list(head)) + skb_shinfo(head)->frag_list = frag; + else + tail->next = frag; + head->truesize += frag->truesize; + head->data_len += frag->len; + head->len += frag->len; + TIPC_SKB_CB(head)->tail = frag; + } + if (fragid == LAST_FRAGMENT) { + *buf = head; + TIPC_SKB_CB(head)->tail = NULL; + *headbuf = NULL; + return 1; + } + *buf = NULL; + return 0; +out_free: + pr_warn_ratelimited("Unable to build fragment list\n"); + kfree_skb(*buf); + return 0; +} diff --git a/net/tipc/msg.h b/net/tipc/msg.h index 76d1269b944..503511903d1 100644 --- a/net/tipc/msg.h +++ b/net/tipc/msg.h @@ -1,7 +1,7 @@ /* * net/tipc/msg.h: Include file for TIPC message header routines * - * Copyright (c) 2000-2007, Ericsson AB + * Copyright (c) 2000-2007, 2014, Ericsson AB * Copyright (c) 2005-2008, 2010-2011, Wind River Systems * All rights reserved. * @@ -711,4 +711,7 @@ void tipc_msg_init(struct tipc_msg *m, u32 user, u32 type, u32 hsize, u32 destnode); int tipc_msg_build(struct tipc_msg *hdr, struct iovec const *msg_sect, unsigned int len, int max_size, struct sk_buff **buf); + +int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf); + #endif diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c index aff8041dc15..8ce730984aa 100644 --- a/net/tipc/name_distr.c +++ b/net/tipc/name_distr.c @@ -38,34 +38,6 @@ #include "link.h" #include "name_distr.h" -#define ITEM_SIZE sizeof(struct distr_item) - -/** - * struct distr_item - publication info distributed to other nodes - * @type: name sequence type - * @lower: name sequence lower bound - * @upper: name sequence upper bound - * @ref: publishing port reference - * @key: publication key - * - * ===> All fields are stored in network byte order. <=== - * - * First 3 fields identify (name or) name sequence being published. - * Reference field uniquely identifies port that published name sequence. - * Key field uniquely identifies publication, in the event a port has - * multiple publications of the same name sequence. - * - * Note: There is no field that identifies the publishing node because it is - * the same for all items contained within a publication message. - */ -struct distr_item { - __be32 type; - __be32 lower; - __be32 upper; - __be32 ref; - __be32 key; -}; - /** * struct publ_list - list of publications made by this node * @list: circular list of publications @@ -127,7 +99,7 @@ static struct sk_buff *named_prepare_buf(u32 type, u32 size, u32 dest) return buf; } -static void named_cluster_distribute(struct sk_buff *buf) +void named_cluster_distribute(struct sk_buff *buf) { struct sk_buff *buf_copy; struct tipc_node *n_ptr; @@ -135,18 +107,18 @@ static void named_cluster_distribute(struct sk_buff *buf) rcu_read_lock(); list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { - spin_lock_bh(&n_ptr->lock); + tipc_node_lock(n_ptr); l_ptr = n_ptr->active_links[n_ptr->addr & 1]; if (l_ptr) { buf_copy = skb_copy(buf, GFP_ATOMIC); if (!buf_copy) { - spin_unlock_bh(&n_ptr->lock); + tipc_node_unlock(n_ptr); break; } msg_set_destnode(buf_msg(buf_copy), n_ptr->addr); __tipc_link_xmit(l_ptr, buf_copy); } - spin_unlock_bh(&n_ptr->lock); + tipc_node_unlock(n_ptr); } rcu_read_unlock(); @@ -156,7 +128,7 @@ static void named_cluster_distribute(struct sk_buff *buf) /** * tipc_named_publish - tell other nodes about a new publication by this node */ -void tipc_named_publish(struct publication *publ) +struct sk_buff *tipc_named_publish(struct publication *publ) { struct sk_buff *buf; struct distr_item *item; @@ -165,23 +137,23 @@ void tipc_named_publish(struct publication *publ) publ_lists[publ->scope]->size++; if (publ->scope == TIPC_NODE_SCOPE) - return; + return NULL; buf = named_prepare_buf(PUBLICATION, ITEM_SIZE, 0); if (!buf) { pr_warn("Publication distribution failure\n"); - return; + return NULL; } item = (struct distr_item *)msg_data(buf_msg(buf)); publ_to_item(item, publ); - named_cluster_distribute(buf); + return buf; } /** * tipc_named_withdraw - tell other nodes about a withdrawn publication by this node */ -void tipc_named_withdraw(struct publication *publ) +struct sk_buff *tipc_named_withdraw(struct publication *publ) { struct sk_buff *buf; struct distr_item *item; @@ -190,17 +162,17 @@ void tipc_named_withdraw(struct publication *publ) publ_lists[publ->scope]->size--; if (publ->scope == TIPC_NODE_SCOPE) - return; + return NULL; buf = named_prepare_buf(WITHDRAWAL, ITEM_SIZE, 0); if (!buf) { pr_warn("Withdrawal distribution failure\n"); - return; + return NULL; } item = (struct distr_item *)msg_data(buf_msg(buf)); publ_to_item(item, publ); - named_cluster_distribute(buf); + return buf; } /* @@ -239,31 +211,9 @@ static void named_distribute(struct list_head *message_list, u32 node, /** * tipc_named_node_up - tell specified node about all publications by this node */ -void tipc_named_node_up(unsigned long nodearg) +void tipc_named_node_up(u32 max_item_buf, u32 node) { - struct tipc_node *n_ptr; - struct tipc_link *l_ptr; - struct list_head message_list; - u32 node = (u32)nodearg; - u32 max_item_buf = 0; - - /* compute maximum amount of publication data to send per message */ - read_lock_bh(&tipc_net_lock); - n_ptr = tipc_node_find(node); - if (n_ptr) { - tipc_node_lock(n_ptr); - l_ptr = n_ptr->active_links[0]; - if (l_ptr) - max_item_buf = ((l_ptr->max_pkt - INT_H_SIZE) / - ITEM_SIZE) * ITEM_SIZE; - tipc_node_unlock(n_ptr); - } - read_unlock_bh(&tipc_net_lock); - if (!max_item_buf) - return; - - /* create list of publication messages, then send them as a unit */ - INIT_LIST_HEAD(&message_list); + LIST_HEAD(message_list); read_lock_bh(&tipc_nametbl_lock); named_distribute(&message_list, node, &publ_cluster, max_item_buf); diff --git a/net/tipc/name_distr.h b/net/tipc/name_distr.h index 9b312ccfd43..b2eed4ec152 100644 --- a/net/tipc/name_distr.h +++ b/net/tipc/name_distr.h @@ -39,9 +39,38 @@ #include "name_table.h" -void tipc_named_publish(struct publication *publ); -void tipc_named_withdraw(struct publication *publ); -void tipc_named_node_up(unsigned long node); +#define ITEM_SIZE sizeof(struct distr_item) + +/** + * struct distr_item - publication info distributed to other nodes + * @type: name sequence type + * @lower: name sequence lower bound + * @upper: name sequence upper bound + * @ref: publishing port reference + * @key: publication key + * + * ===> All fields are stored in network byte order. <=== + * + * First 3 fields identify (name or) name sequence being published. + * Reference field uniquely identifies port that published name sequence. + * Key field uniquely identifies publication, in the event a port has + * multiple publications of the same name sequence. + * + * Note: There is no field that identifies the publishing node because it is + * the same for all items contained within a publication message. + */ +struct distr_item { + __be32 type; + __be32 lower; + __be32 upper; + __be32 ref; + __be32 key; +}; + +struct sk_buff *tipc_named_publish(struct publication *publ); +struct sk_buff *tipc_named_withdraw(struct publication *publ); +void named_cluster_distribute(struct sk_buff *buf); +void tipc_named_node_up(u32 max_item_buf, u32 node); void tipc_named_rcv(struct sk_buff *buf); void tipc_named_reinit(void); diff --git a/net/tipc/name_table.c b/net/tipc/name_table.c index 042e8e3cabc..9d7d37d9518 100644 --- a/net/tipc/name_table.c +++ b/net/tipc/name_table.c @@ -664,6 +664,7 @@ struct publication *tipc_nametbl_publish(u32 type, u32 lower, u32 upper, u32 scope, u32 port_ref, u32 key) { struct publication *publ; + struct sk_buff *buf = NULL; if (table.local_publ_count >= TIPC_MAX_PUBLICATIONS) { pr_warn("Publication failed, local publication limit reached (%u)\n", @@ -676,9 +677,12 @@ struct publication *tipc_nametbl_publish(u32 type, u32 lower, u32 upper, tipc_own_addr, port_ref, key); if (likely(publ)) { table.local_publ_count++; - tipc_named_publish(publ); + buf = tipc_named_publish(publ); } write_unlock_bh(&tipc_nametbl_lock); + + if (buf) + named_cluster_distribute(buf); return publ; } @@ -688,15 +692,19 @@ struct publication *tipc_nametbl_publish(u32 type, u32 lower, u32 upper, int tipc_nametbl_withdraw(u32 type, u32 lower, u32 ref, u32 key) { struct publication *publ; + struct sk_buff *buf; write_lock_bh(&tipc_nametbl_lock); publ = tipc_nametbl_remove_publ(type, lower, tipc_own_addr, ref, key); if (likely(publ)) { table.local_publ_count--; - tipc_named_withdraw(publ); + buf = tipc_named_withdraw(publ); write_unlock_bh(&tipc_nametbl_lock); list_del_init(&publ->pport_list); kfree(publ); + + if (buf) + named_cluster_distribute(buf); return 1; } write_unlock_bh(&tipc_nametbl_lock); @@ -961,6 +969,7 @@ static void tipc_purge_publications(struct name_seq *seq) list_for_each_entry_safe(publ, safe, &info->zone_list, zone_list) { tipc_nametbl_remove_publ(publ->type, publ->lower, publ->node, publ->ref, publ->key); + kfree(publ); } } @@ -982,7 +991,6 @@ void tipc_nametbl_stop(void) hlist_for_each_entry_safe(seq, safe, seq_head, ns_list) { tipc_purge_publications(seq); } - continue; } kfree(table.types); table.types = NULL; diff --git a/net/tipc/net.c b/net/tipc/net.c index 4c564eb69e1..f64375e7f99 100644 --- a/net/tipc/net.c +++ b/net/tipc/net.c @@ -39,45 +39,41 @@ #include "name_distr.h" #include "subscr.h" #include "port.h" +#include "socket.h" #include "node.h" #include "config.h" /* * The TIPC locking policy is designed to ensure a very fine locking * granularity, permitting complete parallel access to individual - * port and node/link instances. The code consists of three major + * port and node/link instances. The code consists of four major * locking domains, each protected with their own disjunct set of locks. * - * 1: The routing hierarchy. - * Comprises the structures 'zone', 'cluster', 'node', 'link' - * and 'bearer'. The whole hierarchy is protected by a big - * read/write lock, tipc_net_lock, to enssure that nothing is added - * or removed while code is accessing any of these structures. - * This layer must not be called from the two others while they - * hold any of their own locks. - * Neither must it itself do any upcalls to the other two before - * it has released tipc_net_lock and other protective locks. + * 1: The bearer level. + * RTNL lock is used to serialize the process of configuring bearer + * on update side, and RCU lock is applied on read side to make + * bearer instance valid on both paths of message transmission and + * reception. * - * Within the tipc_net_lock domain there are two sub-domains;'node' and - * 'bearer', where local write operations are permitted, - * provided that those are protected by individual spin_locks - * per instance. Code holding tipc_net_lock(read) and a node spin_lock - * is permitted to poke around in both the node itself and its - * subordinate links. I.e, it can update link counters and queues, - * change link state, send protocol messages, and alter the - * "active_links" array in the node; but it can _not_ remove a link - * or a node from the overall structure. - * Correspondingly, individual bearers may change status within a - * tipc_net_lock(read), protected by an individual spin_lock ber bearer - * instance, but it needs tipc_net_lock(write) to remove/add any bearers. + * 2: The node and link level. + * All node instances are saved into two tipc_node_list and node_htable + * lists. The two lists are protected by node_list_lock on write side, + * and they are guarded with RCU lock on read side. Especially node + * instance is destroyed only when TIPC module is removed, and we can + * confirm that there has no any user who is accessing the node at the + * moment. Therefore, Except for iterating the two lists within RCU + * protection, it's no needed to hold RCU that we access node instance + * in other places. * + * In addition, all members in node structure including link instances + * are protected by node spin lock. * - * 2: The transport level of the protocol. - * This consists of the structures port, (and its user level - * representations, such as user_port and tipc_sock), reference and - * tipc_user (port.c, reg.c, socket.c). + * 3: The transport level of the protocol. + * This consists of the structures port, (and its user level + * representations, such as user_port and tipc_sock), reference and + * tipc_user (port.c, reg.c, socket.c). * - * This layer has four different locks: + * This layer has four different locks: * - The tipc_port spin_lock. This is protecting each port instance * from parallel data access and removal. Since we can not place * this lock in the port itself, it has been placed in the @@ -96,7 +92,7 @@ * There are two such lists; 'port_list', which is used for management, * and 'wait_list', which is used to queue ports during congestion. * - * 3: The name table (name_table.c, name_distr.c, subscription.c) + * 4: The name table (name_table.c, name_distr.c, subscription.c) * - There is one big read/write-lock (tipc_nametbl_lock) protecting the * overall name table structure. Nothing must be added/removed to * this structure without holding write access to it. @@ -108,8 +104,6 @@ * - A local spin_lock protecting the queue of subscriber events. */ -DEFINE_RWLOCK(tipc_net_lock); - static void net_route_named_msg(struct sk_buff *buf) { struct tipc_msg *msg = buf_msg(buf); @@ -148,7 +142,7 @@ void tipc_net_route_msg(struct sk_buff *buf) if (msg_mcast(msg)) tipc_port_mcast_rcv(buf, NULL); else if (msg_destport(msg)) - tipc_port_rcv(buf); + tipc_sk_rcv(buf); else net_route_named_msg(buf); return; @@ -171,22 +165,25 @@ void tipc_net_route_msg(struct sk_buff *buf) tipc_link_xmit(buf, dnode, msg_link_selector(msg)); } -void tipc_net_start(u32 addr) +int tipc_net_start(u32 addr) { char addr_string[16]; + int res; - write_lock_bh(&tipc_net_lock); tipc_own_addr = addr; tipc_named_reinit(); tipc_port_reinit(); - tipc_bclink_init(); - write_unlock_bh(&tipc_net_lock); + res = tipc_bclink_init(); + if (res) + return res; tipc_nametbl_publish(TIPC_CFG_SRV, tipc_own_addr, tipc_own_addr, TIPC_ZONE_SCOPE, 0, tipc_own_addr); + pr_info("Started in network mode\n"); pr_info("Own node address %s, network identity %u\n", tipc_addr_string_fill(addr_string, tipc_own_addr), tipc_net_id); + return 0; } void tipc_net_stop(void) @@ -195,11 +192,11 @@ void tipc_net_stop(void) return; tipc_nametbl_withdraw(TIPC_CFG_SRV, tipc_own_addr, 0, tipc_own_addr); - write_lock_bh(&tipc_net_lock); + rtnl_lock(); tipc_bearer_stop(); tipc_bclink_stop(); tipc_node_stop(); - write_unlock_bh(&tipc_net_lock); + rtnl_unlock(); pr_info("Left network mode\n"); } diff --git a/net/tipc/net.h b/net/tipc/net.h index 079daadb3f7..c6c2b46f7c2 100644 --- a/net/tipc/net.h +++ b/net/tipc/net.h @@ -37,11 +37,9 @@ #ifndef _TIPC_NET_H #define _TIPC_NET_H -extern rwlock_t tipc_net_lock; - void tipc_net_route_msg(struct sk_buff *buf); -void tipc_net_start(u32 addr); +int tipc_net_start(u32 addr); void tipc_net_stop(void); #endif diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c index 3aaf73de9e2..ad844d36534 100644 --- a/net/tipc/netlink.c +++ b/net/tipc/netlink.c @@ -47,7 +47,7 @@ static int handle_cmd(struct sk_buff *skb, struct genl_info *info) int hdr_space = nlmsg_total_size(GENL_HDRLEN + TIPC_GENL_HDRLEN); u16 cmd; - if ((req_userhdr->cmd & 0xC000) && (!capable(CAP_NET_ADMIN))) + if ((req_userhdr->cmd & 0xC000) && (!netlink_capable(skb, CAP_NET_ADMIN))) cmd = TIPC_CMD_NOT_NET_ADMIN; else cmd = req_userhdr->cmd; diff --git a/net/tipc/node.c b/net/tipc/node.c index 1d3a4999a70..5b44c3041be 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -108,7 +108,7 @@ struct tipc_node *tipc_node_create(u32 addr) break; } list_add_tail_rcu(&n_ptr->list, &temp_node->list); - n_ptr->block_setup = WAIT_PEER_DOWN; + n_ptr->action_flags = TIPC_WAIT_PEER_LINKS_DOWN; n_ptr->signature = INVALID_NODE_SIG; tipc_num_nodes++; @@ -144,11 +144,13 @@ void tipc_node_stop(void) void tipc_node_link_up(struct tipc_node *n_ptr, struct tipc_link *l_ptr) { struct tipc_link **active = &n_ptr->active_links[0]; + u32 addr = n_ptr->addr; n_ptr->working_links++; - + tipc_nametbl_publish(TIPC_LINK_STATE, addr, addr, TIPC_NODE_SCOPE, + l_ptr->bearer_id, addr); pr_info("Established link <%s> on network plane %c\n", - l_ptr->name, l_ptr->b_ptr->net_plane); + l_ptr->name, l_ptr->net_plane); if (!active[0]) { active[0] = active[1] = l_ptr; @@ -203,16 +205,18 @@ static void node_select_active_links(struct tipc_node *n_ptr) void tipc_node_link_down(struct tipc_node *n_ptr, struct tipc_link *l_ptr) { struct tipc_link **active; + u32 addr = n_ptr->addr; n_ptr->working_links--; + tipc_nametbl_withdraw(TIPC_LINK_STATE, addr, l_ptr->bearer_id, addr); if (!tipc_link_is_active(l_ptr)) { pr_info("Lost standby link <%s> on network plane %c\n", - l_ptr->name, l_ptr->b_ptr->net_plane); + l_ptr->name, l_ptr->net_plane); return; } pr_info("Lost link <%s> on network plane %c\n", - l_ptr->name, l_ptr->b_ptr->net_plane); + l_ptr->name, l_ptr->net_plane); active = &n_ptr->active_links[0]; if (active[0] == l_ptr) @@ -239,7 +243,7 @@ int tipc_node_is_up(struct tipc_node *n_ptr) void tipc_node_attach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr) { - n_ptr->links[l_ptr->b_ptr->identity] = l_ptr; + n_ptr->links[l_ptr->bearer_id] = l_ptr; spin_lock_bh(&node_list_lock); tipc_num_links++; spin_unlock_bh(&node_list_lock); @@ -263,26 +267,12 @@ void tipc_node_detach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr) static void node_established_contact(struct tipc_node *n_ptr) { - tipc_k_signal((Handler)tipc_named_node_up, n_ptr->addr); + n_ptr->action_flags |= TIPC_NOTIFY_NODE_UP; n_ptr->bclink.oos_state = 0; n_ptr->bclink.acked = tipc_bclink_get_last_sent(); tipc_bclink_add_node(n_ptr->addr); } -static void node_name_purge_complete(unsigned long node_addr) -{ - struct tipc_node *n_ptr; - - read_lock_bh(&tipc_net_lock); - n_ptr = tipc_node_find(node_addr); - if (n_ptr) { - tipc_node_lock(n_ptr); - n_ptr->block_setup &= ~WAIT_NAMES_GONE; - tipc_node_unlock(n_ptr); - } - read_unlock_bh(&tipc_net_lock); -} - static void node_lost_contact(struct tipc_node *n_ptr) { char addr_string[16]; @@ -296,10 +286,9 @@ static void node_lost_contact(struct tipc_node *n_ptr) kfree_skb_list(n_ptr->bclink.deferred_head); n_ptr->bclink.deferred_size = 0; - if (n_ptr->bclink.reasm_head) { - kfree_skb(n_ptr->bclink.reasm_head); - n_ptr->bclink.reasm_head = NULL; - n_ptr->bclink.reasm_tail = NULL; + if (n_ptr->bclink.reasm_buf) { + kfree_skb(n_ptr->bclink.reasm_buf); + n_ptr->bclink.reasm_buf = NULL; } tipc_bclink_remove_node(n_ptr->addr); @@ -318,12 +307,13 @@ static void node_lost_contact(struct tipc_node *n_ptr) tipc_link_reset_fragments(l_ptr); } - /* Notify subscribers */ - tipc_nodesub_notify(n_ptr); + n_ptr->action_flags &= ~TIPC_WAIT_OWN_LINKS_DOWN; - /* Prevent re-contact with node until cleanup is done */ - n_ptr->block_setup = WAIT_PEER_DOWN | WAIT_NAMES_GONE; - tipc_k_signal((Handler)node_name_purge_complete, n_ptr->addr); + /* Notify subscribers and prevent re-contact with node until + * cleanup is done. + */ + n_ptr->action_flags |= TIPC_WAIT_PEER_LINKS_DOWN | + TIPC_NOTIFY_NODE_DOWN; } struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space) @@ -436,3 +426,63 @@ struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) rcu_read_unlock(); return buf; } + +/** + * tipc_node_get_linkname - get the name of a link + * + * @bearer_id: id of the bearer + * @node: peer node address + * @linkname: link name output buffer + * + * Returns 0 on success + */ +int tipc_node_get_linkname(u32 bearer_id, u32 addr, char *linkname, size_t len) +{ + struct tipc_link *link; + struct tipc_node *node = tipc_node_find(addr); + + if ((bearer_id >= MAX_BEARERS) || !node) + return -EINVAL; + tipc_node_lock(node); + link = node->links[bearer_id]; + if (link) { + strncpy(linkname, link->name, len); + tipc_node_unlock(node); + return 0; + } + tipc_node_unlock(node); + return -EINVAL; +} + +void tipc_node_unlock(struct tipc_node *node) +{ + LIST_HEAD(nsub_list); + struct tipc_link *link; + int pkt_sz = 0; + u32 addr = 0; + + if (likely(!node->action_flags)) { + spin_unlock_bh(&node->lock); + return; + } + + if (node->action_flags & TIPC_NOTIFY_NODE_DOWN) { + list_replace_init(&node->nsub, &nsub_list); + node->action_flags &= ~TIPC_NOTIFY_NODE_DOWN; + } + if (node->action_flags & TIPC_NOTIFY_NODE_UP) { + link = node->active_links[0]; + node->action_flags &= ~TIPC_NOTIFY_NODE_UP; + if (link) { + pkt_sz = ((link->max_pkt - INT_H_SIZE) / ITEM_SIZE) * + ITEM_SIZE; + addr = node->addr; + } + } + spin_unlock_bh(&node->lock); + + if (!list_empty(&nsub_list)) + tipc_nodesub_notify(&nsub_list); + if (pkt_sz) + tipc_named_node_up(pkt_sz, addr); +} diff --git a/net/tipc/node.h b/net/tipc/node.h index 7cbb8cec1a9..9087063793f 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -47,62 +47,73 @@ */ #define INVALID_NODE_SIG 0x10000 -/* Flags used to block (re)establishment of contact with a neighboring node */ -#define WAIT_PEER_DOWN 0x0001 /* wait to see that peer's links are down */ -#define WAIT_NAMES_GONE 0x0002 /* wait for peer's publications to be purged */ -#define WAIT_NODE_DOWN 0x0004 /* wait until peer node is declared down */ +/* Flags used to take different actions according to flag type + * TIPC_WAIT_PEER_LINKS_DOWN: wait to see that peer's links are down + * TIPC_WAIT_OWN_LINKS_DOWN: wait until peer node is declared down + * TIPC_NOTIFY_NODE_DOWN: notify node is down + * TIPC_NOTIFY_NODE_UP: notify node is up + */ +enum { + TIPC_WAIT_PEER_LINKS_DOWN = (1 << 1), + TIPC_WAIT_OWN_LINKS_DOWN = (1 << 2), + TIPC_NOTIFY_NODE_DOWN = (1 << 3), + TIPC_NOTIFY_NODE_UP = (1 << 4) +}; + +/** + * struct tipc_node_bclink - TIPC node bclink structure + * @acked: sequence # of last outbound b'cast message acknowledged by node + * @last_in: sequence # of last in-sequence b'cast message received from node + * @last_sent: sequence # of last b'cast message sent by node + * @oos_state: state tracker for handling OOS b'cast messages + * @deferred_size: number of OOS b'cast messages in deferred queue + * @deferred_head: oldest OOS b'cast message received from node + * @deferred_tail: newest OOS b'cast message received from node + * @reasm_buf: broadcast reassembly queue head from node + * @recv_permitted: true if node is allowed to receive b'cast messages + */ +struct tipc_node_bclink { + u32 acked; + u32 last_in; + u32 last_sent; + u32 oos_state; + u32 deferred_size; + struct sk_buff *deferred_head; + struct sk_buff *deferred_tail; + struct sk_buff *reasm_buf; + bool recv_permitted; +}; /** * struct tipc_node - TIPC node structure * @addr: network address of node * @lock: spinlock governing access to structure * @hash: links to adjacent nodes in unsorted hash chain - * @list: links to adjacent nodes in sorted list of cluster's nodes - * @nsub: list of "node down" subscriptions monitoring node * @active_links: pointers to active links to node * @links: pointers to all links to node + * @action_flags: bit mask of different types of node actions + * @bclink: broadcast-related info + * @list: links to adjacent nodes in sorted list of cluster's nodes * @working_links: number of working links to node (both active and standby) - * @block_setup: bit mask of conditions preventing link establishment to node * @link_cnt: number of links to node * @signature: node instance identifier - * @bclink: broadcast-related info + * @nsub: list of "node down" subscriptions monitoring node * @rcu: rcu struct for tipc_node - * @acked: sequence # of last outbound b'cast message acknowledged by node - * @last_in: sequence # of last in-sequence b'cast message received from node - * @last_sent: sequence # of last b'cast message sent by node - * @oos_state: state tracker for handling OOS b'cast messages - * @deferred_size: number of OOS b'cast messages in deferred queue - * @deferred_head: oldest OOS b'cast message received from node - * @deferred_tail: newest OOS b'cast message received from node - * @reasm_head: broadcast reassembly queue head from node - * @reasm_tail: last broadcast fragment received from node - * @recv_permitted: true if node is allowed to receive b'cast messages */ struct tipc_node { u32 addr; spinlock_t lock; struct hlist_node hash; - struct list_head list; - struct list_head nsub; struct tipc_link *active_links[2]; struct tipc_link *links[MAX_BEARERS]; + unsigned int action_flags; + struct tipc_node_bclink bclink; + struct list_head list; int link_cnt; int working_links; - int block_setup; u32 signature; + struct list_head nsub; struct rcu_head rcu; - struct { - u32 acked; - u32 last_in; - u32 last_sent; - u32 oos_state; - u32 deferred_size; - struct sk_buff *deferred_head; - struct sk_buff *deferred_tail; - struct sk_buff *reasm_head; - struct sk_buff *reasm_tail; - bool recv_permitted; - } bclink; }; extern struct list_head tipc_node_list; @@ -118,15 +129,18 @@ int tipc_node_active_links(struct tipc_node *n_ptr); int tipc_node_is_up(struct tipc_node *n_ptr); struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space); struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space); +int tipc_node_get_linkname(u32 bearer_id, u32 node, char *linkname, size_t len); +void tipc_node_unlock(struct tipc_node *node); -static inline void tipc_node_lock(struct tipc_node *n_ptr) +static inline void tipc_node_lock(struct tipc_node *node) { - spin_lock_bh(&n_ptr->lock); + spin_lock_bh(&node->lock); } -static inline void tipc_node_unlock(struct tipc_node *n_ptr) +static inline bool tipc_node_blocked(struct tipc_node *node) { - spin_unlock_bh(&n_ptr->lock); + return (node->action_flags & (TIPC_WAIT_PEER_LINKS_DOWN | + TIPC_NOTIFY_NODE_DOWN | TIPC_WAIT_OWN_LINKS_DOWN)); } #endif diff --git a/net/tipc/node_subscr.c b/net/tipc/node_subscr.c index 8a7384c04ad..7c59ab1d6ec 100644 --- a/net/tipc/node_subscr.c +++ b/net/tipc/node_subscr.c @@ -81,14 +81,13 @@ void tipc_nodesub_unsubscribe(struct tipc_node_subscr *node_sub) * * Note: node is locked by caller */ -void tipc_nodesub_notify(struct tipc_node *node) +void tipc_nodesub_notify(struct list_head *nsub_list) { - struct tipc_node_subscr *ns; + struct tipc_node_subscr *ns, *safe; - list_for_each_entry(ns, &node->nsub, nodesub_list) { + list_for_each_entry_safe(ns, safe, nsub_list, nodesub_list) { if (ns->handle_node_down) { - tipc_k_signal((Handler)ns->handle_node_down, - (unsigned long)ns->usr_handle); + ns->handle_node_down(ns->usr_handle); ns->handle_node_down = NULL; } } diff --git a/net/tipc/node_subscr.h b/net/tipc/node_subscr.h index c95d20727de..d91b8cc81e3 100644 --- a/net/tipc/node_subscr.h +++ b/net/tipc/node_subscr.h @@ -58,6 +58,6 @@ struct tipc_node_subscr { void tipc_nodesub_subscribe(struct tipc_node_subscr *node_sub, u32 addr, void *usr_handle, net_ev_handler handle_down); void tipc_nodesub_unsubscribe(struct tipc_node_subscr *node_sub); -void tipc_nodesub_notify(struct tipc_node *node); +void tipc_nodesub_notify(struct list_head *nsub_list); #endif diff --git a/net/tipc/port.c b/net/tipc/port.c index 5c14c7801ee..5fd7acce01e 100644 --- a/net/tipc/port.c +++ b/net/tipc/port.c @@ -165,7 +165,7 @@ void tipc_port_mcast_rcv(struct sk_buff *buf, struct tipc_port_list *dp) msg_set_destnode(msg, tipc_own_addr); if (dp->count == 1) { msg_set_destport(msg, dp->ports[0]); - tipc_port_rcv(buf); + tipc_sk_rcv(buf); tipc_port_list_free(dp); return; } @@ -180,7 +180,7 @@ void tipc_port_mcast_rcv(struct sk_buff *buf, struct tipc_port_list *dp) if ((index == 0) && (cnt != 0)) item = item->next; msg_set_destport(buf_msg(b), item->ports[index]); - tipc_port_rcv(b); + tipc_sk_rcv(b); } } exit: @@ -343,7 +343,7 @@ int tipc_reject_msg(struct sk_buff *buf, u32 err) /* send returned message & dispose of rejected message */ src_node = msg_prevnode(msg); if (in_own_node(src_node)) - tipc_port_rcv(rbuf); + tipc_sk_rcv(rbuf); else tipc_link_xmit(rbuf, src_node, msg_link_selector(rmsg)); exit: @@ -754,37 +754,6 @@ int tipc_port_shutdown(u32 ref) return tipc_port_disconnect(ref); } -/** - * tipc_port_rcv - receive message from lower layer and deliver to port user - */ -int tipc_port_rcv(struct sk_buff *buf) -{ - struct tipc_port *p_ptr; - struct tipc_msg *msg = buf_msg(buf); - u32 destport = msg_destport(msg); - u32 dsz = msg_data_sz(msg); - u32 err; - - /* forward unresolved named message */ - if (unlikely(!destport)) { - tipc_net_route_msg(buf); - return dsz; - } - - /* validate destination & pass to port, otherwise reject message */ - p_ptr = tipc_port_lock(destport); - if (likely(p_ptr)) { - err = tipc_sk_rcv(&tipc_port_to_sock(p_ptr)->sk, buf); - tipc_port_unlock(p_ptr); - if (likely(!err)) - return dsz; - } else { - err = TIPC_ERR_NO_PORT; - } - - return tipc_reject_msg(buf, err); -} - /* * tipc_port_iovec_rcv: Concatenate and deliver sectioned * message for this node. @@ -798,7 +767,7 @@ static int tipc_port_iovec_rcv(struct tipc_port *sender, res = tipc_msg_build(&sender->phdr, msg_sect, len, MAX_MSG_SIZE, &buf); if (likely(buf)) - tipc_port_rcv(buf); + tipc_sk_rcv(buf); return res; } diff --git a/net/tipc/port.h b/net/tipc/port.h index a00397393bd..cf4ca5b1d9a 100644 --- a/net/tipc/port.h +++ b/net/tipc/port.h @@ -42,9 +42,10 @@ #include "msg.h" #include "node_subscr.h" -#define TIPC_FLOW_CONTROL_WIN 512 -#define CONN_OVERLOAD_LIMIT ((TIPC_FLOW_CONTROL_WIN * 2 + 1) * \ - SKB_TRUESIZE(TIPC_MAX_USER_MSG_SIZE)) +#define TIPC_CONNACK_INTV 256 +#define TIPC_FLOWCTRL_WIN (TIPC_CONNACK_INTV * 2) +#define TIPC_CONN_OVERLOAD_LIMIT ((TIPC_FLOWCTRL_WIN * 2 + 1) * \ + SKB_TRUESIZE(TIPC_MAX_USER_MSG_SIZE)) /** * struct tipc_port - TIPC port structure @@ -134,7 +135,6 @@ int tipc_port_peer_msg(struct tipc_port *p_ptr, struct tipc_msg *msg); /* * TIPC messaging routines */ -int tipc_port_rcv(struct sk_buff *buf); int tipc_send(struct tipc_port *port, struct iovec const *msg_sect, @@ -187,7 +187,7 @@ static inline void tipc_port_unlock(struct tipc_port *p_ptr) static inline int tipc_port_congested(struct tipc_port *p_ptr) { - return (p_ptr->sent - p_ptr->acked) >= (TIPC_FLOW_CONTROL_WIN * 2); + return ((p_ptr->sent - p_ptr->acked) >= TIPC_FLOWCTRL_WIN); } diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 3c0256962f7..08d87fc80b1 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -1,5 +1,5 @@ /* - * net/tipc/socket.c: TIPC socket API +* net/tipc/socket.c: TIPC socket API * * Copyright (c) 2001-2007, 2012-2014, Ericsson AB * Copyright (c) 2004-2008, 2010-2013, Wind River Systems @@ -36,6 +36,7 @@ #include "core.h" #include "port.h" +#include "node.h" #include <linux/export.h> @@ -44,7 +45,7 @@ #define CONN_TIMEOUT_DEFAULT 8000 /* default connect timeout = 8s */ -static int backlog_rcv(struct sock *sk, struct sk_buff *skb); +static int tipc_backlog_rcv(struct sock *sk, struct sk_buff *skb); static void tipc_data_ready(struct sock *sk); static void tipc_write_space(struct sock *sk); static int tipc_release(struct socket *sock); @@ -195,11 +196,12 @@ static int tipc_sk_create(struct net *net, struct socket *sock, sock->state = state; sock_init_data(sock, sk); - sk->sk_backlog_rcv = backlog_rcv; + sk->sk_backlog_rcv = tipc_backlog_rcv; sk->sk_rcvbuf = sysctl_tipc_rmem[1]; sk->sk_data_ready = tipc_data_ready; sk->sk_write_space = tipc_write_space; - tipc_sk(sk)->conn_timeout = CONN_TIMEOUT_DEFAULT; + tsk->conn_timeout = CONN_TIMEOUT_DEFAULT; + atomic_set(&tsk->dupl_rcvcnt, 0); tipc_port_unlock(port); if (sock->state == SS_READY) { @@ -983,10 +985,11 @@ static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, return 0; } -static int tipc_wait_for_rcvmsg(struct socket *sock, long timeo) +static int tipc_wait_for_rcvmsg(struct socket *sock, long *timeop) { struct sock *sk = sock->sk; DEFINE_WAIT(wait); + long timeo = *timeop; int err; for (;;) { @@ -1011,6 +1014,7 @@ static int tipc_wait_for_rcvmsg(struct socket *sock, long timeo) break; } finish_wait(sk_sleep(sk), &wait); + *timeop = timeo; return err; } @@ -1054,7 +1058,7 @@ static int tipc_recvmsg(struct kiocb *iocb, struct socket *sock, restart: /* Look for a message in receive queue; wait if necessary */ - res = tipc_wait_for_rcvmsg(sock, timeo); + res = tipc_wait_for_rcvmsg(sock, &timeo); if (res) goto exit; @@ -1100,7 +1104,7 @@ restart: /* Consume received message (optional) */ if (likely(!(flags & MSG_PEEK))) { if ((sock->state != SS_READY) && - (++port->conn_unacked >= TIPC_FLOW_CONTROL_WIN)) + (++port->conn_unacked >= TIPC_CONNACK_INTV)) tipc_acknowledge(port->ref, port->conn_unacked); advance_rx_queue(sk); } @@ -1152,7 +1156,7 @@ static int tipc_recv_stream(struct kiocb *iocb, struct socket *sock, restart: /* Look for a message in receive queue; wait if necessary */ - res = tipc_wait_for_rcvmsg(sock, timeo); + res = tipc_wait_for_rcvmsg(sock, &timeo); if (res) goto exit; @@ -1209,7 +1213,7 @@ restart: /* Consume received message (optional) */ if (likely(!(flags & MSG_PEEK))) { - if (unlikely(++port->conn_unacked >= TIPC_FLOW_CONTROL_WIN)) + if (unlikely(++port->conn_unacked >= TIPC_CONNACK_INTV)) tipc_acknowledge(port->ref, port->conn_unacked); advance_rx_queue(sk); } @@ -1415,7 +1419,7 @@ static u32 filter_rcv(struct sock *sk, struct sk_buff *buf) } /** - * backlog_rcv - handle incoming message from backlog queue + * tipc_backlog_rcv - handle incoming message from backlog queue * @sk: socket * @buf: message * @@ -1423,47 +1427,73 @@ static u32 filter_rcv(struct sock *sk, struct sk_buff *buf) * * Returns 0 */ -static int backlog_rcv(struct sock *sk, struct sk_buff *buf) +static int tipc_backlog_rcv(struct sock *sk, struct sk_buff *buf) { u32 res; + struct tipc_sock *tsk = tipc_sk(sk); res = filter_rcv(sk, buf); - if (res) + if (unlikely(res)) tipc_reject_msg(buf, res); + + if (atomic_read(&tsk->dupl_rcvcnt) < TIPC_CONN_OVERLOAD_LIMIT) + atomic_add(buf->truesize, &tsk->dupl_rcvcnt); + return 0; } /** * tipc_sk_rcv - handle incoming message - * @sk: socket receiving message - * @buf: message - * - * Called with port lock already taken. - * - * Returns TIPC error status code (TIPC_OK if message is not to be rejected) + * @buf: buffer containing arriving message + * Consumes buffer + * Returns 0 if success, or errno: -EHOSTUNREACH */ -u32 tipc_sk_rcv(struct sock *sk, struct sk_buff *buf) +int tipc_sk_rcv(struct sk_buff *buf) { - u32 res; + struct tipc_sock *tsk; + struct tipc_port *port; + struct sock *sk; + u32 dport = msg_destport(buf_msg(buf)); + int err = TIPC_OK; + uint limit; - /* - * Process message if socket is unlocked; otherwise add to backlog queue - * - * This code is based on sk_receive_skb(), but must be distinct from it - * since a TIPC-specific filter/reject mechanism is utilized - */ + /* Forward unresolved named message */ + if (unlikely(!dport)) { + tipc_net_route_msg(buf); + return 0; + } + + /* Validate destination */ + port = tipc_port_lock(dport); + if (unlikely(!port)) { + err = TIPC_ERR_NO_PORT; + goto exit; + } + + tsk = tipc_port_to_sock(port); + sk = &tsk->sk; + + /* Queue message */ bh_lock_sock(sk); + if (!sock_owned_by_user(sk)) { - res = filter_rcv(sk, buf); + err = filter_rcv(sk, buf); } else { - if (sk_add_backlog(sk, buf, rcvbuf_limit(sk, buf))) - res = TIPC_ERR_OVERLOAD; - else - res = TIPC_OK; + if (sk->sk_backlog.len == 0) + atomic_set(&tsk->dupl_rcvcnt, 0); + limit = rcvbuf_limit(sk, buf) + atomic_read(&tsk->dupl_rcvcnt); + if (sk_add_backlog(sk, buf, limit)) + err = TIPC_ERR_OVERLOAD; } + bh_unlock_sock(sk); + tipc_port_unlock(port); - return res; + if (likely(!err)) + return 0; +exit: + tipc_reject_msg(buf, err); + return -EHOSTUNREACH; } static int tipc_wait_for_connect(struct socket *sock, long *timeo_p) @@ -1905,6 +1935,28 @@ static int tipc_getsockopt(struct socket *sock, int lvl, int opt, return put_user(sizeof(value), ol); } +int tipc_ioctl(struct socket *sk, unsigned int cmd, unsigned long arg) +{ + struct tipc_sioc_ln_req lnr; + void __user *argp = (void __user *)arg; + + switch (cmd) { + case SIOCGETLINKNAME: + if (copy_from_user(&lnr, argp, sizeof(lnr))) + return -EFAULT; + if (!tipc_node_get_linkname(lnr.bearer_id, lnr.peer, + lnr.linkname, TIPC_MAX_LINK_NAME)) { + if (copy_to_user(argp, &lnr, sizeof(lnr))) + return -EFAULT; + return 0; + } + return -EADDRNOTAVAIL; + break; + default: + return -ENOIOCTLCMD; + } +} + /* Protocol switches for the various types of TIPC sockets */ static const struct proto_ops msg_ops = { @@ -1917,7 +1969,7 @@ static const struct proto_ops msg_ops = { .accept = sock_no_accept, .getname = tipc_getname, .poll = tipc_poll, - .ioctl = sock_no_ioctl, + .ioctl = tipc_ioctl, .listen = sock_no_listen, .shutdown = tipc_shutdown, .setsockopt = tipc_setsockopt, @@ -1938,7 +1990,7 @@ static const struct proto_ops packet_ops = { .accept = tipc_accept, .getname = tipc_getname, .poll = tipc_poll, - .ioctl = sock_no_ioctl, + .ioctl = tipc_ioctl, .listen = tipc_listen, .shutdown = tipc_shutdown, .setsockopt = tipc_setsockopt, @@ -1959,7 +2011,7 @@ static const struct proto_ops stream_ops = { .accept = tipc_accept, .getname = tipc_getname, .poll = tipc_poll, - .ioctl = sock_no_ioctl, + .ioctl = tipc_ioctl, .listen = tipc_listen, .shutdown = tipc_shutdown, .setsockopt = tipc_setsockopt, diff --git a/net/tipc/socket.h b/net/tipc/socket.h index 74e5c7f195a..3afcd2a70b3 100644 --- a/net/tipc/socket.h +++ b/net/tipc/socket.h @@ -44,12 +44,14 @@ * @port: port - interacts with 'sk' and with the rest of the TIPC stack * @peer_name: the peer of the connection, if any * @conn_timeout: the time we can wait for an unresponded setup request + * @dupl_rcvcnt: number of bytes counted twice, in both backlog and rcv queue */ struct tipc_sock { struct sock sk; struct tipc_port port; unsigned int conn_timeout; + atomic_t dupl_rcvcnt; }; static inline struct tipc_sock *tipc_sk(const struct sock *sk) @@ -67,6 +69,6 @@ static inline void tipc_sock_wakeup(struct tipc_sock *tsk) tsk->sk.sk_write_space(&tsk->sk); } -u32 tipc_sk_rcv(struct sock *sk, struct sk_buff *buf); +int tipc_sk_rcv(struct sk_buff *buf); #endif diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index bb7e8ba821f..7b9114e0a5b 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -1492,10 +1492,14 @@ static int unix_dgram_sendmsg(struct kiocb *kiocb, struct socket *sock, if (len > sk->sk_sndbuf - 32) goto out; - if (len > SKB_MAX_ALLOC) + if (len > SKB_MAX_ALLOC) { data_len = min_t(size_t, len - SKB_MAX_ALLOC, MAX_SKB_FRAGS * PAGE_SIZE); + data_len = PAGE_ALIGN(data_len); + + BUILD_BUG_ON(SKB_MAX_ALLOC < PAGE_SIZE); + } skb = sock_alloc_send_pskb(sk, len - data_len, data_len, msg->msg_flags & MSG_DONTWAIT, &err, @@ -1670,6 +1674,8 @@ static int unix_stream_sendmsg(struct kiocb *kiocb, struct socket *sock, data_len = max_t(int, 0, size - SKB_MAX_HEAD(0)); + data_len = min_t(size_t, size, PAGE_ALIGN(data_len)); + skb = sock_alloc_send_pskb(sk, size - data_len, data_len, msg->msg_flags & MSG_DONTWAIT, &err, get_order(UNIX_SKB_FRAGS_SZ)); diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 5adfd94c5b8..85d232bed87 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -1925,9 +1925,23 @@ static struct miscdevice vsock_device = { .fops = &vsock_device_ops, }; -static int __vsock_core_init(void) +int __vsock_core_init(const struct vsock_transport *t, struct module *owner) { - int err; + int err = mutex_lock_interruptible(&vsock_register_mutex); + + if (err) + return err; + + if (transport) { + err = -EBUSY; + goto err_busy; + } + + /* Transport must be the owner of the protocol so that it can't + * unload while there are open sockets. + */ + vsock_proto.owner = owner; + transport = t; vsock_init_tables(); @@ -1951,36 +1965,19 @@ static int __vsock_core_init(void) goto err_unregister_proto; } + mutex_unlock(&vsock_register_mutex); return 0; err_unregister_proto: proto_unregister(&vsock_proto); err_misc_deregister: misc_deregister(&vsock_device); - return err; -} - -int vsock_core_init(const struct vsock_transport *t) -{ - int retval = mutex_lock_interruptible(&vsock_register_mutex); - if (retval) - return retval; - - if (transport) { - retval = -EBUSY; - goto out; - } - - transport = t; - retval = __vsock_core_init(); - if (retval) - transport = NULL; - -out: + transport = NULL; +err_busy: mutex_unlock(&vsock_register_mutex); - return retval; + return err; } -EXPORT_SYMBOL_GPL(vsock_core_init); +EXPORT_SYMBOL_GPL(__vsock_core_init); void vsock_core_exit(void) { @@ -2000,5 +1997,5 @@ EXPORT_SYMBOL_GPL(vsock_core_exit); MODULE_AUTHOR("VMware, Inc."); MODULE_DESCRIPTION("VMware Virtual Socket Family"); -MODULE_VERSION("1.0.0.0-k"); +MODULE_VERSION("1.0.1.0-k"); MODULE_LICENSE("GPL v2"); diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c index 3bb2cdc13b4..c51e8f7b865 100644 --- a/net/xfrm/xfrm_output.c +++ b/net/xfrm/xfrm_output.c @@ -199,6 +199,7 @@ int xfrm_output(struct sk_buff *skb) return xfrm_output2(skb); } +EXPORT_SYMBOL_GPL(xfrm_output); int xfrm_inner_extract_output(struct xfrm_state *x, struct sk_buff *skb) { @@ -213,6 +214,7 @@ int xfrm_inner_extract_output(struct xfrm_state *x, struct sk_buff *skb) return -EAFNOSUPPORT; return inner_mode->afinfo->extract_output(x, skb); } +EXPORT_SYMBOL_GPL(xfrm_inner_extract_output); void xfrm_local_error(struct sk_buff *skb, int mtu) { @@ -233,7 +235,4 @@ void xfrm_local_error(struct sk_buff *skb, int mtu) afinfo->local_error(skb, mtu); xfrm_state_put_afinfo(afinfo); } - -EXPORT_SYMBOL_GPL(xfrm_output); -EXPORT_SYMBOL_GPL(xfrm_inner_extract_output); EXPORT_SYMBOL_GPL(xfrm_local_error); diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index c08fbd11cef..a8ef5108e0d 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -769,7 +769,7 @@ EXPORT_SYMBOL(xfrm_policy_byid); #ifdef CONFIG_SECURITY_NETWORK_XFRM static inline int -xfrm_policy_flush_secctx_check(struct net *net, u8 type, struct xfrm_audit *audit_info) +xfrm_policy_flush_secctx_check(struct net *net, u8 type, bool task_valid) { int dir, err = 0; @@ -783,10 +783,7 @@ xfrm_policy_flush_secctx_check(struct net *net, u8 type, struct xfrm_audit *audi continue; err = security_xfrm_policy_delete(pol->security); if (err) { - xfrm_audit_policy_delete(pol, 0, - audit_info->loginuid, - audit_info->sessionid, - audit_info->secid); + xfrm_audit_policy_delete(pol, 0, task_valid); return err; } } @@ -800,9 +797,7 @@ xfrm_policy_flush_secctx_check(struct net *net, u8 type, struct xfrm_audit *audi pol->security); if (err) { xfrm_audit_policy_delete(pol, 0, - audit_info->loginuid, - audit_info->sessionid, - audit_info->secid); + task_valid); return err; } } @@ -812,19 +807,19 @@ xfrm_policy_flush_secctx_check(struct net *net, u8 type, struct xfrm_audit *audi } #else static inline int -xfrm_policy_flush_secctx_check(struct net *net, u8 type, struct xfrm_audit *audit_info) +xfrm_policy_flush_secctx_check(struct net *net, u8 type, bool task_valid) { return 0; } #endif -int xfrm_policy_flush(struct net *net, u8 type, struct xfrm_audit *audit_info) +int xfrm_policy_flush(struct net *net, u8 type, bool task_valid) { int dir, err = 0, cnt = 0; write_lock_bh(&net->xfrm.xfrm_policy_lock); - err = xfrm_policy_flush_secctx_check(net, type, audit_info); + err = xfrm_policy_flush_secctx_check(net, type, task_valid); if (err) goto out; @@ -841,9 +836,7 @@ int xfrm_policy_flush(struct net *net, u8 type, struct xfrm_audit *audit_info) write_unlock_bh(&net->xfrm.xfrm_policy_lock); cnt++; - xfrm_audit_policy_delete(pol, 1, audit_info->loginuid, - audit_info->sessionid, - audit_info->secid); + xfrm_audit_policy_delete(pol, 1, task_valid); xfrm_policy_kill(pol); @@ -862,10 +855,7 @@ int xfrm_policy_flush(struct net *net, u8 type, struct xfrm_audit *audit_info) write_unlock_bh(&net->xfrm.xfrm_policy_lock); cnt++; - xfrm_audit_policy_delete(pol, 1, - audit_info->loginuid, - audit_info->sessionid, - audit_info->secid); + xfrm_audit_policy_delete(pol, 1, task_valid); xfrm_policy_kill(pol); write_lock_bh(&net->xfrm.xfrm_policy_lock); @@ -2783,21 +2773,19 @@ static struct notifier_block xfrm_dev_notifier = { static int __net_init xfrm_statistics_init(struct net *net) { int rv; - - if (snmp_mib_init((void __percpu **)net->mib.xfrm_statistics, - sizeof(struct linux_xfrm_mib), - __alignof__(struct linux_xfrm_mib)) < 0) + net->mib.xfrm_statistics = alloc_percpu(struct linux_xfrm_mib); + if (!net->mib.xfrm_statistics) return -ENOMEM; rv = xfrm_proc_init(net); if (rv < 0) - snmp_mib_free((void __percpu **)net->mib.xfrm_statistics); + free_percpu(net->mib.xfrm_statistics); return rv; } static void xfrm_statistics_fini(struct net *net) { xfrm_proc_fini(net); - snmp_mib_free((void __percpu **)net->mib.xfrm_statistics); + free_percpu(net->mib.xfrm_statistics); } #else static int __net_init xfrm_statistics_init(struct net *net) @@ -2862,21 +2850,14 @@ out_byidx: static void xfrm_policy_fini(struct net *net) { - struct xfrm_audit audit_info; unsigned int sz; int dir; flush_work(&net->xfrm.policy_hash_work); #ifdef CONFIG_XFRM_SUB_POLICY - audit_info.loginuid = INVALID_UID; - audit_info.sessionid = (unsigned int)-1; - audit_info.secid = 0; - xfrm_policy_flush(net, XFRM_POLICY_TYPE_SUB, &audit_info); + xfrm_policy_flush(net, XFRM_POLICY_TYPE_SUB, false); #endif - audit_info.loginuid = INVALID_UID; - audit_info.sessionid = (unsigned int)-1; - audit_info.secid = 0; - xfrm_policy_flush(net, XFRM_POLICY_TYPE_MAIN, &audit_info); + xfrm_policy_flush(net, XFRM_POLICY_TYPE_MAIN, false); WARN_ON(!list_empty(&net->xfrm.policy_all)); @@ -2991,15 +2972,14 @@ static void xfrm_audit_common_policyinfo(struct xfrm_policy *xp, } } -void xfrm_audit_policy_add(struct xfrm_policy *xp, int result, - kuid_t auid, unsigned int sessionid, u32 secid) +void xfrm_audit_policy_add(struct xfrm_policy *xp, int result, bool task_valid) { struct audit_buffer *audit_buf; audit_buf = xfrm_audit_start("SPD-add"); if (audit_buf == NULL) return; - xfrm_audit_helper_usrinfo(auid, sessionid, secid, audit_buf); + xfrm_audit_helper_usrinfo(task_valid, audit_buf); audit_log_format(audit_buf, " res=%u", result); xfrm_audit_common_policyinfo(xp, audit_buf); audit_log_end(audit_buf); @@ -3007,14 +2987,14 @@ void xfrm_audit_policy_add(struct xfrm_policy *xp, int result, EXPORT_SYMBOL_GPL(xfrm_audit_policy_add); void xfrm_audit_policy_delete(struct xfrm_policy *xp, int result, - kuid_t auid, unsigned int sessionid, u32 secid) + bool task_valid) { struct audit_buffer *audit_buf; audit_buf = xfrm_audit_start("SPD-delete"); if (audit_buf == NULL) return; - xfrm_audit_helper_usrinfo(auid, sessionid, secid, audit_buf); + xfrm_audit_helper_usrinfo(task_valid, audit_buf); audit_log_format(audit_buf, " res=%u", result); xfrm_audit_common_policyinfo(xp, audit_buf); audit_log_end(audit_buf); diff --git a/net/xfrm/xfrm_proc.c b/net/xfrm/xfrm_proc.c index fc5abd0b456..9c4fbd8935f 100644 --- a/net/xfrm/xfrm_proc.c +++ b/net/xfrm/xfrm_proc.c @@ -54,8 +54,7 @@ static int xfrm_statistics_seq_show(struct seq_file *seq, void *v) int i; for (i = 0; xfrm_mib_list[i].name; i++) seq_printf(seq, "%-24s\t%lu\n", xfrm_mib_list[i].name, - snmp_fold_field((void __percpu **) - net->mib.xfrm_statistics, + snmp_fold_field(net->mib.xfrm_statistics, xfrm_mib_list[i].entry)); return 0; } diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 8e9c781a6bb..0ab54134bb4 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -463,9 +463,7 @@ expired: if (!err) km_state_expired(x, 1, 0); - xfrm_audit_state_delete(x, err ? 0 : 1, - audit_get_loginuid(current), - audit_get_sessionid(current), 0); + xfrm_audit_state_delete(x, err ? 0 : 1, true); out: spin_unlock(&x->lock); @@ -562,7 +560,7 @@ EXPORT_SYMBOL(xfrm_state_delete); #ifdef CONFIG_SECURITY_NETWORK_XFRM static inline int -xfrm_state_flush_secctx_check(struct net *net, u8 proto, struct xfrm_audit *audit_info) +xfrm_state_flush_secctx_check(struct net *net, u8 proto, bool task_valid) { int i, err = 0; @@ -572,10 +570,7 @@ xfrm_state_flush_secctx_check(struct net *net, u8 proto, struct xfrm_audit *audi hlist_for_each_entry(x, net->xfrm.state_bydst+i, bydst) { if (xfrm_id_proto_match(x->id.proto, proto) && (err = security_xfrm_state_delete(x)) != 0) { - xfrm_audit_state_delete(x, 0, - audit_info->loginuid, - audit_info->sessionid, - audit_info->secid); + xfrm_audit_state_delete(x, 0, task_valid); return err; } } @@ -585,18 +580,18 @@ xfrm_state_flush_secctx_check(struct net *net, u8 proto, struct xfrm_audit *audi } #else static inline int -xfrm_state_flush_secctx_check(struct net *net, u8 proto, struct xfrm_audit *audit_info) +xfrm_state_flush_secctx_check(struct net *net, u8 proto, bool task_valid) { return 0; } #endif -int xfrm_state_flush(struct net *net, u8 proto, struct xfrm_audit *audit_info) +int xfrm_state_flush(struct net *net, u8 proto, bool task_valid) { int i, err = 0, cnt = 0; spin_lock_bh(&net->xfrm.xfrm_state_lock); - err = xfrm_state_flush_secctx_check(net, proto, audit_info); + err = xfrm_state_flush_secctx_check(net, proto, task_valid); if (err) goto out; @@ -612,9 +607,7 @@ restart: err = xfrm_state_delete(x); xfrm_audit_state_delete(x, err ? 0 : 1, - audit_info->loginuid, - audit_info->sessionid, - audit_info->secid); + task_valid); xfrm_state_put(x); if (!err) cnt++; @@ -2128,14 +2121,10 @@ out_bydst: void xfrm_state_fini(struct net *net) { - struct xfrm_audit audit_info; unsigned int sz; flush_work(&net->xfrm.state_hash_work); - audit_info.loginuid = INVALID_UID; - audit_info.sessionid = (unsigned int)-1; - audit_info.secid = 0; - xfrm_state_flush(net, IPSEC_PROTO_ANY, &audit_info); + xfrm_state_flush(net, IPSEC_PROTO_ANY, false); flush_work(&net->xfrm.state_gc_work); WARN_ON(!list_empty(&net->xfrm.state_all)); @@ -2198,30 +2187,28 @@ static void xfrm_audit_helper_pktinfo(struct sk_buff *skb, u16 family, } } -void xfrm_audit_state_add(struct xfrm_state *x, int result, - kuid_t auid, unsigned int sessionid, u32 secid) +void xfrm_audit_state_add(struct xfrm_state *x, int result, bool task_valid) { struct audit_buffer *audit_buf; audit_buf = xfrm_audit_start("SAD-add"); if (audit_buf == NULL) return; - xfrm_audit_helper_usrinfo(auid, sessionid, secid, audit_buf); + xfrm_audit_helper_usrinfo(task_valid, audit_buf); xfrm_audit_helper_sainfo(x, audit_buf); audit_log_format(audit_buf, " res=%u", result); audit_log_end(audit_buf); } EXPORT_SYMBOL_GPL(xfrm_audit_state_add); -void xfrm_audit_state_delete(struct xfrm_state *x, int result, - kuid_t auid, unsigned int sessionid, u32 secid) +void xfrm_audit_state_delete(struct xfrm_state *x, int result, bool task_valid) { struct audit_buffer *audit_buf; audit_buf = xfrm_audit_start("SAD-delete"); if (audit_buf == NULL) return; - xfrm_audit_helper_usrinfo(auid, sessionid, secid, audit_buf); + xfrm_audit_helper_usrinfo(task_valid, audit_buf); xfrm_audit_helper_sainfo(x, audit_buf); audit_log_format(audit_buf, " res=%u", result); audit_log_end(audit_buf); diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 8f131c10a6f..412d9dc3a87 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -597,9 +597,6 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, struct xfrm_state *x; int err; struct km_event c; - kuid_t loginuid = audit_get_loginuid(current); - unsigned int sessionid = audit_get_sessionid(current); - u32 sid; err = verify_newsa_info(p, attrs); if (err) @@ -615,8 +612,7 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, else err = xfrm_state_update(x); - security_task_getsecid(current, &sid); - xfrm_audit_state_add(x, err ? 0 : 1, loginuid, sessionid, sid); + xfrm_audit_state_add(x, err ? 0 : 1, true); if (err < 0) { x->km.state = XFRM_STATE_DEAD; @@ -676,9 +672,6 @@ static int xfrm_del_sa(struct sk_buff *skb, struct nlmsghdr *nlh, int err = -ESRCH; struct km_event c; struct xfrm_usersa_id *p = nlmsg_data(nlh); - kuid_t loginuid = audit_get_loginuid(current); - unsigned int sessionid = audit_get_sessionid(current); - u32 sid; x = xfrm_user_state_lookup(net, p, attrs, &err); if (x == NULL) @@ -703,8 +696,7 @@ static int xfrm_del_sa(struct sk_buff *skb, struct nlmsghdr *nlh, km_state_notify(x, &c); out: - security_task_getsecid(current, &sid); - xfrm_audit_state_delete(x, err ? 0 : 1, loginuid, sessionid, sid); + xfrm_audit_state_delete(x, err ? 0 : 1, true); xfrm_state_put(x); return err; } @@ -955,6 +947,20 @@ static struct sk_buff *xfrm_state_netlink(struct sk_buff *in_skb, return skb; } +/* A wrapper for nlmsg_multicast() checking that nlsk is still available. + * Must be called with RCU read lock. + */ +static inline int xfrm_nlmsg_multicast(struct net *net, struct sk_buff *skb, + u32 pid, unsigned int group) +{ + struct sock *nlsk = rcu_dereference(net->xfrm.nlsk); + + if (nlsk) + return nlmsg_multicast(nlsk, skb, pid, group, GFP_ATOMIC); + else + return -1; +} + static inline size_t xfrm_spdinfo_msgsize(void) { return NLMSG_ALIGN(4) @@ -1414,9 +1420,6 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, struct km_event c; int err; int excl; - kuid_t loginuid = audit_get_loginuid(current); - unsigned int sessionid = audit_get_sessionid(current); - u32 sid; err = verify_newpolicy_info(p); if (err) @@ -1435,8 +1438,7 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, * a type XFRM_MSG_UPDPOLICY - JHS */ excl = nlh->nlmsg_type == XFRM_MSG_NEWPOLICY; err = xfrm_policy_insert(p->dir, xp, excl); - security_task_getsecid(current, &sid); - xfrm_audit_policy_add(xp, err ? 0 : 1, loginuid, sessionid, sid); + xfrm_audit_policy_add(xp, err ? 0 : 1, true); if (err) { security_xfrm_policy_free(xp->security); @@ -1673,13 +1675,7 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, NETLINK_CB(skb).portid); } } else { - kuid_t loginuid = audit_get_loginuid(current); - unsigned int sessionid = audit_get_sessionid(current); - u32 sid; - - security_task_getsecid(current, &sid); - xfrm_audit_policy_delete(xp, err ? 0 : 1, loginuid, sessionid, - sid); + xfrm_audit_policy_delete(xp, err ? 0 : 1, true); if (err != 0) goto out; @@ -1704,13 +1700,9 @@ static int xfrm_flush_sa(struct sk_buff *skb, struct nlmsghdr *nlh, struct net *net = sock_net(skb->sk); struct km_event c; struct xfrm_usersa_flush *p = nlmsg_data(nlh); - struct xfrm_audit audit_info; int err; - audit_info.loginuid = audit_get_loginuid(current); - audit_info.sessionid = audit_get_sessionid(current); - security_task_getsecid(current, &audit_info.secid); - err = xfrm_state_flush(net, p->proto, &audit_info); + err = xfrm_state_flush(net, p->proto, true); if (err) { if (err == -ESRCH) /* empty table */ return 0; @@ -1894,16 +1886,12 @@ static int xfrm_flush_policy(struct sk_buff *skb, struct nlmsghdr *nlh, struct km_event c; u8 type = XFRM_POLICY_TYPE_MAIN; int err; - struct xfrm_audit audit_info; err = copy_from_user_policy_type(&type, attrs); if (err) return err; - audit_info.loginuid = audit_get_loginuid(current); - audit_info.sessionid = audit_get_sessionid(current); - security_task_getsecid(current, &audit_info.secid); - err = xfrm_policy_flush(net, type, &audit_info); + err = xfrm_policy_flush(net, type, true); if (err) { if (err == -ESRCH) /* empty table */ return 0; @@ -1969,14 +1957,8 @@ static int xfrm_add_pol_expire(struct sk_buff *skb, struct nlmsghdr *nlh, err = 0; if (up->hard) { - kuid_t loginuid = audit_get_loginuid(current); - unsigned int sessionid = audit_get_sessionid(current); - u32 sid; - - security_task_getsecid(current, &sid); xfrm_policy_delete(xp, p->dir); - xfrm_audit_policy_delete(xp, 1, loginuid, sessionid, sid); - + xfrm_audit_policy_delete(xp, 1, true); } else { // reset the timers here? WARN(1, "Dont know what to do with soft policy expire\n"); @@ -2012,13 +1994,8 @@ static int xfrm_add_sa_expire(struct sk_buff *skb, struct nlmsghdr *nlh, km_state_expired(x, ue->hard, nlh->nlmsg_pid); if (ue->hard) { - kuid_t loginuid = audit_get_loginuid(current); - unsigned int sessionid = audit_get_sessionid(current); - u32 sid; - - security_task_getsecid(current, &sid); __xfrm_state_delete(x); - xfrm_audit_state_delete(x, 1, loginuid, sessionid, sid); + xfrm_audit_state_delete(x, 1, true); } err = 0; out: @@ -2265,7 +2242,7 @@ static int xfrm_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, if (build_migrate(skb, m, num_migrate, k, sel, dir, type) < 0) BUG(); - return nlmsg_multicast(net->xfrm.nlsk, skb, 0, XFRMNLGRP_MIGRATE, GFP_ATOMIC); + return xfrm_nlmsg_multicast(net, skb, 0, XFRMNLGRP_MIGRATE); } #else static int xfrm_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, @@ -2377,7 +2354,7 @@ static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) link = &xfrm_dispatch[type]; /* All operations require privileges, even GET */ - if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) + if (!netlink_net_capable(skb, CAP_NET_ADMIN)) return -EPERM; if ((type == (XFRM_MSG_GETSA - XFRM_MSG_BASE) || @@ -2456,7 +2433,7 @@ static int xfrm_exp_state_notify(struct xfrm_state *x, const struct km_event *c) return -EMSGSIZE; } - return nlmsg_multicast(net->xfrm.nlsk, skb, 0, XFRMNLGRP_EXPIRE, GFP_ATOMIC); + return xfrm_nlmsg_multicast(net, skb, 0, XFRMNLGRP_EXPIRE); } static int xfrm_aevent_state_notify(struct xfrm_state *x, const struct km_event *c) @@ -2471,7 +2448,7 @@ static int xfrm_aevent_state_notify(struct xfrm_state *x, const struct km_event if (build_aevent(skb, x, c) < 0) BUG(); - return nlmsg_multicast(net->xfrm.nlsk, skb, 0, XFRMNLGRP_AEVENTS, GFP_ATOMIC); + return xfrm_nlmsg_multicast(net, skb, 0, XFRMNLGRP_AEVENTS); } static int xfrm_notify_sa_flush(const struct km_event *c) @@ -2497,7 +2474,7 @@ static int xfrm_notify_sa_flush(const struct km_event *c) nlmsg_end(skb, nlh); - return nlmsg_multicast(net->xfrm.nlsk, skb, 0, XFRMNLGRP_SA, GFP_ATOMIC); + return xfrm_nlmsg_multicast(net, skb, 0, XFRMNLGRP_SA); } static inline size_t xfrm_sa_len(struct xfrm_state *x) @@ -2584,7 +2561,7 @@ static int xfrm_notify_sa(struct xfrm_state *x, const struct km_event *c) nlmsg_end(skb, nlh); - return nlmsg_multicast(net->xfrm.nlsk, skb, 0, XFRMNLGRP_SA, GFP_ATOMIC); + return xfrm_nlmsg_multicast(net, skb, 0, XFRMNLGRP_SA); out_free_skb: kfree_skb(skb); @@ -2675,7 +2652,7 @@ static int xfrm_send_acquire(struct xfrm_state *x, struct xfrm_tmpl *xt, if (build_acquire(skb, x, xt, xp) < 0) BUG(); - return nlmsg_multicast(net->xfrm.nlsk, skb, 0, XFRMNLGRP_ACQUIRE, GFP_ATOMIC); + return xfrm_nlmsg_multicast(net, skb, 0, XFRMNLGRP_ACQUIRE); } /* User gives us xfrm_user_policy_info followed by an array of 0 @@ -2789,7 +2766,7 @@ static int xfrm_exp_policy_notify(struct xfrm_policy *xp, int dir, const struct if (build_polexpire(skb, xp, dir, c) < 0) BUG(); - return nlmsg_multicast(net->xfrm.nlsk, skb, 0, XFRMNLGRP_EXPIRE, GFP_ATOMIC); + return xfrm_nlmsg_multicast(net, skb, 0, XFRMNLGRP_EXPIRE); } static int xfrm_notify_policy(struct xfrm_policy *xp, int dir, const struct km_event *c) @@ -2851,7 +2828,7 @@ static int xfrm_notify_policy(struct xfrm_policy *xp, int dir, const struct km_e nlmsg_end(skb, nlh); - return nlmsg_multicast(net->xfrm.nlsk, skb, 0, XFRMNLGRP_POLICY, GFP_ATOMIC); + return xfrm_nlmsg_multicast(net, skb, 0, XFRMNLGRP_POLICY); out_free_skb: kfree_skb(skb); @@ -2879,7 +2856,7 @@ static int xfrm_notify_policy_flush(const struct km_event *c) nlmsg_end(skb, nlh); - return nlmsg_multicast(net->xfrm.nlsk, skb, 0, XFRMNLGRP_POLICY, GFP_ATOMIC); + return xfrm_nlmsg_multicast(net, skb, 0, XFRMNLGRP_POLICY); out_free_skb: kfree_skb(skb); @@ -2948,7 +2925,7 @@ static int xfrm_send_report(struct net *net, u8 proto, if (build_report(skb, proto, sel, addr) < 0) BUG(); - return nlmsg_multicast(net->xfrm.nlsk, skb, 0, XFRMNLGRP_REPORT, GFP_ATOMIC); + return xfrm_nlmsg_multicast(net, skb, 0, XFRMNLGRP_REPORT); } static inline size_t xfrm_mapping_msgsize(void) @@ -3000,7 +2977,7 @@ static int xfrm_send_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, if (build_mapping(skb, x, ipaddr, sport) < 0) BUG(); - return nlmsg_multicast(net->xfrm.nlsk, skb, 0, XFRMNLGRP_MAPPING, GFP_ATOMIC); + return xfrm_nlmsg_multicast(net, skb, 0, XFRMNLGRP_MAPPING); } static bool xfrm_is_alive(const struct km_event *c) |