diff options
Diffstat (limited to 'net')
-rw-r--r-- | net/dsa/slave.c | 3 | ||||
-rw-r--r-- | net/ipv4/igmp.c | 4 | ||||
-rw-r--r-- | net/ipv4/ip_output.c | 7 | ||||
-rw-r--r-- | net/ipv4/ip_tunnel_core.c | 2 | ||||
-rw-r--r-- | net/ipv4/ipmr.c | 7 | ||||
-rw-r--r-- | net/ipv4/raw.c | 2 | ||||
-rw-r--r-- | net/ipv4/route.c | 4 | ||||
-rw-r--r-- | net/ipv4/tcp_input.c | 29 | ||||
-rw-r--r-- | net/ipv4/tcp_ipv4.c | 29 | ||||
-rw-r--r-- | net/ipv4/xfrm4_mode_tunnel.c | 2 | ||||
-rw-r--r-- | net/ipv6/fib6_rules.c | 9 | ||||
-rw-r--r-- | net/ipv6/ip6_output.c | 6 | ||||
-rw-r--r-- | net/ipv6/output_core.c | 14 | ||||
-rw-r--r-- | net/ipv6/udp_offload.c | 4 | ||||
-rw-r--r-- | net/netfilter/ipvs/ip_vs_xmit.c | 5 | ||||
-rw-r--r-- | net/sctp/socket.c | 5 | ||||
-rw-r--r-- | net/sctp/sysctl.c | 4 | ||||
-rw-r--r-- | net/tipc/bcast.c | 38 | ||||
-rw-r--r-- | net/tipc/bcast.h | 4 | ||||
-rw-r--r-- | net/tipc/discover.c | 1 | ||||
-rw-r--r-- | net/tipc/link.c | 219 | ||||
-rw-r--r-- | net/tipc/link.h | 9 | ||||
-rw-r--r-- | net/tipc/msg.h | 36 | ||||
-rw-r--r-- | net/tipc/name_distr.c | 2 | ||||
-rw-r--r-- | net/tipc/node.c | 90 | ||||
-rw-r--r-- | net/tipc/node.h | 12 |
26 files changed, 314 insertions, 233 deletions
diff --git a/net/dsa/slave.c b/net/dsa/slave.c index f0af7aa331c1..39555f3f263b 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -830,12 +830,13 @@ static bool dsa_slave_dev_check(struct net_device *dev) static int dsa_slave_master_changed(struct net_device *dev) { struct net_device *master = netdev_master_upper_dev_get(dev); + struct dsa_slave_priv *p = netdev_priv(dev); int err = 0; if (master && master->rtnl_link_ops && !strcmp(master->rtnl_link_ops->kind, "bridge")) err = dsa_slave_bridge_port_join(dev, master); - else + else if (dsa_port_is_bridged(p)) err = dsa_slave_bridge_port_leave(dev); return err; diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index ad3f866085de..ad09213ac5b2 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -370,7 +370,7 @@ static struct sk_buff *igmpv3_newpack(struct net_device *dev, unsigned int mtu) pip->saddr = fl4.saddr; pip->protocol = IPPROTO_IGMP; pip->tot_len = 0; /* filled in later */ - ip_select_ident(skb, NULL); + ip_select_ident(net, 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, NULL); + ip_select_ident(net, skb, NULL); ((u8 *)&iph[1])[0] = IPOPT_RA; ((u8 *)&iph[1])[1] = 4; ((u8 *)&iph[1])[2] = 0; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 90b49e88e84a..8259e777b249 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, sk); + ip_select_ident(sock_net(sk), skb, sk); if (opt && opt->opt.optlen) { iph->ihl += opt->opt.optlen>>2; @@ -430,7 +430,8 @@ packet_routed: ip_options_build(skb, &inet_opt->opt, inet->inet_daddr, rt, 0); } - ip_select_ident_segs(skb, sk, skb_shinfo(skb)->gso_segs ?: 1); + ip_select_ident_segs(sock_net(sk), skb, sk, + skb_shinfo(skb)->gso_segs ?: 1); /* TODO : should we use skb->sk here instead of sk ? */ skb->priority = sk->sk_priority; @@ -1379,7 +1380,7 @@ 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, sk); + ip_select_ident(net, skb, sk); if (opt) { iph->ihl += opt->optlen>>2; diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 88c386cf7d85..8c4dcc46acd2 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, skb_shinfo(skb)->gso_segs ?: 1); + __ip_select_ident(sock_net(sk), iph, skb_shinfo(skb)->gso_segs ?: 1); err = ip_local_out_sk(sk, skb); if (unlikely(net_xmit_eval(err))) diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 5b188832800f..c688cd1b2110 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1642,7 +1642,8 @@ static struct notifier_block ip_mr_notifier = { * important for multicast video. */ -static void ip_encap(struct sk_buff *skb, __be32 saddr, __be32 daddr) +static void ip_encap(struct net *net, struct sk_buff *skb, + __be32 saddr, __be32 daddr) { struct iphdr *iph; const struct iphdr *old_iph = ip_hdr(skb); @@ -1661,7 +1662,7 @@ 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, NULL); + ip_select_ident(net, skb, NULL); ip_send_check(iph); memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); @@ -1758,7 +1759,7 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt, * What do we do with netfilter? -- RR */ if (vif->flags & VIFF_TUNNEL) { - ip_encap(skb, vif->local, vif->remote); + ip_encap(net, skb, vif->local, vif->remote); /* FIXME: extra output firewall step used to be here. --RR */ vif->dev->stats.tx_packets++; vif->dev->stats.tx_bytes += skb->len; diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 923cf538fce1..56946f47d446 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -404,7 +404,7 @@ 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, NULL); + ip_select_ident(net, 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 649c8a3f0189..be8703d02ef0 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -482,7 +482,7 @@ u32 ip_idents_reserve(u32 hash, int segs) } EXPORT_SYMBOL(ip_idents_reserve); -void __ip_select_ident(struct iphdr *iph, int segs) +void __ip_select_ident(struct net *net, struct iphdr *iph, int segs) { static u32 ip_idents_hashrnd __read_mostly; u32 hash, id; @@ -491,7 +491,7 @@ void __ip_select_ident(struct iphdr *iph, int segs) hash = jhash_3words((__force u32)iph->daddr, (__force u32)iph->saddr, - iph->protocol, + iph->protocol ^ net_hash_mix(net), ip_idents_hashrnd); id = ip_idents_reserve(hash, segs); iph->id = htons(id); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 023196f7ec37..18b80e8bc533 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5987,6 +5987,35 @@ struct request_sock *inet_reqsk_alloc(const struct request_sock_ops *ops, } EXPORT_SYMBOL(inet_reqsk_alloc); +/* + * Return true if a syncookie should be sent + */ +static bool tcp_syn_flood_action(struct sock *sk, + const struct sk_buff *skb, + const char *proto) +{ + const char *msg = "Dropping request"; + bool want_cookie = false; + struct listen_sock *lopt; + +#ifdef CONFIG_SYN_COOKIES + if (sysctl_tcp_syncookies) { + msg = "Sending cookies"; + want_cookie = true; + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPREQQFULLDOCOOKIES); + } else +#endif + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPREQQFULLDROP); + + lopt = inet_csk(sk)->icsk_accept_queue.listen_opt; + if (!lopt->synflood_warned && sysctl_tcp_syncookies != 2) { + lopt->synflood_warned = 1; + pr_info("%s: Possible SYN flooding on port %d. %s. Check SNMP counters.\n", + proto, ntohs(tcp_hdr(skb)->dest), msg); + } + return want_cookie; +} + int tcp_conn_request(struct request_sock_ops *rsk_ops, const struct tcp_request_sock_ops *af_ops, struct sock *sk, struct sk_buff *skb) diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index e073517b2cc7..5aababa20a21 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -856,35 +856,6 @@ static void tcp_v4_reqsk_destructor(struct request_sock *req) kfree(inet_rsk(req)->opt); } -/* - * Return true if a syncookie should be sent - */ -bool tcp_syn_flood_action(struct sock *sk, - const struct sk_buff *skb, - const char *proto) -{ - const char *msg = "Dropping request"; - bool want_cookie = false; - struct listen_sock *lopt; - -#ifdef CONFIG_SYN_COOKIES - if (sysctl_tcp_syncookies) { - msg = "Sending cookies"; - want_cookie = true; - NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPREQQFULLDOCOOKIES); - } else -#endif - NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPREQQFULLDROP); - - lopt = inet_csk(sk)->icsk_accept_queue.listen_opt; - if (!lopt->synflood_warned && sysctl_tcp_syncookies != 2) { - lopt->synflood_warned = 1; - pr_info("%s: Possible SYN flooding on port %d. %s. Check SNMP counters.\n", - proto, ntohs(tcp_hdr(skb)->dest), msg); - } - return want_cookie; -} -EXPORT_SYMBOL(tcp_syn_flood_action); #ifdef CONFIG_TCP_MD5SIG /* diff --git a/net/ipv4/xfrm4_mode_tunnel.c b/net/ipv4/xfrm4_mode_tunnel.c index 91771a7c802f..35feda676464 100644 --- a/net/ipv4/xfrm4_mode_tunnel.c +++ b/net/ipv4/xfrm4_mode_tunnel.c @@ -63,7 +63,7 @@ static int xfrm4_mode_tunnel_output(struct xfrm_state *x, struct sk_buff *skb) top_iph->saddr = x->props.saddr.a4; top_iph->daddr = x->id.daddr.a4; - ip_select_ident(skb, NULL); + ip_select_ident(dev_net(dst->dev), skb, NULL); return 0; } diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index 27ca79682efb..273eb26cd6d4 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -299,19 +299,16 @@ static int __net_init fib6_rules_net_init(struct net *net) ops = fib_rules_register(&fib6_rules_ops_template, net); if (IS_ERR(ops)) return PTR_ERR(ops); - net->ipv6.fib6_rules_ops = ops; - - err = fib_default_rule_add(net->ipv6.fib6_rules_ops, 0, - RT6_TABLE_LOCAL, 0); + err = fib_default_rule_add(ops, 0, RT6_TABLE_LOCAL, 0); if (err) goto out_fib6_rules_ops; - err = fib_default_rule_add(net->ipv6.fib6_rules_ops, - 0x7FFE, RT6_TABLE_MAIN, 0); + err = fib_default_rule_add(ops, 0x7FFE, RT6_TABLE_MAIN, 0); if (err) goto out_fib6_rules_ops; + net->ipv6.fib6_rules_ops = ops; out: return err; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 7e80b61b51ff..b06ad00048d5 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -628,7 +628,7 @@ int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) skb_reset_network_header(skb); memcpy(skb_network_header(skb), tmp_hdr, hlen); - ipv6_select_ident(fh, rt); + ipv6_select_ident(net, fh, rt); fh->nexthdr = nexthdr; fh->reserved = 0; fh->frag_off = htons(IP6_MF); @@ -775,7 +775,7 @@ slow_path: fh->nexthdr = nexthdr; fh->reserved = 0; if (!frag_id) { - ipv6_select_ident(fh, rt); + ipv6_select_ident(net, fh, rt); frag_id = fh->identification; } else fh->identification = frag_id; @@ -1079,7 +1079,7 @@ static inline int ip6_ufo_append_data(struct sock *sk, skb_shinfo(skb)->gso_size = (mtu - fragheaderlen - sizeof(struct frag_hdr)) & ~7; skb_shinfo(skb)->gso_type = SKB_GSO_UDP; - ipv6_select_ident(&fhdr, rt); + ipv6_select_ident(sock_net(sk), &fhdr, rt); skb_shinfo(skb)->ip6_frag_id = fhdr.identification; append: diff --git a/net/ipv6/output_core.c b/net/ipv6/output_core.c index 74581f706c4d..4016a6ef9d61 100644 --- a/net/ipv6/output_core.c +++ b/net/ipv6/output_core.c @@ -9,13 +9,14 @@ #include <net/addrconf.h> #include <net/secure_seq.h> -static u32 __ipv6_select_ident(u32 hashrnd, struct in6_addr *dst, - struct in6_addr *src) +static u32 __ipv6_select_ident(struct net *net, u32 hashrnd, + struct in6_addr *dst, struct in6_addr *src) { u32 hash, id; hash = __ipv6_addr_jhash(dst, hashrnd); hash = __ipv6_addr_jhash(src, hash); + hash ^= net_hash_mix(net); /* Treat id of 0 as unset and if we get 0 back from ip_idents_reserve, * set the hight order instead thus minimizing possible future @@ -36,7 +37,7 @@ static u32 __ipv6_select_ident(u32 hashrnd, struct in6_addr *dst, * * The network header must be set before calling this. */ -void ipv6_proxy_select_ident(struct sk_buff *skb) +void ipv6_proxy_select_ident(struct net *net, struct sk_buff *skb) { static u32 ip6_proxy_idents_hashrnd __read_mostly; struct in6_addr buf[2]; @@ -53,20 +54,21 @@ void ipv6_proxy_select_ident(struct sk_buff *skb) net_get_random_once(&ip6_proxy_idents_hashrnd, sizeof(ip6_proxy_idents_hashrnd)); - id = __ipv6_select_ident(ip6_proxy_idents_hashrnd, + id = __ipv6_select_ident(net, ip6_proxy_idents_hashrnd, &addrs[1], &addrs[0]); skb_shinfo(skb)->ip6_frag_id = htonl(id); } EXPORT_SYMBOL_GPL(ipv6_proxy_select_ident); -void ipv6_select_ident(struct frag_hdr *fhdr, struct rt6_info *rt) +void ipv6_select_ident(struct net *net, struct frag_hdr *fhdr, + struct rt6_info *rt) { static u32 ip6_idents_hashrnd __read_mostly; u32 id; net_get_random_once(&ip6_idents_hashrnd, sizeof(ip6_idents_hashrnd)); - id = __ipv6_select_ident(ip6_idents_hashrnd, &rt->rt6i_dst.addr, + id = __ipv6_select_ident(net, ip6_idents_hashrnd, &rt->rt6i_dst.addr, &rt->rt6i_src.addr); fhdr->identification = htonl(id); } diff --git a/net/ipv6/udp_offload.c b/net/ipv6/udp_offload.c index be2c0ba82c85..7441e1e63893 100644 --- a/net/ipv6/udp_offload.c +++ b/net/ipv6/udp_offload.c @@ -54,7 +54,7 @@ static struct sk_buff *udp6_ufo_fragment(struct sk_buff *skb, /* Set the IPv6 fragment id if not set yet */ if (!skb_shinfo(skb)->ip6_frag_id) - ipv6_proxy_select_ident(skb); + ipv6_proxy_select_ident(dev_net(skb->dev), skb); segs = NULL; goto out; @@ -113,7 +113,7 @@ static struct sk_buff *udp6_ufo_fragment(struct sk_buff *skb, fptr->nexthdr = nexthdr; fptr->reserved = 0; if (!skb_shinfo(skb)->ip6_frag_id) - ipv6_proxy_select_ident(skb); + ipv6_proxy_select_ident(dev_net(skb->dev), skb); fptr->identification = skb_shinfo(skb)->ip6_frag_id; /* Fragment the skb. ipv6 header and the remaining fields of the diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index f35c15b0de6b..bf02932b7188 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -924,7 +924,8 @@ int ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, struct ip_vs_protocol *pp, struct ip_vs_iphdr *ipvsh) { - struct netns_ipvs *ipvs = net_ipvs(skb_net(skb)); + struct net *net = skb_net(skb); + struct netns_ipvs *ipvs = net_ipvs(net); struct rtable *rt; /* Route to the other host */ __be32 saddr; /* Source for tunnel */ struct net_device *tdev; /* Device to other host */ @@ -991,7 +992,7 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, iph->daddr = cp->daddr.ip; iph->saddr = saddr; iph->ttl = ttl; - ip_select_ident(skb, NULL); + ip_select_ident(net, skb, NULL); /* Another hack: avoid icmp_send in ip_fragment */ skb->ignore_df = 1; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index f1a65398f311..f09de7fac2e6 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -102,11 +102,6 @@ static int sctp_autobind(struct sock *sk); static void sctp_sock_migrate(struct sock *, struct sock *, struct sctp_association *, sctp_socket_type_t); -extern struct kmem_cache *sctp_bucket_cachep; -extern long sysctl_sctp_mem[3]; -extern int sysctl_sctp_rmem[3]; -extern int sysctl_sctp_wmem[3]; - static int sctp_memory_pressure; static atomic_long_t sctp_memory_allocated; struct percpu_counter sctp_sockets_allocated; diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c index 2e9ada10fd84..26d50c565f54 100644 --- a/net/sctp/sysctl.c +++ b/net/sctp/sysctl.c @@ -58,10 +58,6 @@ static unsigned long max_autoclose_max = (MAX_SCHEDULE_TIMEOUT / HZ > UINT_MAX) ? UINT_MAX : MAX_SCHEDULE_TIMEOUT / HZ; -extern long sysctl_sctp_mem[3]; -extern int sysctl_sctp_rmem[3]; -extern int sysctl_sctp_wmem[3]; - static int proc_sctp_do_hmac_alg(struct ctl_table *ctl, int write, void __user *buffer, size_t *lenp, loff_t *ppos); diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c index 5aff0844d4d3..ae558dd7f8ee 100644 --- a/net/tipc/bcast.c +++ b/net/tipc/bcast.c @@ -62,21 +62,8 @@ static void tipc_bclink_lock(struct net *net) static void tipc_bclink_unlock(struct net *net) { struct tipc_net *tn = net_generic(net, tipc_net_id); - struct tipc_node *node = NULL; - if (likely(!tn->bclink->flags)) { - spin_unlock_bh(&tn->bclink->lock); - return; - } - - if (tn->bclink->flags & TIPC_BCLINK_RESET) { - tn->bclink->flags &= ~TIPC_BCLINK_RESET; - node = tipc_bclink_retransmit_to(net); - } spin_unlock_bh(&tn->bclink->lock); - - if (node) - tipc_link_reset_all(node); } void tipc_bclink_input(struct net *net) @@ -91,13 +78,6 @@ uint tipc_bclink_get_mtu(void) return MAX_PKT_DEFAULT_MCAST; } -void tipc_bclink_set_flags(struct net *net, unsigned int flags) -{ - struct tipc_net *tn = net_generic(net, tipc_net_id); - - tn->bclink->flags |= flags; -} - static u32 bcbuf_acks(struct sk_buff *buf) { return (u32)(unsigned long)TIPC_SKB_CB(buf)->handle; @@ -156,7 +136,6 @@ static void bclink_update_last_sent(struct tipc_node *node, u32 seqno) seqno : node->bclink.last_sent; } - /** * tipc_bclink_retransmit_to - get most recent node to request retransmission * @@ -350,13 +329,12 @@ static void bclink_peek_nack(struct net *net, struct tipc_msg *msg) return; tipc_node_lock(n_ptr); - if (n_ptr->bclink.recv_permitted && (n_ptr->bclink.last_in != n_ptr->bclink.last_sent) && (n_ptr->bclink.last_in == msg_bcgap_after(msg))) n_ptr->bclink.oos_state = 2; - tipc_node_unlock(n_ptr); + tipc_node_put(n_ptr); } /* tipc_bclink_xmit - deliver buffer chain to all nodes in cluster @@ -476,17 +454,18 @@ void tipc_bclink_rcv(struct net *net, struct sk_buff *buf) goto unlock; if (msg_destnode(msg) == tn->own_addr) { tipc_bclink_acknowledge(node, msg_bcast_ack(msg)); - tipc_node_unlock(node); tipc_bclink_lock(net); bcl->stats.recv_nacks++; tn->bclink->retransmit_to = node; bclink_retransmit_pkt(tn, msg_bcgap_after(msg), msg_bcgap_to(msg)); tipc_bclink_unlock(net); + tipc_node_unlock(node); } else { tipc_node_unlock(node); bclink_peek_nack(net, msg); } + tipc_node_put(node); goto exit; } @@ -523,11 +502,13 @@ receive: tipc_bclink_unlock(net); tipc_node_unlock(node); } else if (msg_user(msg) == MSG_FRAGMENTER) { - tipc_buf_append(&node->bclink.reasm_buf, &buf); - if (unlikely(!buf && !node->bclink.reasm_buf)) - goto unlock; tipc_bclink_lock(net); bclink_accept_pkt(node, seqno); + tipc_buf_append(&node->bclink.reasm_buf, &buf); + if (unlikely(!buf && !node->bclink.reasm_buf)) { + tipc_bclink_unlock(net); + goto unlock; + } bcl->stats.recv_fragments++; if (buf) { bcl->stats.recv_fragmented++; @@ -589,6 +570,7 @@ receive: unlock: tipc_node_unlock(node); + tipc_node_put(node); exit: kfree_skb(buf); } @@ -829,7 +811,7 @@ int tipc_nl_add_bc_link(struct net *net, struct tipc_nl_msg *msg) prop = nla_nest_start(msg->skb, TIPC_NLA_LINK_PROP); if (!prop) goto attr_msg_full; - if (nla_put_u32(msg->skb, TIPC_NLA_PROP_WIN, bcl->queue_limit[0])) + if (nla_put_u32(msg->skb, TIPC_NLA_PROP_WIN, bcl->window)) goto prop_msg_full; nla_nest_end(msg->skb, prop); diff --git a/net/tipc/bcast.h b/net/tipc/bcast.h index 43f397fbac55..4bdc12277d33 100644 --- a/net/tipc/bcast.h +++ b/net/tipc/bcast.h @@ -55,7 +55,6 @@ struct tipc_bcbearer_pair { struct tipc_bearer *secondary; }; -#define TIPC_BCLINK_RESET 1 #define BCBEARER MAX_BEARERS /** @@ -86,7 +85,6 @@ struct tipc_bcbearer { * @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 * @@ -96,7 +94,6 @@ struct tipc_bclink { spinlock_t lock; struct tipc_link link; struct tipc_node node; - unsigned int flags; struct sk_buff_head arrvq; struct sk_buff_head inputq; struct tipc_node_map bcast_nodes; @@ -117,7 +114,6 @@ static inline int tipc_nmap_equal(struct tipc_node_map *nm_a, int tipc_bclink_init(struct net *net); void tipc_bclink_stop(struct net *net); -void tipc_bclink_set_flags(struct net *tn, unsigned int flags); void tipc_bclink_add_node(struct net *net, u32 addr); void tipc_bclink_remove_node(struct net *net, u32 addr); struct tipc_node *tipc_bclink_retransmit_to(struct net *tn); diff --git a/net/tipc/discover.c b/net/tipc/discover.c index 169f3dd038b9..967e292f53c8 100644 --- a/net/tipc/discover.c +++ b/net/tipc/discover.c @@ -260,6 +260,7 @@ void tipc_disc_rcv(struct net *net, struct sk_buff *buf, } } tipc_node_unlock(node); + tipc_node_put(node); } /** diff --git a/net/tipc/link.c b/net/tipc/link.c index 8c98c4d00ad6..514466efc25c 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -139,6 +139,13 @@ static void tipc_link_put(struct tipc_link *l_ptr) kref_put(&l_ptr->ref, tipc_link_release); } +static struct tipc_link *tipc_parallel_link(struct tipc_link *l) +{ + if (l->owner->active_links[0] != l) + return l->owner->active_links[0]; + return l->owner->active_links[1]; +} + static void link_init_max_pkt(struct tipc_link *l_ptr) { struct tipc_node *node = l_ptr->owner; @@ -310,7 +317,6 @@ struct tipc_link *tipc_link_create(struct tipc_node *n_ptr, link_init_max_pkt(l_ptr); l_ptr->priority = b_ptr->priority; tipc_link_set_queue_limits(l_ptr, b_ptr->window); - l_ptr->next_out_no = 1; __skb_queue_head_init(&l_ptr->transmq); __skb_queue_head_init(&l_ptr->backlogq); @@ -368,28 +374,43 @@ void tipc_link_delete_list(struct net *net, unsigned int bearer_id, } /** - * link_schedule_user - schedule user for wakeup after congestion + * link_schedule_user - schedule a message sender for wakeup after congestion * @link: congested link - * @oport: sending port - * @chain_sz: size of buffer chain that was attempted sent - * @imp: importance of message attempted sent + * @list: message that was attempted sent * Create pseudo msg to send back to user when congestion abates + * Only consumes message if there is an error */ -static bool link_schedule_user(struct tipc_link *link, u32 oport, - uint chain_sz, uint imp) +static int link_schedule_user(struct tipc_link *link, struct sk_buff_head *list) { - struct sk_buff *buf; + struct tipc_msg *msg = buf_msg(skb_peek(list)); + int imp = msg_importance(msg); + u32 oport = msg_origport(msg); + u32 addr = link_own_addr(link); + struct sk_buff *skb; - buf = tipc_msg_create(SOCK_WAKEUP, 0, INT_H_SIZE, 0, - link_own_addr(link), link_own_addr(link), - oport, 0, 0); - if (!buf) - return false; - TIPC_SKB_CB(buf)->chain_sz = chain_sz; - TIPC_SKB_CB(buf)->chain_imp = imp; - skb_queue_tail(&link->wakeupq, buf); + /* This really cannot happen... */ + if (unlikely(imp > TIPC_CRITICAL_IMPORTANCE)) { + pr_warn("%s<%s>, send queue full", link_rst_msg, link->name); + tipc_link_reset(link); + goto err; + } + /* Non-blocking sender: */ + if (TIPC_SKB_CB(skb_peek(list))->wakeup_pending) + return -ELINKCONG; + + /* Create and schedule wakeup pseudo message */ + skb = tipc_msg_create(SOCK_WAKEUP, 0, INT_H_SIZE, 0, + addr, addr, oport, 0, 0); + if (!skb) + goto err; + TIPC_SKB_CB(skb)->chain_sz = skb_queue_len(list); + TIPC_SKB_CB(skb)->chain_imp = imp; + skb_queue_tail(&link->wakeupq, skb); link->stats.link_congs++; - return true; + return -ELINKCONG; +err: + __skb_queue_purge(list); + return -ENOBUFS; } /** @@ -398,19 +419,22 @@ static bool link_schedule_user(struct tipc_link *link, u32 oport, * Move a number of waiting users, as permitted by available space in * the send queue, from link wait queue to node wait queue for wakeup */ -void link_prepare_wakeup(struct tipc_link *link) +void link_prepare_wakeup(struct tipc_link *l) { - uint pend_qsz = skb_queue_len(&link->backlogq); + int pnd[TIPC_SYSTEM_IMPORTANCE + 1] = {0,}; + int imp, lim; struct sk_buff *skb, *tmp; - skb_queue_walk_safe(&link->wakeupq, skb, tmp) { - if (pend_qsz >= link->queue_limit[TIPC_SKB_CB(skb)->chain_imp]) + skb_queue_walk_safe(&l->wakeupq, skb, tmp) { + imp = TIPC_SKB_CB(skb)->chain_imp; + lim = l->window + l->backlog[imp].limit; + pnd[imp] += TIPC_SKB_CB(skb)->chain_sz; + if ((pnd[imp] + l->backlog[imp].len) >= lim) break; - pend_qsz += TIPC_SKB_CB(skb)->chain_sz; - skb_unlink(skb, &link->wakeupq); - skb_queue_tail(&link->inputq, skb); - link->owner->inputq = &link->inputq; - link->owner->action_flags |= TIPC_MSG_EVT; + skb_unlink(skb, &l->wakeupq); + skb_queue_tail(&l->inputq, skb); + l->owner->inputq = &l->inputq; + l->owner->action_flags |= TIPC_MSG_EVT; } } @@ -424,6 +448,16 @@ void tipc_link_reset_fragments(struct tipc_link *l_ptr) l_ptr->reasm_buf = NULL; } +static void tipc_link_purge_backlog(struct tipc_link *l) +{ + __skb_queue_purge(&l->backlogq); + l->backlog[TIPC_LOW_IMPORTANCE].len = 0; + l->backlog[TIPC_MEDIUM_IMPORTANCE].len = 0; + l->backlog[TIPC_HIGH_IMPORTANCE].len = 0; + l->backlog[TIPC_CRITICAL_IMPORTANCE].len = 0; + l->backlog[TIPC_SYSTEM_IMPORTANCE].len = 0; +} + /** * tipc_link_purge_queues - purge all pkt queues associated with link * @l_ptr: pointer to link @@ -432,7 +466,7 @@ void tipc_link_purge_queues(struct tipc_link *l_ptr) { __skb_queue_purge(&l_ptr->deferdq); __skb_queue_purge(&l_ptr->transmq); - __skb_queue_purge(&l_ptr->backlogq); + tipc_link_purge_backlog(l_ptr); tipc_link_reset_fragments(l_ptr); } @@ -466,13 +500,13 @@ void tipc_link_reset(struct tipc_link *l_ptr) /* Clean up all queues, except inputq: */ __skb_queue_purge(&l_ptr->transmq); - __skb_queue_purge(&l_ptr->backlogq); __skb_queue_purge(&l_ptr->deferdq); if (!owner->inputq) owner->inputq = &l_ptr->inputq; skb_queue_splice_init(&l_ptr->wakeupq, owner->inputq); if (!skb_queue_empty(owner->inputq)) owner->action_flags |= TIPC_MSG_EVT; + tipc_link_purge_backlog(l_ptr); l_ptr->rcv_unacked = 0; l_ptr->checkpoint = 1; l_ptr->next_out_no = 1; @@ -696,48 +730,15 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) } } -/* tipc_link_cong: determine return value and how to treat the - * sent buffer during link congestion. - * - For plain, errorless user data messages we keep the buffer and - * return -ELINKONG. - * - For all other messages we discard the buffer and return -EHOSTUNREACH - * - For TIPC internal messages we also reset the link - */ -static int tipc_link_cong(struct tipc_link *link, struct sk_buff_head *list) -{ - struct sk_buff *skb = skb_peek(list); - struct tipc_msg *msg = buf_msg(skb); - int imp = msg_importance(msg); - u32 oport = msg_tot_origport(msg); - - if (unlikely(imp > TIPC_CRITICAL_IMPORTANCE)) { - pr_warn("%s<%s>, send queue full", link_rst_msg, link->name); - tipc_link_reset(link); - goto drop; - } - if (unlikely(msg_errcode(msg))) - goto drop; - if (unlikely(msg_reroute_cnt(msg))) - goto drop; - if (TIPC_SKB_CB(skb)->wakeup_pending) - return -ELINKCONG; - if (link_schedule_user(link, oport, skb_queue_len(list), imp)) - return -ELINKCONG; -drop: - __skb_queue_purge(list); - return -EHOSTUNREACH; -} - /** * __tipc_link_xmit(): same as tipc_link_xmit, but destlink is known & locked * @link: link to use * @list: chain of buffers containing message * - * Consumes the buffer chain, except when returning -ELINKCONG - * Returns 0 if success, otherwise errno: -ELINKCONG, -EMSGSIZE (plain socket - * user data messages) or -EHOSTUNREACH (all other messages/senders) - * Only the socket functions tipc_send_stream() and tipc_send_packet() need - * to act on the return value, since they may need to do more send attempts. + * Consumes the buffer chain, except when returning -ELINKCONG, + * since the caller then may want to make more send attempts. + * Returns 0 if success, or errno: -ELINKCONG, -EMSGSIZE or -ENOBUFS + * Messages at TIPC_SYSTEM_IMPORTANCE are always accepted */ int __tipc_link_xmit(struct net *net, struct tipc_link *link, struct sk_buff_head *list) @@ -754,16 +755,14 @@ int __tipc_link_xmit(struct net *net, struct tipc_link *link, struct sk_buff_head *backlogq = &link->backlogq; struct sk_buff *skb, *tmp; - /* Match queue limit against msg importance: */ - if (unlikely(skb_queue_len(backlogq) >= link->queue_limit[imp])) - return tipc_link_cong(link, list); + /* Match backlog limit against msg importance: */ + if (unlikely(link->backlog[imp].len >= link->backlog[imp].limit)) + return link_schedule_user(link, list); - /* Has valid packet limit been used ? */ if (unlikely(msg_size(msg) > mtu)) { __skb_queue_purge(list); return -EMSGSIZE; } - /* Prepare each packet for sending, and add to relevant queue: */ skb_queue_walk_safe(list, skb, tmp) { __skb_unlink(skb, list); @@ -786,8 +785,10 @@ int __tipc_link_xmit(struct net *net, struct tipc_link *link, if (tipc_msg_make_bundle(&skb, mtu, link->addr)) { link->stats.sent_bundled++; link->stats.sent_bundles++; + imp = msg_importance(buf_msg(skb)); } __skb_queue_tail(backlogq, skb); + link->backlog[imp].len++; seqno++; } link->next_out_no = seqno; @@ -808,13 +809,25 @@ static int __tipc_link_xmit_skb(struct tipc_link *link, struct sk_buff *skb) return __tipc_link_xmit(link->owner->net, link, &head); } +/* tipc_link_xmit_skb(): send single buffer to destination + * Buffers sent via this functon are generally TIPC_SYSTEM_IMPORTANCE + * messages, which will not be rejected + * The only exception is datagram messages rerouted after secondary + * lookup, which are rare and safe to dispose of anyway. + * TODO: Return real return value, and let callers use + * tipc_wait_for_sendpkt() where applicable + */ int tipc_link_xmit_skb(struct net *net, struct sk_buff *skb, u32 dnode, u32 selector) { struct sk_buff_head head; + int rc; skb2list(skb, &head); - return tipc_link_xmit(net, &head, dnode, selector); + rc = tipc_link_xmit(net, &head, dnode, selector); + if (rc == -ELINKCONG) + kfree_skb(skb); + return 0; } /** @@ -841,6 +854,7 @@ int tipc_link_xmit(struct net *net, struct sk_buff_head *list, u32 dnode, if (link) rc = __tipc_link_xmit(net, link, list); tipc_node_unlock(node); + tipc_node_put(node); } if (link) return rc; @@ -914,6 +928,7 @@ void tipc_link_push_packets(struct tipc_link *link) if (!skb) break; msg = buf_msg(skb); + link->backlog[msg_importance(msg)].len--; msg_set_ack(msg, ack); msg_set_bcast_ack(msg, link->owner->bclink.last_in); link->rcv_unacked = 0; @@ -966,7 +981,6 @@ static void link_retransmit_failure(struct tipc_link *l_ptr, (unsigned long) TIPC_SKB_CB(buf)->handle); n_ptr = tipc_bclink_retransmit_to(net); - tipc_node_lock(n_ptr); tipc_addr_string_fill(addr_string, n_ptr->addr); pr_info("Broadcast link info for %s\n", addr_string); @@ -978,9 +992,7 @@ static void link_retransmit_failure(struct tipc_link *l_ptr, n_ptr->bclink.oos_state, n_ptr->bclink.last_sent); - tipc_node_unlock(n_ptr); - - tipc_bclink_set_flags(net, TIPC_BCLINK_RESET); + n_ptr->action_flags |= TIPC_BCAST_RESET; l_ptr->stale_count = 0; } } @@ -1019,6 +1031,32 @@ void tipc_link_retransmit(struct tipc_link *l_ptr, struct sk_buff *skb, } } +/* link_synch(): check if all packets arrived before the synch + * point have been consumed + * Returns true if the parallel links are synched, otherwise false + */ +static bool link_synch(struct tipc_link *l) +{ + unsigned int post_synch; + struct tipc_link *pl; + + pl = tipc_parallel_link(l); + if (pl == l) + goto synched; + + /* Was last pre-synch packet added to input queue ? */ + if (less_eq(pl->next_in_no, l->synch_point)) + return false; + + /* Is it still in the input queue ? */ + post_synch = mod(pl->next_in_no - l->synch_point) - 1; + if (skb_queue_len(&pl->inputq) > post_synch) + return false; +synched: + l->flags &= ~LINK_SYNCHING; + return true; +} + static void link_retrieve_defq(struct tipc_link *link, struct sk_buff_head *list) { @@ -1079,8 +1117,8 @@ void tipc_rcv(struct net *net, struct sk_buff *skb, struct tipc_bearer *b_ptr) n_ptr = tipc_node_find(net, msg_prevnode(msg)); if (unlikely(!n_ptr)) goto discard; - tipc_node_lock(n_ptr); + tipc_node_lock(n_ptr); /* Locate unicast link endpoint that should handle message */ l_ptr = n_ptr->links[b_ptr->identity]; if (unlikely(!l_ptr)) @@ -1149,6 +1187,14 @@ void tipc_rcv(struct net *net, struct sk_buff *skb, struct tipc_bearer *b_ptr) skb = NULL; goto unlock; } + /* Synchronize with parallel link if applicable */ + if (unlikely((l_ptr->flags & LINK_SYNCHING) && !msg_dup(msg))) { + link_handle_out_of_seq_msg(l_ptr, skb); + if (link_synch(l_ptr)) + link_retrieve_defq(l_ptr, &head); + skb = NULL; + goto unlock; + } l_ptr->next_in_no++; if (unlikely(!skb_queue_empty(&l_ptr->deferdq))) link_retrieve_defq(l_ptr, &head); @@ -1160,6 +1206,7 @@ void tipc_rcv(struct net *net, struct sk_buff *skb, struct tipc_bearer *b_ptr) skb = NULL; unlock: tipc_node_unlock(n_ptr); + tipc_node_put(n_ptr); discard: if (unlikely(skb)) kfree_skb(skb); @@ -1224,6 +1271,10 @@ static void tipc_link_input(struct tipc_link *link, struct sk_buff *skb) switch (msg_user(msg)) { case CHANGEOVER_PROTOCOL: + if (msg_dup(msg)) { + link->flags |= LINK_SYNCHING; + link->synch_point = msg_seqno(msg_get_wrapped(msg)); + } if (!tipc_link_tunnel_rcv(node, &skb)) break; if (msg_user(buf_msg(skb)) != MSG_BUNDLER) { @@ -1610,6 +1661,7 @@ void tipc_link_failover_send_queue(struct tipc_link *l_ptr) tipc_msg_init(link_own_addr(l_ptr), &tunnel_hdr, CHANGEOVER_PROTOCOL, ORIGINAL_MSG, INT_H_SIZE, l_ptr->addr); skb_queue_splice_tail_init(&l_ptr->backlogq, &l_ptr->transmq); + tipc_link_purge_backlog(l_ptr); msgcount = skb_queue_len(&l_ptr->transmq); msg_set_bearer_id(&tunnel_hdr, l_ptr->peer_bearer_id); msg_set_msgcnt(&tunnel_hdr, msgcount); @@ -1817,11 +1869,11 @@ void tipc_link_set_queue_limits(struct tipc_link *l, u32 win) int max_bulk = TIPC_MAX_PUBLICATIONS / (l->max_pkt / ITEM_SIZE); l->window = win; - l->queue_limit[TIPC_LOW_IMPORTANCE] = win / 2; - l->queue_limit[TIPC_MEDIUM_IMPORTANCE] = win; - l->queue_limit[TIPC_HIGH_IMPORTANCE] = win / 2 * 3; - l->queue_limit[TIPC_CRITICAL_IMPORTANCE] = win * 2; - l->queue_limit[TIPC_SYSTEM_IMPORTANCE] = max_bulk; + l->backlog[TIPC_LOW_IMPORTANCE].limit = win / 2; + l->backlog[TIPC_MEDIUM_IMPORTANCE].limit = win; + l->backlog[TIPC_HIGH_IMPORTANCE].limit = win / 2 * 3; + l->backlog[TIPC_CRITICAL_IMPORTANCE].limit = win * 2; + l->backlog[TIPC_SYSTEM_IMPORTANCE].limit = max_bulk; } /* tipc_link_find_owner - locate owner node of link by link's name @@ -2120,7 +2172,7 @@ static int __tipc_nl_add_link(struct net *net, struct tipc_nl_msg *msg, if (nla_put_u32(msg->skb, TIPC_NLA_PROP_TOL, link->tolerance)) goto prop_msg_full; if (nla_put_u32(msg->skb, TIPC_NLA_PROP_WIN, - link->queue_limit[TIPC_LOW_IMPORTANCE])) + link->window)) goto prop_msg_full; if (nla_put_u32(msg->skb, TIPC_NLA_PROP_PRIO, link->priority)) goto prop_msg_full; @@ -2186,7 +2238,6 @@ int tipc_nl_link_dump(struct sk_buff *skb, struct netlink_callback *cb) msg.seq = cb->nlh->nlmsg_seq; rcu_read_lock(); - if (prev_node) { node = tipc_node_find(net, prev_node); if (!node) { @@ -2199,6 +2250,7 @@ int tipc_nl_link_dump(struct sk_buff *skb, struct netlink_callback *cb) cb->prev_seq = 1; goto out; } + tipc_node_put(node); list_for_each_entry_continue_rcu(node, &tn->node_list, list) { @@ -2206,6 +2258,7 @@ int tipc_nl_link_dump(struct sk_buff *skb, struct netlink_callback *cb) err = __tipc_nl_add_node_links(net, &msg, node, &prev_link); tipc_node_unlock(node); + tipc_node_put(node); if (err) goto out; diff --git a/net/tipc/link.h b/net/tipc/link.h index eec3ecf2d450..d2b5663643da 100644 --- a/net/tipc/link.h +++ b/net/tipc/link.h @@ -60,6 +60,7 @@ */ #define LINK_STARTED 0x0001 #define LINK_STOPPED 0x0002 +#define LINK_SYNCHING 0x0004 /* Starting value for maximum packet size negotiation on unicast links * (unless bearer MTU is less) @@ -118,7 +119,7 @@ struct tipc_stats { * @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) + * @backlog_limit: backlog queue congestion thresholds (indexed by importance) * @exp_msg_count: # of tunnelled messages expected during link changeover * @reset_checkpoint: seq # of last acknowledged message at time of link reset * @max_pkt: current maximum packet size for this link @@ -166,11 +167,11 @@ struct tipc_link { struct tipc_msg *pmsg; u32 priority; char net_plane; - u32 queue_limit[15]; /* queue_limit[0]==window limit */ /* Changeover */ u32 exp_msg_count; u32 reset_checkpoint; + u32 synch_point; /* Max packet negotiation */ u32 max_pkt; @@ -180,6 +181,10 @@ struct tipc_link { /* Sending */ struct sk_buff_head transmq; struct sk_buff_head backlogq; + struct { + u16 len; + u16 limit; + } backlog[5]; u32 next_out_no; u32 window; u32 last_retransmitted; diff --git a/net/tipc/msg.h b/net/tipc/msg.h index bd3969a80dd4..d273207ede28 100644 --- a/net/tipc/msg.h +++ b/net/tipc/msg.h @@ -240,6 +240,15 @@ static inline void msg_set_size(struct tipc_msg *m, u32 sz) m->hdr[0] = htonl((msg_word(m, 0) & ~0x1ffff) | sz); } +static inline unchar *msg_data(struct tipc_msg *m) +{ + return ((unchar *)m) + msg_hdr_sz(m); +} + +static inline struct tipc_msg *msg_get_wrapped(struct tipc_msg *m) +{ + return (struct tipc_msg *)msg_data(m); +} /* * Word 1 @@ -372,6 +381,8 @@ static inline void msg_set_prevnode(struct tipc_msg *m, u32 a) static inline u32 msg_origport(struct tipc_msg *m) { + if (msg_user(m) == MSG_FRAGMENTER) + m = msg_get_wrapped(m); return msg_word(m, 4); } @@ -467,16 +478,6 @@ static inline void msg_set_nameupper(struct tipc_msg *m, u32 n) msg_set_word(m, 10, n); } -static inline unchar *msg_data(struct tipc_msg *m) -{ - return ((unchar *)m) + msg_hdr_sz(m); -} - -static inline struct tipc_msg *msg_get_wrapped(struct tipc_msg *m) -{ - return (struct tipc_msg *)msg_data(m); -} - /* * Constants and routines used to read and write TIPC internal message headers */ @@ -553,6 +554,14 @@ static inline void msg_set_node_capabilities(struct tipc_msg *m, u32 n) msg_set_bits(m, 1, 15, 0x1fff, n); } +static inline bool msg_dup(struct tipc_msg *m) +{ + if (likely(msg_user(m) != CHANGEOVER_PROTOCOL)) + return false; + if (msg_type(m) != DUPLICATE_MSG) + return false; + return true; +} /* * Word 2 @@ -753,13 +762,6 @@ static inline void msg_set_link_tolerance(struct tipc_msg *m, u32 n) msg_set_bits(m, 9, 0, 0xffff, n); } -static inline u32 msg_tot_origport(struct tipc_msg *m) -{ - if ((msg_user(m) == MSG_FRAGMENTER) && (msg_type(m) == FIRST_FRAGMENT)) - return msg_origport(msg_get_wrapped(m)); - return msg_origport(m); -} - struct sk_buff *tipc_buf_acquire(u32 size); bool tipc_msg_validate(struct sk_buff *skb); bool tipc_msg_reverse(u32 own_addr, struct sk_buff *buf, u32 *dnode, diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c index 506aaa565da7..41e7b7e4dda0 100644 --- a/net/tipc/name_distr.c +++ b/net/tipc/name_distr.c @@ -244,6 +244,7 @@ static void tipc_publ_subscribe(struct net *net, struct publication *publ, tipc_node_lock(node); list_add_tail(&publ->nodesub_list, &node->publ_list); tipc_node_unlock(node); + tipc_node_put(node); } static void tipc_publ_unsubscribe(struct net *net, struct publication *publ, @@ -258,6 +259,7 @@ static void tipc_publ_unsubscribe(struct net *net, struct publication *publ, tipc_node_lock(node); list_del_init(&publ->nodesub_list); tipc_node_unlock(node); + tipc_node_put(node); } /** diff --git a/net/tipc/node.c b/net/tipc/node.c index 26d1de1bf34d..3e4f04897c03 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -42,6 +42,7 @@ static void node_lost_contact(struct tipc_node *n_ptr); static void node_established_contact(struct tipc_node *n_ptr); +static void tipc_node_delete(struct tipc_node *node); struct tipc_sock_conn { u32 port; @@ -67,6 +68,23 @@ static unsigned int tipc_hashfn(u32 addr) return addr & (NODE_HTABLE_SIZE - 1); } +static void tipc_node_kref_release(struct kref *kref) +{ + struct tipc_node *node = container_of(kref, struct tipc_node, kref); + + tipc_node_delete(node); +} + +void tipc_node_put(struct tipc_node *node) +{ + kref_put(&node->kref, tipc_node_kref_release); +} + +static void tipc_node_get(struct tipc_node *node) +{ + kref_get(&node->kref); +} + /* * tipc_node_find - locate specified node object, if it exists */ @@ -82,6 +100,7 @@ struct tipc_node *tipc_node_find(struct net *net, u32 addr) hlist_for_each_entry_rcu(node, &tn->node_htable[tipc_hashfn(addr)], hash) { if (node->addr == addr) { + tipc_node_get(node); rcu_read_unlock(); return node; } @@ -106,6 +125,7 @@ struct tipc_node *tipc_node_create(struct net *net, u32 addr) } n_ptr->addr = addr; n_ptr->net = net; + kref_init(&n_ptr->kref); spin_lock_init(&n_ptr->lock); INIT_HLIST_NODE(&n_ptr->hash); INIT_LIST_HEAD(&n_ptr->list); @@ -120,16 +140,17 @@ struct tipc_node *tipc_node_create(struct net *net, u32 addr) list_add_tail_rcu(&n_ptr->list, &temp_node->list); n_ptr->action_flags = TIPC_WAIT_PEER_LINKS_DOWN; n_ptr->signature = INVALID_NODE_SIG; + tipc_node_get(n_ptr); exit: spin_unlock_bh(&tn->node_list_lock); return n_ptr; } -static void tipc_node_delete(struct tipc_net *tn, struct tipc_node *n_ptr) +static void tipc_node_delete(struct tipc_node *node) { - list_del_rcu(&n_ptr->list); - hlist_del_rcu(&n_ptr->hash); - kfree_rcu(n_ptr, rcu); + list_del_rcu(&node->list); + hlist_del_rcu(&node->hash); + kfree_rcu(node, rcu); } void tipc_node_stop(struct net *net) @@ -139,7 +160,7 @@ void tipc_node_stop(struct net *net) spin_lock_bh(&tn->node_list_lock); list_for_each_entry_safe(node, t_node, &tn->node_list, list) - tipc_node_delete(tn, node); + tipc_node_put(node); spin_unlock_bh(&tn->node_list_lock); } @@ -147,6 +168,7 @@ int tipc_node_add_conn(struct net *net, u32 dnode, u32 port, u32 peer_port) { struct tipc_node *node; struct tipc_sock_conn *conn; + int err = 0; if (in_own_node(net, dnode)) return 0; @@ -157,8 +179,10 @@ int tipc_node_add_conn(struct net *net, u32 dnode, u32 port, u32 peer_port) return -EHOSTUNREACH; } conn = kmalloc(sizeof(*conn), GFP_ATOMIC); - if (!conn) - return -EHOSTUNREACH; + if (!conn) { + err = -EHOSTUNREACH; + goto exit; + } conn->peer_node = dnode; conn->port = port; conn->peer_port = peer_port; @@ -166,7 +190,9 @@ int tipc_node_add_conn(struct net *net, u32 dnode, u32 port, u32 peer_port) tipc_node_lock(node); list_add_tail(&conn->list, &node->conn_sks); tipc_node_unlock(node); - return 0; +exit: + tipc_node_put(node); + return err; } void tipc_node_remove_conn(struct net *net, u32 dnode, u32 port) @@ -189,6 +215,7 @@ void tipc_node_remove_conn(struct net *net, u32 dnode, u32 port) kfree(conn); } tipc_node_unlock(node); + tipc_node_put(node); } /** @@ -417,19 +444,25 @@ int tipc_node_get_linkname(struct net *net, u32 bearer_id, u32 addr, char *linkname, size_t len) { struct tipc_link *link; + int err = -EINVAL; struct tipc_node *node = tipc_node_find(net, addr); - if ((bearer_id >= MAX_BEARERS) || !node) - return -EINVAL; + if (!node) + return err; + + if (bearer_id >= MAX_BEARERS) + goto exit; + tipc_node_lock(node); link = node->links[bearer_id]; if (link) { strncpy(linkname, link->name, len); - tipc_node_unlock(node); - return 0; + err = 0; } +exit: tipc_node_unlock(node); - return -EINVAL; + tipc_node_put(node); + return err; } void tipc_node_unlock(struct tipc_node *node) @@ -459,7 +492,7 @@ void tipc_node_unlock(struct tipc_node *node) TIPC_NOTIFY_NODE_DOWN | TIPC_NOTIFY_NODE_UP | TIPC_NOTIFY_LINK_DOWN | TIPC_NOTIFY_LINK_UP | TIPC_WAKEUP_BCAST_USERS | TIPC_BCAST_MSG_EVT | - TIPC_NAMED_MSG_EVT); + TIPC_NAMED_MSG_EVT | TIPC_BCAST_RESET); spin_unlock_bh(&node->lock); @@ -488,6 +521,9 @@ void tipc_node_unlock(struct tipc_node *node) if (flags & TIPC_BCAST_MSG_EVT) tipc_bclink_input(net); + + if (flags & TIPC_BCAST_RESET) + tipc_link_reset_all(node); } /* Caller should hold node lock for the passed node */ @@ -542,17 +578,21 @@ int tipc_nl_node_dump(struct sk_buff *skb, struct netlink_callback *cb) msg.seq = cb->nlh->nlmsg_seq; rcu_read_lock(); - - if (last_addr && !tipc_node_find(net, last_addr)) { - rcu_read_unlock(); - /* We never set seq or call nl_dump_check_consistent() this - * means that setting prev_seq here will cause the consistence - * check to fail in the netlink callback handler. Resulting in - * the NLMSG_DONE message having the NLM_F_DUMP_INTR flag set if - * the node state changed while we released the lock. - */ - cb->prev_seq = 1; - return -EPIPE; + if (last_addr) { + node = tipc_node_find(net, last_addr); + if (!node) { + rcu_read_unlock(); + /* We never set seq or call nl_dump_check_consistent() + * this means that setting prev_seq here will cause the + * consistence check to fail in the netlink callback + * handler. Resulting in the NLMSG_DONE message having + * the NLM_F_DUMP_INTR flag set if the node state + * changed while we released the lock. + */ + cb->prev_seq = 1; + return -EPIPE; + } + tipc_node_put(node); } list_for_each_entry_rcu(node, &tn->node_list, list) { diff --git a/net/tipc/node.h b/net/tipc/node.h index e89ac04ec2c3..02d5c20dc551 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -64,7 +64,8 @@ enum { TIPC_NOTIFY_LINK_UP = (1 << 6), TIPC_NOTIFY_LINK_DOWN = (1 << 7), TIPC_NAMED_MSG_EVT = (1 << 8), - TIPC_BCAST_MSG_EVT = (1 << 9) + TIPC_BCAST_MSG_EVT = (1 << 9), + TIPC_BCAST_RESET = (1 << 10) }; /** @@ -93,6 +94,7 @@ struct tipc_node_bclink { /** * struct tipc_node - TIPC node structure * @addr: network address of node + * @ref: reference counter to node object * @lock: spinlock governing access to structure * @net: the applicable net namespace * @hash: links to adjacent nodes in unsorted hash chain @@ -114,6 +116,7 @@ struct tipc_node_bclink { */ struct tipc_node { u32 addr; + struct kref kref; spinlock_t lock; struct net *net; struct hlist_node hash; @@ -136,6 +139,7 @@ struct tipc_node { }; struct tipc_node *tipc_node_find(struct net *net, u32 addr); +void tipc_node_put(struct tipc_node *node); struct tipc_node *tipc_node_create(struct net *net, u32 addr); void tipc_node_stop(struct net *net); void tipc_node_attach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr); @@ -170,10 +174,12 @@ static inline uint tipc_node_get_mtu(struct net *net, u32 addr, u32 selector) node = tipc_node_find(net, addr); - if (likely(node)) + if (likely(node)) { mtu = node->act_mtus[selector & 1]; - else + tipc_node_put(node); + } else { mtu = MAX_MSG_SIZE; + } return mtu; } |