diff options
Diffstat (limited to 'net/tipc')
39 files changed, 4415 insertions, 4063 deletions
diff --git a/net/tipc/Kconfig b/net/tipc/Kconfig index c890848f9d56..c25a3a149dc4 100644 --- a/net/tipc/Kconfig +++ b/net/tipc/Kconfig @@ -20,21 +20,17 @@ menuconfig TIPC If in doubt, say N. -config TIPC_PORTS - int "Maximum number of ports in a node" - depends on TIPC - range 127 65535 - default "8191" - help - Specifies how many ports can be supported by a node. - Can range from 127 to 65535 ports; default is 8191. - - Setting this to a smaller value saves some memory, - setting it to higher allows for more ports. - config TIPC_MEDIA_IB bool "InfiniBand media type support" depends on TIPC && INFINIBAND_IPOIB help Saying Y here will enable support for running TIPC on IP-over-InfiniBand devices. +config TIPC_MEDIA_UDP + bool "IP/UDP media type support" + depends on TIPC + select NET_UDP_TUNNEL + help + Saying Y here will enable support for running TIPC over IP/UDP + bool + default y diff --git a/net/tipc/Makefile b/net/tipc/Makefile index 333e4592772c..57e460be4692 100644 --- a/net/tipc/Makefile +++ b/net/tipc/Makefile @@ -4,11 +4,12 @@ obj-$(CONFIG_TIPC) := tipc.o -tipc-y += addr.o bcast.o bearer.o config.o \ +tipc-y += addr.o bcast.o bearer.o \ core.o link.o discover.o msg.o \ name_distr.o subscr.o name_table.o net.o \ - netlink.o node.o socket.o log.o eth_media.o \ - server.o + netlink.o netlink_compat.o node.o socket.o eth_media.o \ + server.o socket.o +tipc-$(CONFIG_TIPC_MEDIA_UDP) += udp_media.o tipc-$(CONFIG_TIPC_MEDIA_IB) += ib_media.o tipc-$(CONFIG_SYSCTL) += sysctl.o diff --git a/net/tipc/addr.c b/net/tipc/addr.c index 357b74b26f9e..ba7daa864d44 100644 --- a/net/tipc/addr.c +++ b/net/tipc/addr.c @@ -34,8 +34,58 @@ * POSSIBILITY OF SUCH DAMAGE. */ -#include "core.h" +#include <linux/kernel.h> #include "addr.h" +#include "core.h" + +u32 tipc_own_addr(struct net *net) +{ + struct tipc_net *tn = net_generic(net, tipc_net_id); + + return tn->own_addr; +} + +/** + * in_own_cluster - test for cluster inclusion; <0.0.0> always matches + */ +int in_own_cluster(struct net *net, u32 addr) +{ + return in_own_cluster_exact(net, addr) || !addr; +} + +int in_own_cluster_exact(struct net *net, u32 addr) +{ + struct tipc_net *tn = net_generic(net, tipc_net_id); + + return !((addr ^ tn->own_addr) >> 12); +} + +/** + * in_own_node - test for node inclusion; <0.0.0> always matches + */ +int in_own_node(struct net *net, u32 addr) +{ + struct tipc_net *tn = net_generic(net, tipc_net_id); + + return (addr == tn->own_addr) || !addr; +} + +/** + * addr_domain - convert 2-bit scope value to equivalent message lookup domain + * + * Needed when address of a named message must be looked up a second time + * after a network hop. + */ +u32 addr_domain(struct net *net, u32 sc) +{ + struct tipc_net *tn = net_generic(net, tipc_net_id); + + if (likely(sc == TIPC_NODE_SCOPE)) + return tn->own_addr; + if (sc == TIPC_CLUSTER_SCOPE) + return tipc_cluster_mask(tn->own_addr); + return tipc_zone_mask(tn->own_addr); +} /** * tipc_addr_domain_valid - validates a network domain address diff --git a/net/tipc/addr.h b/net/tipc/addr.h index a74acf9ee804..7ba6d5c8ae40 100644 --- a/net/tipc/addr.h +++ b/net/tipc/addr.h @@ -37,7 +37,10 @@ #ifndef _TIPC_ADDR_H #define _TIPC_ADDR_H -#include "core.h" +#include <linux/types.h> +#include <linux/tipc.h> +#include <net/net_namespace.h> +#include <net/netns/generic.h> #define TIPC_ZONE_MASK 0xff000000u #define TIPC_CLUSTER_MASK 0xfffff000u @@ -52,42 +55,11 @@ static inline u32 tipc_cluster_mask(u32 addr) return addr & TIPC_CLUSTER_MASK; } -static inline int in_own_cluster_exact(u32 addr) -{ - return !((addr ^ tipc_own_addr) >> 12); -} - -/** - * in_own_node - test for node inclusion; <0.0.0> always matches - */ -static inline int in_own_node(u32 addr) -{ - return (addr == tipc_own_addr) || !addr; -} - -/** - * in_own_cluster - test for cluster inclusion; <0.0.0> always matches - */ -static inline int in_own_cluster(u32 addr) -{ - return in_own_cluster_exact(addr) || !addr; -} - -/** - * addr_domain - convert 2-bit scope value to equivalent message lookup domain - * - * Needed when address of a named message must be looked up a second time - * after a network hop. - */ -static inline u32 addr_domain(u32 sc) -{ - if (likely(sc == TIPC_NODE_SCOPE)) - return tipc_own_addr; - if (sc == TIPC_CLUSTER_SCOPE) - return tipc_cluster_mask(tipc_own_addr); - return tipc_zone_mask(tipc_own_addr); -} - +u32 tipc_own_addr(struct net *net); +int in_own_cluster(struct net *net, u32 addr); +int in_own_cluster_exact(struct net *net, u32 addr); +int in_own_node(struct net *net, u32 addr); +u32 addr_domain(struct net *net, u32 sc); int tipc_addr_domain_valid(u32); int tipc_addr_node_valid(u32 addr); int tipc_in_scope(u32 domain, u32 addr); diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c index a9e174fc0f91..c5cbdcb1f0b5 100644 --- a/net/tipc/bcast.c +++ b/net/tipc/bcast.c @@ -1,7 +1,7 @@ /* * net/tipc/bcast.c: TIPC broadcast code * - * Copyright (c) 2004-2006, 2014, Ericsson AB + * Copyright (c) 2004-2006, 2014-2015, Ericsson AB * Copyright (c) 2004, Intel Corporation. * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. @@ -35,77 +35,14 @@ * POSSIBILITY OF SUCH DAMAGE. */ -#include "core.h" -#include "link.h" #include "socket.h" #include "msg.h" #include "bcast.h" #include "name_distr.h" +#include "core.h" #define MAX_PKT_DEFAULT_MCAST 1500 /* bcast link max packet size (fixed) */ #define BCLINK_WIN_DEFAULT 20 /* bcast link window size (default) */ -#define BCBEARER MAX_BEARERS - -/** - * struct tipc_bcbearer_pair - a pair of bearers used by broadcast link - * @primary: pointer to primary bearer - * @secondary: pointer to secondary bearer - * - * Bearers must have same priority and same set of reachable destinations - * to be paired. - */ - -struct tipc_bcbearer_pair { - struct tipc_bearer *primary; - struct tipc_bearer *secondary; -}; - -/** - * struct tipc_bcbearer - bearer used by broadcast link - * @bearer: (non-standard) broadcast bearer structure - * @media: (non-standard) broadcast media structure - * @bpairs: array of bearer pairs - * @bpairs_temp: temporary array of bearer pairs used by tipc_bcbearer_sort() - * @remains: temporary node map used by tipc_bcbearer_send() - * @remains_new: temporary node map used tipc_bcbearer_send() - * - * 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 "bclink_lock". - */ -struct tipc_bcbearer { - struct tipc_bearer bearer; - struct tipc_media media; - struct tipc_bcbearer_pair bpairs[MAX_BEARERS]; - struct tipc_bcbearer_pair bpairs_temp[TIPC_MAX_LINK_PRI + 1]; - struct tipc_node_map remains; - struct tipc_node_map remains_new; -}; - -/** - * 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 *bcbearer; -static struct tipc_bclink *bclink; -static struct tipc_link *bcl; const char tipc_bclink_name[] = "broadcast-link"; @@ -115,28 +52,25 @@ static void tipc_nmap_diff(struct tipc_node_map *nm_a, 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) +static void tipc_bclink_lock(struct net *net) { - spin_lock_bh(&bclink->lock); + struct tipc_net *tn = net_generic(net, tipc_net_id); + + spin_lock_bh(&tn->bclink->lock); } -static void tipc_bclink_unlock(void) +static void tipc_bclink_unlock(struct net *net) { - struct tipc_node *node = NULL; + struct tipc_net *tn = net_generic(net, tipc_net_id); - if (likely(!bclink->flags)) { - spin_unlock_bh(&bclink->lock); - return; - } + spin_unlock_bh(&tn->bclink->lock); +} - if (bclink->flags & TIPC_BCLINK_RESET) { - bclink->flags &= ~TIPC_BCLINK_RESET; - node = tipc_bclink_retransmit_to(); - } - spin_unlock_bh(&bclink->lock); +void tipc_bclink_input(struct net *net) +{ + struct tipc_net *tn = net_generic(net, tipc_net_id); - if (node) - tipc_link_reset_all(node); + tipc_sk_mcast_rcv(net, &tn->bclink->arrvq, &tn->bclink->inputq); } uint tipc_bclink_get_mtu(void) @@ -144,11 +78,6 @@ uint tipc_bclink_get_mtu(void) return MAX_PKT_DEFAULT_MCAST; } -void tipc_bclink_set_flags(unsigned int flags) -{ - bclink->flags |= flags; -} - static u32 bcbuf_acks(struct sk_buff *buf) { return (u32)(unsigned long)TIPC_SKB_CB(buf)->handle; @@ -164,31 +93,41 @@ static void bcbuf_decr_acks(struct sk_buff *buf) bcbuf_set_acks(buf, bcbuf_acks(buf) - 1); } -void tipc_bclink_add_node(u32 addr) +void tipc_bclink_add_node(struct net *net, u32 addr) { - tipc_bclink_lock(); - tipc_nmap_add(&bclink->bcast_nodes, addr); - tipc_bclink_unlock(); + struct tipc_net *tn = net_generic(net, tipc_net_id); + + tipc_bclink_lock(net); + tipc_nmap_add(&tn->bclink->bcast_nodes, addr); + tipc_bclink_unlock(net); } -void tipc_bclink_remove_node(u32 addr) +void tipc_bclink_remove_node(struct net *net, u32 addr) { - tipc_bclink_lock(); - tipc_nmap_remove(&bclink->bcast_nodes, addr); - tipc_bclink_unlock(); + struct tipc_net *tn = net_generic(net, tipc_net_id); + + tipc_bclink_lock(net); + tipc_nmap_remove(&tn->bclink->bcast_nodes, addr); + tipc_bclink_unlock(net); } -static void bclink_set_last_sent(void) +static void bclink_set_last_sent(struct net *net) { - if (bcl->next_out) - bcl->fsm_msg_cnt = mod(buf_seqno(bcl->next_out) - 1); + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_link *bcl = tn->bcl; + struct sk_buff *skb = skb_peek(&bcl->backlogq); + + if (skb) + bcl->fsm_msg_cnt = mod(buf_seqno(skb) - 1); else bcl->fsm_msg_cnt = mod(bcl->next_out_no - 1); } -u32 tipc_bclink_get_last_sent(void) +u32 tipc_bclink_get_last_sent(struct net *net) { - return bcl->fsm_msg_cnt; + struct tipc_net *tn = net_generic(net, tipc_net_id); + + return tn->bcl->fsm_msg_cnt; } static void bclink_update_last_sent(struct tipc_node *node, u32 seqno) @@ -197,15 +136,16 @@ 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 * * Called with bclink_lock locked */ -struct tipc_node *tipc_bclink_retransmit_to(void) +struct tipc_node *tipc_bclink_retransmit_to(struct net *net) { - return bclink->retransmit_to; + struct tipc_net *tn = net_generic(net, tipc_net_id); + + return tn->bclink->retransmit_to; } /** @@ -215,11 +155,12 @@ struct tipc_node *tipc_bclink_retransmit_to(void) * * Called with bclink_lock locked */ -static void bclink_retransmit_pkt(u32 after, u32 to) +static void bclink_retransmit_pkt(struct tipc_net *tn, u32 after, u32 to) { struct sk_buff *skb; + struct tipc_link *bcl = tn->bcl; - skb_queue_walk(&bcl->outqueue, skb) { + skb_queue_walk(&bcl->transmq, skb) { if (more(buf_seqno(skb), after)) { tipc_link_retransmit(bcl, skb, mod(to - after)); break; @@ -232,13 +173,11 @@ static void bclink_retransmit_pkt(u32 after, u32 to) * * Called with no locks taken */ -void tipc_bclink_wakeup_users(void) +void tipc_bclink_wakeup_users(struct net *net) { - struct sk_buff *skb; - - while ((skb = skb_dequeue(&bclink->link.waiting_sks))) - tipc_sk_rcv(skb); + struct tipc_net *tn = net_generic(net, tipc_net_id); + tipc_sk_rcv(net, &tn->bclink->link.wakeupq); } /** @@ -251,12 +190,17 @@ void tipc_bclink_wakeup_users(void) void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked) { struct sk_buff *skb, *tmp; - struct sk_buff *next; unsigned int released = 0; + struct net *net = n_ptr->net; + struct tipc_net *tn = net_generic(net, tipc_net_id); + + if (unlikely(!n_ptr->bclink.recv_permitted)) + return; + + tipc_bclink_lock(net); - tipc_bclink_lock(); /* Bail out if tx queue is empty (no clean up is required) */ - skb = skb_peek(&bcl->outqueue); + skb = skb_peek(&tn->bcl->transmq); if (!skb) goto exit; @@ -267,43 +211,35 @@ void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked) * acknowledge sent messages only (if other nodes still exist) * or both sent and unsent messages (otherwise) */ - if (bclink->bcast_nodes.count) - acked = bcl->fsm_msg_cnt; + if (tn->bclink->bcast_nodes.count) + acked = tn->bcl->fsm_msg_cnt; else - acked = bcl->next_out_no; + acked = tn->bcl->next_out_no; } else { /* * Bail out if specified sequence number does not correspond * to a message that has been sent and not yet acknowledged */ if (less(acked, buf_seqno(skb)) || - less(bcl->fsm_msg_cnt, acked) || + less(tn->bcl->fsm_msg_cnt, acked) || less_eq(acked, n_ptr->bclink.acked)) goto exit; } /* Skip over packets that node has previously acknowledged */ - skb_queue_walk(&bcl->outqueue, skb) { + skb_queue_walk(&tn->bcl->transmq, skb) { if (more(buf_seqno(skb), n_ptr->bclink.acked)) break; } /* Update packets that node is now acknowledging */ - skb_queue_walk_from_safe(&bcl->outqueue, skb, tmp) { + skb_queue_walk_from_safe(&tn->bcl->transmq, skb, tmp) { if (more(buf_seqno(skb), acked)) break; - - next = tipc_skb_queue_next(&bcl->outqueue, skb); - if (skb != bcl->next_out) { - bcbuf_decr_acks(skb); - } else { - bcbuf_set_acks(skb, 0); - bcl->next_out = next; - bclink_set_last_sent(); - } - + bcbuf_decr_acks(skb); + bclink_set_last_sent(net); if (bcbuf_acks(skb) == 0) { - __skb_unlink(skb, &bcl->outqueue); + __skb_unlink(skb, &tn->bcl->transmq); kfree_skb(skb); released = 1; } @@ -311,15 +247,14 @@ void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked) n_ptr->bclink.acked = acked; /* Try resolving broadcast link congestion, if necessary */ - if (unlikely(bcl->next_out)) { - tipc_link_push_packets(bcl); - bclink_set_last_sent(); + if (unlikely(skb_peek(&tn->bcl->backlogq))) { + tipc_link_push_packets(tn->bcl); + bclink_set_last_sent(net); } - if (unlikely(released && !skb_queue_empty(&bcl->waiting_sks))) + if (unlikely(released && !skb_queue_empty(&tn->bcl->wakeupq))) n_ptr->action_flags |= TIPC_WAKEUP_BCAST_USERS; - exit: - tipc_bclink_unlock(); + tipc_bclink_unlock(net); } /** @@ -327,9 +262,12 @@ exit: * * RCU and node lock set */ -void tipc_bclink_update_link_state(struct tipc_node *n_ptr, u32 last_sent) +void tipc_bclink_update_link_state(struct tipc_node *n_ptr, + u32 last_sent) { struct sk_buff *buf; + struct net *net = n_ptr->net; + struct tipc_net *tn = net_generic(net, tipc_net_id); /* Ignore "stale" link state info */ if (less_eq(last_sent, n_ptr->bclink.last_in)) @@ -356,21 +294,21 @@ void tipc_bclink_update_link_state(struct tipc_node *n_ptr, u32 last_sent) buf = tipc_buf_acquire(INT_H_SIZE); if (buf) { struct tipc_msg *msg = buf_msg(buf); - struct sk_buff *skb = skb_peek(&n_ptr->bclink.deferred_queue); + struct sk_buff *skb = skb_peek(&n_ptr->bclink.deferdq); u32 to = skb ? buf_seqno(skb) - 1 : n_ptr->bclink.last_sent; - tipc_msg_init(msg, BCAST_PROTOCOL, STATE_MSG, + tipc_msg_init(tn->own_addr, msg, BCAST_PROTOCOL, STATE_MSG, INT_H_SIZE, n_ptr->addr); msg_set_non_seq(msg, 1); - msg_set_mc_netid(msg, tipc_net_id); + msg_set_mc_netid(msg, tn->net_id); msg_set_bcast_ack(msg, n_ptr->bclink.last_in); msg_set_bcgap_after(msg, n_ptr->bclink.last_in); msg_set_bcgap_to(msg, to); - tipc_bclink_lock(); - tipc_bearer_send(MAX_BEARERS, buf, NULL); - bcl->stats.sent_nacks++; - tipc_bclink_unlock(); + tipc_bclink_lock(net); + tipc_bearer_send(net, MAX_BEARERS, buf, NULL); + tn->bcl->stats.sent_nacks++; + tipc_bclink_unlock(net); kfree_skb(buf); n_ptr->bclink.oos_state++; @@ -383,34 +321,39 @@ 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. */ -static void bclink_peek_nack(struct tipc_msg *msg) +static void bclink_peek_nack(struct net *net, struct tipc_msg *msg) { - struct tipc_node *n_ptr = tipc_node_find(msg_destnode(msg)); + struct tipc_node *n_ptr = tipc_node_find(net, msg_destnode(msg)); if (unlikely(!n_ptr)) 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 - broadcast buffer chain to all nodes in cluster +/* tipc_bclink_xmit - deliver buffer chain to all nodes in cluster * and to identified node local sockets + * @net: the applicable net namespace * @list: chain of buffers containing message * Consumes the buffer chain, except when returning -ELINKCONG * Returns 0 if success, otherwise errno: -ELINKCONG,-EHOSTUNREACH,-EMSGSIZE */ -int tipc_bclink_xmit(struct sk_buff_head *list) +int tipc_bclink_xmit(struct net *net, struct sk_buff_head *list) { + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_link *bcl = tn->bcl; + struct tipc_bclink *bclink = tn->bclink; int rc = 0; int bc = 0; struct sk_buff *skb; + struct sk_buff_head arrvq; + struct sk_buff_head inputq; /* Prepare clone of message for local node */ skb = tipc_msg_reassemble(list); @@ -418,33 +361,35 @@ int tipc_bclink_xmit(struct sk_buff_head *list) __skb_queue_purge(list); return -EHOSTUNREACH; } - - /* Broadcast to all other nodes */ + /* Broadcast to all nodes */ if (likely(bclink)) { - tipc_bclink_lock(); + tipc_bclink_lock(net); if (likely(bclink->bcast_nodes.count)) { - rc = __tipc_link_xmit(bcl, list); + rc = __tipc_link_xmit(net, bcl, list); if (likely(!rc)) { - u32 len = skb_queue_len(&bcl->outqueue); + u32 len = skb_queue_len(&bcl->transmq); - bclink_set_last_sent(); + bclink_set_last_sent(net); bcl->stats.queue_sz_counts++; bcl->stats.accu_queue_sz += len; } bc = 1; } - tipc_bclink_unlock(); + tipc_bclink_unlock(net); } if (unlikely(!bc)) __skb_queue_purge(list); - /* Deliver message clone */ - if (likely(!rc)) - tipc_sk_mcast_rcv(skb); - else + if (unlikely(rc)) { kfree_skb(skb); - + return rc; + } + /* Deliver message clone */ + __skb_queue_head_init(&arrvq); + skb_queue_head_init(&inputq); + __skb_queue_tail(&arrvq, skb); + tipc_sk_mcast_rcv(net, &arrvq, &inputq); return rc; } @@ -455,19 +400,21 @@ int tipc_bclink_xmit(struct sk_buff_head *list) */ static void bclink_accept_pkt(struct tipc_node *node, u32 seqno) { + struct tipc_net *tn = net_generic(node->net, tipc_net_id); + bclink_update_last_sent(node, seqno); node->bclink.last_in = seqno; node->bclink.oos_state = 0; - bcl->stats.recv_info++; + tn->bcl->stats.recv_info++; /* * Unicast an ACK periodically, ensuring that * all nodes in the cluster don't ACK at the same time */ - if (((seqno - tipc_own_addr) % TIPC_MIN_LINK_WIN) == 0) { + if (((seqno - tn->own_addr) % TIPC_MIN_LINK_WIN) == 0) { tipc_link_proto_xmit(node->active_links[node->addr & 1], - STATE_MSG, 0, 0, 0, 0, 0); - bcl->stats.sent_acks++; + STATE_MSG, 0, 0, 0, 0); + tn->bcl->stats.sent_acks++; } } @@ -476,19 +423,24 @@ static void bclink_accept_pkt(struct tipc_node *node, u32 seqno) * * RCU is locked, no other locks set */ -void tipc_bclink_rcv(struct sk_buff *buf) +void tipc_bclink_rcv(struct net *net, struct sk_buff *buf) { + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_link *bcl = tn->bcl; struct tipc_msg *msg = buf_msg(buf); struct tipc_node *node; u32 next_in; u32 seqno; int deferred = 0; + int pos = 0; + struct sk_buff *iskb; + struct sk_buff_head *arrvq, *inputq; /* Screen out unwanted broadcast messages */ - if (msg_mc_netid(msg) != tipc_net_id) + if (msg_mc_netid(msg) != tn->net_id) goto exit; - node = tipc_node_find(msg_prevnode(msg)); + node = tipc_node_find(net, msg_prevnode(msg)); if (unlikely(!node)) goto exit; @@ -500,71 +452,76 @@ void tipc_bclink_rcv(struct sk_buff *buf) if (unlikely(msg_user(msg) == BCAST_PROTOCOL)) { if (msg_type(msg) != STATE_MSG) goto unlock; - if (msg_destnode(msg) == tipc_own_addr) { + if (msg_destnode(msg) == tn->own_addr) { tipc_bclink_acknowledge(node, msg_bcast_ack(msg)); - tipc_node_unlock(node); - tipc_bclink_lock(); + tipc_bclink_lock(net); bcl->stats.recv_nacks++; - bclink->retransmit_to = node; - bclink_retransmit_pkt(msg_bcgap_after(msg), + tn->bclink->retransmit_to = node; + bclink_retransmit_pkt(tn, msg_bcgap_after(msg), msg_bcgap_to(msg)); - tipc_bclink_unlock(); + tipc_bclink_unlock(net); + tipc_node_unlock(node); } else { tipc_node_unlock(node); - bclink_peek_nack(msg); + bclink_peek_nack(net, msg); } + tipc_node_put(node); goto exit; } /* Handle in-sequence broadcast message */ seqno = msg_seqno(msg); next_in = mod(node->bclink.last_in + 1); + arrvq = &tn->bclink->arrvq; + inputq = &tn->bclink->inputq; if (likely(seqno == next_in)) { receive: /* Deliver message to destination */ if (likely(msg_isdata(msg))) { - tipc_bclink_lock(); + tipc_bclink_lock(net); bclink_accept_pkt(node, seqno); - tipc_bclink_unlock(); + spin_lock_bh(&inputq->lock); + __skb_queue_tail(arrvq, buf); + spin_unlock_bh(&inputq->lock); + node->action_flags |= TIPC_BCAST_MSG_EVT; + tipc_bclink_unlock(net); tipc_node_unlock(node); - if (likely(msg_mcast(msg))) - tipc_sk_mcast_rcv(buf); - else - kfree_skb(buf); } else if (msg_user(msg) == MSG_BUNDLER) { - tipc_bclink_lock(); + tipc_bclink_lock(net); bclink_accept_pkt(node, seqno); bcl->stats.recv_bundles++; bcl->stats.recv_bundled += msg_msgcnt(msg); - tipc_bclink_unlock(); + pos = 0; + while (tipc_msg_extract(buf, &iskb, &pos)) { + spin_lock_bh(&inputq->lock); + __skb_queue_tail(arrvq, iskb); + spin_unlock_bh(&inputq->lock); + } + node->action_flags |= TIPC_BCAST_MSG_EVT; + tipc_bclink_unlock(net); tipc_node_unlock(node); - tipc_link_bundle_rcv(buf); } else if (msg_user(msg) == MSG_FRAGMENTER) { + tipc_bclink_lock(net); + bclink_accept_pkt(node, seqno); tipc_buf_append(&node->bclink.reasm_buf, &buf); - if (unlikely(!buf && !node->bclink.reasm_buf)) + if (unlikely(!buf && !node->bclink.reasm_buf)) { + tipc_bclink_unlock(net); goto unlock; - tipc_bclink_lock(); - bclink_accept_pkt(node, seqno); + } bcl->stats.recv_fragments++; if (buf) { bcl->stats.recv_fragmented++; msg = buf_msg(buf); - tipc_bclink_unlock(); + tipc_bclink_unlock(net); goto receive; } - tipc_bclink_unlock(); - tipc_node_unlock(node); - } else if (msg_user(msg) == NAME_DISTRIBUTOR) { - tipc_bclink_lock(); - bclink_accept_pkt(node, seqno); - tipc_bclink_unlock(); + tipc_bclink_unlock(net); tipc_node_unlock(node); - tipc_named_rcv(buf); } else { - tipc_bclink_lock(); + tipc_bclink_lock(net); bclink_accept_pkt(node, seqno); - tipc_bclink_unlock(); + tipc_bclink_unlock(net); tipc_node_unlock(node); kfree_skb(buf); } @@ -578,41 +535,42 @@ receive: if (node->bclink.last_in == node->bclink.last_sent) goto unlock; - if (skb_queue_empty(&node->bclink.deferred_queue)) { + if (skb_queue_empty(&node->bclink.deferdq)) { node->bclink.oos_state = 1; goto unlock; } - msg = buf_msg(skb_peek(&node->bclink.deferred_queue)); + msg = buf_msg(skb_peek(&node->bclink.deferdq)); seqno = msg_seqno(msg); next_in = mod(next_in + 1); if (seqno != next_in) goto unlock; /* Take in-sequence message from deferred queue & deliver it */ - buf = __skb_dequeue(&node->bclink.deferred_queue); + buf = __skb_dequeue(&node->bclink.deferdq); goto receive; } /* Handle out-of-sequence broadcast message */ if (less(next_in, seqno)) { - deferred = tipc_link_defer_pkt(&node->bclink.deferred_queue, + deferred = tipc_link_defer_pkt(&node->bclink.deferdq, buf); bclink_update_last_sent(node, seqno); buf = NULL; } - tipc_bclink_lock(); + tipc_bclink_lock(net); if (deferred) bcl->stats.deferred_recv++; else bcl->stats.duplicates++; - tipc_bclink_unlock(); + tipc_bclink_unlock(net); unlock: tipc_node_unlock(node); + tipc_node_put(node); exit: kfree_skb(buf); } @@ -620,7 +578,7 @@ exit: u32 tipc_bclink_acks_missing(struct tipc_node *n_ptr) { return (n_ptr->bclink.recv_permitted && - (tipc_bclink_get_last_sent() != n_ptr->bclink.acked)); + (tipc_bclink_get_last_sent(n_ptr->net) != n_ptr->bclink.acked)); } @@ -633,11 +591,15 @@ u32 tipc_bclink_acks_missing(struct tipc_node *n_ptr) * Returns 0 (packet sent successfully) under all circumstances, * since the broadcast link's pseudo-bearer never blocks */ -static int tipc_bcbearer_send(struct sk_buff *buf, struct tipc_bearer *unused1, +static int tipc_bcbearer_send(struct net *net, struct sk_buff *buf, + struct tipc_bearer *unused1, struct tipc_media_addr *unused2) { int bp_index; struct tipc_msg *msg = buf_msg(buf); + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_bcbearer *bcbearer = tn->bcbearer; + struct tipc_bclink *bclink = tn->bclink; /* Prepare broadcast link message for reliable transmission, * if first time trying to send it; @@ -647,9 +609,8 @@ static int tipc_bcbearer_send(struct sk_buff *buf, struct tipc_bearer *unused1, if (likely(!msg_non_seq(buf_msg(buf)))) { bcbuf_set_acks(buf, bclink->bcast_nodes.count); msg_set_non_seq(msg, 1); - msg_set_mc_netid(msg, tipc_net_id); - bcl->stats.sent_info++; - + msg_set_mc_netid(msg, tn->net_id); + tn->bcl->stats.sent_info++; if (WARN_ON(!bclink->bcast_nodes.count)) { dump_stack(); return 0; @@ -677,13 +638,14 @@ 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->identity, buf, &b->bcast_addr); + tipc_bearer_send(net, b->identity, buf, &b->bcast_addr); } else { /* Avoid concurrent buffer access */ tbuf = pskb_copy_for_clone(buf, GFP_ATOMIC); if (!tbuf) break; - tipc_bearer_send(b->identity, tbuf, &b->bcast_addr); + tipc_bearer_send(net, b->identity, tbuf, + &b->bcast_addr); kfree_skb(tbuf); /* Bearer keeps a clone */ } if (bcbearer->remains_new.count == 0) @@ -698,15 +660,18 @@ 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(struct tipc_node_map *nm_ptr, u32 node, bool action) +void tipc_bcbearer_sort(struct net *net, struct tipc_node_map *nm_ptr, + u32 node, bool action) { + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_bcbearer *bcbearer = tn->bcbearer; struct tipc_bcbearer_pair *bp_temp = bcbearer->bpairs_temp; struct tipc_bcbearer_pair *bp_curr; struct tipc_bearer *b; int b_index; int pri; - tipc_bclink_lock(); + tipc_bclink_lock(net); if (action) tipc_nmap_add(nm_ptr, node); @@ -718,7 +683,7 @@ void tipc_bcbearer_sort(struct tipc_node_map *nm_ptr, u32 node, bool action) rcu_read_lock(); for (b_index = 0; b_index < MAX_BEARERS; b_index++) { - b = rcu_dereference_rtnl(bearer_list[b_index]); + b = rcu_dereference_rtnl(tn->bearer_list[b_index]); if (!b || !b->nodes.count) continue; @@ -753,7 +718,7 @@ void tipc_bcbearer_sort(struct tipc_node_map *nm_ptr, u32 node, bool action) bp_curr++; } - tipc_bclink_unlock(); + tipc_bclink_unlock(net); } static int __tipc_nl_add_bc_link_stat(struct sk_buff *skb, @@ -807,19 +772,21 @@ msg_full: return -EMSGSIZE; } -int tipc_nl_add_bc_link(struct tipc_nl_msg *msg) +int tipc_nl_add_bc_link(struct net *net, struct tipc_nl_msg *msg) { int err; void *hdr; struct nlattr *attrs; struct nlattr *prop; + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_link *bcl = tn->bcl; if (!bcl) return 0; - tipc_bclink_lock(); + tipc_bclink_lock(net); - hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_v2_family, + hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, NLM_F_MULTI, TIPC_NL_LINK_GET); if (!hdr) return -EMSGSIZE; @@ -844,7 +811,7 @@ int tipc_nl_add_bc_link(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); @@ -852,7 +819,7 @@ int tipc_nl_add_bc_link(struct tipc_nl_msg *msg) if (err) goto attr_msg_full; - tipc_bclink_unlock(); + tipc_bclink_unlock(net); nla_nest_end(msg->skb, attrs); genlmsg_end(msg->skb, hdr); @@ -863,79 +830,49 @@ prop_msg_full: attr_msg_full: nla_nest_cancel(msg->skb, attrs); msg_full: - tipc_bclink_unlock(); + tipc_bclink_unlock(net); genlmsg_cancel(msg->skb, hdr); return -EMSGSIZE; } -int tipc_bclink_stats(char *buf, const u32 buf_size) +int tipc_bclink_reset_stats(struct net *net) { - int ret; - struct tipc_stats *s; + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_link *bcl = tn->bcl; if (!bcl) - return 0; - - tipc_bclink_lock(); - - s = &bcl->stats; - - ret = tipc_snprintf(buf, buf_size, "Link <%s>\n" - " Window:%u packets\n", - bcl->name, bcl->queue_limit[0]); - ret += tipc_snprintf(buf + ret, buf_size - ret, - " RX packets:%u fragments:%u/%u bundles:%u/%u\n", - s->recv_info, s->recv_fragments, - s->recv_fragmented, s->recv_bundles, - s->recv_bundled); - ret += tipc_snprintf(buf + ret, buf_size - ret, - " TX packets:%u fragments:%u/%u bundles:%u/%u\n", - s->sent_info, s->sent_fragments, - s->sent_fragmented, s->sent_bundles, - s->sent_bundled); - ret += tipc_snprintf(buf + ret, buf_size - ret, - " RX naks:%u defs:%u dups:%u\n", - s->recv_nacks, s->deferred_recv, s->duplicates); - ret += tipc_snprintf(buf + ret, buf_size - ret, - " TX naks:%u acks:%u dups:%u\n", - s->sent_nacks, s->sent_acks, s->retransmitted); - ret += tipc_snprintf(buf + ret, buf_size - ret, - " Congestion link:%u Send queue max:%u avg:%u\n", - s->link_congs, s->max_queue_sz, - s->queue_sz_counts ? - (s->accu_queue_sz / s->queue_sz_counts) : 0); - - tipc_bclink_unlock(); - return ret; -} - -int tipc_bclink_reset_stats(void) -{ - if (!bcl) return -ENOPROTOOPT; - tipc_bclink_lock(); + tipc_bclink_lock(net); memset(&bcl->stats, 0, sizeof(bcl->stats)); - tipc_bclink_unlock(); + tipc_bclink_unlock(net); return 0; } -int tipc_bclink_set_queue_limits(u32 limit) +int tipc_bclink_set_queue_limits(struct net *net, u32 limit) { + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_link *bcl = tn->bcl; + if (!bcl) return -ENOPROTOOPT; if ((limit < TIPC_MIN_LINK_WIN) || (limit > TIPC_MAX_LINK_WIN)) return -EINVAL; - tipc_bclink_lock(); + tipc_bclink_lock(net); tipc_link_set_queue_limits(bcl, limit); - tipc_bclink_unlock(); + tipc_bclink_unlock(net); return 0; } -int tipc_bclink_init(void) +int tipc_bclink_init(struct net *net) { + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_bcbearer *bcbearer; + struct tipc_bclink *bclink; + struct tipc_link *bcl; + bcbearer = kzalloc(sizeof(*bcbearer), GFP_ATOMIC); if (!bcbearer) return -ENOMEM; @@ -952,32 +889,42 @@ int tipc_bclink_init(void) sprintf(bcbearer->media.name, "tipc-broadcast"); spin_lock_init(&bclink->lock); - __skb_queue_head_init(&bcl->outqueue); - __skb_queue_head_init(&bcl->deferred_queue); - skb_queue_head_init(&bcl->waiting_sks); + __skb_queue_head_init(&bcl->transmq); + __skb_queue_head_init(&bcl->backlogq); + __skb_queue_head_init(&bcl->deferdq); + skb_queue_head_init(&bcl->wakeupq); bcl->next_out_no = 1; spin_lock_init(&bclink->node.lock); - __skb_queue_head_init(&bclink->node.waiting_sks); + __skb_queue_head_init(&bclink->arrvq); + skb_queue_head_init(&bclink->inputq); bcl->owner = &bclink->node; - bcl->max_pkt = MAX_PKT_DEFAULT_MCAST; + bcl->owner->net = net; + bcl->mtu = MAX_PKT_DEFAULT_MCAST; tipc_link_set_queue_limits(bcl, BCLINK_WIN_DEFAULT); bcl->bearer_id = MAX_BEARERS; - rcu_assign_pointer(bearer_list[MAX_BEARERS], &bcbearer->bearer); + rcu_assign_pointer(tn->bearer_list[MAX_BEARERS], &bcbearer->bearer); bcl->state = WORKING_WORKING; + bcl->pmsg = (struct tipc_msg *)&bcl->proto_msg; + msg_set_prevnode(bcl->pmsg, tn->own_addr); strlcpy(bcl->name, tipc_bclink_name, TIPC_MAX_LINK_NAME); + tn->bcbearer = bcbearer; + tn->bclink = bclink; + tn->bcl = bcl; return 0; } -void tipc_bclink_stop(void) +void tipc_bclink_stop(struct net *net) { - tipc_bclink_lock(); - tipc_link_purge_queues(bcl); - tipc_bclink_unlock(); + struct tipc_net *tn = net_generic(net, tipc_net_id); + + tipc_bclink_lock(net); + tipc_link_purge_queues(tn->bcl); + tipc_bclink_unlock(net); - RCU_INIT_POINTER(bearer_list[BCBEARER], NULL); + RCU_INIT_POINTER(tn->bearer_list[BCBEARER], NULL); synchronize_net(); - kfree(bcbearer); - kfree(bclink); + kfree(tn->bcbearer); + kfree(tn->bclink); } /** @@ -1037,50 +984,3 @@ static void tipc_nmap_diff(struct tipc_node_map *nm_a, } } } - -/** - * tipc_port_list_add - add a port to a port list, ensuring no duplicates - */ -void tipc_port_list_add(struct tipc_port_list *pl_ptr, u32 port) -{ - struct tipc_port_list *item = pl_ptr; - int i; - int item_sz = PLSIZE; - int cnt = pl_ptr->count; - - for (; ; cnt -= item_sz, item = item->next) { - if (cnt < PLSIZE) - item_sz = cnt; - for (i = 0; i < item_sz; i++) - if (item->ports[i] == port) - return; - if (i < PLSIZE) { - item->ports[i] = port; - pl_ptr->count++; - return; - } - if (!item->next) { - item->next = kmalloc(sizeof(*item), GFP_ATOMIC); - if (!item->next) { - pr_warn("Incomplete multicast delivery, no memory\n"); - return; - } - item->next->next = NULL; - } - } -} - -/** - * tipc_port_list_free - free dynamically created entries in port_list chain - * - */ -void tipc_port_list_free(struct tipc_port_list *pl_ptr) -{ - struct tipc_port_list *item; - struct tipc_port_list *next; - - for (item = pl_ptr->next; item; item = next) { - next = item->next; - kfree(item); - } -} diff --git a/net/tipc/bcast.h b/net/tipc/bcast.h index 644d79129fba..4bdc12277d33 100644 --- a/net/tipc/bcast.h +++ b/net/tipc/bcast.h @@ -1,7 +1,7 @@ /* * net/tipc/bcast.h: Include file for TIPC broadcast code * - * Copyright (c) 2003-2006, 2014, Ericsson AB + * Copyright (c) 2003-2006, 2014-2015, Ericsson AB * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * @@ -37,39 +37,70 @@ #ifndef _TIPC_BCAST_H #define _TIPC_BCAST_H -#include "netlink.h" - -#define MAX_NODES 4096 -#define WSIZE 32 -#define TIPC_BCLINK_RESET 1 +#include <linux/tipc_config.h> +#include "link.h" +#include "node.h" /** - * struct tipc_node_map - set of node identifiers - * @count: # of nodes in set - * @map: bitmap of node identifiers that are in the set + * struct tipc_bcbearer_pair - a pair of bearers used by broadcast link + * @primary: pointer to primary bearer + * @secondary: pointer to secondary bearer + * + * Bearers must have same priority and same set of reachable destinations + * to be paired. */ -struct tipc_node_map { - u32 count; - u32 map[MAX_NODES / WSIZE]; + +struct tipc_bcbearer_pair { + struct tipc_bearer *primary; + struct tipc_bearer *secondary; }; -#define PLSIZE 32 +#define BCBEARER MAX_BEARERS /** - * struct tipc_port_list - set of node local destination ports - * @count: # of ports in set (only valid for first entry in list) - * @next: pointer to next entry in list - * @ports: array of port references + * struct tipc_bcbearer - bearer used by broadcast link + * @bearer: (non-standard) broadcast bearer structure + * @media: (non-standard) broadcast media structure + * @bpairs: array of bearer pairs + * @bpairs_temp: temporary array of bearer pairs used by tipc_bcbearer_sort() + * @remains: temporary node map used by tipc_bcbearer_send() + * @remains_new: temporary node map used tipc_bcbearer_send() + * + * 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 "bclink_lock". */ -struct tipc_port_list { - int count; - struct tipc_port_list *next; - u32 ports[PLSIZE]; +struct tipc_bcbearer { + struct tipc_bearer bearer; + struct tipc_media media; + struct tipc_bcbearer_pair bpairs[MAX_BEARERS]; + struct tipc_bcbearer_pair bpairs_temp[TIPC_MAX_LINK_PRI + 1]; + struct tipc_node_map remains; + struct tipc_node_map remains_new; }; +/** + * 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 + * @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; + struct sk_buff_head arrvq; + struct sk_buff_head inputq; + struct tipc_node_map bcast_nodes; + struct tipc_node *retransmit_to; +}; struct tipc_node; - extern const char tipc_bclink_name[]; /** @@ -81,27 +112,25 @@ static inline int tipc_nmap_equal(struct tipc_node_map *nm_a, return !memcmp(nm_a, nm_b, sizeof(*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); - -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); +int tipc_bclink_init(struct net *net); +void tipc_bclink_stop(struct net *net); +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); void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked); -void tipc_bclink_rcv(struct sk_buff *buf); -u32 tipc_bclink_get_last_sent(void); +void tipc_bclink_rcv(struct net *net, struct sk_buff *buf); +u32 tipc_bclink_get_last_sent(struct net *net); u32 tipc_bclink_acks_missing(struct tipc_node *n_ptr); -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(struct tipc_node_map *nm_ptr, u32 node, bool action); +void tipc_bclink_update_link_state(struct tipc_node *node, + u32 last_sent); +int tipc_bclink_reset_stats(struct net *net); +int tipc_bclink_set_queue_limits(struct net *net, u32 limit); +void tipc_bcbearer_sort(struct net *net, struct tipc_node_map *nm_ptr, + u32 node, bool action); uint tipc_bclink_get_mtu(void); -int tipc_bclink_xmit(struct sk_buff_head *list); -void tipc_bclink_wakeup_users(void); -int tipc_nl_add_bc_link(struct tipc_nl_msg *msg); +int tipc_bclink_xmit(struct net *net, struct sk_buff_head *list); +void tipc_bclink_wakeup_users(struct net *net); +int tipc_nl_add_bc_link(struct net *net, struct tipc_nl_msg *msg); +void tipc_bclink_input(struct net *net); #endif diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 463db5b15b8b..3613e72e858e 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -34,11 +34,12 @@ * POSSIBILITY OF SUCH DAMAGE. */ +#include <net/sock.h> #include "core.h" -#include "config.h" #include "bearer.h" #include "link.h" #include "discover.h" +#include "bcast.h" #define MAX_ADDR_STR 60 @@ -47,6 +48,9 @@ static struct tipc_media * const media_info_array[] = { #ifdef CONFIG_TIPC_MEDIA_IB &ib_media_info, #endif +#ifdef CONFIG_TIPC_MEDIA_UDP + &udp_media_info, +#endif NULL }; @@ -67,9 +71,8 @@ static const struct nla_policy tipc_nl_media_policy[TIPC_NLA_MEDIA_MAX + 1] = { [TIPC_NLA_MEDIA_PROP] = { .type = NLA_NESTED } }; -struct tipc_bearer __rcu *bearer_list[MAX_BEARERS + 1]; - -static void bearer_disable(struct tipc_bearer *b_ptr, bool shutting_down); +static void bearer_disable(struct net *net, struct tipc_bearer *b_ptr, + bool shutting_down); /** * tipc_media_find - locates specified media object by name @@ -111,38 +114,18 @@ void tipc_media_addr_printf(char *buf, int len, struct tipc_media_addr *a) m_ptr = media_find_id(a->media_id); if (m_ptr && !m_ptr->addr2str(a, addr_str, sizeof(addr_str))) - ret = tipc_snprintf(buf, len, "%s(%s)", m_ptr->name, addr_str); + ret = scnprintf(buf, len, "%s(%s)", m_ptr->name, addr_str); else { u32 i; - ret = tipc_snprintf(buf, len, "UNKNOWN(%u)", a->media_id); + ret = scnprintf(buf, len, "UNKNOWN(%u)", a->media_id); for (i = 0; i < sizeof(a->value); i++) - ret += tipc_snprintf(buf - ret, len + ret, + ret += scnprintf(buf - ret, len + ret, "-%02x", a->value[i]); } } /** - * tipc_media_get_names - record names of registered media in buffer - */ -struct sk_buff *tipc_media_get_names(void) -{ - struct sk_buff *buf; - int i; - - buf = tipc_cfg_reply_alloc(MAX_MEDIA * TLV_SPACE(TIPC_MAX_MEDIA_NAME)); - if (!buf) - return NULL; - - for (i = 0; media_info_array[i] != NULL; i++) { - tipc_cfg_append_tlv(buf, TIPC_TLV_MEDIA_NAME, - media_info_array[i]->name, - strlen(media_info_array[i]->name) + 1); - } - return buf; -} - -/** * bearer_name_validate - validate & (optionally) deconstruct bearer name * @name: ptr to bearer name string * @name_parts: ptr to area for bearer name components (or NULL if not needed) @@ -190,68 +173,43 @@ static int bearer_name_validate(const char *name, /** * tipc_bearer_find - locates bearer object with matching bearer name */ -struct tipc_bearer *tipc_bearer_find(const char *name) +struct tipc_bearer *tipc_bearer_find(struct net *net, const char *name) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_bearer *b_ptr; u32 i; for (i = 0; i < MAX_BEARERS; i++) { - b_ptr = rtnl_dereference(bearer_list[i]); + b_ptr = rtnl_dereference(tn->bearer_list[i]); if (b_ptr && (!strcmp(b_ptr->name, name))) return b_ptr; } return NULL; } -/** - * tipc_bearer_get_names - record names of bearers in buffer - */ -struct sk_buff *tipc_bearer_get_names(void) -{ - struct sk_buff *buf; - struct tipc_bearer *b; - int i, j; - - buf = tipc_cfg_reply_alloc(MAX_BEARERS * TLV_SPACE(TIPC_MAX_BEARER_NAME)); - if (!buf) - return NULL; - - for (i = 0; media_info_array[i] != NULL; i++) { - for (j = 0; j < MAX_BEARERS; j++) { - b = rtnl_dereference(bearer_list[j]); - if (!b) - continue; - if (b->media == media_info_array[i]) { - tipc_cfg_append_tlv(buf, TIPC_TLV_BEARER_NAME, - b->name, - strlen(b->name) + 1); - } - } - } - return buf; -} - -void tipc_bearer_add_dest(u32 bearer_id, u32 dest) +void tipc_bearer_add_dest(struct net *net, u32 bearer_id, u32 dest) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_bearer *b_ptr; rcu_read_lock(); - b_ptr = rcu_dereference_rtnl(bearer_list[bearer_id]); + b_ptr = rcu_dereference_rtnl(tn->bearer_list[bearer_id]); if (b_ptr) { - tipc_bcbearer_sort(&b_ptr->nodes, dest, true); + tipc_bcbearer_sort(net, &b_ptr->nodes, dest, true); tipc_disc_add_dest(b_ptr->link_req); } rcu_read_unlock(); } -void tipc_bearer_remove_dest(u32 bearer_id, u32 dest) +void tipc_bearer_remove_dest(struct net *net, u32 bearer_id, u32 dest) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_bearer *b_ptr; rcu_read_lock(); - b_ptr = rcu_dereference_rtnl(bearer_list[bearer_id]); + b_ptr = rcu_dereference_rtnl(tn->bearer_list[bearer_id]); if (b_ptr) { - tipc_bcbearer_sort(&b_ptr->nodes, dest, false); + tipc_bcbearer_sort(net, &b_ptr->nodes, dest, false); tipc_disc_remove_dest(b_ptr->link_req); } rcu_read_unlock(); @@ -260,8 +218,11 @@ void tipc_bearer_remove_dest(u32 bearer_id, u32 dest) /** * tipc_enable_bearer - enable bearer with the given name */ -int tipc_enable_bearer(const char *name, u32 disc_domain, u32 priority) +static int tipc_enable_bearer(struct net *net, const char *name, + u32 disc_domain, u32 priority, + struct nlattr *attr[]) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_bearer *b_ptr; struct tipc_media *m_ptr; struct tipc_bearer_names b_names; @@ -271,7 +232,7 @@ int tipc_enable_bearer(const char *name, u32 disc_domain, u32 priority) u32 i; int res = -EINVAL; - if (!tipc_own_addr) { + if (!tn->own_addr) { pr_warn("Bearer <%s> rejected, not supported in standalone mode\n", name); return -ENOPROTOOPT; @@ -281,11 +242,11 @@ int tipc_enable_bearer(const char *name, u32 disc_domain, u32 priority) return -EINVAL; } if (tipc_addr_domain_valid(disc_domain) && - (disc_domain != tipc_own_addr)) { - if (tipc_in_scope(disc_domain, tipc_own_addr)) { - disc_domain = tipc_own_addr & TIPC_CLUSTER_MASK; + (disc_domain != tn->own_addr)) { + if (tipc_in_scope(disc_domain, tn->own_addr)) { + disc_domain = tn->own_addr & TIPC_CLUSTER_MASK; res = 0; /* accept any node in own cluster */ - } else if (in_own_cluster_exact(disc_domain)) + } else if (in_own_cluster_exact(net, disc_domain)) res = 0; /* accept specified node in own cluster */ } if (res) { @@ -313,7 +274,7 @@ restart: bearer_id = MAX_BEARERS; with_this_prio = 1; for (i = MAX_BEARERS; i-- != 0; ) { - b_ptr = rtnl_dereference(bearer_list[i]); + b_ptr = rtnl_dereference(tn->bearer_list[i]); if (!b_ptr) { bearer_id = i; continue; @@ -347,7 +308,7 @@ restart: strcpy(b_ptr->name, name); b_ptr->media = m_ptr; - res = m_ptr->enable_media(b_ptr); + res = m_ptr->enable_media(net, b_ptr, attr); if (res) { pr_warn("Bearer <%s> rejected, enable failure (%d)\n", name, -res); @@ -361,15 +322,15 @@ restart: b_ptr->net_plane = bearer_id + 'A'; b_ptr->priority = priority; - res = tipc_disc_create(b_ptr, &b_ptr->bcast_addr); + res = tipc_disc_create(net, b_ptr, &b_ptr->bcast_addr); if (res) { - bearer_disable(b_ptr, false); + bearer_disable(net, b_ptr, false); pr_warn("Bearer <%s> rejected, discovery object creation failed\n", name); return -EINVAL; } - rcu_assign_pointer(bearer_list[bearer_id], b_ptr); + rcu_assign_pointer(tn->bearer_list[bearer_id], b_ptr); pr_info("Enabled bearer <%s>, discovery domain %s, priority %u\n", name, @@ -380,11 +341,11 @@ restart: /** * tipc_reset_bearer - Reset all links established over this bearer */ -static int tipc_reset_bearer(struct tipc_bearer *b_ptr) +static int tipc_reset_bearer(struct net *net, struct tipc_bearer *b_ptr) { pr_info("Resetting bearer <%s>\n", b_ptr->name); - tipc_link_reset_list(b_ptr->identity); - tipc_disc_reset(b_ptr); + tipc_link_reset_list(net, b_ptr->identity); + tipc_disc_reset(net, b_ptr); return 0; } @@ -393,49 +354,36 @@ static int tipc_reset_bearer(struct tipc_bearer *b_ptr) * * Note: This routine assumes caller holds RTNL lock. */ -static void bearer_disable(struct tipc_bearer *b_ptr, bool shutting_down) +static void bearer_disable(struct net *net, struct tipc_bearer *b_ptr, + bool shutting_down) { + struct tipc_net *tn = net_generic(net, tipc_net_id); u32 i; pr_info("Disabling bearer <%s>\n", b_ptr->name); b_ptr->media->disable_media(b_ptr); - tipc_link_delete_list(b_ptr->identity, shutting_down); + tipc_link_delete_list(net, b_ptr->identity, shutting_down); if (b_ptr->link_req) tipc_disc_delete(b_ptr->link_req); for (i = 0; i < MAX_BEARERS; i++) { - if (b_ptr == rtnl_dereference(bearer_list[i])) { - RCU_INIT_POINTER(bearer_list[i], NULL); + if (b_ptr == rtnl_dereference(tn->bearer_list[i])) { + RCU_INIT_POINTER(tn->bearer_list[i], NULL); break; } } kfree_rcu(b_ptr, rcu); } -int tipc_disable_bearer(const char *name) -{ - struct tipc_bearer *b_ptr; - int res; - - b_ptr = tipc_bearer_find(name); - if (b_ptr == NULL) { - pr_warn("Attempt to disable unknown bearer <%s>\n", name); - res = -EINVAL; - } else { - bearer_disable(b_ptr, false); - res = 0; - } - return res; -} - -int tipc_enable_l2_media(struct tipc_bearer *b) +int tipc_enable_l2_media(struct net *net, struct tipc_bearer *b, + struct nlattr *attr[]) { struct net_device *dev; char *driver_name = strchr((const char *)b->name, ':') + 1; /* Find device with specified name */ - dev = dev_get_by_name(&init_net, driver_name); + dev = dev_get_by_name(net, driver_name); if (!dev) return -ENODEV; @@ -474,8 +422,8 @@ void tipc_disable_l2_media(struct tipc_bearer *b) * @b_ptr: the bearer through which the packet is to be sent * @dest: peer destination address */ -int tipc_l2_send_msg(struct sk_buff *buf, struct tipc_bearer *b, - struct tipc_media_addr *dest) +int tipc_l2_send_msg(struct net *net, struct sk_buff *buf, + struct tipc_bearer *b, struct tipc_media_addr *dest) { struct sk_buff *clone; struct net_device *dev; @@ -511,15 +459,16 @@ 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(u32 bearer_id, struct sk_buff *buf, +void tipc_bearer_send(struct net *net, u32 bearer_id, struct sk_buff *buf, struct tipc_media_addr *dest) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_bearer *b_ptr; rcu_read_lock(); - b_ptr = rcu_dereference_rtnl(bearer_list[bearer_id]); + b_ptr = rcu_dereference_rtnl(tn->bearer_list[bearer_id]); if (likely(b_ptr)) - b_ptr->media->send_msg(buf, b_ptr, dest); + b_ptr->media->send_msg(net, buf, b_ptr, dest); rcu_read_unlock(); } @@ -539,17 +488,12 @@ static int tipc_l2_rcv_msg(struct sk_buff *buf, struct net_device *dev, { struct tipc_bearer *b_ptr; - if (!net_eq(dev_net(dev), &init_net)) { - kfree_skb(buf); - return NET_RX_DROP; - } - rcu_read_lock(); b_ptr = rcu_dereference_rtnl(dev->tipc_ptr); if (likely(b_ptr)) { if (likely(buf->pkt_type <= PACKET_BROADCAST)) { buf->next = NULL; - tipc_rcv(buf, b_ptr); + tipc_rcv(dev_net(dev), buf, b_ptr); rcu_read_unlock(); return NET_RX_SUCCESS; } @@ -572,11 +516,9 @@ static int tipc_l2_rcv_msg(struct sk_buff *buf, struct net_device *dev, static int tipc_l2_device_event(struct notifier_block *nb, unsigned long evt, void *ptr) { - struct tipc_bearer *b_ptr; struct net_device *dev = netdev_notifier_info_to_dev(ptr); - - if (!net_eq(dev_net(dev), &init_net)) - return NOTIFY_DONE; + struct net *net = dev_net(dev); + struct tipc_bearer *b_ptr; b_ptr = rtnl_dereference(dev->tipc_ptr); if (!b_ptr) @@ -590,16 +532,16 @@ static int tipc_l2_device_event(struct notifier_block *nb, unsigned long evt, break; case NETDEV_DOWN: case NETDEV_CHANGEMTU: - tipc_reset_bearer(b_ptr); + tipc_reset_bearer(net, b_ptr); break; case NETDEV_CHANGEADDR: b_ptr->media->raw2addr(b_ptr, &b_ptr->addr, (char *)dev->dev_addr); - tipc_reset_bearer(b_ptr); + tipc_reset_bearer(net, b_ptr); break; case NETDEV_UNREGISTER: case NETDEV_CHANGENAME: - bearer_disable(b_ptr, false); + bearer_disable(dev_net(dev), b_ptr, false); break; } return NOTIFY_OK; @@ -632,16 +574,17 @@ void tipc_bearer_cleanup(void) dev_remove_pack(&tipc_packet_type); } -void tipc_bearer_stop(void) +void tipc_bearer_stop(struct net *net) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_bearer *b_ptr; u32 i; for (i = 0; i < MAX_BEARERS; i++) { - b_ptr = rtnl_dereference(bearer_list[i]); + b_ptr = rtnl_dereference(tn->bearer_list[i]); if (b_ptr) { - bearer_disable(b_ptr, true); - bearer_list[i] = NULL; + bearer_disable(net, b_ptr, true); + tn->bearer_list[i] = NULL; } } } @@ -654,7 +597,7 @@ static int __tipc_nl_add_bearer(struct tipc_nl_msg *msg, struct nlattr *attrs; struct nlattr *prop; - hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_v2_family, + hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, NLM_F_MULTI, TIPC_NL_BEARER_GET); if (!hdr) return -EMSGSIZE; @@ -698,6 +641,8 @@ int tipc_nl_bearer_dump(struct sk_buff *skb, struct netlink_callback *cb) int i = cb->args[0]; struct tipc_bearer *bearer; struct tipc_nl_msg msg; + struct net *net = sock_net(skb->sk); + struct tipc_net *tn = net_generic(net, tipc_net_id); if (i == MAX_BEARERS) return 0; @@ -708,7 +653,7 @@ int tipc_nl_bearer_dump(struct sk_buff *skb, struct netlink_callback *cb) rtnl_lock(); for (i = 0; i < MAX_BEARERS; i++) { - bearer = rtnl_dereference(bearer_list[i]); + bearer = rtnl_dereference(tn->bearer_list[i]); if (!bearer) continue; @@ -730,6 +675,7 @@ int tipc_nl_bearer_get(struct sk_buff *skb, struct genl_info *info) struct tipc_bearer *bearer; struct tipc_nl_msg msg; struct nlattr *attrs[TIPC_NLA_BEARER_MAX + 1]; + struct net *net = genl_info_net(info); if (!info->attrs[TIPC_NLA_BEARER]) return -EINVAL; @@ -753,7 +699,7 @@ int tipc_nl_bearer_get(struct sk_buff *skb, struct genl_info *info) msg.seq = info->snd_seq; rtnl_lock(); - bearer = tipc_bearer_find(name); + bearer = tipc_bearer_find(net, name); if (!bearer) { err = -EINVAL; goto err_out; @@ -778,6 +724,7 @@ int tipc_nl_bearer_disable(struct sk_buff *skb, struct genl_info *info) char *name; struct tipc_bearer *bearer; struct nlattr *attrs[TIPC_NLA_BEARER_MAX + 1]; + struct net *net = sock_net(skb->sk); if (!info->attrs[TIPC_NLA_BEARER]) return -EINVAL; @@ -794,13 +741,13 @@ int tipc_nl_bearer_disable(struct sk_buff *skb, struct genl_info *info) name = nla_data(attrs[TIPC_NLA_BEARER_NAME]); rtnl_lock(); - bearer = tipc_bearer_find(name); + bearer = tipc_bearer_find(net, name); if (!bearer) { rtnl_unlock(); return -EINVAL; } - bearer_disable(bearer, false); + bearer_disable(net, bearer, false); rtnl_unlock(); return 0; @@ -811,11 +758,13 @@ int tipc_nl_bearer_enable(struct sk_buff *skb, struct genl_info *info) int err; char *bearer; struct nlattr *attrs[TIPC_NLA_BEARER_MAX + 1]; + struct net *net = sock_net(skb->sk); + struct tipc_net *tn = net_generic(net, tipc_net_id); u32 domain; u32 prio; prio = TIPC_MEDIA_LINK_PRI; - domain = tipc_own_addr & TIPC_CLUSTER_MASK; + domain = tn->own_addr & TIPC_CLUSTER_MASK; if (!info->attrs[TIPC_NLA_BEARER]) return -EINVAL; @@ -847,7 +796,7 @@ int tipc_nl_bearer_enable(struct sk_buff *skb, struct genl_info *info) } rtnl_lock(); - err = tipc_enable_bearer(bearer, domain, prio); + err = tipc_enable_bearer(net, bearer, domain, prio, attrs); if (err) { rtnl_unlock(); return err; @@ -863,6 +812,7 @@ int tipc_nl_bearer_set(struct sk_buff *skb, struct genl_info *info) char *name; struct tipc_bearer *b; struct nlattr *attrs[TIPC_NLA_BEARER_MAX + 1]; + struct net *net = genl_info_net(info); if (!info->attrs[TIPC_NLA_BEARER]) return -EINVAL; @@ -878,7 +828,7 @@ int tipc_nl_bearer_set(struct sk_buff *skb, struct genl_info *info) name = nla_data(attrs[TIPC_NLA_BEARER_NAME]); rtnl_lock(); - b = tipc_bearer_find(name); + b = tipc_bearer_find(net, name); if (!b) { rtnl_unlock(); return -EINVAL; @@ -913,7 +863,7 @@ static int __tipc_nl_add_media(struct tipc_nl_msg *msg, struct nlattr *attrs; struct nlattr *prop; - hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_v2_family, + hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, NLM_F_MULTI, TIPC_NL_MEDIA_GET); if (!hdr) return -EMSGSIZE; diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h index 2c1230ac5dfe..5cad243ee8fc 100644 --- a/net/tipc/bearer.h +++ b/net/tipc/bearer.h @@ -37,26 +37,39 @@ #ifndef _TIPC_BEARER_H #define _TIPC_BEARER_H -#include "bcast.h" #include "netlink.h" #include <net/genetlink.h> #define MAX_BEARERS 2 -#define MAX_MEDIA 2 +#define MAX_MEDIA 3 +#define MAX_NODES 4096 +#define WSIZE 32 /* 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 32 +#define TIPC_MEDIA_INFO_SIZE 32 #define TIPC_MEDIA_TYPE_OFFSET 3 +#define TIPC_MEDIA_ADDR_OFFSET 4 /* * Identifiers of supported TIPC media types */ #define TIPC_MEDIA_TYPE_ETH 1 #define TIPC_MEDIA_TYPE_IB 2 +#define TIPC_MEDIA_TYPE_UDP 3 + +/** + * struct tipc_node_map - set of node identifiers + * @count: # of nodes in set + * @map: bitmap of node identifiers that are in the set + */ +struct tipc_node_map { + u32 count; + u32 map[MAX_NODES / WSIZE]; +}; /** * struct tipc_media_addr - destination address used by TIPC bearers @@ -65,7 +78,7 @@ * @broadcast: non-zero if address is a broadcast address */ struct tipc_media_addr { - u8 value[TIPC_MEDIA_ADDR_SIZE]; + u8 value[TIPC_MEDIA_INFO_SIZE]; u8 media_id; u8 broadcast; }; @@ -89,10 +102,11 @@ struct tipc_bearer; * @name: media name */ struct tipc_media { - int (*send_msg)(struct sk_buff *buf, + int (*send_msg)(struct net *net, struct sk_buff *buf, struct tipc_bearer *b_ptr, struct tipc_media_addr *dest); - int (*enable_media)(struct tipc_bearer *b_ptr); + int (*enable_media)(struct net *net, struct tipc_bearer *b_ptr, + struct nlattr *attr[]); void (*disable_media)(struct tipc_bearer *b_ptr); int (*addr2str)(struct tipc_media_addr *addr, char *strbuf, @@ -157,17 +171,11 @@ struct tipc_bearer_names { char if_name[TIPC_MAX_IF_NAME]; }; -struct tipc_link; - -extern struct tipc_bearer __rcu *bearer_list[]; - /* * TIPC routines available to supported media types */ -void tipc_rcv(struct sk_buff *skb, struct tipc_bearer *tb_ptr); -int tipc_enable_bearer(const char *bearer_name, u32 disc_domain, u32 priority); -int tipc_disable_bearer(const char *name); +void tipc_rcv(struct net *net, struct sk_buff *skb, struct tipc_bearer *b_ptr); /* * Routines made available to TIPC by supported media types @@ -177,6 +185,9 @@ extern struct tipc_media eth_media_info; #ifdef CONFIG_TIPC_MEDIA_IB extern struct tipc_media ib_media_info; #endif +#ifdef CONFIG_TIPC_MEDIA_UDP +extern struct tipc_media udp_media_info; +#endif int tipc_nl_bearer_disable(struct sk_buff *skb, struct genl_info *info); int tipc_nl_bearer_enable(struct sk_buff *skb, struct genl_info *info); @@ -191,21 +202,20 @@ int tipc_nl_media_set(struct sk_buff *skb, struct genl_info *info); 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); -int tipc_enable_l2_media(struct tipc_bearer *b); +int tipc_enable_l2_media(struct net *net, struct tipc_bearer *b, + struct nlattr *attrs[]); 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); +int tipc_l2_send_msg(struct net *net, 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(u32 bearer_id, u32 dest); -void tipc_bearer_remove_dest(u32 bearer_id, u32 dest); -struct tipc_bearer *tipc_bearer_find(const char *name); +void tipc_bearer_add_dest(struct net *net, u32 bearer_id, u32 dest); +void tipc_bearer_remove_dest(struct net *net, u32 bearer_id, u32 dest); +struct tipc_bearer *tipc_bearer_find(struct net *net, 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(u32 bearer_id, struct sk_buff *buf, +void tipc_bearer_stop(struct net *net); +void tipc_bearer_send(struct net *net, 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 deleted file mode 100644 index 876f4c6a2631..000000000000 --- a/net/tipc/config.c +++ /dev/null @@ -1,342 +0,0 @@ -/* - * net/tipc/config.c: TIPC configuration management code - * - * Copyright (c) 2002-2006, Ericsson AB - * Copyright (c) 2004-2007, 2010-2013, 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" -#include "socket.h" -#include "name_table.h" -#include "config.h" -#include "server.h" - -#define REPLY_TRUNCATED "<truncated>\n" - -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 */ - -struct sk_buff *tipc_cfg_reply_alloc(int payload_size) -{ - struct sk_buff *buf; - - buf = alloc_skb(rep_headroom + payload_size, GFP_ATOMIC); - if (buf) - skb_reserve(buf, rep_headroom); - return buf; -} - -int tipc_cfg_append_tlv(struct sk_buff *buf, int tlv_type, - void *tlv_data, int tlv_data_size) -{ - struct tlv_desc *tlv = (struct tlv_desc *)skb_tail_pointer(buf); - int new_tlv_space = TLV_SPACE(tlv_data_size); - - if (skb_tailroom(buf) < new_tlv_space) - return 0; - skb_put(buf, new_tlv_space); - tlv->tlv_type = htons(tlv_type); - tlv->tlv_len = htons(TLV_LENGTH(tlv_data_size)); - if (tlv_data_size && tlv_data) - memcpy(TLV_DATA(tlv), tlv_data, tlv_data_size); - return 1; -} - -static struct sk_buff *tipc_cfg_reply_unsigned_type(u16 tlv_type, u32 value) -{ - struct sk_buff *buf; - __be32 value_net; - - buf = tipc_cfg_reply_alloc(TLV_SPACE(sizeof(value))); - if (buf) { - value_net = htonl(value); - tipc_cfg_append_tlv(buf, tlv_type, &value_net, - sizeof(value_net)); - } - return buf; -} - -static struct sk_buff *tipc_cfg_reply_unsigned(u32 value) -{ - return tipc_cfg_reply_unsigned_type(TIPC_TLV_UNSIGNED, value); -} - -struct sk_buff *tipc_cfg_reply_string_type(u16 tlv_type, char *string) -{ - struct sk_buff *buf; - int string_len = strlen(string) + 1; - - buf = tipc_cfg_reply_alloc(TLV_SPACE(string_len)); - if (buf) - tipc_cfg_append_tlv(buf, tlv_type, string, string_len); - return buf; -} - -static struct sk_buff *tipc_show_stats(void) -{ - struct sk_buff *buf; - struct tlv_desc *rep_tlv; - char *pb; - int pb_len; - int str_len; - u32 value; - - if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) - return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - - value = ntohl(*(u32 *)TLV_DATA(req_tlv_area)); - if (value != 0) - return tipc_cfg_reply_error_string("unsupported argument"); - - buf = tipc_cfg_reply_alloc(TLV_SPACE(ULTRA_STRING_MAX_LEN)); - if (buf == NULL) - return NULL; - - rep_tlv = (struct tlv_desc *)buf->data; - pb = TLV_DATA(rep_tlv); - pb_len = ULTRA_STRING_MAX_LEN; - - str_len = tipc_snprintf(pb, pb_len, "TIPC version " TIPC_MOD_VER "\n"); - str_len += 1; /* for "\0" */ - skb_put(buf, TLV_SPACE(str_len)); - TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); - - return buf; -} - -static struct sk_buff *cfg_enable_bearer(void) -{ - struct tipc_bearer_config *args; - - if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_BEARER_CONFIG)) - return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - - args = (struct tipc_bearer_config *)TLV_DATA(req_tlv_area); - if (tipc_enable_bearer(args->name, - ntohl(args->disc_domain), - ntohl(args->priority))) - return tipc_cfg_reply_error_string("unable to enable bearer"); - - return tipc_cfg_reply_none(); -} - -static struct sk_buff *cfg_disable_bearer(void) -{ - if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_BEARER_NAME)) - return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - - if (tipc_disable_bearer((char *)TLV_DATA(req_tlv_area))) - return tipc_cfg_reply_error_string("unable to disable bearer"); - - return tipc_cfg_reply_none(); -} - -static struct sk_buff *cfg_set_own_addr(void) -{ - u32 addr; - - if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) - return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - - addr = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); - if (addr == tipc_own_addr) - return tipc_cfg_reply_none(); - if (!tipc_addr_node_valid(addr)) - return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (node address)"); - if (tipc_own_addr) - return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (cannot change node address once assigned)"); - 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) -{ - u32 value; - - if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) - return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); - if (value == tipc_max_ports) - return tipc_cfg_reply_none(); - if (value < 127 || value > 65535) - return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (max ports must be 127-65535)"); - return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (cannot change max ports while TIPC is active)"); -} - -static struct sk_buff *cfg_set_netid(void) -{ - u32 value; - - if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) - return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); - if (value == tipc_net_id) - return tipc_cfg_reply_none(); - if (value < 1 || value > 9999) - return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (network id must be 1-9999)"); - if (tipc_own_addr) - return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (cannot change network id once TIPC has joined a network)"); - tipc_net_id = value; - return tipc_cfg_reply_none(); -} - -struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area, - int request_space, int reply_headroom) -{ - struct sk_buff *rep_tlv_buf; - - rtnl_lock(); - - /* Save request and reply details in a well-known location */ - req_tlv_area = request_area; - req_tlv_space = request_space; - rep_headroom = reply_headroom; - - /* Check command authorization */ - if (likely(in_own_node(orig_node))) { - /* command is permitted */ - } else { - rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (cannot be done remotely)"); - goto exit; - } - - /* Call appropriate processing routine */ - switch (cmd) { - case TIPC_CMD_NOOP: - rep_tlv_buf = tipc_cfg_reply_none(); - break; - case TIPC_CMD_GET_NODES: - rep_tlv_buf = tipc_node_get_nodes(req_tlv_area, req_tlv_space); - break; - case TIPC_CMD_GET_LINKS: - rep_tlv_buf = tipc_node_get_links(req_tlv_area, req_tlv_space); - break; - case TIPC_CMD_SHOW_LINK_STATS: - rep_tlv_buf = tipc_link_cmd_show_stats(req_tlv_area, req_tlv_space); - break; - case TIPC_CMD_RESET_LINK_STATS: - rep_tlv_buf = tipc_link_cmd_reset_stats(req_tlv_area, req_tlv_space); - break; - case TIPC_CMD_SHOW_NAME_TABLE: - rep_tlv_buf = tipc_nametbl_get(req_tlv_area, req_tlv_space); - break; - case TIPC_CMD_GET_BEARER_NAMES: - rep_tlv_buf = tipc_bearer_get_names(); - break; - case TIPC_CMD_GET_MEDIA_NAMES: - rep_tlv_buf = tipc_media_get_names(); - break; - case TIPC_CMD_SHOW_PORTS: - rep_tlv_buf = tipc_sk_socks_show(); - break; - case TIPC_CMD_SHOW_STATS: - rep_tlv_buf = tipc_show_stats(); - break; - case TIPC_CMD_SET_LINK_TOL: - case TIPC_CMD_SET_LINK_PRI: - case TIPC_CMD_SET_LINK_WINDOW: - rep_tlv_buf = tipc_link_cmd_config(req_tlv_area, req_tlv_space, cmd); - break; - case TIPC_CMD_ENABLE_BEARER: - rep_tlv_buf = cfg_enable_bearer(); - break; - case TIPC_CMD_DISABLE_BEARER: - rep_tlv_buf = cfg_disable_bearer(); - break; - case TIPC_CMD_SET_NODE_ADDR: - rep_tlv_buf = cfg_set_own_addr(); - break; - case TIPC_CMD_SET_MAX_PORTS: - rep_tlv_buf = cfg_set_max_ports(); - break; - case TIPC_CMD_SET_NETID: - rep_tlv_buf = cfg_set_netid(); - break; - case TIPC_CMD_GET_MAX_PORTS: - rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_ports); - break; - case TIPC_CMD_GET_NETID: - rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_net_id); - break; - case TIPC_CMD_NOT_NET_ADMIN: - rep_tlv_buf = - tipc_cfg_reply_error_string(TIPC_CFG_NOT_NET_ADMIN); - break; - case TIPC_CMD_SET_MAX_ZONES: - case TIPC_CMD_GET_MAX_ZONES: - case TIPC_CMD_SET_MAX_SLAVES: - case TIPC_CMD_GET_MAX_SLAVES: - case TIPC_CMD_SET_MAX_CLUSTERS: - case TIPC_CMD_GET_MAX_CLUSTERS: - case TIPC_CMD_SET_MAX_NODES: - case TIPC_CMD_GET_MAX_NODES: - case TIPC_CMD_SET_MAX_SUBSCR: - case TIPC_CMD_GET_MAX_SUBSCR: - case TIPC_CMD_SET_MAX_PUBL: - case TIPC_CMD_GET_MAX_PUBL: - case TIPC_CMD_SET_LOG_SIZE: - case TIPC_CMD_SET_REMOTE_MNG: - case TIPC_CMD_GET_REMOTE_MNG: - case TIPC_CMD_DUMP_LOG: - rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (obsolete command)"); - break; - default: - rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (unknown command)"); - break; - } - - WARN_ON(rep_tlv_buf->len > TLV_SPACE(ULTRA_STRING_MAX_LEN)); - - /* Append an error message if we cannot return all requested data */ - if (rep_tlv_buf->len == TLV_SPACE(ULTRA_STRING_MAX_LEN)) { - if (*(rep_tlv_buf->data + ULTRA_STRING_MAX_LEN) != '\0') - sprintf(rep_tlv_buf->data + rep_tlv_buf->len - - sizeof(REPLY_TRUNCATED) - 1, REPLY_TRUNCATED); - } - - /* Return reply buffer */ -exit: - rtnl_unlock(); - return rep_tlv_buf; -} diff --git a/net/tipc/config.h b/net/tipc/config.h deleted file mode 100644 index 47b1bf181612..000000000000 --- a/net/tipc/config.h +++ /dev/null @@ -1,67 +0,0 @@ -/* - * net/tipc/config.h: Include file for TIPC configuration service code - * - * Copyright (c) 2003-2006, 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. - */ - -#ifndef _TIPC_CONFIG_H -#define _TIPC_CONFIG_H - -/* ---------------------------------------------------------------------- */ - -#include "link.h" - -struct sk_buff *tipc_cfg_reply_alloc(int payload_size); -int tipc_cfg_append_tlv(struct sk_buff *buf, int tlv_type, - void *tlv_data, int tlv_data_size); -struct sk_buff *tipc_cfg_reply_string_type(u16 tlv_type, char *string); - -static inline struct sk_buff *tipc_cfg_reply_none(void) -{ - return tipc_cfg_reply_alloc(0); -} - -static inline struct sk_buff *tipc_cfg_reply_error_string(char *string) -{ - return tipc_cfg_reply_string_type(TIPC_TLV_ERROR_STRING, string); -} - -static inline struct sk_buff *tipc_cfg_reply_ultra_string(char *string) -{ - return tipc_cfg_reply_string_type(TIPC_TLV_ULTRA_STRING, string); -} - -struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, - const void *req_tlv_area, int req_tlv_space, - int headroom); -#endif diff --git a/net/tipc/core.c b/net/tipc/core.c index a5737b8407dd..be1c9fa60b09 100644 --- a/net/tipc/core.c +++ b/net/tipc/core.c @@ -34,82 +34,88 @@ * POSSIBILITY OF SUCH DAMAGE. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include "core.h" #include "name_table.h" #include "subscr.h" -#include "config.h" +#include "bearer.h" +#include "net.h" #include "socket.h" #include <linux/module.h> -/* global variables used by multiple sub-systems within TIPC */ -int tipc_random __read_mostly; - /* configurable TIPC parameters */ -u32 tipc_own_addr __read_mostly; -int tipc_max_ports __read_mostly; int tipc_net_id __read_mostly; int sysctl_tipc_rmem[3] __read_mostly; /* min/default/max */ -/** - * tipc_buf_acquire - creates a TIPC message buffer - * @size: message size (including TIPC header) - * - * Returns a new buffer with data pointers set to the specified size. - * - * NOTE: Headroom is reserved to allow prepending of a data link header. - * There may also be unrequested tailroom present at the buffer's end. - */ -struct sk_buff *tipc_buf_acquire(u32 size) +static int __net_init tipc_init_net(struct net *net) { - struct sk_buff *skb; - unsigned int buf_size = (BUF_HEADROOM + size + 3) & ~3u; - - skb = alloc_skb_fclone(buf_size, GFP_ATOMIC); - if (skb) { - skb_reserve(skb, BUF_HEADROOM); - skb_put(skb, size); - skb->next = NULL; - } - return skb; + struct tipc_net *tn = net_generic(net, tipc_net_id); + int err; + + tn->net_id = 4711; + tn->own_addr = 0; + get_random_bytes(&tn->random, sizeof(int)); + INIT_LIST_HEAD(&tn->node_list); + spin_lock_init(&tn->node_list_lock); + + err = tipc_sk_rht_init(net); + if (err) + goto out_sk_rht; + + err = tipc_nametbl_init(net); + if (err) + goto out_nametbl; + + err = tipc_subscr_start(net); + if (err) + goto out_subscr; + return 0; + +out_subscr: + tipc_nametbl_stop(net); +out_nametbl: + tipc_sk_rht_destroy(net); +out_sk_rht: + return err; } -/** - * tipc_core_stop - switch TIPC from SINGLE NODE to NOT RUNNING mode - */ -static void tipc_core_stop(void) +static void __net_exit tipc_exit_net(struct net *net) { - tipc_net_stop(); - tipc_bearer_cleanup(); - tipc_netlink_stop(); - tipc_subscr_stop(); - tipc_nametbl_stop(); - tipc_sk_ref_table_stop(); - tipc_socket_stop(); - tipc_unregister_sysctl(); + tipc_subscr_stop(net); + tipc_net_stop(net); + tipc_nametbl_stop(net); + tipc_sk_rht_destroy(net); } -/** - * tipc_core_start - switch TIPC from NOT RUNNING to SINGLE NODE mode - */ -static int tipc_core_start(void) +static struct pernet_operations tipc_net_ops = { + .init = tipc_init_net, + .exit = tipc_exit_net, + .id = &tipc_net_id, + .size = sizeof(struct tipc_net), +}; + +static int __init tipc_init(void) { int err; - get_random_bytes(&tipc_random, sizeof(tipc_random)); - - err = tipc_sk_ref_table_init(tipc_max_ports, tipc_random); - if (err) - goto out_reftbl; + pr_info("Activated (version " TIPC_MOD_VER ")\n"); - err = tipc_nametbl_init(); - if (err) - goto out_nametbl; + 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] = TIPC_CONN_OVERLOAD_LIMIT; err = tipc_netlink_start(); if (err) goto out_netlink; + err = tipc_netlink_compat_start(); + if (err) + goto out_netlink_compat; + err = tipc_socket_init(); if (err) goto out_socket; @@ -118,58 +124,40 @@ static int tipc_core_start(void) if (err) goto out_sysctl; - err = tipc_subscr_start(); + err = register_pernet_subsys(&tipc_net_ops); if (err) - goto out_subscr; + goto out_pernet; err = tipc_bearer_setup(); if (err) goto out_bearer; + pr_info("Started in single node mode\n"); return 0; out_bearer: - tipc_subscr_stop(); -out_subscr: + unregister_pernet_subsys(&tipc_net_ops); +out_pernet: tipc_unregister_sysctl(); out_sysctl: tipc_socket_stop(); out_socket: + tipc_netlink_compat_stop(); +out_netlink_compat: tipc_netlink_stop(); out_netlink: - tipc_nametbl_stop(); -out_nametbl: - tipc_sk_ref_table_stop(); -out_reftbl: + pr_err("Unable to start in single node mode\n"); return err; } -static int __init tipc_init(void) -{ - int res; - - pr_info("Activated (version " TIPC_MOD_VER ")\n"); - - tipc_own_addr = 0; - tipc_max_ports = CONFIG_TIPC_PORTS; - tipc_net_id = 4711; - - 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] = TIPC_CONN_OVERLOAD_LIMIT; - - res = tipc_core_start(); - if (res) - pr_err("Unable to start in single node mode\n"); - else - pr_info("Started in single node mode\n"); - return res; -} - static void __exit tipc_exit(void) { - tipc_core_stop(); + tipc_bearer_cleanup(); + unregister_pernet_subsys(&tipc_net_ops); + tipc_netlink_stop(); + tipc_netlink_compat_stop(); + tipc_socket_stop(); + tipc_unregister_sysctl(); + pr_info("Deactivated\n"); } diff --git a/net/tipc/core.h b/net/tipc/core.h index 84602137ce20..3dc68c7a966d 100644 --- a/net/tipc/core.h +++ b/net/tipc/core.h @@ -37,8 +37,6 @@ #ifndef _TIPC_CORE_H #define _TIPC_CORE_H -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - #include <linux/tipc.h> #include <linux/tipc_config.h> #include <linux/tipc_netlink.h> @@ -59,47 +57,54 @@ #include <linux/vmalloc.h> #include <linux/rtnetlink.h> #include <linux/etherdevice.h> +#include <net/netns/generic.h> +#include <linux/rhashtable.h> + +#include "node.h" +#include "bearer.h" +#include "bcast.h" +#include "netlink.h" +#include "link.h" +#include "node.h" +#include "msg.h" #define TIPC_MOD_VER "2.0.0" -#define ULTRA_STRING_MAX_LEN 32768 -#define TIPC_MAX_SUBSCRIPTIONS 65535 -#define TIPC_MAX_PUBLICATIONS 65535 +extern int tipc_net_id __read_mostly; +extern int sysctl_tipc_rmem[3] __read_mostly; +extern int sysctl_tipc_named_timeout __read_mostly; -struct tipc_msg; /* msg.h */ +struct tipc_net { + u32 own_addr; + int net_id; + int random; -int tipc_snprintf(char *buf, int len, const char *fmt, ...); + /* Node table and node list */ + spinlock_t node_list_lock; + struct hlist_head node_htable[NODE_HTABLE_SIZE]; + struct list_head node_list; + u32 num_nodes; + u32 num_links; -/* - * TIPC-specific error codes - */ -#define ELINKCONG EAGAIN /* link congestion <=> resource unavailable */ + /* Bearer list */ + struct tipc_bearer __rcu *bearer_list[MAX_BEARERS + 1]; -/* - * Global configuration variables - */ -extern u32 tipc_own_addr __read_mostly; -extern int tipc_max_ports __read_mostly; -extern int tipc_net_id __read_mostly; -extern int sysctl_tipc_rmem[3] __read_mostly; -extern int sysctl_tipc_named_timeout __read_mostly; + /* Broadcast link */ + struct tipc_bcbearer *bcbearer; + struct tipc_bclink *bclink; + struct tipc_link *bcl; -/* - * Other global variables - */ -extern int tipc_random __read_mostly; + /* Socket hash table */ + struct rhashtable sk_rht; -/* - * Routines available to privileged subsystems - */ -int tipc_netlink_start(void); -void tipc_netlink_stop(void); -int tipc_socket_init(void); -void tipc_socket_stop(void); -int tipc_sock_create_local(int type, struct socket **res); -void tipc_sock_release_local(struct socket *sock); -int tipc_sock_accept_local(struct socket *sock, struct socket **newsock, - int flags); + /* Name table */ + spinlock_t nametbl_lock; + struct name_table *nametbl; + + /* Topology subscription server */ + struct tipc_server *topsrv; + atomic_t subscription_count; +}; #ifdef CONFIG_SYSCTL int tipc_register_sysctl(void); @@ -108,102 +113,4 @@ void tipc_unregister_sysctl(void); #define tipc_register_sysctl() 0 #define tipc_unregister_sysctl() #endif - -/* - * TIPC timer code - */ -typedef void (*Handler) (unsigned long); - -/** - * k_init_timer - initialize a timer - * @timer: pointer to timer structure - * @routine: pointer to routine to invoke when timer expires - * @argument: value to pass to routine when timer expires - * - * Timer must be initialized before use (and terminated when no longer needed). - */ -static inline void k_init_timer(struct timer_list *timer, Handler routine, - unsigned long argument) -{ - setup_timer(timer, routine, argument); -} - -/** - * k_start_timer - start a timer - * @timer: pointer to timer structure - * @msec: time to delay (in ms) - * - * Schedules a previously initialized timer for later execution. - * If timer is already running, the new timeout overrides the previous request. - * - * To ensure the timer doesn't expire before the specified delay elapses, - * the amount of delay is rounded up when converting to the jiffies - * then an additional jiffy is added to account for the fact that - * the starting time may be in the middle of the current jiffy. - */ -static inline void k_start_timer(struct timer_list *timer, unsigned long msec) -{ - mod_timer(timer, jiffies + msecs_to_jiffies(msec) + 1); -} - -/** - * k_cancel_timer - cancel a timer - * @timer: pointer to timer structure - * - * Cancels a previously initialized timer. - * Can be called safely even if the timer is already inactive. - * - * WARNING: Must not be called when holding locks required by the timer's - * timeout routine, otherwise deadlock can occur on SMP systems! - */ -static inline void k_cancel_timer(struct timer_list *timer) -{ - del_timer_sync(timer); -} - -/** - * k_term_timer - terminate a timer - * @timer: pointer to timer structure - * - * Prevents further use of a previously initialized timer. - * - * WARNING: Caller must ensure timer isn't currently running. - * - * (Do not "enhance" this routine to automatically cancel an active timer, - * otherwise deadlock can arise when a timeout routine calls k_term_timer.) - */ -static inline void k_term_timer(struct timer_list *timer) -{ -} - -/* - * TIPC message buffer code - * - * TIPC message buffer headroom reserves space for the worst-case - * link-level device header (in case the message is sent off-node). - * - * Note: Headroom should be a multiple of 4 to ensure the TIPC header fields - * are word aligned for quicker access - */ -#define BUF_HEADROOM LL_MAX_HEADER - -struct tipc_skb_cb { - void *handle; - struct sk_buff *tail; - bool deferred; - bool wakeup_pending; - bool bundling; - u16 chain_sz; - u16 chain_imp; -}; - -#define TIPC_SKB_CB(__skb) ((struct tipc_skb_cb *)&((__skb)->cb[0])) - -static inline struct tipc_msg *buf_msg(struct sk_buff *skb) -{ - return (struct tipc_msg *)skb->data; -} - -struct sk_buff *tipc_buf_acquire(u32 size); - #endif diff --git a/net/tipc/discover.c b/net/tipc/discover.c index aa722a42ef8b..967e292f53c8 100644 --- a/net/tipc/discover.c +++ b/net/tipc/discover.c @@ -1,7 +1,7 @@ /* * net/tipc/discover.c * - * Copyright (c) 2003-2006, 2014, Ericsson AB + * Copyright (c) 2003-2006, 2014-2015, Ericsson AB * Copyright (c) 2005-2006, 2010-2011, Wind River Systems * All rights reserved. * @@ -38,15 +38,19 @@ #include "link.h" #include "discover.h" -#define TIPC_LINK_REQ_INIT 125 /* min delay during bearer start up */ -#define TIPC_LINK_REQ_FAST 1000 /* max delay if bearer has no links */ -#define TIPC_LINK_REQ_SLOW 60000 /* max delay if bearer has links */ -#define TIPC_LINK_REQ_INACTIVE 0xffffffff /* indicates no timer in use */ - +/* min delay during bearer start up */ +#define TIPC_LINK_REQ_INIT msecs_to_jiffies(125) +/* max delay if bearer has no links */ +#define TIPC_LINK_REQ_FAST msecs_to_jiffies(1000) +/* max delay if bearer has links */ +#define TIPC_LINK_REQ_SLOW msecs_to_jiffies(60000) +/* indicates no timer in use */ +#define TIPC_LINK_REQ_INACTIVE 0xffffffff /** * struct tipc_link_req - information about an ongoing link setup request * @bearer_id: identity of bearer issuing requests + * @net: network namespace instance * @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) @@ -58,31 +62,36 @@ struct tipc_link_req { u32 bearer_id; struct tipc_media_addr dest; + struct net *net; u32 domain; int num_nodes; spinlock_t lock; struct sk_buff *buf; struct timer_list timer; - unsigned int timer_intv; + unsigned long timer_intv; }; /** * tipc_disc_init_msg - initialize a link setup message + * @net: the applicable net namespace * @type: message type (request or response) * @b_ptr: ptr to bearer issuing message */ -static void tipc_disc_init_msg(struct sk_buff *buf, u32 type, +static void tipc_disc_init_msg(struct net *net, struct sk_buff *buf, u32 type, struct tipc_bearer *b_ptr) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_msg *msg; u32 dest_domain = b_ptr->domain; msg = buf_msg(buf); - tipc_msg_init(msg, LINK_CONFIG, type, INT_H_SIZE, dest_domain); + tipc_msg_init(tn->own_addr, msg, LINK_CONFIG, type, + MAX_H_SIZE, dest_domain); msg_set_non_seq(msg, 1); - msg_set_node_sig(msg, tipc_random); + msg_set_node_sig(msg, tn->random); + msg_set_node_capabilities(msg, 0); msg_set_dest_domain(msg, dest_domain); - msg_set_bc_netid(msg, tipc_net_id); + msg_set_bc_netid(msg, tn->net_id); b_ptr->media->addr2msg(msg_media_addr(msg), &b_ptr->addr); } @@ -107,11 +116,14 @@ static void disc_dupl_alert(struct tipc_bearer *b_ptr, u32 node_addr, /** * tipc_disc_rcv - handle incoming discovery message (request or response) + * @net: the applicable net namespace * @buf: buffer containing message * @bearer: bearer that message arrived on */ -void tipc_disc_rcv(struct sk_buff *buf, struct tipc_bearer *bearer) +void tipc_disc_rcv(struct net *net, struct sk_buff *buf, + struct tipc_bearer *bearer) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_node *node; struct tipc_link *link; struct tipc_media_addr maddr; @@ -122,6 +134,7 @@ void tipc_disc_rcv(struct sk_buff *buf, struct tipc_bearer *bearer) u32 net_id = msg_bc_netid(msg); u32 mtyp = msg_type(msg); u32 signature = msg_node_sig(msg); + u16 caps = msg_node_capabilities(msg); bool addr_match = false; bool sign_match = false; bool link_up = false; @@ -133,7 +146,7 @@ void tipc_disc_rcv(struct sk_buff *buf, struct tipc_bearer *bearer) kfree_skb(buf); /* Ensure message from node is valid and communication is permitted */ - if (net_id != tipc_net_id) + if (net_id != tn->net_id) return; if (maddr.broadcast) return; @@ -142,24 +155,21 @@ void tipc_disc_rcv(struct sk_buff *buf, struct tipc_bearer *bearer) if (!tipc_addr_node_valid(onode)) return; - if (in_own_node(onode)) { + if (in_own_node(net, onode)) { if (memcmp(&maddr, &bearer->addr, sizeof(maddr))) - disc_dupl_alert(bearer, tipc_own_addr, &maddr); + disc_dupl_alert(bearer, tn->own_addr, &maddr); return; } - if (!tipc_in_scope(ddom, tipc_own_addr)) + if (!tipc_in_scope(ddom, tn->own_addr)) return; if (!tipc_in_scope(bearer->domain, onode)) return; - /* Locate, or if necessary, create, node: */ - node = tipc_node_find(onode); - if (!node) - node = tipc_node_create(onode); + node = tipc_node_create(net, onode); if (!node) return; - tipc_node_lock(node); + node->capabilities = caps; link = node->links[bearer->identity]; /* Prepare to validate requesting node's signature and media address */ @@ -242,14 +252,15 @@ void tipc_disc_rcv(struct sk_buff *buf, struct tipc_bearer *bearer) /* Send response, if necessary */ if (respond && (mtyp == DSC_REQ_MSG)) { - rbuf = tipc_buf_acquire(INT_H_SIZE); + rbuf = tipc_buf_acquire(MAX_H_SIZE); if (rbuf) { - tipc_disc_init_msg(rbuf, DSC_RESP_MSG, bearer); - tipc_bearer_send(bearer->identity, rbuf, &maddr); + tipc_disc_init_msg(net, rbuf, DSC_RESP_MSG, bearer); + tipc_bearer_send(net, bearer->identity, rbuf, &maddr); kfree_skb(rbuf); } } tipc_node_unlock(node); + tipc_node_put(node); } /** @@ -265,7 +276,7 @@ static void disc_update(struct tipc_link_req *req) if ((req->timer_intv == TIPC_LINK_REQ_INACTIVE) || (req->timer_intv > TIPC_LINK_REQ_FAST)) { req->timer_intv = TIPC_LINK_REQ_INIT; - k_start_timer(&req->timer, req->timer_intv); + mod_timer(&req->timer, jiffies + req->timer_intv); } } } @@ -295,12 +306,13 @@ void tipc_disc_remove_dest(struct tipc_link_req *req) /** * disc_timeout - send a periodic link setup request - * @req: ptr to link request structure + * @data: ptr to link request structure * * Called whenever a link setup request timer associated with a bearer expires. */ -static void disc_timeout(struct tipc_link_req *req) +static void disc_timeout(unsigned long data) { + struct tipc_link_req *req = (struct tipc_link_req *)data; int max_delay; spin_lock_bh(&req->lock); @@ -318,7 +330,7 @@ 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_id, req->buf, &req->dest); + tipc_bearer_send(req->net, req->bearer_id, req->buf, &req->dest); req->timer_intv *= 2; @@ -329,44 +341,46 @@ static void disc_timeout(struct tipc_link_req *req) if (req->timer_intv > max_delay) req->timer_intv = max_delay; - k_start_timer(&req->timer, req->timer_intv); + mod_timer(&req->timer, jiffies + req->timer_intv); exit: spin_unlock_bh(&req->lock); } /** * tipc_disc_create - create object to send periodic link setup requests + * @net: the applicable net namespace * @b_ptr: ptr to bearer issuing requests * @dest: destination address for request messages * @dest_domain: network domain to which links can be established * * Returns 0 if successful, otherwise -errno. */ -int tipc_disc_create(struct tipc_bearer *b_ptr, struct tipc_media_addr *dest) +int tipc_disc_create(struct net *net, struct tipc_bearer *b_ptr, + struct tipc_media_addr *dest) { struct tipc_link_req *req; req = kmalloc(sizeof(*req), GFP_ATOMIC); if (!req) return -ENOMEM; - - req->buf = tipc_buf_acquire(INT_H_SIZE); + req->buf = tipc_buf_acquire(MAX_H_SIZE); if (!req->buf) { kfree(req); return -ENOMEM; } - tipc_disc_init_msg(req->buf, DSC_REQ_MSG, b_ptr); + tipc_disc_init_msg(net, req->buf, DSC_REQ_MSG, b_ptr); memcpy(&req->dest, dest, sizeof(*dest)); + req->net = net; 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); + setup_timer(&req->timer, disc_timeout, (unsigned long)req); + mod_timer(&req->timer, jiffies + req->timer_intv); b_ptr->link_req = req; - tipc_bearer_send(req->bearer_id, req->buf, &req->dest); + tipc_bearer_send(net, req->bearer_id, req->buf, &req->dest); return 0; } @@ -376,28 +390,29 @@ int tipc_disc_create(struct tipc_bearer *b_ptr, struct tipc_media_addr *dest) */ void tipc_disc_delete(struct tipc_link_req *req) { - k_cancel_timer(&req->timer); - k_term_timer(&req->timer); + del_timer_sync(&req->timer); kfree_skb(req->buf); kfree(req); } /** * tipc_disc_reset - reset object to send periodic link setup requests + * @net: the applicable net namespace * @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) +void tipc_disc_reset(struct net *net, 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); + tipc_disc_init_msg(net, req->buf, DSC_REQ_MSG, b_ptr); + req->net = net; 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); + mod_timer(&req->timer, jiffies + req->timer_intv); + tipc_bearer_send(net, 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 515b57392f4d..c9b12770c5ed 100644 --- a/net/tipc/discover.h +++ b/net/tipc/discover.h @@ -39,11 +39,13 @@ struct tipc_link_req; -int tipc_disc_create(struct tipc_bearer *b_ptr, struct tipc_media_addr *dest); +int tipc_disc_create(struct net *net, 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_reset(struct net *net, 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); +void tipc_disc_rcv(struct net *net, struct sk_buff *buf, + struct tipc_bearer *b_ptr); #endif diff --git a/net/tipc/eth_media.c b/net/tipc/eth_media.c index 5e1426f1751f..f69a2fde9f4a 100644 --- a/net/tipc/eth_media.c +++ b/net/tipc/eth_media.c @@ -37,8 +37,6 @@ #include "core.h" #include "bearer.h" -#define ETH_ADDR_OFFSET 4 /* MAC addr position inside address field */ - /* Convert Ethernet address (media address format) to string */ static int tipc_eth_addr2str(struct tipc_media_addr *addr, char *strbuf, int bufsz) @@ -53,9 +51,9 @@ static int tipc_eth_addr2str(struct tipc_media_addr *addr, /* Convert from media address format to discovery message addr format */ static int tipc_eth_addr2msg(char *msg, struct tipc_media_addr *addr) { - memset(msg, 0, TIPC_MEDIA_ADDR_SIZE); + memset(msg, 0, TIPC_MEDIA_INFO_SIZE); msg[TIPC_MEDIA_TYPE_OFFSET] = TIPC_MEDIA_TYPE_ETH; - memcpy(msg + ETH_ADDR_OFFSET, addr->value, ETH_ALEN); + memcpy(msg + TIPC_MEDIA_ADDR_OFFSET, addr->value, ETH_ALEN); return 0; } @@ -79,7 +77,7 @@ static int tipc_eth_msg2addr(struct tipc_bearer *b, char *msg) { /* Skip past preamble: */ - msg += ETH_ADDR_OFFSET; + msg += TIPC_MEDIA_ADDR_OFFSET; return tipc_eth_raw2addr(b, addr, msg); } diff --git a/net/tipc/ib_media.c b/net/tipc/ib_media.c index 8522eef9c136..e8c16718e3fa 100644 --- a/net/tipc/ib_media.c +++ b/net/tipc/ib_media.c @@ -57,7 +57,7 @@ static int tipc_ib_addr2str(struct tipc_media_addr *a, char *str_buf, /* Convert from media address format to discovery message addr format */ static int tipc_ib_addr2msg(char *msg, struct tipc_media_addr *addr) { - memset(msg, 0, TIPC_MEDIA_ADDR_SIZE); + memset(msg, 0, TIPC_MEDIA_INFO_SIZE); memcpy(msg, addr->value, INFINIBAND_ALEN); return 0; } diff --git a/net/tipc/link.c b/net/tipc/link.c index 23bcc1132365..a6b30df6ec02 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -1,7 +1,7 @@ /* * net/tipc/link.c: TIPC link code * - * Copyright (c) 1996-2007, 2012-2014, Ericsson AB + * Copyright (c) 1996-2007, 2012-2015, Ericsson AB * Copyright (c) 2004-2007, 2010-2013, Wind River Systems * All rights reserved. * @@ -35,12 +35,12 @@ */ #include "core.h" +#include "subscr.h" #include "link.h" #include "bcast.h" #include "socket.h" #include "name_distr.h" #include "discover.h" -#include "config.h" #include "netlink.h" #include <linux/pkt_sched.h> @@ -89,32 +89,23 @@ static const struct nla_policy tipc_nl_prop_policy[TIPC_NLA_PROP_MAX + 1] = { #define TIMEOUT_EVT 560817u /* link timer expired */ /* - * The following two 'message types' is really just implementation - * data conveniently stored in the message header. - * They must not be considered part of the protocol + * State value stored in 'failover_pkts' */ -#define OPEN_MSG 0 -#define CLOSED_MSG 1 +#define FIRST_FAILOVER 0xffffu -/* - * State value stored in 'exp_msg_count' - */ -#define START_CHANGEOVER 100000u - -static void link_handle_out_of_seq_msg(struct tipc_link *l_ptr, - struct sk_buff *buf); -static void tipc_link_proto_rcv(struct tipc_link *l_ptr, struct sk_buff *buf); -static int tipc_link_tunnel_rcv(struct tipc_node *n_ptr, - struct sk_buff **buf); -static void link_set_supervision_props(struct tipc_link *l_ptr, u32 tolerance); +static void link_handle_out_of_seq_msg(struct tipc_link *link, + struct sk_buff *skb); +static void tipc_link_proto_rcv(struct tipc_link *link, + struct sk_buff *skb); +static void link_set_supervision_props(struct tipc_link *l_ptr, u32 tol); static void link_state_event(struct tipc_link *l_ptr, u32 event); static void link_reset_statistics(struct tipc_link *l_ptr); static void link_print(struct tipc_link *l_ptr, const char *str); static void tipc_link_sync_xmit(struct tipc_link *l); static void tipc_link_sync_rcv(struct tipc_node *n, struct sk_buff *buf); -static int tipc_link_input(struct tipc_link *l, struct sk_buff *buf); -static int tipc_link_prepare_input(struct tipc_link *l, struct sk_buff **buf); - +static void tipc_link_input(struct tipc_link *l, struct sk_buff *skb); +static bool tipc_data_input(struct tipc_link *l, struct sk_buff *skb); +static bool tipc_link_failover_rcv(struct tipc_link *l, struct sk_buff **skb); /* * Simple link routines */ @@ -123,30 +114,26 @@ static unsigned int align(unsigned int i) return (i + 3) & ~3u; } -static void link_init_max_pkt(struct tipc_link *l_ptr) +static void tipc_link_release(struct kref *kref) { - struct tipc_bearer *b_ptr; - u32 max_pkt; - - 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(); + kfree(container_of(kref, struct tipc_link, ref)); +} - if (max_pkt > MAX_MSG_SIZE) - max_pkt = MAX_MSG_SIZE; +static void tipc_link_get(struct tipc_link *l_ptr) +{ + kref_get(&l_ptr->ref); +} - l_ptr->max_pkt_target = max_pkt; - if (l_ptr->max_pkt_target < MAX_PKT_DEFAULT) - l_ptr->max_pkt = l_ptr->max_pkt_target; - else - l_ptr->max_pkt = MAX_PKT_DEFAULT; +static void tipc_link_put(struct tipc_link *l_ptr) +{ + kref_put(&l_ptr->ref, tipc_link_release); +} - l_ptr->max_pkt_probes = 0; +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]; } /* @@ -169,17 +156,18 @@ int tipc_link_is_active(struct tipc_link *l_ptr) * link_timeout - handle expiration of link timer * @l_ptr: pointer to link */ -static void link_timeout(struct tipc_link *l_ptr) +static void link_timeout(unsigned long data) { + struct tipc_link *l_ptr = (struct tipc_link *)data; struct sk_buff *skb; tipc_node_lock(l_ptr->owner); /* update counters used in statistical profiling of send traffic */ - l_ptr->stats.accu_queue_sz += skb_queue_len(&l_ptr->outqueue); + l_ptr->stats.accu_queue_sz += skb_queue_len(&l_ptr->transmq); l_ptr->stats.queue_sz_counts++; - skb = skb_peek(&l_ptr->outqueue); + skb = skb_peek(&l_ptr->transmq); if (skb) { struct tipc_msg *msg = buf_msg(skb); u32 length = msg_size(msg); @@ -211,15 +199,17 @@ static void link_timeout(struct tipc_link *l_ptr) /* do all other link processing performed on a periodic basis */ link_state_event(l_ptr, TIMEOUT_EVT); - if (l_ptr->next_out) + if (skb_queue_len(&l_ptr->backlogq)) tipc_link_push_packets(l_ptr); tipc_node_unlock(l_ptr->owner); + tipc_link_put(l_ptr); } -static void link_set_timer(struct tipc_link *l_ptr, u32 time) +static void link_set_timer(struct tipc_link *link, unsigned long time) { - k_start_timer(&l_ptr->timer, time); + if (!mod_timer(&link->timer, jiffies + time)) + tipc_link_get(link); } /** @@ -234,6 +224,7 @@ struct tipc_link *tipc_link_create(struct tipc_node *n_ptr, struct tipc_bearer *b_ptr, const struct tipc_media_addr *media_addr) { + struct tipc_net *tn = net_generic(n_ptr->net, tipc_net_id); struct tipc_link *l_ptr; struct tipc_msg *msg; char *if_name; @@ -259,12 +250,12 @@ struct tipc_link *tipc_link_create(struct tipc_node *n_ptr, pr_warn("Link creation failed, no memory\n"); return NULL; } - + kref_init(&l_ptr->ref); l_ptr->addr = peer; if_name = strchr(b_ptr->name, ':') + 1; sprintf(l_ptr->name, "%u.%u.%u:%s-%u.%u.%u:unknown", - tipc_zone(tipc_own_addr), tipc_cluster(tipc_own_addr), - tipc_node(tipc_own_addr), + tipc_zone(tn->own_addr), tipc_cluster(tn->own_addr), + tipc_node(tn->own_addr), if_name, tipc_zone(peer), tipc_cluster(peer), tipc_node(peer)); /* note: peer i/f name is updated by reset/activate message */ @@ -278,89 +269,104 @@ struct tipc_link *tipc_link_create(struct tipc_node *n_ptr, l_ptr->pmsg = (struct tipc_msg *)&l_ptr->proto_msg; msg = l_ptr->pmsg; - tipc_msg_init(msg, LINK_PROTOCOL, RESET_MSG, INT_H_SIZE, l_ptr->addr); + tipc_msg_init(tn->own_addr, msg, LINK_PROTOCOL, RESET_MSG, INT_H_SIZE, + l_ptr->addr); msg_set_size(msg, sizeof(l_ptr->proto_msg)); - msg_set_session(msg, (tipc_random & 0xffff)); + msg_set_session(msg, (tn->random & 0xffff)); msg_set_bearer_id(msg, b_ptr->identity); strcpy((char *)msg_data(msg), if_name); - + l_ptr->net_plane = b_ptr->net_plane; + l_ptr->advertised_mtu = b_ptr->mtu; + l_ptr->mtu = l_ptr->advertised_mtu; 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; - __skb_queue_head_init(&l_ptr->outqueue); - __skb_queue_head_init(&l_ptr->deferred_queue); - skb_queue_head_init(&l_ptr->waiting_sks); - + __skb_queue_head_init(&l_ptr->transmq); + __skb_queue_head_init(&l_ptr->backlogq); + __skb_queue_head_init(&l_ptr->deferdq); + skb_queue_head_init(&l_ptr->wakeupq); + skb_queue_head_init(&l_ptr->inputq); + skb_queue_head_init(&l_ptr->namedq); link_reset_statistics(l_ptr); - tipc_node_attach_link(n_ptr, l_ptr); - - k_init_timer(&l_ptr->timer, (Handler)link_timeout, - (unsigned long)l_ptr); - + setup_timer(&l_ptr->timer, link_timeout, (unsigned long)l_ptr); link_state_event(l_ptr, STARTING_EVT); return l_ptr; } -void tipc_link_delete_list(unsigned int bearer_id, bool shutting_down) +/** + * tipc_link_delete - Delete a link + * @l: link to be deleted + */ +void tipc_link_delete(struct tipc_link *l) { - struct tipc_link *l_ptr; - struct tipc_node *n_ptr; + tipc_link_reset(l); + if (del_timer(&l->timer)) + tipc_link_put(l); + l->flags |= LINK_STOPPED; + /* Delete link now, or when timer is finished: */ + tipc_link_reset_fragments(l); + tipc_node_detach_link(l->owner, l); + tipc_link_put(l); +} + +void tipc_link_delete_list(struct net *net, unsigned int bearer_id, + bool shutting_down) +{ + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_link *link; + struct tipc_node *node; rcu_read_lock(); - list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { - 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); - tipc_node_unlock(n_ptr); - - /* Nobody else can access this link now: */ - del_timer_sync(&l_ptr->timer); - kfree(l_ptr); - } else { - /* Detach/delete when failover is finished: */ - l_ptr->flags |= LINK_STOPPED; - tipc_node_unlock(n_ptr); - del_timer_sync(&l_ptr->timer); - } - continue; - } - tipc_node_unlock(n_ptr); + list_for_each_entry_rcu(node, &tn->node_list, list) { + tipc_node_lock(node); + link = node->links[bearer_id]; + if (link) + tipc_link_delete(link); + tipc_node_unlock(node); } rcu_read_unlock(); } /** - * 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, tipc_own_addr, - tipc_own_addr, 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->waiting_sks, 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; } /** @@ -369,17 +375,22 @@ 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 */ -static void link_prepare_wakeup(struct tipc_link *link) +void link_prepare_wakeup(struct tipc_link *l) { - uint pend_qsz = skb_queue_len(&link->outqueue); + int pnd[TIPC_SYSTEM_IMPORTANCE + 1] = {0,}; + int imp, lim; struct sk_buff *skb, *tmp; - skb_queue_walk_safe(&link->waiting_sks, 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->waiting_sks); - skb_queue_tail(&link->owner->waiting_sks, skb); + skb_unlink(skb, &l->wakeupq); + skb_queue_tail(&l->inputq, skb); + l->owner->inputq = &l->inputq; + l->owner->action_flags |= TIPC_MSG_EVT; } } @@ -393,31 +404,42 @@ 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 */ void tipc_link_purge_queues(struct tipc_link *l_ptr) { - __skb_queue_purge(&l_ptr->deferred_queue); - __skb_queue_purge(&l_ptr->outqueue); + __skb_queue_purge(&l_ptr->deferdq); + __skb_queue_purge(&l_ptr->transmq); + tipc_link_purge_backlog(l_ptr); tipc_link_reset_fragments(l_ptr); } void tipc_link_reset(struct tipc_link *l_ptr) { u32 prev_state = l_ptr->state; - u32 checkpoint = l_ptr->next_in_no; int was_active_link = tipc_link_is_active(l_ptr); struct tipc_node *owner = l_ptr->owner; + struct tipc_link *pl = tipc_parallel_link(l_ptr); msg_set_session(l_ptr->pmsg, ((msg_session(l_ptr->pmsg) + 1) & 0xffff)); /* Link is down, accept any session */ l_ptr->peer_session = INVALID_SESSION; - /* Prepare for max packet size negotiation */ - link_init_max_pkt(l_ptr); + /* Prepare for renewed mtu size negotiation */ + l_ptr->mtu = l_ptr->advertised_mtu; l_ptr->state = RESET_UNKNOWN; @@ -425,22 +447,28 @@ 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->bearer_id, l_ptr->addr); - - if (was_active_link && tipc_node_active_links(l_ptr->owner)) { - l_ptr->reset_checkpoint = checkpoint; - l_ptr->exp_msg_count = START_CHANGEOVER; - } - - /* Clean up all queues: */ - __skb_queue_purge(&l_ptr->outqueue); - __skb_queue_purge(&l_ptr->deferred_queue); - if (!skb_queue_empty(&l_ptr->waiting_sks)) { - skb_queue_splice_init(&l_ptr->waiting_sks, &owner->waiting_sks); - owner->action_flags |= TIPC_WAKEUP_USERS; + tipc_bearer_remove_dest(owner->net, l_ptr->bearer_id, l_ptr->addr); + + if (was_active_link && tipc_node_is_up(l_ptr->owner) && (pl != l_ptr)) { + l_ptr->flags |= LINK_FAILINGOVER; + l_ptr->failover_checkpt = l_ptr->next_in_no; + pl->failover_pkts = FIRST_FAILOVER; + pl->failover_checkpt = l_ptr->next_in_no; + pl->failover_skb = l_ptr->reasm_buf; + } else { + kfree_skb(l_ptr->reasm_buf); } - l_ptr->next_out = NULL; - l_ptr->unacked_window = 0; + /* Clean up all queues, except inputq: */ + __skb_queue_purge(&l_ptr->transmq); + __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->reasm_buf = NULL; + l_ptr->rcv_unacked = 0; l_ptr->checkpoint = 1; l_ptr->next_out_no = 1; l_ptr->fsm_msg_cnt = 0; @@ -448,13 +476,14 @@ void tipc_link_reset(struct tipc_link *l_ptr) link_reset_statistics(l_ptr); } -void tipc_link_reset_list(unsigned int bearer_id) +void tipc_link_reset_list(struct net *net, unsigned int bearer_id) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_link *l_ptr; struct tipc_node *n_ptr; rcu_read_lock(); - list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { + list_for_each_entry_rcu(n_ptr, &tn->node_list, list) { tipc_node_lock(n_ptr); l_ptr = n_ptr->links[bearer_id]; if (l_ptr) @@ -464,11 +493,14 @@ void tipc_link_reset_list(unsigned int bearer_id) rcu_read_unlock(); } -static void link_activate(struct tipc_link *l_ptr) +static void link_activate(struct tipc_link *link) { - 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->bearer_id, l_ptr->addr); + struct tipc_node *node = link->owner; + + link->next_in_no = 1; + link->stats.recv_info = 1; + tipc_node_link_up(node, link); + tipc_bearer_add_dest(node->net, link->bearer_id, link->addr); } /** @@ -479,7 +511,7 @@ static void link_activate(struct tipc_link *l_ptr) static void link_state_event(struct tipc_link *l_ptr, unsigned int event) { struct tipc_link *other; - u32 cont_intv = l_ptr->continuity_interval; + unsigned long cont_intv = l_ptr->cont_intv; if (l_ptr->flags & LINK_STOPPED) return; @@ -487,8 +519,7 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) if (!(l_ptr->flags & LINK_STARTED) && (event != STARTING_EVT)) return; /* Not yet. */ - /* Check whether changeover is going on */ - if (l_ptr->exp_msg_count) { + if (l_ptr->flags & LINK_FAILINGOVER) { if (event == TIMEOUT_EVT) link_set_timer(l_ptr, cont_intv); return; @@ -505,11 +536,7 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) l_ptr->checkpoint = l_ptr->next_in_no; if (tipc_bclink_acks_missing(l_ptr->owner)) { tipc_link_proto_xmit(l_ptr, STATE_MSG, - 0, 0, 0, 0, 0); - l_ptr->fsm_msg_cnt++; - } else if (l_ptr->max_pkt < l_ptr->max_pkt_target) { - tipc_link_proto_xmit(l_ptr, STATE_MSG, - 1, 0, 0, 0, 0); + 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; } link_set_timer(l_ptr, cont_intv); @@ -517,23 +544,23 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) } l_ptr->state = WORKING_UNKNOWN; l_ptr->fsm_msg_cnt = 0; - tipc_link_proto_xmit(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, STATE_MSG, 1, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv / 4); break; case RESET_MSG: - pr_info("%s<%s>, requested by peer\n", link_rst_msg, - l_ptr->name); + pr_debug("%s<%s>, requested by peer\n", + link_rst_msg, l_ptr->name); tipc_link_reset(l_ptr); l_ptr->state = RESET_RESET; l_ptr->fsm_msg_cnt = 0; tipc_link_proto_xmit(l_ptr, ACTIVATE_MSG, - 0, 0, 0, 0, 0); + 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; default: - pr_err("%s%u in WW state\n", link_unk_evt, event); + pr_debug("%s%u in WW state\n", link_unk_evt, event); } break; case WORKING_UNKNOWN: @@ -545,13 +572,13 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) link_set_timer(l_ptr, cont_intv); break; case RESET_MSG: - pr_info("%s<%s>, requested by peer while probing\n", - link_rst_msg, l_ptr->name); + pr_debug("%s<%s>, requested by peer while probing\n", + link_rst_msg, l_ptr->name); tipc_link_reset(l_ptr); l_ptr->state = RESET_RESET; l_ptr->fsm_msg_cnt = 0; tipc_link_proto_xmit(l_ptr, ACTIVATE_MSG, - 0, 0, 0, 0, 0); + 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; @@ -562,23 +589,23 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) l_ptr->checkpoint = l_ptr->next_in_no; if (tipc_bclink_acks_missing(l_ptr->owner)) { tipc_link_proto_xmit(l_ptr, STATE_MSG, - 0, 0, 0, 0, 0); + 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; } link_set_timer(l_ptr, cont_intv); } else if (l_ptr->fsm_msg_cnt < l_ptr->abort_limit) { tipc_link_proto_xmit(l_ptr, STATE_MSG, - 1, 0, 0, 0, 0); + 1, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv / 4); } else { /* Link has failed */ - pr_warn("%s<%s>, peer not responding\n", - link_rst_msg, l_ptr->name); + pr_debug("%s<%s>, peer not responding\n", + link_rst_msg, l_ptr->name); tipc_link_reset(l_ptr); l_ptr->state = RESET_UNKNOWN; l_ptr->fsm_msg_cnt = 0; tipc_link_proto_xmit(l_ptr, RESET_MSG, - 0, 0, 0, 0, 0); + 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); } @@ -598,7 +625,7 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) l_ptr->state = WORKING_WORKING; l_ptr->fsm_msg_cnt = 0; link_activate(l_ptr); - tipc_link_proto_xmit(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, STATE_MSG, 1, 0, 0, 0); l_ptr->fsm_msg_cnt++; if (l_ptr->owner->working_links == 1) tipc_link_sync_xmit(l_ptr); @@ -608,15 +635,17 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) l_ptr->state = RESET_RESET; l_ptr->fsm_msg_cnt = 0; tipc_link_proto_xmit(l_ptr, ACTIVATE_MSG, - 1, 0, 0, 0, 0); + 1, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; case STARTING_EVT: l_ptr->flags |= LINK_STARTED; - /* fall through */ + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + break; case TIMEOUT_EVT: - tipc_link_proto_xmit(l_ptr, RESET_MSG, 0, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, RESET_MSG, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; @@ -634,7 +663,7 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) l_ptr->state = WORKING_WORKING; l_ptr->fsm_msg_cnt = 0; link_activate(l_ptr); - tipc_link_proto_xmit(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, STATE_MSG, 1, 0, 0, 0); l_ptr->fsm_msg_cnt++; if (l_ptr->owner->working_links == 1) tipc_link_sync_xmit(l_ptr); @@ -644,7 +673,7 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) break; case TIMEOUT_EVT: tipc_link_proto_xmit(l_ptr, ACTIVATE_MSG, - 0, 0, 0, 0, 0); + 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; @@ -657,99 +686,65 @@ 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); - uint imp = tipc_msg_tot_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 tipc_link *link, struct sk_buff_head *list) +int __tipc_link_xmit(struct net *net, struct tipc_link *link, + struct sk_buff_head *list) { struct tipc_msg *msg = buf_msg(skb_peek(list)); - uint psz = msg_size(msg); - uint sndlim = link->queue_limit[0]; - uint imp = tipc_msg_tot_importance(msg); - uint mtu = link->max_pkt; + unsigned int maxwin = link->window; + unsigned int imp = msg_importance(msg); + uint mtu = link->mtu; uint ack = mod(link->next_in_no - 1); uint seqno = link->next_out_no; uint bc_last_in = link->owner->bclink.last_in; struct tipc_media_addr *addr = &link->media_addr; - struct sk_buff_head *outqueue = &link->outqueue; + struct sk_buff_head *transmq = &link->transmq; + struct sk_buff_head *backlogq = &link->backlogq; struct sk_buff *skb, *tmp; - /* Match queue limits against msg importance: */ - if (unlikely(skb_queue_len(outqueue) >= 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(psz > mtu)) { + if (unlikely(msg_size(msg) > mtu)) { __skb_queue_purge(list); return -EMSGSIZE; } - - /* Prepare each packet for sending, and add to outqueue: */ + /* Prepare each packet for sending, and add to relevant queue: */ skb_queue_walk_safe(list, skb, tmp) { __skb_unlink(skb, list); msg = buf_msg(skb); - msg_set_word(msg, 2, ((ack << 16) | mod(seqno))); + msg_set_seqno(msg, seqno); + msg_set_ack(msg, ack); msg_set_bcast_ack(msg, bc_last_in); - if (skb_queue_len(outqueue) < sndlim) { - __skb_queue_tail(outqueue, skb); - tipc_bearer_send(link->bearer_id, skb, addr); - link->next_out = NULL; - link->unacked_window = 0; - } else if (tipc_msg_bundle(outqueue, skb, mtu)) { + if (likely(skb_queue_len(transmq) < maxwin)) { + __skb_queue_tail(transmq, skb); + tipc_bearer_send(net, link->bearer_id, skb, addr); + link->rcv_unacked = 0; + seqno++; + continue; + } + if (tipc_msg_bundle(skb_peek_tail(backlogq), skb, mtu)) { link->stats.sent_bundled++; continue; - } else if (tipc_msg_make_bundle(outqueue, skb, mtu, - link->addr)) { + } + if (tipc_msg_make_bundle(&skb, mtu, link->addr)) { link->stats.sent_bundled++; link->stats.sent_bundles++; - if (!link->next_out) - link->next_out = skb_peek_tail(outqueue); - } else { - __skb_queue_tail(outqueue, skb); - if (!link->next_out) - link->next_out = skb; + imp = msg_importance(buf_msg(skb)); } + __skb_queue_tail(backlogq, skb); + link->backlog[imp].len++; seqno++; } link->next_out_no = seqno; @@ -758,7 +753,7 @@ int __tipc_link_xmit(struct tipc_link *link, struct sk_buff_head *list) static void skb2list(struct sk_buff *skb, struct sk_buff_head *list) { - __skb_queue_head_init(list); + skb_queue_head_init(list); __skb_queue_tail(list, skb); } @@ -767,19 +762,33 @@ static int __tipc_link_xmit_skb(struct tipc_link *link, struct sk_buff *skb) struct sk_buff_head head; skb2list(skb, &head); - return __tipc_link_xmit(link, &head); + return __tipc_link_xmit(link->owner->net, link, &head); } -int tipc_link_xmit_skb(struct sk_buff *skb, u32 dnode, u32 selector) +/* 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(&head, dnode, selector); + rc = tipc_link_xmit(net, &head, dnode, selector); + if (rc == -ELINKCONG) + kfree_skb(skb); + return 0; } /** * tipc_link_xmit() is the general link level function for message sending + * @net: the applicable net namespace * @list: chain of buffers containing message * @dsz: amount of user data to be sent * @dnode: address of destination node @@ -787,33 +796,31 @@ int tipc_link_xmit_skb(struct sk_buff *skb, u32 dnode, u32 selector) * Consumes the buffer chain, except when returning -ELINKCONG * Returns 0 if success, otherwise errno: -ELINKCONG,-EHOSTUNREACH,-EMSGSIZE */ -int tipc_link_xmit(struct sk_buff_head *list, u32 dnode, u32 selector) +int tipc_link_xmit(struct net *net, struct sk_buff_head *list, u32 dnode, + u32 selector) { struct tipc_link *link = NULL; struct tipc_node *node; int rc = -EHOSTUNREACH; - node = tipc_node_find(dnode); + node = tipc_node_find(net, dnode); if (node) { tipc_node_lock(node); link = node->active_links[selector & 1]; if (link) - rc = __tipc_link_xmit(link, list); + rc = __tipc_link_xmit(net, link, list); tipc_node_unlock(node); + tipc_node_put(node); } - if (link) return rc; - if (likely(in_own_node(dnode))) { - /* As a node local message chain never contains more than one - * buffer, we just need to dequeue one SKB buffer from the - * head list. - */ - return tipc_sk_rcv(__skb_dequeue(list)); + if (likely(in_own_node(net, dnode))) { + tipc_sk_rcv(net, list); + return 0; } - __skb_queue_purge(list); + __skb_queue_purge(list); return rc; } @@ -835,7 +842,8 @@ static void tipc_link_sync_xmit(struct tipc_link *link) return; msg = buf_msg(skb); - tipc_msg_init(msg, BCAST_PROTOCOL, STATE_MSG, INT_H_SIZE, link->addr); + tipc_msg_init(link_own_addr(link), msg, BCAST_PROTOCOL, STATE_MSG, + INT_H_SIZE, link->addr); msg_set_last_bcast(msg, link->owner->bclink.acked); __tipc_link_xmit_skb(link, skb); } @@ -857,14 +865,6 @@ static void tipc_link_sync_rcv(struct tipc_node *n, struct sk_buff *buf) kfree_skb(buf); } -struct sk_buff *tipc_skb_queue_next(const struct sk_buff_head *list, - const struct sk_buff *skb) -{ - if (skb_queue_is_last(list, skb)) - return NULL; - return skb->next; -} - /* * tipc_link_push_packets - push unsent packets to bearer * @@ -873,29 +873,24 @@ struct sk_buff *tipc_skb_queue_next(const struct sk_buff_head *list, * * Called with node locked */ -void tipc_link_push_packets(struct tipc_link *l_ptr) +void tipc_link_push_packets(struct tipc_link *link) { - struct sk_buff_head *outqueue = &l_ptr->outqueue; - struct sk_buff *skb = l_ptr->next_out; + struct sk_buff *skb; struct tipc_msg *msg; - u32 next, first; + unsigned int ack = mod(link->next_in_no - 1); - skb_queue_walk_from(outqueue, skb) { - msg = buf_msg(skb); - next = msg_seqno(msg); - first = buf_seqno(skb_peek(outqueue)); - - 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); - if (msg_user(msg) == MSG_BUNDLER) - TIPC_SKB_CB(skb)->bundling = false; - tipc_bearer_send(l_ptr->bearer_id, skb, - &l_ptr->media_addr); - l_ptr->next_out = tipc_skb_queue_next(outqueue, skb); - } else { + while (skb_queue_len(&link->transmq) < link->window) { + skb = __skb_dequeue(&link->backlogq); + 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; + __skb_queue_tail(&link->transmq, skb); + tipc_bearer_send(link->owner->net, link->bearer_id, + skb, &link->media_addr); } } @@ -923,6 +918,7 @@ static void link_retransmit_failure(struct tipc_link *l_ptr, struct sk_buff *buf) { struct tipc_msg *msg = buf_msg(buf); + struct net *net = l_ptr->owner->net; pr_warn("Retransmission failure on link <%s>\n", l_ptr->name); @@ -940,8 +936,7 @@ static void link_retransmit_failure(struct tipc_link *l_ptr, pr_cont("Outstanding acks: %lu\n", (unsigned long) TIPC_SKB_CB(buf)->handle); - n_ptr = tipc_bclink_retransmit_to(); - tipc_node_lock(n_ptr); + n_ptr = tipc_bclink_retransmit_to(net); tipc_addr_string_fill(addr_string, n_ptr->addr); pr_info("Broadcast link info for %s\n", addr_string); @@ -953,9 +948,7 @@ 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(TIPC_BCLINK_RESET); + n_ptr->action_flags |= TIPC_BCAST_RESET; l_ptr->stale_count = 0; } } @@ -981,96 +974,70 @@ void tipc_link_retransmit(struct tipc_link *l_ptr, struct sk_buff *skb, l_ptr->stale_count = 1; } - skb_queue_walk_from(&l_ptr->outqueue, skb) { - if (!retransmits || skb == l_ptr->next_out) + skb_queue_walk_from(&l_ptr->transmq, skb) { + if (!retransmits) break; msg = buf_msg(skb); 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->bearer_id, skb, &l_ptr->media_addr); + tipc_bearer_send(l_ptr->owner->net, l_ptr->bearer_id, skb, + &l_ptr->media_addr); retransmits--; l_ptr->stats.retransmitted++; } } -static void link_retrieve_defq(struct tipc_link *link, - struct sk_buff_head *list) -{ - u32 seq_no; - - if (skb_queue_empty(&link->deferred_queue)) - return; - - seq_no = buf_seqno(skb_peek(&link->deferred_queue)); - if (seq_no == mod(link->next_in_no)) - skb_queue_splice_tail_init(&link->deferred_queue, list); -} - -/** - * link_recv_buf_validate - validate basic format of received message - * - * This routine ensures a TIPC message has an acceptable header, and at least - * as much data as the header indicates it should. The routine also ensures - * that the entire message header is stored in the main fragment of the message - * buffer, to simplify future access to message header fields. - * - * Note: Having extra info present in the message header or data areas is OK. - * TIPC will ignore the excess, under the assumption that it is optional info - * introduced by a later release of the protocol. +/* 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 int link_recv_buf_validate(struct sk_buff *buf) +static bool link_synch(struct tipc_link *l) { - static u32 min_data_hdr_size[8] = { - SHORT_H_SIZE, MCAST_H_SIZE, NAMED_H_SIZE, BASIC_H_SIZE, - MAX_H_SIZE, MAX_H_SIZE, MAX_H_SIZE, MAX_H_SIZE - }; + unsigned int post_synch; + struct tipc_link *pl; - struct tipc_msg *msg; - u32 tipc_hdr[2]; - u32 size; - u32 hdr_size; - u32 min_hdr_size; + pl = tipc_parallel_link(l); + if (pl == l) + goto synched; - /* If this packet comes from the defer queue, the skb has already - * been validated - */ - if (unlikely(TIPC_SKB_CB(buf)->deferred)) - return 1; - - if (unlikely(buf->len < MIN_H_SIZE)) - return 0; - - msg = skb_header_pointer(buf, 0, sizeof(tipc_hdr), tipc_hdr); - if (msg == NULL) - return 0; + /* Was last pre-synch packet added to input queue ? */ + if (less_eq(pl->next_in_no, l->synch_point)) + return false; - if (unlikely(msg_version(msg) != TIPC_VERSION)) - return 0; + /* 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; +} - size = msg_size(msg); - hdr_size = msg_hdr_sz(msg); - min_hdr_size = msg_isdata(msg) ? - min_data_hdr_size[msg_type(msg)] : INT_H_SIZE; +static void link_retrieve_defq(struct tipc_link *link, + struct sk_buff_head *list) +{ + u32 seq_no; - if (unlikely((hdr_size < min_hdr_size) || - (size < hdr_size) || - (buf->len < size) || - (size - hdr_size > TIPC_MAX_USER_MSG_SIZE))) - return 0; + if (skb_queue_empty(&link->deferdq)) + return; - return pskb_may_pull(buf, hdr_size); + seq_no = buf_seqno(skb_peek(&link->deferdq)); + if (seq_no == mod(link->next_in_no)) + skb_queue_splice_tail_init(&link->deferdq, list); } /** * tipc_rcv - process TIPC packets/messages arriving from off-node + * @net: the applicable net namespace * @skb: TIPC packet * @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 *skb, struct tipc_bearer *b_ptr) +void tipc_rcv(struct net *net, struct sk_buff *skb, struct tipc_bearer *b_ptr) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct sk_buff_head head; struct tipc_node *n_ptr; struct tipc_link *l_ptr; @@ -1084,39 +1051,34 @@ void tipc_rcv(struct sk_buff *skb, struct tipc_bearer *b_ptr) while ((skb = __skb_dequeue(&head))) { /* Ensure message is well-formed */ - if (unlikely(!link_recv_buf_validate(skb))) - goto discard; - - /* Ensure message data is a single contiguous unit */ - if (unlikely(skb_linearize(skb))) + if (unlikely(!tipc_msg_validate(skb))) goto discard; /* Handle arrival of a non-unicast link message */ msg = buf_msg(skb); - if (unlikely(msg_non_seq(msg))) { if (msg_user(msg) == LINK_CONFIG) - tipc_disc_rcv(skb, b_ptr); + tipc_disc_rcv(net, skb, b_ptr); else - tipc_bclink_rcv(skb); + tipc_bclink_rcv(net, skb); continue; } /* Discard unicast link messages destined for another node */ if (unlikely(!msg_short(msg) && - (msg_destnode(msg) != tipc_own_addr))) + (msg_destnode(msg) != tn->own_addr))) goto discard; /* Locate neighboring node that sent message */ - n_ptr = tipc_node_find(msg_prevnode(msg)); + 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)) - goto unlock_discard; + goto unlock; /* Verify that communication with node is currently allowed */ if ((n_ptr->action_flags & TIPC_WAIT_PEER_LINKS_DOWN) && @@ -1127,42 +1089,39 @@ void tipc_rcv(struct sk_buff *skb, struct tipc_bearer *b_ptr) n_ptr->action_flags &= ~TIPC_WAIT_PEER_LINKS_DOWN; if (tipc_node_blocked(n_ptr)) - goto unlock_discard; + goto unlock; /* Validate message sequence number info */ seq_no = msg_seqno(msg); ackd = msg_ack(msg); /* Release acked messages */ - if (n_ptr->bclink.recv_permitted) + if (unlikely(n_ptr->bclink.acked != msg_bcast_ack(msg))) tipc_bclink_acknowledge(n_ptr, msg_bcast_ack(msg)); released = 0; - skb_queue_walk_safe(&l_ptr->outqueue, skb1, tmp) { - if (skb1 == l_ptr->next_out || - more(buf_seqno(skb1), ackd)) + skb_queue_walk_safe(&l_ptr->transmq, skb1, tmp) { + if (more(buf_seqno(skb1), ackd)) break; - __skb_unlink(skb1, &l_ptr->outqueue); + __skb_unlink(skb1, &l_ptr->transmq); kfree_skb(skb1); released = 1; } /* Try sending any messages link endpoint has pending */ - if (unlikely(l_ptr->next_out)) + if (unlikely(skb_queue_len(&l_ptr->backlogq))) tipc_link_push_packets(l_ptr); - if (released && !skb_queue_empty(&l_ptr->waiting_sks)) { + if (released && !skb_queue_empty(&l_ptr->wakeupq)) link_prepare_wakeup(l_ptr); - l_ptr->owner->action_flags |= TIPC_WAKEUP_USERS; - } /* Process the incoming packet */ if (unlikely(!link_working_working(l_ptr))) { if (msg_user(msg) == LINK_PROTOCOL) { tipc_link_proto_rcv(l_ptr, skb); link_retrieve_defq(l_ptr, &head); - tipc_node_unlock(n_ptr); - continue; + skb = NULL; + goto unlock; } /* Traffic message. Conditionally activate link */ @@ -1171,116 +1130,137 @@ void tipc_rcv(struct sk_buff *skb, struct tipc_bearer *b_ptr) if (link_working_working(l_ptr)) { /* Re-insert buffer in front of queue */ __skb_queue_head(&head, skb); - tipc_node_unlock(n_ptr); - continue; + skb = NULL; + goto unlock; } - goto unlock_discard; + goto unlock; } /* Link is now in state WORKING_WORKING */ if (unlikely(seq_no != mod(l_ptr->next_in_no))) { link_handle_out_of_seq_msg(l_ptr, skb); link_retrieve_defq(l_ptr, &head); - tipc_node_unlock(n_ptr); - continue; + 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->deferred_queue))) + if (unlikely(!skb_queue_empty(&l_ptr->deferdq))) link_retrieve_defq(l_ptr, &head); - - if (unlikely(++l_ptr->unacked_window >= TIPC_MIN_LINK_WIN)) { + if (unlikely(++l_ptr->rcv_unacked >= TIPC_MIN_LINK_WIN)) { l_ptr->stats.sent_acks++; - tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); - } - - if (tipc_link_prepare_input(l_ptr, &skb)) { - tipc_node_unlock(n_ptr); - continue; + tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, 0, 0); } + tipc_link_input(l_ptr, skb); + skb = NULL; +unlock: tipc_node_unlock(n_ptr); - - if (tipc_link_input(l_ptr, skb) != 0) - goto discard; - continue; -unlock_discard: - tipc_node_unlock(n_ptr); + tipc_node_put(n_ptr); discard: - kfree_skb(skb); + if (unlikely(skb)) + kfree_skb(skb); } } -/** - * tipc_link_prepare_input - process TIPC link messages - * - * returns nonzero if the message was consumed +/* tipc_data_input - deliver data and name distr msgs to upper layer * + * Consumes buffer if message is of right type * Node lock must be held */ -static int tipc_link_prepare_input(struct tipc_link *l, struct sk_buff **buf) +static bool tipc_data_input(struct tipc_link *link, struct sk_buff *skb) { - struct tipc_node *n; - struct tipc_msg *msg; - int res = -EINVAL; + struct tipc_node *node = link->owner; + struct tipc_msg *msg = buf_msg(skb); + u32 dport = msg_destport(msg); - n = l->owner; - msg = buf_msg(*buf); switch (msg_user(msg)) { - case CHANGEOVER_PROTOCOL: - if (tipc_link_tunnel_rcv(n, buf)) - res = 0; - break; - case MSG_FRAGMENTER: - l->stats.recv_fragments++; - if (tipc_buf_append(&l->reasm_buf, buf)) { - l->stats.recv_fragmented++; - res = 0; - } else if (!l->reasm_buf) { - tipc_link_reset(l); + case TIPC_LOW_IMPORTANCE: + case TIPC_MEDIUM_IMPORTANCE: + case TIPC_HIGH_IMPORTANCE: + case TIPC_CRITICAL_IMPORTANCE: + case CONN_MANAGER: + if (tipc_skb_queue_tail(&link->inputq, skb, dport)) { + node->inputq = &link->inputq; + node->action_flags |= TIPC_MSG_EVT; } - break; - case MSG_BUNDLER: - l->stats.recv_bundles++; - l->stats.recv_bundled += msg_msgcnt(msg); - res = 0; - break; + return true; case NAME_DISTRIBUTOR: - n->bclink.recv_permitted = true; - res = 0; - break; + node->bclink.recv_permitted = true; + node->namedq = &link->namedq; + skb_queue_tail(&link->namedq, skb); + if (skb_queue_len(&link->namedq) == 1) + node->action_flags |= TIPC_NAMED_MSG_EVT; + return true; + case MSG_BUNDLER: + case TUNNEL_PROTOCOL: + case MSG_FRAGMENTER: case BCAST_PROTOCOL: - tipc_link_sync_rcv(n, *buf); - break; + return false; default: - res = 0; - } - return res; + pr_warn("Dropping received illegal msg type\n"); + kfree_skb(skb); + return false; + }; } -/** - * tipc_link_input - Deliver message too higher layers + +/* tipc_link_input - process packet that has passed link protocol check + * + * Consumes buffer + * Node lock must be held */ -static int tipc_link_input(struct tipc_link *l, struct sk_buff *buf) +static void tipc_link_input(struct tipc_link *link, struct sk_buff *skb) { - struct tipc_msg *msg = buf_msg(buf); - int res = 0; + struct tipc_node *node = link->owner; + struct tipc_msg *msg = buf_msg(skb); + struct sk_buff *iskb; + int pos = 0; + + if (likely(tipc_data_input(link, skb))) + return; switch (msg_user(msg)) { - case TIPC_LOW_IMPORTANCE: - case TIPC_MEDIUM_IMPORTANCE: - case TIPC_HIGH_IMPORTANCE: - case TIPC_CRITICAL_IMPORTANCE: - case CONN_MANAGER: - tipc_sk_rcv(buf); + case TUNNEL_PROTOCOL: + if (msg_dup(msg)) { + link->flags |= LINK_SYNCHING; + link->synch_point = msg_seqno(msg_get_wrapped(msg)); + kfree_skb(skb); + break; + } + if (!tipc_link_failover_rcv(link, &skb)) + break; + if (msg_user(buf_msg(skb)) != MSG_BUNDLER) { + tipc_data_input(link, skb); + break; + } + case MSG_BUNDLER: + link->stats.recv_bundles++; + link->stats.recv_bundled += msg_msgcnt(msg); + + while (tipc_msg_extract(skb, &iskb, &pos)) + tipc_data_input(link, iskb); break; - case NAME_DISTRIBUTOR: - tipc_named_rcv(buf); + case MSG_FRAGMENTER: + link->stats.recv_fragments++; + if (tipc_buf_append(&link->reasm_buf, &skb)) { + link->stats.recv_fragmented++; + tipc_data_input(link, skb); + } else if (!link->reasm_buf) { + tipc_link_reset(link); + } break; - case MSG_BUNDLER: - tipc_link_bundle_rcv(buf); + case BCAST_PROTOCOL: + tipc_link_sync_rcv(node, skb); break; default: - res = -EINVAL; - } - return res; + break; + }; } /** @@ -1348,11 +1328,10 @@ static void link_handle_out_of_seq_msg(struct tipc_link *l_ptr, return; } - if (tipc_link_defer_pkt(&l_ptr->deferred_queue, buf)) { + if (tipc_link_defer_pkt(&l_ptr->deferdq, buf)) { l_ptr->stats.deferred_recv++; - TIPC_SKB_CB(buf)->deferred = true; - if ((skb_queue_len(&l_ptr->deferred_queue) % 16) == 1) - tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); + if ((skb_queue_len(&l_ptr->deferdq) % TIPC_MIN_LINK_WIN) == 1) + tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, 0, 0); } else { l_ptr->stats.duplicates++; } @@ -1362,15 +1341,15 @@ static void link_handle_out_of_seq_msg(struct tipc_link *l_ptr, * Send protocol message to the other endpoint. */ void tipc_link_proto_xmit(struct tipc_link *l_ptr, u32 msg_typ, int probe_msg, - u32 gap, u32 tolerance, u32 priority, u32 ack_mtu) + u32 gap, u32 tolerance, u32 priority) { struct sk_buff *buf = NULL; struct tipc_msg *msg = l_ptr->pmsg; u32 msg_size = sizeof(l_ptr->proto_msg); int r_flag; - /* Don't send protocol message during link changeover */ - if (l_ptr->exp_msg_count) + /* Don't send protocol message during link failover */ + if (l_ptr->flags & LINK_FAILINGOVER) return; /* Abort non-RESET send if communication with node is prohibited */ @@ -1381,18 +1360,18 @@ void tipc_link_proto_xmit(struct tipc_link *l_ptr, u32 msg_typ, int probe_msg, msg_set_type(msg, msg_typ); 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()); + msg_set_last_bcast(msg, tipc_bclink_get_last_sent(l_ptr->owner->net)); if (msg_typ == STATE_MSG) { u32 next_sent = mod(l_ptr->next_out_no); if (!tipc_link_is_up(l_ptr)) return; - if (l_ptr->next_out) - next_sent = buf_seqno(l_ptr->next_out); + if (skb_queue_len(&l_ptr->backlogq)) + next_sent = buf_seqno(skb_peek(&l_ptr->backlogq)); msg_set_next_sent(msg, next_sent); - if (!skb_queue_empty(&l_ptr->deferred_queue)) { - u32 rec = buf_seqno(skb_peek(&l_ptr->deferred_queue)); + if (!skb_queue_empty(&l_ptr->deferdq)) { + u32 rec = buf_seqno(skb_peek(&l_ptr->deferdq)); gap = mod(rec - mod(l_ptr->next_in_no)); } msg_set_seq_gap(msg, gap); @@ -1400,35 +1379,20 @@ void tipc_link_proto_xmit(struct tipc_link *l_ptr, u32 msg_typ, int probe_msg, l_ptr->stats.sent_nacks++; msg_set_link_tolerance(msg, tolerance); msg_set_linkprio(msg, priority); - msg_set_max_pkt(msg, ack_mtu); + msg_set_max_pkt(msg, l_ptr->mtu); msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); msg_set_probe(msg, probe_msg != 0); - if (probe_msg) { - u32 mtu = l_ptr->max_pkt; - - if ((mtu < l_ptr->max_pkt_target) && - link_working_working(l_ptr) && - l_ptr->fsm_msg_cnt) { - msg_size = (mtu + (l_ptr->max_pkt_target - mtu)/2 + 2) & ~3; - if (l_ptr->max_pkt_probes == 10) { - l_ptr->max_pkt_target = (msg_size - 4); - l_ptr->max_pkt_probes = 0; - msg_size = (mtu + (l_ptr->max_pkt_target - mtu)/2 + 2) & ~3; - } - l_ptr->max_pkt_probes++; - } - + if (probe_msg) l_ptr->stats.sent_probes++; - } l_ptr->stats.sent_states++; } else { /* RESET_MSG or ACTIVATE_MSG */ - msg_set_ack(msg, mod(l_ptr->reset_checkpoint - 1)); + msg_set_ack(msg, mod(l_ptr->failover_checkpt - 1)); msg_set_seq_gap(msg, 0); msg_set_next_sent(msg, 1); msg_set_probe(msg, 0); msg_set_link_tolerance(msg, l_ptr->tolerance); msg_set_linkprio(msg, l_ptr->priority); - msg_set_max_pkt(msg, l_ptr->max_pkt_target); + msg_set_max_pkt(msg, l_ptr->advertised_mtu); } r_flag = (l_ptr->owner->working_links > tipc_link_is_up(l_ptr)); @@ -1444,9 +1408,9 @@ 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->bearer_id, buf, &l_ptr->media_addr); - l_ptr->unacked_window = 0; + tipc_bearer_send(l_ptr->owner->net, l_ptr->bearer_id, buf, + &l_ptr->media_addr); + l_ptr->rcv_unacked = 0; kfree_skb(buf); } @@ -1455,20 +1419,18 @@ void tipc_link_proto_xmit(struct tipc_link *l_ptr, u32 msg_typ, int probe_msg, * Note that network plane id propagates through the network, and may * change at any time. The node with lowest address rules */ -static void tipc_link_proto_rcv(struct tipc_link *l_ptr, struct sk_buff *buf) +static void tipc_link_proto_rcv(struct tipc_link *l_ptr, + struct sk_buff *buf) { u32 rec_gap = 0; - u32 max_pkt_info; - u32 max_pkt_ack; u32 msg_tol; struct tipc_msg *msg = buf_msg(buf); - /* Discard protocol message during link changeover */ - if (l_ptr->exp_msg_count) + if (l_ptr->flags & LINK_FAILINGOVER) goto exit; if (l_ptr->net_plane != msg_net_plane(msg)) - if (tipc_own_addr > msg_prevnode(msg)) + if (link_own_addr(l_ptr) > msg_prevnode(msg)) l_ptr->net_plane = msg_net_plane(msg); switch (msg_type(msg)) { @@ -1503,15 +1465,8 @@ static void tipc_link_proto_rcv(struct tipc_link *l_ptr, struct sk_buff *buf) if (msg_linkprio(msg) > l_ptr->priority) l_ptr->priority = msg_linkprio(msg); - max_pkt_info = msg_max_pkt(msg); - if (max_pkt_info) { - if (max_pkt_info < l_ptr->max_pkt_target) - l_ptr->max_pkt_target = max_pkt_info; - if (l_ptr->max_pkt > l_ptr->max_pkt_target) - l_ptr->max_pkt = l_ptr->max_pkt_target; - } else { - l_ptr->max_pkt = l_ptr->max_pkt_target; - } + if (l_ptr->mtu > msg_max_pkt(msg)) + l_ptr->mtu = msg_max_pkt(msg); /* Synchronize broadcast link info, if not done previously */ if (!tipc_node_is_up(l_ptr->owner)) { @@ -1535,9 +1490,9 @@ static void tipc_link_proto_rcv(struct tipc_link *l_ptr, struct sk_buff *buf) if (msg_linkprio(msg) && (msg_linkprio(msg) != l_ptr->priority)) { - pr_warn("%s<%s>, priority change %u->%u\n", - link_rst_msg, l_ptr->name, l_ptr->priority, - msg_linkprio(msg)); + pr_debug("%s<%s>, priority change %u->%u\n", + link_rst_msg, l_ptr->name, + l_ptr->priority, msg_linkprio(msg)); l_ptr->priority = msg_linkprio(msg); tipc_link_reset(l_ptr); /* Enforce change to take effect */ break; @@ -1556,18 +1511,8 @@ static void tipc_link_proto_rcv(struct tipc_link *l_ptr, struct sk_buff *buf) mod(l_ptr->next_in_no)); } - max_pkt_ack = msg_max_pkt(msg); - if (max_pkt_ack > l_ptr->max_pkt) { - l_ptr->max_pkt = max_pkt_ack; - l_ptr->max_pkt_probes = 0; - } - - max_pkt_ack = 0; - if (msg_probe(msg)) { + if (msg_probe(msg)) l_ptr->stats.recv_probes++; - if (msg_size(msg) > sizeof(l_ptr->proto_msg)) - max_pkt_ack = msg_size(msg); - } /* Protocol message before retransmits, reduce loss risk */ if (l_ptr->owner->bclink.recv_permitted) @@ -1575,12 +1520,12 @@ static void tipc_link_proto_rcv(struct tipc_link *l_ptr, struct sk_buff *buf) msg_last_bcast(msg)); if (rec_gap || (msg_probe(msg))) { - tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, rec_gap, 0, - 0, max_pkt_ack); + tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, + rec_gap, 0, 0); } if (msg_seq_gap(msg)) { l_ptr->stats.recv_nacks++; - tipc_link_retransmit(l_ptr, skb_peek(&l_ptr->outqueue), + tipc_link_retransmit(l_ptr, skb_peek(&l_ptr->transmq), msg_seq_gap(msg)); } break; @@ -1627,7 +1572,7 @@ static void tipc_link_tunnel_xmit(struct tipc_link *l_ptr, */ void tipc_link_failover_send_queue(struct tipc_link *l_ptr) { - u32 msgcount = skb_queue_len(&l_ptr->outqueue); + int msgcount; struct tipc_link *tunnel = l_ptr->owner->active_links[0]; struct tipc_msg tunnel_hdr; struct sk_buff *skb; @@ -1636,12 +1581,15 @@ void tipc_link_failover_send_queue(struct tipc_link *l_ptr) if (!tunnel) return; - tipc_msg_init(&tunnel_hdr, CHANGEOVER_PROTOCOL, - ORIGINAL_MSG, INT_H_SIZE, l_ptr->addr); + tipc_msg_init(link_own_addr(l_ptr), &tunnel_hdr, TUNNEL_PROTOCOL, + FAILOVER_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); - if (skb_queue_empty(&l_ptr->outqueue)) { + if (skb_queue_empty(&l_ptr->transmq)) { skb = tipc_buf_acquire(INT_H_SIZE); if (skb) { skb_copy_to_linear_data(skb, &tunnel_hdr, INT_H_SIZE); @@ -1657,7 +1605,7 @@ void tipc_link_failover_send_queue(struct tipc_link *l_ptr) split_bundles = (l_ptr->owner->active_links[0] != l_ptr->owner->active_links[1]); - skb_queue_walk(&l_ptr->outqueue, skb) { + skb_queue_walk(&l_ptr->transmq, skb) { struct tipc_msg *msg = buf_msg(skb); if ((msg_user(msg) == MSG_BUNDLER) && split_bundles) { @@ -1688,245 +1636,151 @@ void tipc_link_failover_send_queue(struct tipc_link *l_ptr) * and sequence order is preserved per sender/receiver socket pair. * Owner node is locked. */ -void tipc_link_dup_queue_xmit(struct tipc_link *l_ptr, - struct tipc_link *tunnel) +void tipc_link_dup_queue_xmit(struct tipc_link *link, + struct tipc_link *tnl) { struct sk_buff *skb; - struct tipc_msg tunnel_hdr; - - tipc_msg_init(&tunnel_hdr, CHANGEOVER_PROTOCOL, - DUPLICATE_MSG, INT_H_SIZE, l_ptr->addr); - msg_set_msgcnt(&tunnel_hdr, skb_queue_len(&l_ptr->outqueue)); - msg_set_bearer_id(&tunnel_hdr, l_ptr->peer_bearer_id); - skb_queue_walk(&l_ptr->outqueue, skb) { + struct tipc_msg tnl_hdr; + struct sk_buff_head *queue = &link->transmq; + int mcnt; + + tipc_msg_init(link_own_addr(link), &tnl_hdr, TUNNEL_PROTOCOL, + SYNCH_MSG, INT_H_SIZE, link->addr); + mcnt = skb_queue_len(&link->transmq) + skb_queue_len(&link->backlogq); + msg_set_msgcnt(&tnl_hdr, mcnt); + msg_set_bearer_id(&tnl_hdr, link->peer_bearer_id); + +tunnel_queue: + skb_queue_walk(queue, skb) { struct sk_buff *outskb; struct tipc_msg *msg = buf_msg(skb); - u32 length = msg_size(msg); + u32 len = msg_size(msg); - if (msg_user(msg) == MSG_BUNDLER) - msg_set_type(msg, CLOSED_MSG); - msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); /* Update */ - msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); - msg_set_size(&tunnel_hdr, length + INT_H_SIZE); - outskb = tipc_buf_acquire(length + INT_H_SIZE); + msg_set_ack(msg, mod(link->next_in_no - 1)); + msg_set_bcast_ack(msg, link->owner->bclink.last_in); + msg_set_size(&tnl_hdr, len + INT_H_SIZE); + outskb = tipc_buf_acquire(len + INT_H_SIZE); if (outskb == NULL) { pr_warn("%sunable to send duplicate msg\n", link_co_err); return; } - skb_copy_to_linear_data(outskb, &tunnel_hdr, INT_H_SIZE); - skb_copy_to_linear_data_offset(outskb, INT_H_SIZE, skb->data, - length); - __tipc_link_xmit_skb(tunnel, outskb); - if (!tipc_link_is_up(l_ptr)) + skb_copy_to_linear_data(outskb, &tnl_hdr, INT_H_SIZE); + skb_copy_to_linear_data_offset(outskb, INT_H_SIZE, + skb->data, len); + __tipc_link_xmit_skb(tnl, outskb); + if (!tipc_link_is_up(link)) return; } -} - -/** - * buf_extract - extracts embedded TIPC message from another message - * @skb: encapsulating message buffer - * @from_pos: offset to extract from - * - * Returns a new message buffer containing an embedded message. The - * encapsulating message itself is left unchanged. - */ -static struct sk_buff *buf_extract(struct sk_buff *skb, u32 from_pos) -{ - struct tipc_msg *msg = (struct tipc_msg *)(skb->data + from_pos); - u32 size = msg_size(msg); - struct sk_buff *eb; - - eb = tipc_buf_acquire(size); - if (eb) - skb_copy_to_linear_data(eb, msg, size); - return eb; -} - - - -/* tipc_link_dup_rcv(): Receive a tunnelled DUPLICATE_MSG packet. - * Owner node is locked. - */ -static void tipc_link_dup_rcv(struct tipc_link *l_ptr, - struct sk_buff *t_buf) -{ - struct sk_buff *buf; - - if (!tipc_link_is_up(l_ptr)) - return; - - buf = buf_extract(t_buf, INT_H_SIZE); - if (buf == NULL) { - pr_warn("%sfailed to extract inner dup pkt\n", link_co_err); + if (queue == &link->backlogq) return; - } - - /* Add buffer to deferred queue, if applicable: */ - link_handle_out_of_seq_msg(l_ptr, buf); + queue = &link->backlogq; + goto tunnel_queue; } -/* tipc_link_failover_rcv(): Receive a tunnelled ORIGINAL_MSG packet +/* tipc_link_failover_rcv(): Receive a tunnelled FAILOVER_MSG packet * Owner node is locked. */ -static struct sk_buff *tipc_link_failover_rcv(struct tipc_link *l_ptr, - struct sk_buff *t_buf) +static bool tipc_link_failover_rcv(struct tipc_link *link, + struct sk_buff **skb) { - struct tipc_msg *t_msg = buf_msg(t_buf); - struct sk_buff *buf = NULL; - struct tipc_msg *msg; + struct tipc_msg *msg = buf_msg(*skb); + struct sk_buff *iskb = NULL; + struct tipc_link *pl = NULL; + int bearer_id = msg_bearer_id(msg); + int pos = 0; - if (tipc_link_is_up(l_ptr)) - tipc_link_reset(l_ptr); - - /* First failover packet? */ - if (l_ptr->exp_msg_count == START_CHANGEOVER) - l_ptr->exp_msg_count = msg_msgcnt(t_msg); - - /* Should there be an inner packet? */ - if (l_ptr->exp_msg_count) { - l_ptr->exp_msg_count--; - buf = buf_extract(t_buf, INT_H_SIZE); - if (buf == NULL) { - pr_warn("%sno inner failover pkt\n", link_co_err); - goto exit; - } - msg = buf_msg(buf); - - if (less(msg_seqno(msg), l_ptr->reset_checkpoint)) { - kfree_skb(buf); - buf = NULL; - goto exit; - } - if (msg_user(msg) == MSG_FRAGMENTER) { - l_ptr->stats.recv_fragments++; - tipc_buf_append(&l_ptr->reasm_buf, &buf); - } - } -exit: - if ((l_ptr->exp_msg_count == 0) && (l_ptr->flags & LINK_STOPPED)) { - tipc_node_detach_link(l_ptr->owner, l_ptr); - kfree(l_ptr); + if (msg_type(msg) != FAILOVER_MSG) { + pr_warn("%sunknown tunnel pkt received\n", link_co_err); + goto exit; } - return buf; -} + if (bearer_id >= MAX_BEARERS) + goto exit; -/* tipc_link_tunnel_rcv(): Receive a tunnelled packet, sent - * via other link as result of a failover (ORIGINAL_MSG) or - * a new active link (DUPLICATE_MSG). Failover packets are - * returned to the active link for delivery upwards. - * Owner node is locked. - */ -static int tipc_link_tunnel_rcv(struct tipc_node *n_ptr, - struct sk_buff **buf) -{ - struct sk_buff *t_buf = *buf; - struct tipc_link *l_ptr; - struct tipc_msg *t_msg = buf_msg(t_buf); - u32 bearer_id = msg_bearer_id(t_msg); + if (bearer_id == link->bearer_id) + goto exit; - *buf = NULL; + pl = link->owner->links[bearer_id]; + if (pl && tipc_link_is_up(pl)) + tipc_link_reset(pl); - if (bearer_id >= MAX_BEARERS) + if (link->failover_pkts == FIRST_FAILOVER) + link->failover_pkts = msg_msgcnt(msg); + + /* Should we expect an inner packet? */ + if (!link->failover_pkts) goto exit; - l_ptr = n_ptr->links[bearer_id]; - if (!l_ptr) + if (!tipc_msg_extract(*skb, &iskb, &pos)) { + pr_warn("%sno inner failover pkt\n", link_co_err); + *skb = NULL; goto exit; + } + link->failover_pkts--; + *skb = NULL; - if (msg_type(t_msg) == DUPLICATE_MSG) - tipc_link_dup_rcv(l_ptr, t_buf); - else if (msg_type(t_msg) == ORIGINAL_MSG) - *buf = tipc_link_failover_rcv(l_ptr, t_buf); - else - pr_warn("%sunknown tunnel pkt received\n", link_co_err); + /* Was this packet already delivered? */ + if (less(buf_seqno(iskb), link->failover_checkpt)) { + kfree_skb(iskb); + iskb = NULL; + goto exit; + } + if (msg_user(buf_msg(iskb)) == MSG_FRAGMENTER) { + link->stats.recv_fragments++; + tipc_buf_append(&link->failover_skb, &iskb); + } exit: - kfree_skb(t_buf); - return *buf != NULL; + if (!link->failover_pkts && pl) + pl->flags &= ~LINK_FAILINGOVER; + kfree_skb(*skb); + *skb = iskb; + return *skb; } -/* - * Bundler functionality: - */ -void tipc_link_bundle_rcv(struct sk_buff *buf) +static void link_set_supervision_props(struct tipc_link *l_ptr, u32 tol) { - u32 msgcount = msg_msgcnt(buf_msg(buf)); - u32 pos = INT_H_SIZE; - struct sk_buff *obuf; - struct tipc_msg *omsg; - - while (msgcount--) { - obuf = buf_extract(buf, pos); - if (obuf == NULL) { - pr_warn("Link unable to unbundle message(s)\n"); - break; - } - omsg = buf_msg(obuf); - pos += align(msg_size(omsg)); - if (msg_isdata(omsg)) { - if (unlikely(msg_type(omsg) == TIPC_MCAST_MSG)) - tipc_sk_mcast_rcv(obuf); - else - tipc_sk_rcv(obuf); - } else if (msg_user(omsg) == CONN_MANAGER) { - tipc_sk_rcv(obuf); - } else if (msg_user(omsg) == NAME_DISTRIBUTOR) { - tipc_named_rcv(obuf); - } else { - pr_warn("Illegal bundled msg: %u\n", msg_user(omsg)); - kfree_skb(obuf); - } - } - kfree_skb(buf); -} + unsigned long intv = ((tol / 4) > 500) ? 500 : tol / 4; -static void link_set_supervision_props(struct tipc_link *l_ptr, u32 tolerance) -{ - if ((tolerance < TIPC_MIN_LINK_TOL) || (tolerance > TIPC_MAX_LINK_TOL)) + if ((tol < TIPC_MIN_LINK_TOL) || (tol > TIPC_MAX_LINK_TOL)) return; - l_ptr->tolerance = tolerance; - l_ptr->continuity_interval = - ((tolerance / 4) > 500) ? 500 : tolerance / 4; - l_ptr->abort_limit = tolerance / (l_ptr->continuity_interval / 4); + l_ptr->tolerance = tol; + l_ptr->cont_intv = msecs_to_jiffies(intv); + l_ptr->abort_limit = tol / (jiffies_to_msecs(l_ptr->cont_intv) / 4); } -void tipc_link_set_queue_limits(struct tipc_link *l_ptr, u32 window) +void tipc_link_set_queue_limits(struct tipc_link *l, u32 win) { - /* Data messages from this node, inclusive FIRST_FRAGM */ - l_ptr->queue_limit[TIPC_LOW_IMPORTANCE] = window; - l_ptr->queue_limit[TIPC_MEDIUM_IMPORTANCE] = (window / 3) * 4; - l_ptr->queue_limit[TIPC_HIGH_IMPORTANCE] = (window / 3) * 5; - l_ptr->queue_limit[TIPC_CRITICAL_IMPORTANCE] = (window / 3) * 6; - /* Transiting data messages,inclusive FIRST_FRAGM */ - l_ptr->queue_limit[TIPC_LOW_IMPORTANCE + 4] = 300; - l_ptr->queue_limit[TIPC_MEDIUM_IMPORTANCE + 4] = 600; - l_ptr->queue_limit[TIPC_HIGH_IMPORTANCE + 4] = 900; - l_ptr->queue_limit[TIPC_CRITICAL_IMPORTANCE + 4] = 1200; - l_ptr->queue_limit[CONN_MANAGER] = 1200; - l_ptr->queue_limit[CHANGEOVER_PROTOCOL] = 2500; - l_ptr->queue_limit[NAME_DISTRIBUTOR] = 3000; - /* FRAGMENT and LAST_FRAGMENT packets */ - l_ptr->queue_limit[MSG_FRAGMENTER] = 4000; + int max_bulk = TIPC_MAX_PUBLICATIONS / (l->mtu / ITEM_SIZE); + + l->window = win; + 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 + * @net: the applicable net namespace * @name: pointer to link name string * @bearer_id: pointer to index in 'node->links' array where the link was found. * * Returns pointer to node owning the link, or 0 if no matching link is found. */ -static struct tipc_node *tipc_link_find_owner(const char *link_name, +static struct tipc_node *tipc_link_find_owner(struct net *net, + const char *link_name, unsigned int *bearer_id) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_link *l_ptr; struct tipc_node *n_ptr; - struct tipc_node *found_node = 0; + struct tipc_node *found_node = NULL; int i; *bearer_id = 0; rcu_read_lock(); - list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { + list_for_each_entry_rcu(n_ptr, &tn->node_list, list) { tipc_node_lock(n_ptr); for (i = 0; i < MAX_BEARERS; i++) { l_ptr = n_ptr->links[i]; @@ -1946,148 +1800,6 @@ static struct tipc_node *tipc_link_find_owner(const char *link_name, } /** - * link_value_is_valid -- validate proposed link tolerance/priority/window - * - * @cmd: value type (TIPC_CMD_SET_LINK_*) - * @new_value: the new value - * - * Returns 1 if value is within range, 0 if not. - */ -static int link_value_is_valid(u16 cmd, u32 new_value) -{ - switch (cmd) { - case TIPC_CMD_SET_LINK_TOL: - return (new_value >= TIPC_MIN_LINK_TOL) && - (new_value <= TIPC_MAX_LINK_TOL); - case TIPC_CMD_SET_LINK_PRI: - return (new_value <= TIPC_MAX_LINK_PRI); - case TIPC_CMD_SET_LINK_WINDOW: - return (new_value >= TIPC_MIN_LINK_WIN) && - (new_value <= TIPC_MAX_LINK_WIN); - } - return 0; -} - -/** - * link_cmd_set_value - change priority/tolerance/window for link/bearer/media - * @name: ptr to link, bearer, or media name - * @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 RTNL lock to ensure link/bearer/media is not deleted. - * - * Returns 0 if value updated and negative value on error. - */ -static int link_cmd_set_value(const char *name, u32 new_value, u16 cmd) -{ - struct tipc_node *node; - struct tipc_link *l_ptr; - struct tipc_bearer *b_ptr; - struct tipc_media *m_ptr; - int bearer_id; - int res = 0; - - node = tipc_link_find_owner(name, &bearer_id); - if (node) { - tipc_node_lock(node); - l_ptr = node->links[bearer_id]; - - if (l_ptr) { - switch (cmd) { - case TIPC_CMD_SET_LINK_TOL: - link_set_supervision_props(l_ptr, new_value); - tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, - new_value, 0, 0); - break; - case TIPC_CMD_SET_LINK_PRI: - l_ptr->priority = new_value; - tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, - 0, new_value, 0); - break; - case TIPC_CMD_SET_LINK_WINDOW: - tipc_link_set_queue_limits(l_ptr, new_value); - break; - default: - res = -EINVAL; - break; - } - } - tipc_node_unlock(node); - return res; - } - - b_ptr = tipc_bearer_find(name); - if (b_ptr) { - switch (cmd) { - case TIPC_CMD_SET_LINK_TOL: - b_ptr->tolerance = new_value; - break; - case TIPC_CMD_SET_LINK_PRI: - b_ptr->priority = new_value; - break; - case TIPC_CMD_SET_LINK_WINDOW: - b_ptr->window = new_value; - break; - default: - res = -EINVAL; - break; - } - return res; - } - - m_ptr = tipc_media_find(name); - if (!m_ptr) - return -ENODEV; - switch (cmd) { - case TIPC_CMD_SET_LINK_TOL: - m_ptr->tolerance = new_value; - break; - case TIPC_CMD_SET_LINK_PRI: - m_ptr->priority = new_value; - break; - case TIPC_CMD_SET_LINK_WINDOW: - m_ptr->window = new_value; - break; - default: - res = -EINVAL; - break; - } - return res; -} - -struct sk_buff *tipc_link_cmd_config(const void *req_tlv_area, int req_tlv_space, - u16 cmd) -{ - struct tipc_link_config *args; - u32 new_value; - int res; - - if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_CONFIG)) - return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - - args = (struct tipc_link_config *)TLV_DATA(req_tlv_area); - new_value = ntohl(args->value); - - if (!link_value_is_valid(cmd, new_value)) - return tipc_cfg_reply_error_string( - "cannot change, value invalid"); - - if (!strcmp(args->name, tipc_bclink_name)) { - if ((cmd == TIPC_CMD_SET_LINK_WINDOW) && - (tipc_bclink_set_queue_limits(new_value) == 0)) - return tipc_cfg_reply_none(); - return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (cannot change setting on broadcast link)"); - } - - res = link_cmd_set_value(args->name, new_value, cmd); - if (res) - return tipc_cfg_reply_error_string("cannot change link setting"); - - return tipc_cfg_reply_none(); -} - -/** * link_reset_statistics - reset link statistics * @l_ptr: pointer to link */ @@ -2098,207 +1810,13 @@ static void link_reset_statistics(struct tipc_link *l_ptr) l_ptr->stats.recv_info = l_ptr->next_in_no; } -struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_space) -{ - char *link_name; - struct tipc_link *l_ptr; - struct tipc_node *node; - unsigned int bearer_id; - - if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_NAME)) - return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - - link_name = (char *)TLV_DATA(req_tlv_area); - if (!strcmp(link_name, tipc_bclink_name)) { - if (tipc_bclink_reset_stats()) - return tipc_cfg_reply_error_string("link not found"); - return tipc_cfg_reply_none(); - } - node = tipc_link_find_owner(link_name, &bearer_id); - 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); - return tipc_cfg_reply_error_string("link not found"); - } - link_reset_statistics(l_ptr); - tipc_node_unlock(node); - return tipc_cfg_reply_none(); -} - -/** - * percent - convert count to a percentage of total (rounding up or down) - */ -static u32 percent(u32 count, u32 total) -{ - return (count * 100 + (total / 2)) / total; -} - -/** - * tipc_link_stats - print link statistics - * @name: link name - * @buf: print buffer area - * @buf_size: size of print buffer area - * - * Returns length of print buffer data string (or 0 if error) - */ -static int tipc_link_stats(const char *name, char *buf, const u32 buf_size) -{ - struct tipc_link *l; - struct tipc_stats *s; - struct tipc_node *node; - char *status; - u32 profile_total = 0; - unsigned int bearer_id; - int ret; - - if (!strcmp(name, tipc_bclink_name)) - return tipc_bclink_stats(buf, buf_size); - - node = tipc_link_find_owner(name, &bearer_id); - if (!node) - return 0; - - tipc_node_lock(node); - - l = node->links[bearer_id]; - if (!l) { - tipc_node_unlock(node); - return 0; - } - - s = &l->stats; - - if (tipc_link_is_active(l)) - status = "ACTIVE"; - else if (tipc_link_is_up(l)) - status = "STANDBY"; - else - status = "DEFUNCT"; - - ret = tipc_snprintf(buf, buf_size, "Link <%s>\n" - " %s MTU:%u Priority:%u Tolerance:%u ms" - " Window:%u packets\n", - l->name, status, l->max_pkt, l->priority, - l->tolerance, l->queue_limit[0]); - - ret += tipc_snprintf(buf + ret, buf_size - ret, - " RX packets:%u fragments:%u/%u bundles:%u/%u\n", - l->next_in_no - s->recv_info, s->recv_fragments, - s->recv_fragmented, s->recv_bundles, - s->recv_bundled); - - ret += tipc_snprintf(buf + ret, buf_size - ret, - " TX packets:%u fragments:%u/%u bundles:%u/%u\n", - l->next_out_no - s->sent_info, s->sent_fragments, - s->sent_fragmented, s->sent_bundles, - s->sent_bundled); - - profile_total = s->msg_length_counts; - if (!profile_total) - profile_total = 1; - - ret += tipc_snprintf(buf + ret, buf_size - ret, - " TX profile sample:%u packets average:%u octets\n" - " 0-64:%u%% -256:%u%% -1024:%u%% -4096:%u%% " - "-16384:%u%% -32768:%u%% -66000:%u%%\n", - s->msg_length_counts, - s->msg_lengths_total / profile_total, - percent(s->msg_length_profile[0], profile_total), - percent(s->msg_length_profile[1], profile_total), - percent(s->msg_length_profile[2], profile_total), - percent(s->msg_length_profile[3], profile_total), - percent(s->msg_length_profile[4], profile_total), - percent(s->msg_length_profile[5], profile_total), - percent(s->msg_length_profile[6], profile_total)); - - ret += tipc_snprintf(buf + ret, buf_size - ret, - " RX states:%u probes:%u naks:%u defs:%u" - " dups:%u\n", s->recv_states, s->recv_probes, - s->recv_nacks, s->deferred_recv, s->duplicates); - - ret += tipc_snprintf(buf + ret, buf_size - ret, - " TX states:%u probes:%u naks:%u acks:%u" - " dups:%u\n", s->sent_states, s->sent_probes, - s->sent_nacks, s->sent_acks, s->retransmitted); - - ret += tipc_snprintf(buf + ret, buf_size - ret, - " Congestion link:%u Send queue" - " max:%u avg:%u\n", s->link_congs, - s->max_queue_sz, s->queue_sz_counts ? - (s->accu_queue_sz / s->queue_sz_counts) : 0); - - tipc_node_unlock(node); - return ret; -} - -struct sk_buff *tipc_link_cmd_show_stats(const void *req_tlv_area, int req_tlv_space) -{ - struct sk_buff *buf; - struct tlv_desc *rep_tlv; - int str_len; - int pb_len; - char *pb; - - if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_NAME)) - return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - - buf = tipc_cfg_reply_alloc(TLV_SPACE(ULTRA_STRING_MAX_LEN)); - if (!buf) - return NULL; - - rep_tlv = (struct tlv_desc *)buf->data; - pb = TLV_DATA(rep_tlv); - pb_len = ULTRA_STRING_MAX_LEN; - str_len = tipc_link_stats((char *)TLV_DATA(req_tlv_area), - pb, pb_len); - if (!str_len) { - kfree_skb(buf); - return tipc_cfg_reply_error_string("link not found"); - } - str_len += 1; /* for "\0" */ - skb_put(buf, TLV_SPACE(str_len)); - TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); - - return buf; -} - -/** - * tipc_link_get_max_pkt - get maximum packet size to use when sending to destination - * @dest: network address of destination node - * @selector: used to select from set of active links - * - * If no active link can be found, uses default maximum packet size. - */ -u32 tipc_link_get_max_pkt(u32 dest, u32 selector) -{ - struct tipc_node *n_ptr; - struct tipc_link *l_ptr; - u32 res = MAX_PKT_DEFAULT; - - if (dest == tipc_own_addr) - return MAX_MSG_SIZE; - - n_ptr = tipc_node_find(dest); - if (n_ptr) { - tipc_node_lock(n_ptr); - l_ptr = n_ptr->active_links[selector & 1]; - if (l_ptr) - res = l_ptr->max_pkt; - tipc_node_unlock(n_ptr); - } - return res; -} - static void link_print(struct tipc_link *l_ptr, const char *str) { + struct tipc_net *tn = net_generic(l_ptr->owner->net, tipc_net_id); struct tipc_bearer *b_ptr; rcu_read_lock(); - b_ptr = rcu_dereference_rtnl(bearer_list[l_ptr->bearer_id]); + b_ptr = rcu_dereference_rtnl(tn->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(); @@ -2362,6 +1880,7 @@ int tipc_nl_link_set(struct sk_buff *skb, struct genl_info *info) struct tipc_link *link; struct tipc_node *node; struct nlattr *attrs[TIPC_NLA_LINK_MAX + 1]; + struct net *net = sock_net(skb->sk); if (!info->attrs[TIPC_NLA_LINK]) return -EINVAL; @@ -2377,7 +1896,7 @@ int tipc_nl_link_set(struct sk_buff *skb, struct genl_info *info) name = nla_data(attrs[TIPC_NLA_LINK_NAME]); - node = tipc_link_find_owner(name, &bearer_id); + node = tipc_link_find_owner(net, name, &bearer_id); if (!node) return -EINVAL; @@ -2404,14 +1923,14 @@ int tipc_nl_link_set(struct sk_buff *skb, struct genl_info *info) tol = nla_get_u32(props[TIPC_NLA_PROP_TOL]); link_set_supervision_props(link, tol); - tipc_link_proto_xmit(link, STATE_MSG, 0, 0, tol, 0, 0); + tipc_link_proto_xmit(link, STATE_MSG, 0, 0, tol, 0); } if (props[TIPC_NLA_PROP_PRIO]) { u32 prio; prio = nla_get_u32(props[TIPC_NLA_PROP_PRIO]); link->priority = prio; - tipc_link_proto_xmit(link, STATE_MSG, 0, 0, 0, prio, 0); + tipc_link_proto_xmit(link, STATE_MSG, 0, 0, 0, prio); } if (props[TIPC_NLA_PROP_WIN]) { u32 win; @@ -2493,14 +2012,16 @@ msg_full: } /* Caller should hold appropriate locks to protect the link */ -static int __tipc_nl_add_link(struct tipc_nl_msg *msg, struct tipc_link *link) +static int __tipc_nl_add_link(struct net *net, struct tipc_nl_msg *msg, + struct tipc_link *link) { int err; void *hdr; struct nlattr *attrs; struct nlattr *prop; + struct tipc_net *tn = net_generic(net, tipc_net_id); - hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_v2_family, + hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, NLM_F_MULTI, TIPC_NL_LINK_GET); if (!hdr) return -EMSGSIZE; @@ -2512,9 +2033,9 @@ static int __tipc_nl_add_link(struct tipc_nl_msg *msg, struct tipc_link *link) if (nla_put_string(msg->skb, TIPC_NLA_LINK_NAME, link->name)) goto attr_msg_full; if (nla_put_u32(msg->skb, TIPC_NLA_LINK_DEST, - tipc_cluster_mask(tipc_own_addr))) + tipc_cluster_mask(tn->own_addr))) goto attr_msg_full; - if (nla_put_u32(msg->skb, TIPC_NLA_LINK_MTU, link->max_pkt)) + if (nla_put_u32(msg->skb, TIPC_NLA_LINK_MTU, link->mtu)) goto attr_msg_full; if (nla_put_u32(msg->skb, TIPC_NLA_LINK_RX, link->next_in_no)) goto attr_msg_full; @@ -2536,7 +2057,7 @@ static int __tipc_nl_add_link(struct tipc_nl_msg *msg, struct tipc_link *link) 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; @@ -2562,9 +2083,8 @@ msg_full: } /* Caller should hold node lock */ -static int __tipc_nl_add_node_links(struct tipc_nl_msg *msg, - struct tipc_node *node, - u32 *prev_link) +static int __tipc_nl_add_node_links(struct net *net, struct tipc_nl_msg *msg, + struct tipc_node *node, u32 *prev_link) { u32 i; int err; @@ -2575,7 +2095,7 @@ static int __tipc_nl_add_node_links(struct tipc_nl_msg *msg, if (!node->links[i]) continue; - err = __tipc_nl_add_link(msg, node->links[i]); + err = __tipc_nl_add_link(net, msg, node->links[i]); if (err) return err; } @@ -2586,6 +2106,8 @@ static int __tipc_nl_add_node_links(struct tipc_nl_msg *msg, int tipc_nl_link_dump(struct sk_buff *skb, struct netlink_callback *cb) { + struct net *net = sock_net(skb->sk); + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_node *node; struct tipc_nl_msg msg; u32 prev_node = cb->args[0]; @@ -2601,9 +2123,8 @@ 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(prev_node); + node = tipc_node_find(net, prev_node); if (!node) { /* We never set seq or call nl_dump_check_consistent() * this means that setting prev_seq here will cause the @@ -2614,24 +2135,29 @@ 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, &tipc_node_list, list) { + list_for_each_entry_continue_rcu(node, &tn->node_list, + list) { tipc_node_lock(node); - err = __tipc_nl_add_node_links(&msg, node, &prev_link); + err = __tipc_nl_add_node_links(net, &msg, node, + &prev_link); tipc_node_unlock(node); + tipc_node_put(node); if (err) goto out; prev_node = node->addr; } } else { - err = tipc_nl_add_bc_link(&msg); + err = tipc_nl_add_bc_link(net, &msg); if (err) goto out; - list_for_each_entry_rcu(node, &tipc_node_list, list) { + list_for_each_entry_rcu(node, &tn->node_list, list) { tipc_node_lock(node); - err = __tipc_nl_add_node_links(&msg, node, &prev_link); + err = __tipc_nl_add_node_links(net, &msg, node, + &prev_link); tipc_node_unlock(node); if (err) goto out; @@ -2652,6 +2178,7 @@ out: int tipc_nl_link_get(struct sk_buff *skb, struct genl_info *info) { + struct net *net = genl_info_net(info); struct sk_buff *ans_skb; struct tipc_nl_msg msg; struct tipc_link *link; @@ -2664,7 +2191,7 @@ int tipc_nl_link_get(struct sk_buff *skb, struct genl_info *info) return -EINVAL; name = nla_data(info->attrs[TIPC_NLA_LINK_NAME]); - node = tipc_link_find_owner(name, &bearer_id); + node = tipc_link_find_owner(net, name, &bearer_id); if (!node) return -EINVAL; @@ -2683,7 +2210,7 @@ int tipc_nl_link_get(struct sk_buff *skb, struct genl_info *info) goto err_out; } - err = __tipc_nl_add_link(&msg, link); + err = __tipc_nl_add_link(net, &msg, link); if (err) goto err_out; @@ -2706,6 +2233,7 @@ int tipc_nl_link_reset_stats(struct sk_buff *skb, struct genl_info *info) struct tipc_link *link; struct tipc_node *node; struct nlattr *attrs[TIPC_NLA_LINK_MAX + 1]; + struct net *net = sock_net(skb->sk); if (!info->attrs[TIPC_NLA_LINK]) return -EINVAL; @@ -2722,13 +2250,13 @@ int tipc_nl_link_reset_stats(struct sk_buff *skb, struct genl_info *info) link_name = nla_data(attrs[TIPC_NLA_LINK_NAME]); if (strcmp(link_name, tipc_bclink_name) == 0) { - err = tipc_bclink_reset_stats(); + err = tipc_bclink_reset_stats(net); if (err) return err; return 0; } - node = tipc_link_find_owner(link_name, &bearer_id); + node = tipc_link_find_owner(net, link_name, &bearer_id); if (!node) return -EINVAL; diff --git a/net/tipc/link.h b/net/tipc/link.h index 55812e87ca1e..b5b4e3554d4e 100644 --- a/net/tipc/link.h +++ b/net/tipc/link.h @@ -41,6 +41,10 @@ #include "msg.h" #include "node.h" +/* TIPC-specific error codes +*/ +#define ELINKCONG EAGAIN /* link congestion <=> resource unavailable */ + /* Out-of-range value for link sequence numbers */ #define INVALID_LINK_SEQ 0x10000 @@ -54,8 +58,10 @@ /* Link endpoint execution states */ -#define LINK_STARTED 0x0001 -#define LINK_STOPPED 0x0002 +#define LINK_STARTED 0x0001 +#define LINK_STOPPED 0x0002 +#define LINK_SYNCHING 0x0004 +#define LINK_FAILINGOVER 0x0008 /* Starting value for maximum packet size negotiation on unicast links * (unless bearer MTU is less) @@ -99,13 +105,14 @@ struct tipc_stats { * @media_addr: media address to use when sending messages over link * @timer: link timer * @owner: pointer to peer node + * @refcnt: reference counter for permanent references (owner node & timer) * @flags: execution state flags for link endpoint instance * @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 * @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] + * @cont_intv: link continuity testing interval * @abort_limit: # of unacknowledged continuity probes needed to reset link * @state: current state of link FSM * @fsm_msg_cnt: # of protocol messages link FSM has sent in current state @@ -113,21 +120,23 @@ 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 - * @max_pkt_target: desired maximum packet size for this link - * @max_pkt_probes: # of probes based on current (max_pkt, max_pkt_target) - * @outqueue: outbound message queue + * @mtu: current maximum packet size for this link + * @advertised_mtu: advertised own mtu when link is being established + * @transmitq: queue for sent, non-acked messages + * @backlogq: queue for messages waiting to be sent * @next_out_no: next sequence number to use for outbound messages * @last_retransmitted: sequence number of most recently retransmitted message * @stale_count: # of identical retransmit requests made by peer * @next_in_no: next sequence number to expect for inbound messages * @deferred_queue: deferred queue saved OOS b'cast message received from node * @unacked_window: # of inbound messages rx'd without ack'ing back to peer + * @inputq: buffer queue for messages to be delivered upwards + * @namedq: buffer queue for name table messages to be delivered upwards * @next_out: ptr to first unsent outbound message in queue - * @waiting_sks: linked list of sockets waiting for link congestion to abate + * @wakeupq: linked list of wakeup msgs waiting for link congestion to abate * @long_msg_seq_no: next identifier to use for outbound fragmented messages * @reasm_buf: head of partially reassembled inbound message fragments * @stats: collects statistics regarding link activity @@ -138,6 +147,7 @@ struct tipc_link { struct tipc_media_addr media_addr; struct timer_list timer; struct tipc_node *owner; + struct kref ref; /* Management and link supervision data */ unsigned int flags; @@ -146,7 +156,7 @@ struct tipc_link { u32 peer_bearer_id; u32 bearer_id; u32 tolerance; - u32 continuity_interval; + unsigned long cont_intv; u32 abort_limit; int state; u32 fsm_msg_cnt; @@ -157,34 +167,40 @@ struct tipc_link { struct tipc_msg *pmsg; u32 priority; char net_plane; - u32 queue_limit[15]; /* queue_limit[0]==window limit */ + u16 synch_point; - /* Changeover */ - u32 exp_msg_count; - u32 reset_checkpoint; + /* Failover */ + u16 failover_pkts; + u16 failover_checkpt; + struct sk_buff *failover_skb; /* Max packet negotiation */ - u32 max_pkt; - u32 max_pkt_target; - u32 max_pkt_probes; + u16 mtu; + u16 advertised_mtu; /* Sending */ - struct sk_buff_head outqueue; + 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; u32 stale_count; /* Reception */ u32 next_in_no; - struct sk_buff_head deferred_queue; - u32 unacked_window; + u32 rcv_unacked; + struct sk_buff_head deferdq; + struct sk_buff_head inputq; + struct sk_buff_head namedq; /* Congestion handling */ - struct sk_buff *next_out; - struct sk_buff_head waiting_sks; + struct sk_buff_head wakeupq; /* Fragmentation/reassembly */ - u32 long_msg_seq_no; struct sk_buff *reasm_buf; /* Statistics */ @@ -196,30 +212,26 @@ struct tipc_port; struct tipc_link *tipc_link_create(struct tipc_node *n_ptr, struct tipc_bearer *b_ptr, const struct tipc_media_addr *media_addr); -void tipc_link_delete_list(unsigned int bearer_id, bool shutting_down); +void tipc_link_delete(struct tipc_link *link); +void tipc_link_delete_list(struct net *net, unsigned int bearer_id, + bool shutting_down); void tipc_link_failover_send_queue(struct tipc_link *l_ptr); void tipc_link_dup_queue_xmit(struct tipc_link *l_ptr, struct tipc_link *dest); void tipc_link_reset_fragments(struct tipc_link *l_ptr); int tipc_link_is_up(struct tipc_link *l_ptr); int tipc_link_is_active(struct tipc_link *l_ptr); void tipc_link_purge_queues(struct tipc_link *l_ptr); -struct sk_buff *tipc_link_cmd_config(const void *req_tlv_area, - int req_tlv_space, - u16 cmd); -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_skb(struct sk_buff *skb, u32 dest, u32 selector); -int tipc_link_xmit(struct sk_buff_head *list, u32 dest, u32 selector); -int __tipc_link_xmit(struct tipc_link *link, struct sk_buff_head *list); -u32 tipc_link_get_max_pkt(u32 dest, u32 selector); -void tipc_link_bundle_rcv(struct sk_buff *buf); +void tipc_link_reset_list(struct net *net, unsigned int bearer_id); +int tipc_link_xmit_skb(struct net *net, struct sk_buff *skb, u32 dest, + u32 selector); +int tipc_link_xmit(struct net *net, struct sk_buff_head *list, u32 dest, + u32 selector); +int __tipc_link_xmit(struct net *net, struct tipc_link *link, + struct sk_buff_head *list); void tipc_link_proto_xmit(struct tipc_link *l_ptr, u32 msg_typ, int prob, - u32 gap, u32 tolerance, u32 priority, u32 acked_mtu); + u32 gap, u32 tolerance, u32 priority); void tipc_link_push_packets(struct tipc_link *l_ptr); u32 tipc_link_defer_pkt(struct sk_buff_head *list, struct sk_buff *buf); void tipc_link_set_queue_limits(struct tipc_link *l_ptr, u32 window); @@ -233,6 +245,7 @@ int tipc_nl_link_get(struct sk_buff *skb, struct genl_info *info); int tipc_nl_link_set(struct sk_buff *skb, struct genl_info *info); int tipc_nl_link_reset_stats(struct sk_buff *skb, struct genl_info *info); int tipc_nl_parse_link_prop(struct nlattr *prop, struct nlattr *props[]); +void link_prepare_wakeup(struct tipc_link *l); /* * Link sequence number manipulation routines (uses modulo 2**16 arithmetic) @@ -267,6 +280,10 @@ static inline u32 lesser(u32 left, u32 right) return less_eq(left, right) ? left : right; } +static inline u32 link_own_addr(struct tipc_link *l) +{ + return msg_prevnode(l->pmsg); +} /* * Link status checking routines @@ -291,9 +308,4 @@ static inline int link_reset_reset(struct tipc_link *l_ptr) return l_ptr->state == RESET_RESET; } -static inline int link_congested(struct tipc_link *l_ptr) -{ - return skb_queue_len(&l_ptr->outqueue) >= l_ptr->queue_limit[0]; -} - #endif diff --git a/net/tipc/log.c b/net/tipc/log.c deleted file mode 100644 index abef644f27d8..000000000000 --- a/net/tipc/log.c +++ /dev/null @@ -1,55 +0,0 @@ -/* - * net/tipc/log.c: TIPC print buffer routines for debugging - * - * Copyright (c) 1996-2006, Ericsson AB - * Copyright (c) 2005-2007, 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" -#include "config.h" - -/** - * tipc_snprintf - append formatted output to print buffer - * @buf: pointer to print buffer - * @len: buffer length - * @fmt: formatted info to be printed - */ -int tipc_snprintf(char *buf, int len, const char *fmt, ...) -{ - int i; - va_list args; - - va_start(args, fmt); - i = vscnprintf(buf, len, fmt, args); - va_end(args); - return i; -} diff --git a/net/tipc/msg.c b/net/tipc/msg.c index a687b30a699c..c3e96e815418 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, 2014, Ericsson AB + * Copyright (c) 2000-2006, 2014-2015, Ericsson AB * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * @@ -34,6 +34,7 @@ * POSSIBILITY OF SUCH DAMAGE. */ +#include <net/sock.h> #include "core.h" #include "msg.h" #include "addr.h" @@ -46,25 +47,48 @@ static unsigned int align(unsigned int i) return (i + 3) & ~3u; } -void tipc_msg_init(struct tipc_msg *m, u32 user, u32 type, u32 hsize, - u32 destnode) +/** + * tipc_buf_acquire - creates a TIPC message buffer + * @size: message size (including TIPC header) + * + * Returns a new buffer with data pointers set to the specified size. + * + * NOTE: Headroom is reserved to allow prepending of a data link header. + * There may also be unrequested tailroom present at the buffer's end. + */ +struct sk_buff *tipc_buf_acquire(u32 size) +{ + struct sk_buff *skb; + unsigned int buf_size = (BUF_HEADROOM + size + 3) & ~3u; + + skb = alloc_skb_fclone(buf_size, GFP_ATOMIC); + if (skb) { + skb_reserve(skb, BUF_HEADROOM); + skb_put(skb, size); + skb->next = NULL; + } + return skb; +} + +void tipc_msg_init(u32 own_node, struct tipc_msg *m, u32 user, u32 type, + u32 hsize, u32 dnode) { memset(m, 0, hsize); msg_set_version(m); msg_set_user(m, user); msg_set_hdr_sz(m, hsize); msg_set_size(m, hsize); - msg_set_prevnode(m, tipc_own_addr); + msg_set_prevnode(m, own_node); msg_set_type(m, type); if (hsize > SHORT_H_SIZE) { - msg_set_orignode(m, tipc_own_addr); - msg_set_destnode(m, destnode); + msg_set_orignode(m, own_node); + msg_set_destnode(m, dnode); } } -struct sk_buff *tipc_msg_create(uint user, uint type, uint hdr_sz, - uint data_sz, u32 dnode, u32 onode, - u32 dport, u32 oport, int errcode) +struct sk_buff *tipc_msg_create(uint user, uint type, + uint hdr_sz, uint data_sz, u32 dnode, + u32 onode, u32 dport, u32 oport, int errcode) { struct tipc_msg *msg; struct sk_buff *buf; @@ -74,9 +98,8 @@ struct sk_buff *tipc_msg_create(uint user, uint type, uint hdr_sz, return NULL; msg = buf_msg(buf); - tipc_msg_init(msg, user, type, hdr_sz, dnode); + tipc_msg_init(onode, msg, user, type, hdr_sz, dnode); msg_set_size(msg, hdr_sz + data_sz); - msg_set_prevnode(msg, onode); msg_set_origport(msg, oport); msg_set_destport(msg, dport); msg_set_errcode(msg, errcode); @@ -142,6 +165,9 @@ int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf) } if (fragid == LAST_FRAGMENT) { + TIPC_SKB_CB(head)->validated = false; + if (unlikely(!tipc_msg_validate(head))) + goto err; *buf = head; TIPC_SKB_CB(head)->tail = NULL; *headbuf = NULL; @@ -149,7 +175,6 @@ int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf) } *buf = NULL; return 0; - err: pr_warn_ratelimited("Unable to build fragment list\n"); kfree_skb(*buf); @@ -158,20 +183,61 @@ err: return 0; } +/* tipc_msg_validate - validate basic format of received message + * + * This routine ensures a TIPC message has an acceptable header, and at least + * as much data as the header indicates it should. The routine also ensures + * that the entire message header is stored in the main fragment of the message + * buffer, to simplify future access to message header fields. + * + * Note: Having extra info present in the message header or data areas is OK. + * TIPC will ignore the excess, under the assumption that it is optional info + * introduced by a later release of the protocol. + */ +bool tipc_msg_validate(struct sk_buff *skb) +{ + struct tipc_msg *msg; + int msz, hsz; + + if (unlikely(TIPC_SKB_CB(skb)->validated)) + return true; + if (unlikely(!pskb_may_pull(skb, MIN_H_SIZE))) + return false; + + hsz = msg_hdr_sz(buf_msg(skb)); + if (unlikely(hsz < MIN_H_SIZE) || (hsz > MAX_H_SIZE)) + return false; + if (unlikely(!pskb_may_pull(skb, hsz))) + return false; + + msg = buf_msg(skb); + if (unlikely(msg_version(msg) != TIPC_VERSION)) + return false; + + msz = msg_size(msg); + if (unlikely(msz < hsz)) + return false; + if (unlikely((msz - hsz) > TIPC_MAX_USER_MSG_SIZE)) + return false; + if (unlikely(skb->len < msz)) + return false; + + TIPC_SKB_CB(skb)->validated = true; + return true; +} /** * tipc_msg_build - create buffer chain containing specified header and data * @mhdr: Message header, to be prepended to data * @m: User message - * @offset: Posision in iov to start copying from * @dsz: Total length of user data * @pktmax: Max packet size that can be used * @list: Buffer or chain of buffers to be returned to caller * * Returns message data size or errno: -ENOMEM, -EFAULT */ -int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m, int offset, - int dsz, int pktmax, struct sk_buff_head *list) +int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m, + int offset, int dsz, int pktmax, struct sk_buff_head *list) { int mhsz = msg_hdr_sz(mhdr); int msz = mhsz + dsz; @@ -191,26 +257,28 @@ int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m, int offset, skb = tipc_buf_acquire(msz); if (unlikely(!skb)) return -ENOMEM; + skb_orphan(skb); __skb_queue_tail(list, skb); skb_copy_to_linear_data(skb, mhdr, mhsz); pktpos = skb->data + mhsz; - if (!dsz || !memcpy_fromiovecend(pktpos, m->msg_iter.iov, offset, - dsz)) + if (copy_from_iter(pktpos, dsz, &m->msg_iter) == dsz) return dsz; rc = -EFAULT; goto error; } /* Prepare reusable fragment header */ - tipc_msg_init(&pkthdr, MSG_FRAGMENTER, FIRST_FRAGMENT, - INT_H_SIZE, msg_destnode(mhdr)); + tipc_msg_init(msg_prevnode(mhdr), &pkthdr, MSG_FRAGMENTER, + FIRST_FRAGMENT, INT_H_SIZE, msg_destnode(mhdr)); msg_set_size(&pkthdr, pktmax); msg_set_fragm_no(&pkthdr, pktno); + msg_set_importance(&pkthdr, msg_importance(mhdr)); /* Prepare first fragment */ skb = tipc_buf_acquire(pktmax); if (!skb) return -ENOMEM; + skb_orphan(skb); __skb_queue_tail(list, skb); pktpos = skb->data; skb_copy_to_linear_data(skb, &pkthdr, INT_H_SIZE); @@ -224,12 +292,11 @@ int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m, int offset, if (drem < pktrem) pktrem = drem; - if (memcpy_fromiovecend(pktpos, m->msg_iter.iov, offset, pktrem)) { + if (copy_from_iter(pktpos, pktrem, &m->msg_iter) != pktrem) { rc = -EFAULT; goto error; } drem -= pktrem; - offset += pktrem; if (!drem) break; @@ -244,6 +311,7 @@ int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m, int offset, rc = -ENOMEM; goto error; } + skb_orphan(skb); __skb_queue_tail(list, skb); msg_set_type(&pkthdr, FRAGMENT); msg_set_size(&pkthdr, pktsz); @@ -263,33 +331,36 @@ error: /** * tipc_msg_bundle(): Append contents of a buffer to tail of an existing one - * @list: the buffer chain of the existing buffer ("bundle") + * @bskb: the buffer to append to ("bundle") * @skb: buffer to be appended * @mtu: max allowable size for the bundle buffer * Consumes buffer if successful * Returns true if bundling could be performed, otherwise false */ -bool tipc_msg_bundle(struct sk_buff_head *list, struct sk_buff *skb, u32 mtu) +bool tipc_msg_bundle(struct sk_buff *bskb, struct sk_buff *skb, u32 mtu) { - struct sk_buff *bskb = skb_peek_tail(list); - struct tipc_msg *bmsg = buf_msg(bskb); + struct tipc_msg *bmsg; struct tipc_msg *msg = buf_msg(skb); - unsigned int bsz = msg_size(bmsg); + unsigned int bsz; unsigned int msz = msg_size(msg); - u32 start = align(bsz); + u32 start, pad; u32 max = mtu - INT_H_SIZE; - u32 pad = start - bsz; if (likely(msg_user(msg) == MSG_FRAGMENTER)) return false; - if (unlikely(msg_user(msg) == CHANGEOVER_PROTOCOL)) + if (!bskb) + return false; + bmsg = buf_msg(bskb); + bsz = msg_size(bmsg); + start = align(bsz); + pad = start - bsz; + + if (unlikely(msg_user(msg) == TUNNEL_PROTOCOL)) return false; if (unlikely(msg_user(msg) == BCAST_PROTOCOL)) return false; if (likely(msg_user(bmsg) != MSG_BUNDLER)) return false; - if (likely(!TIPC_SKB_CB(bskb)->bundling)) - return false; if (unlikely(skb_tailroom(bskb) < (pad + msz))) return false; if (unlikely(max < (start + msz))) @@ -304,6 +375,46 @@ bool tipc_msg_bundle(struct sk_buff_head *list, struct sk_buff *skb, u32 mtu) } /** + * tipc_msg_extract(): extract bundled inner packet from buffer + * @skb: buffer to be extracted from. + * @iskb: extracted inner buffer, to be returned + * @pos: position in outer message of msg to be extracted. + * Returns position of next msg + * Consumes outer buffer when last packet extracted + * Returns true when when there is an extracted buffer, otherwise false + */ +bool tipc_msg_extract(struct sk_buff *skb, struct sk_buff **iskb, int *pos) +{ + struct tipc_msg *msg; + int imsz, offset; + + *iskb = NULL; + if (unlikely(skb_linearize(skb))) + goto none; + + msg = buf_msg(skb); + offset = msg_hdr_sz(msg) + *pos; + if (unlikely(offset > (msg_size(msg) - MIN_H_SIZE))) + goto none; + + *iskb = skb_clone(skb, GFP_ATOMIC); + if (unlikely(!*iskb)) + goto none; + skb_pull(*iskb, offset); + imsz = msg_size(buf_msg(*iskb)); + skb_trim(*iskb, imsz); + if (unlikely(!tipc_msg_validate(*iskb))) + goto none; + *pos += align(imsz); + return true; +none: + kfree_skb(skb); + kfree_skb(*iskb); + *iskb = NULL; + return false; +} + +/** * tipc_msg_make_bundle(): Create bundle buf and append message to its tail * @list: the buffer chain * @skb: buffer to be appended and replaced @@ -312,18 +423,17 @@ bool tipc_msg_bundle(struct sk_buff_head *list, struct sk_buff *skb, u32 mtu) * Replaces buffer if successful * Returns true if success, otherwise false */ -bool tipc_msg_make_bundle(struct sk_buff_head *list, struct sk_buff *skb, - u32 mtu, u32 dnode) +bool tipc_msg_make_bundle(struct sk_buff **skb, u32 mtu, u32 dnode) { struct sk_buff *bskb; struct tipc_msg *bmsg; - struct tipc_msg *msg = buf_msg(skb); + struct tipc_msg *msg = buf_msg(*skb); u32 msz = msg_size(msg); u32 max = mtu - INT_H_SIZE; if (msg_user(msg) == MSG_FRAGMENTER) return false; - if (msg_user(msg) == CHANGEOVER_PROTOCOL) + if (msg_user(msg) == TUNNEL_PROTOCOL) return false; if (msg_user(msg) == BCAST_PROTOCOL) return false; @@ -336,13 +446,14 @@ bool tipc_msg_make_bundle(struct sk_buff_head *list, struct sk_buff *skb, skb_trim(bskb, INT_H_SIZE); bmsg = buf_msg(bskb); - tipc_msg_init(bmsg, MSG_BUNDLER, 0, INT_H_SIZE, dnode); + tipc_msg_init(msg_prevnode(msg), bmsg, MSG_BUNDLER, 0, + INT_H_SIZE, dnode); msg_set_seqno(bmsg, msg_seqno(msg)); msg_set_ack(bmsg, msg_ack(msg)); msg_set_bcast_ack(bmsg, msg_bcast_ack(msg)); - TIPC_SKB_CB(bskb)->bundling = true; - __skb_queue_tail(list, bskb); - return tipc_msg_bundle(list, skb, mtu); + tipc_msg_bundle(bskb, *skb, mtu); + *skb = bskb; + return true; } /** @@ -353,28 +464,25 @@ bool tipc_msg_make_bundle(struct sk_buff_head *list, struct sk_buff *skb, * Consumes buffer if failure * Returns true if success, otherwise false */ -bool tipc_msg_reverse(struct sk_buff *buf, u32 *dnode, int err) +bool tipc_msg_reverse(u32 own_addr, struct sk_buff *buf, u32 *dnode, + int err) { struct tipc_msg *msg = buf_msg(buf); - uint imp = msg_importance(msg); struct tipc_msg ohdr; uint rdsz = min_t(uint, msg_data_sz(msg), MAX_FORWARD_SIZE); if (skb_linearize(buf)) goto exit; + msg = buf_msg(buf); if (msg_dest_droppable(msg)) goto exit; if (msg_errcode(msg)) goto exit; - memcpy(&ohdr, msg, msg_hdr_sz(msg)); - imp = min_t(uint, imp + 1, TIPC_CRITICAL_IMPORTANCE); - if (msg_isdata(msg)) - msg_set_importance(msg, imp); msg_set_errcode(msg, err); msg_set_origport(msg, msg_destport(&ohdr)); msg_set_destport(msg, msg_origport(&ohdr)); - msg_set_prevnode(msg, tipc_own_addr); + msg_set_prevnode(msg, own_addr); if (!msg_short(msg)) { msg_set_orignode(msg, msg_destnode(&ohdr)); msg_set_destnode(msg, msg_orignode(&ohdr)); @@ -386,43 +494,48 @@ bool tipc_msg_reverse(struct sk_buff *buf, u32 *dnode, int err) return true; exit: kfree_skb(buf); + *dnode = 0; return false; } /** - * tipc_msg_eval: determine fate of message that found no destination - * @buf: the buffer containing the message. - * @dnode: return value: next-hop node, if message to be forwarded - * @err: error code to use, if message to be rejected - * + * tipc_msg_lookup_dest(): try to find new destination for named message + * @skb: the buffer containing the message. + * @dnode: return value: next-hop node, if destination found + * @err: return value: error code to use, if message to be rejected * Does not consume buffer - * Returns 0 (TIPC_OK) if message ok and we can try again, -TIPC error - * code if message to be rejected + * Returns true if a destination is found, false otherwise */ -int tipc_msg_eval(struct sk_buff *buf, u32 *dnode) +bool tipc_msg_lookup_dest(struct net *net, struct sk_buff *skb, + u32 *dnode, int *err) { - struct tipc_msg *msg = buf_msg(buf); + struct tipc_msg *msg = buf_msg(skb); u32 dport; + u32 own_addr = tipc_own_addr(net); - if (msg_type(msg) != TIPC_NAMED_MSG) - return -TIPC_ERR_NO_PORT; - if (skb_linearize(buf)) - return -TIPC_ERR_NO_NAME; - if (msg_data_sz(msg) > MAX_FORWARD_SIZE) - return -TIPC_ERR_NO_NAME; - if (msg_reroute_cnt(msg) > 0) - return -TIPC_ERR_NO_NAME; - - *dnode = addr_domain(msg_lookup_scope(msg)); - dport = tipc_nametbl_translate(msg_nametype(msg), - msg_nameinst(msg), - dnode); + if (!msg_isdata(msg)) + return false; + if (!msg_named(msg)) + return false; + if (msg_errcode(msg)) + return false; + *err = -TIPC_ERR_NO_NAME; + if (skb_linearize(skb)) + return false; + if (msg_reroute_cnt(msg)) + return false; + *dnode = addr_domain(net, msg_lookup_scope(msg)); + dport = tipc_nametbl_translate(net, msg_nametype(msg), + msg_nameinst(msg), dnode); if (!dport) - return -TIPC_ERR_NO_NAME; + return false; msg_incr_reroute_cnt(msg); + if (*dnode != own_addr) + msg_set_prevnode(msg, own_addr); msg_set_destnode(msg, *dnode); msg_set_destport(msg, dport); - return TIPC_OK; + *err = TIPC_OK; + return true; } /* tipc_msg_reassemble() - clone a buffer chain of fragments and diff --git a/net/tipc/msg.h b/net/tipc/msg.h index d5c83d7ecb47..e1d3595e2ee9 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, 2014, Ericsson AB + * Copyright (c) 2000-2007, 2014-2015 Ericsson AB * Copyright (c) 2005-2008, 2010-2011, Wind River Systems * All rights reserved. * @@ -37,7 +37,7 @@ #ifndef _TIPC_MSG_H #define _TIPC_MSG_H -#include "bearer.h" +#include <linux/tipc.h> /* * Constants and routines used to read and write TIPC payload message headers @@ -45,6 +45,7 @@ * Note: Some items are also used with TIPC internal message headers */ #define TIPC_VERSION 2 +struct plist; /* * Payload message users are defined in TIPC's public API: @@ -53,6 +54,8 @@ * - TIPC_HIGH_IMPORTANCE * - TIPC_CRITICAL_IMPORTANCE */ +#define TIPC_SYSTEM_IMPORTANCE 4 + /* * Payload message types @@ -63,6 +66,19 @@ #define TIPC_DIRECT_MSG 3 /* + * Internal message users + */ +#define BCAST_PROTOCOL 5 +#define MSG_BUNDLER 6 +#define LINK_PROTOCOL 7 +#define CONN_MANAGER 8 +#define TUNNEL_PROTOCOL 10 +#define NAME_DISTRIBUTOR 11 +#define MSG_FRAGMENTER 12 +#define LINK_CONFIG 13 +#define SOCK_WAKEUP 14 /* pseudo user */ + +/* * Message header sizes */ #define SHORT_H_SIZE 24 /* In-cluster basic payload message */ @@ -75,13 +91,39 @@ #define MAX_MSG_SIZE (MAX_H_SIZE + TIPC_MAX_USER_MSG_SIZE) -#define TIPC_MEDIA_ADDR_OFFSET 5 +#define TIPC_MEDIA_INFO_OFFSET 5 +/** + * TIPC message buffer code + * + * TIPC message buffer headroom reserves space for the worst-case + * link-level device header (in case the message is sent off-node). + * + * Note: Headroom should be a multiple of 4 to ensure the TIPC header fields + * are word aligned for quicker access + */ +#define BUF_HEADROOM (LL_MAX_HEADER + 48) + +struct tipc_skb_cb { + void *handle; + struct sk_buff *tail; + bool validated; + bool wakeup_pending; + bool bundling; + u16 chain_sz; + u16 chain_imp; +}; + +#define TIPC_SKB_CB(__skb) ((struct tipc_skb_cb *)&((__skb)->cb[0])) struct tipc_msg { __be32 hdr[15]; }; +static inline struct tipc_msg *buf_msg(struct sk_buff *skb) +{ + return (struct tipc_msg *)skb->data; +} static inline u32 msg_word(struct tipc_msg *m, u32 pos) { @@ -143,16 +185,6 @@ static inline void msg_set_user(struct tipc_msg *m, u32 n) msg_set_bits(m, 0, 25, 0xf, n); } -static inline u32 msg_importance(struct tipc_msg *m) -{ - return msg_bits(m, 0, 25, 0xf); -} - -static inline void msg_set_importance(struct tipc_msg *m, u32 i) -{ - msg_set_user(m, i); -} - static inline u32 msg_hdr_sz(struct tipc_msg *m) { return msg_bits(m, 0, 21, 0xf) << 2; @@ -208,6 +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 @@ -309,6 +350,25 @@ static inline void msg_set_seqno(struct tipc_msg *m, u32 n) /* * Words 3-10 */ +static inline u32 msg_importance(struct tipc_msg *m) +{ + if (unlikely(msg_user(m) == MSG_FRAGMENTER)) + return msg_bits(m, 5, 13, 0x7); + if (likely(msg_isdata(m) && !msg_errcode(m))) + return msg_user(m); + return TIPC_SYSTEM_IMPORTANCE; +} + +static inline void msg_set_importance(struct tipc_msg *m, u32 i) +{ + if (unlikely(msg_user(m) == MSG_FRAGMENTER)) + msg_set_bits(m, 5, 13, 0x7, i); + else if (likely(i < TIPC_SYSTEM_IMPORTANCE)) + msg_set_user(m, i); + else + pr_warn("Trying to set illegal importance in message\n"); +} + static inline u32 msg_prevnode(struct tipc_msg *m) { return msg_word(m, 3); @@ -321,6 +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); } @@ -416,35 +478,11 @@ 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 */ /* - * Internal message users - */ -#define BCAST_PROTOCOL 5 -#define MSG_BUNDLER 6 -#define LINK_PROTOCOL 7 -#define CONN_MANAGER 8 -#define ROUTE_DISTRIBUTOR 9 /* obsoleted */ -#define CHANGEOVER_PROTOCOL 10 -#define NAME_DISTRIBUTOR 11 -#define MSG_FRAGMENTER 12 -#define LINK_CONFIG 13 -#define SOCK_WAKEUP 14 /* pseudo user */ - -/* * Connection management protocol message types */ #define CONN_PROBE 0 @@ -474,8 +512,8 @@ static inline struct tipc_msg *msg_get_wrapped(struct tipc_msg *m) /* * Changeover tunnel message types */ -#define DUPLICATE_MSG 0 -#define ORIGINAL_MSG 1 +#define SYNCH_MSG 0 +#define FAILOVER_MSG 1 /* * Config protocol message types @@ -483,7 +521,6 @@ static inline struct tipc_msg *msg_get_wrapped(struct tipc_msg *m) #define DSC_REQ_MSG 0 #define DSC_RESP_MSG 1 - /* * Word 1 */ @@ -507,6 +544,24 @@ static inline void msg_set_node_sig(struct tipc_msg *m, u32 n) msg_set_bits(m, 1, 0, 0xffff, n); } +static inline u32 msg_node_capabilities(struct tipc_msg *m) +{ + return msg_bits(m, 1, 15, 0x1fff); +} + +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) != TUNNEL_PROTOCOL)) + return false; + if (msg_type(m) != SYNCH_MSG) + return false; + return true; +} /* * Word 2 @@ -661,7 +716,7 @@ static inline void msg_set_redundant_link(struct tipc_msg *m, u32 r) static inline char *msg_media_addr(struct tipc_msg *m) { - return (char *)&m->hdr[TIPC_MEDIA_ADDR_OFFSET]; + return (char *)&m->hdr[TIPC_MEDIA_INFO_OFFSET]; } /* @@ -707,41 +762,112 @@ static inline void msg_set_link_tolerance(struct tipc_msg *m, u32 n) msg_set_bits(m, 9, 0, 0xffff, n); } -static inline u32 tipc_msg_tot_importance(struct tipc_msg *m) -{ - if ((msg_user(m) == MSG_FRAGMENTER) && (msg_type(m) == FIRST_FRAGMENT)) - return msg_importance(msg_get_wrapped(m)); - return msg_importance(m); -} - -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); -} - -bool tipc_msg_reverse(struct sk_buff *buf, u32 *dnode, int err); - -int tipc_msg_eval(struct sk_buff *buf, u32 *dnode); - -void tipc_msg_init(struct tipc_msg *m, u32 user, u32 type, u32 hsize, - u32 destnode); - +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, + int err); +void tipc_msg_init(u32 own_addr, struct tipc_msg *m, u32 user, u32 type, + u32 hsize, u32 destnode); struct sk_buff *tipc_msg_create(uint user, uint type, uint hdr_sz, uint data_sz, u32 dnode, u32 onode, u32 dport, u32 oport, int errcode); - int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf); +bool tipc_msg_bundle(struct sk_buff *bskb, struct sk_buff *skb, u32 mtu); + +bool tipc_msg_make_bundle(struct sk_buff **skb, u32 mtu, u32 dnode); +bool tipc_msg_extract(struct sk_buff *skb, struct sk_buff **iskb, int *pos); +int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m, + int offset, int dsz, int mtu, struct sk_buff_head *list); +bool tipc_msg_lookup_dest(struct net *net, struct sk_buff *skb, u32 *dnode, + int *err); +struct sk_buff *tipc_msg_reassemble(struct sk_buff_head *list); -bool tipc_msg_bundle(struct sk_buff_head *list, struct sk_buff *skb, u32 mtu); - -bool tipc_msg_make_bundle(struct sk_buff_head *list, struct sk_buff *skb, - u32 mtu, u32 dnode); - -int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m, int offset, - int dsz, int mtu, struct sk_buff_head *list); +/* tipc_skb_peek(): peek and reserve first buffer in list + * @list: list to be peeked in + * Returns pointer to first buffer in list, if any + */ +static inline struct sk_buff *tipc_skb_peek(struct sk_buff_head *list, + spinlock_t *lock) +{ + struct sk_buff *skb; + + spin_lock_bh(lock); + skb = skb_peek(list); + if (skb) + skb_get(skb); + spin_unlock_bh(lock); + return skb; +} + +/* tipc_skb_peek_port(): find a destination port, ignoring all destinations + * up to and including 'filter'. + * Note: ignoring previously tried destinations minimizes the risk of + * contention on the socket lock + * @list: list to be peeked in + * @filter: last destination to be ignored from search + * Returns a destination port number, of applicable. + */ +static inline u32 tipc_skb_peek_port(struct sk_buff_head *list, u32 filter) +{ + struct sk_buff *skb; + u32 dport = 0; + bool ignore = true; + + spin_lock_bh(&list->lock); + skb_queue_walk(list, skb) { + dport = msg_destport(buf_msg(skb)); + if (!filter || skb_queue_is_last(list, skb)) + break; + if (dport == filter) + ignore = false; + else if (!ignore) + break; + } + spin_unlock_bh(&list->lock); + return dport; +} + +/* tipc_skb_dequeue(): unlink first buffer with dest 'dport' from list + * @list: list to be unlinked from + * @dport: selection criteria for buffer to unlink + */ +static inline struct sk_buff *tipc_skb_dequeue(struct sk_buff_head *list, + u32 dport) +{ + struct sk_buff *_skb, *tmp, *skb = NULL; + + spin_lock_bh(&list->lock); + skb_queue_walk_safe(list, _skb, tmp) { + if (msg_destport(buf_msg(_skb)) == dport) { + __skb_unlink(_skb, list); + skb = _skb; + break; + } + } + spin_unlock_bh(&list->lock); + return skb; +} + +/* tipc_skb_queue_tail(): add buffer to tail of list; + * @list: list to be appended to + * @skb: buffer to append. Always appended + * @dport: the destination port of the buffer + * returns true if dport differs from previous destination + */ +static inline bool tipc_skb_queue_tail(struct sk_buff_head *list, + struct sk_buff *skb, u32 dport) +{ + struct sk_buff *_skb = NULL; + bool rv = false; -struct sk_buff *tipc_msg_reassemble(struct sk_buff_head *list); + spin_lock_bh(&list->lock); + _skb = skb_peek_tail(list); + if (!_skb || (msg_destport(buf_msg(_skb)) != dport) || + (skb_queue_len(list) > 32)) + rv = true; + __skb_queue_tail(list, skb); + spin_unlock_bh(&list->lock); + return rv; +} #endif diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c index ba6083dca95b..41e7b7e4dda0 100644 --- a/net/tipc/name_distr.c +++ b/net/tipc/name_distr.c @@ -68,37 +68,41 @@ static void publ_to_item(struct distr_item *i, struct publication *p) /** * named_prepare_buf - allocate & initialize a publication message */ -static struct sk_buff *named_prepare_buf(u32 type, u32 size, u32 dest) +static struct sk_buff *named_prepare_buf(struct net *net, u32 type, u32 size, + u32 dest) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct sk_buff *buf = tipc_buf_acquire(INT_H_SIZE + size); struct tipc_msg *msg; if (buf != NULL) { msg = buf_msg(buf); - tipc_msg_init(msg, NAME_DISTRIBUTOR, type, INT_H_SIZE, dest); + tipc_msg_init(tn->own_addr, msg, NAME_DISTRIBUTOR, type, + INT_H_SIZE, dest); msg_set_size(msg, INT_H_SIZE + size); } return buf; } -void named_cluster_distribute(struct sk_buff *skb) +void named_cluster_distribute(struct net *net, struct sk_buff *skb) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct sk_buff *oskb; struct tipc_node *node; u32 dnode; rcu_read_lock(); - list_for_each_entry_rcu(node, &tipc_node_list, list) { + list_for_each_entry_rcu(node, &tn->node_list, list) { dnode = node->addr; - if (in_own_node(dnode)) + if (in_own_node(net, dnode)) continue; if (!tipc_node_active_links(node)) continue; - oskb = skb_copy(skb, GFP_ATOMIC); + oskb = pskb_copy(skb, GFP_ATOMIC); if (!oskb) break; msg_set_destnode(buf_msg(oskb), dnode); - tipc_link_xmit_skb(oskb, dnode, dnode); + tipc_link_xmit_skb(net, oskb, dnode, dnode); } rcu_read_unlock(); @@ -108,18 +112,19 @@ void named_cluster_distribute(struct sk_buff *skb) /** * tipc_named_publish - tell other nodes about a new publication by this node */ -struct sk_buff *tipc_named_publish(struct publication *publ) +struct sk_buff *tipc_named_publish(struct net *net, struct publication *publ) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct sk_buff *buf; struct distr_item *item; list_add_tail_rcu(&publ->local_list, - &tipc_nametbl->publ_list[publ->scope]); + &tn->nametbl->publ_list[publ->scope]); if (publ->scope == TIPC_NODE_SCOPE) return NULL; - buf = named_prepare_buf(PUBLICATION, ITEM_SIZE, 0); + buf = named_prepare_buf(net, PUBLICATION, ITEM_SIZE, 0); if (!buf) { pr_warn("Publication distribution failure\n"); return NULL; @@ -133,7 +138,7 @@ struct sk_buff *tipc_named_publish(struct publication *publ) /** * tipc_named_withdraw - tell other nodes about a withdrawn publication by this node */ -struct sk_buff *tipc_named_withdraw(struct publication *publ) +struct sk_buff *tipc_named_withdraw(struct net *net, struct publication *publ) { struct sk_buff *buf; struct distr_item *item; @@ -143,7 +148,7 @@ struct sk_buff *tipc_named_withdraw(struct publication *publ) if (publ->scope == TIPC_NODE_SCOPE) return NULL; - buf = named_prepare_buf(WITHDRAWAL, ITEM_SIZE, 0); + buf = named_prepare_buf(net, WITHDRAWAL, ITEM_SIZE, 0); if (!buf) { pr_warn("Withdrawal distribution failure\n"); return NULL; @@ -160,19 +165,21 @@ struct sk_buff *tipc_named_withdraw(struct publication *publ) * @dnode: node to be updated * @pls: linked list of publication items to be packed into buffer chain */ -static void named_distribute(struct sk_buff_head *list, u32 dnode, - struct list_head *pls) +static void named_distribute(struct net *net, struct sk_buff_head *list, + u32 dnode, struct list_head *pls) { struct publication *publ; struct sk_buff *skb = NULL; struct distr_item *item = NULL; - uint msg_dsz = (tipc_node_get_mtu(dnode, 0) / ITEM_SIZE) * ITEM_SIZE; + uint msg_dsz = (tipc_node_get_mtu(net, dnode, 0) / ITEM_SIZE) * + ITEM_SIZE; uint msg_rem = msg_dsz; list_for_each_entry(publ, pls, local_list) { /* Prepare next buffer: */ if (!skb) { - skb = named_prepare_buf(PUBLICATION, msg_rem, dnode); + skb = named_prepare_buf(net, PUBLICATION, msg_rem, + dnode); if (!skb) { pr_warn("Bulk publication failure\n"); return; @@ -202,30 +209,32 @@ static void named_distribute(struct sk_buff_head *list, u32 dnode, /** * tipc_named_node_up - tell specified node about all publications by this node */ -void tipc_named_node_up(u32 dnode) +void tipc_named_node_up(struct net *net, u32 dnode) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct sk_buff_head head; __skb_queue_head_init(&head); rcu_read_lock(); - named_distribute(&head, dnode, - &tipc_nametbl->publ_list[TIPC_CLUSTER_SCOPE]); - named_distribute(&head, dnode, - &tipc_nametbl->publ_list[TIPC_ZONE_SCOPE]); + named_distribute(net, &head, dnode, + &tn->nametbl->publ_list[TIPC_CLUSTER_SCOPE]); + named_distribute(net, &head, dnode, + &tn->nametbl->publ_list[TIPC_ZONE_SCOPE]); rcu_read_unlock(); - tipc_link_xmit(&head, dnode, dnode); + tipc_link_xmit(net, &head, dnode, dnode); } -static void tipc_publ_subscribe(struct publication *publ, u32 addr) +static void tipc_publ_subscribe(struct net *net, struct publication *publ, + u32 addr) { struct tipc_node *node; - if (in_own_node(addr)) + if (in_own_node(net, addr)) return; - node = tipc_node_find(addr); + node = tipc_node_find(net, addr); if (!node) { pr_warn("Node subscription rejected, unknown node 0x%x\n", addr); @@ -235,19 +244,22 @@ static void tipc_publ_subscribe(struct publication *publ, u32 addr) 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 publication *publ, u32 addr) +static void tipc_publ_unsubscribe(struct net *net, struct publication *publ, + u32 addr) { struct tipc_node *node; - node = tipc_node_find(addr); + node = tipc_node_find(net, addr); if (!node) return; tipc_node_lock(node); list_del_init(&publ->nodesub_list); tipc_node_unlock(node); + tipc_node_put(node); } /** @@ -256,16 +268,17 @@ static void tipc_publ_unsubscribe(struct publication *publ, u32 addr) * Invoked for each publication issued by a newly failed node. * Removes publication structure from name table & deletes it. */ -static void tipc_publ_purge(struct publication *publ, u32 addr) +static void tipc_publ_purge(struct net *net, struct publication *publ, u32 addr) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct publication *p; - spin_lock_bh(&tipc_nametbl_lock); - p = tipc_nametbl_remove_publ(publ->type, publ->lower, + spin_lock_bh(&tn->nametbl_lock); + p = tipc_nametbl_remove_publ(net, publ->type, publ->lower, publ->node, publ->ref, publ->key); if (p) - tipc_publ_unsubscribe(p, addr); - spin_unlock_bh(&tipc_nametbl_lock); + tipc_publ_unsubscribe(net, p, addr); + spin_unlock_bh(&tn->nametbl_lock); if (p != publ) { pr_err("Unable to remove publication from failed node\n" @@ -277,12 +290,12 @@ static void tipc_publ_purge(struct publication *publ, u32 addr) kfree_rcu(p, rcu); } -void tipc_publ_notify(struct list_head *nsub_list, u32 addr) +void tipc_publ_notify(struct net *net, struct list_head *nsub_list, u32 addr) { struct publication *publ, *tmp; list_for_each_entry_safe(publ, tmp, nsub_list, nodesub_list) - tipc_publ_purge(publ, addr); + tipc_publ_purge(net, publ, addr); } /** @@ -292,25 +305,28 @@ void tipc_publ_notify(struct list_head *nsub_list, u32 addr) * tipc_nametbl_lock must be held. * Returns the publication item if successful, otherwise NULL. */ -static bool tipc_update_nametbl(struct distr_item *i, u32 node, u32 dtype) +static bool tipc_update_nametbl(struct net *net, struct distr_item *i, + u32 node, u32 dtype) { struct publication *publ = NULL; if (dtype == PUBLICATION) { - publ = tipc_nametbl_insert_publ(ntohl(i->type), ntohl(i->lower), + publ = tipc_nametbl_insert_publ(net, ntohl(i->type), + ntohl(i->lower), ntohl(i->upper), TIPC_CLUSTER_SCOPE, node, ntohl(i->ref), ntohl(i->key)); if (publ) { - tipc_publ_subscribe(publ, node); + tipc_publ_subscribe(net, publ, node); return true; } } else if (dtype == WITHDRAWAL) { - publ = tipc_nametbl_remove_publ(ntohl(i->type), ntohl(i->lower), + publ = tipc_nametbl_remove_publ(net, ntohl(i->type), + ntohl(i->lower), node, ntohl(i->ref), ntohl(i->key)); if (publ) { - tipc_publ_unsubscribe(publ, node); + tipc_publ_unsubscribe(net, publ, node); kfree_rcu(publ, rcu); return true; } @@ -343,7 +359,7 @@ static void tipc_named_add_backlog(struct distr_item *i, u32 type, u32 node) * tipc_named_process_backlog - try to process any pending name table updates * from the network. */ -void tipc_named_process_backlog(void) +void tipc_named_process_backlog(struct net *net) { struct distr_queue_item *e, *tmp; char addr[16]; @@ -351,7 +367,7 @@ void tipc_named_process_backlog(void) list_for_each_entry_safe(e, tmp, &tipc_dist_queue, next) { if (time_after(e->expires, now)) { - if (!tipc_update_nametbl(&e->i, e->node, e->dtype)) + if (!tipc_update_nametbl(net, &e->i, e->node, e->dtype)) continue; } else { tipc_addr_string_fill(addr, e->node); @@ -367,24 +383,34 @@ void tipc_named_process_backlog(void) } /** - * tipc_named_rcv - process name table update message sent by another node + * tipc_named_rcv - process name table update messages sent by another node */ -void tipc_named_rcv(struct sk_buff *buf) +void tipc_named_rcv(struct net *net, struct sk_buff_head *inputq) { - struct tipc_msg *msg = buf_msg(buf); - struct distr_item *item = (struct distr_item *)msg_data(msg); - u32 count = msg_data_sz(msg) / ITEM_SIZE; - u32 node = msg_orignode(msg); - - spin_lock_bh(&tipc_nametbl_lock); - while (count--) { - if (!tipc_update_nametbl(item, node, msg_type(msg))) - tipc_named_add_backlog(item, msg_type(msg), node); - item++; + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_msg *msg; + struct distr_item *item; + uint count; + u32 node; + struct sk_buff *skb; + int mtype; + + spin_lock_bh(&tn->nametbl_lock); + for (skb = skb_dequeue(inputq); skb; skb = skb_dequeue(inputq)) { + msg = buf_msg(skb); + mtype = msg_type(msg); + item = (struct distr_item *)msg_data(msg); + count = msg_data_sz(msg) / ITEM_SIZE; + node = msg_orignode(msg); + while (count--) { + if (!tipc_update_nametbl(net, item, node, mtype)) + tipc_named_add_backlog(item, mtype, node); + item++; + } + kfree_skb(skb); + tipc_named_process_backlog(net); } - tipc_named_process_backlog(); - spin_unlock_bh(&tipc_nametbl_lock); - kfree_skb(buf); + spin_unlock_bh(&tn->nametbl_lock); } /** @@ -394,17 +420,18 @@ void tipc_named_rcv(struct sk_buff *buf) * All name table entries published by this node are updated to reflect * the node's new network address. */ -void tipc_named_reinit(void) +void tipc_named_reinit(struct net *net) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct publication *publ; int scope; - spin_lock_bh(&tipc_nametbl_lock); + spin_lock_bh(&tn->nametbl_lock); for (scope = TIPC_ZONE_SCOPE; scope <= TIPC_NODE_SCOPE; scope++) - list_for_each_entry_rcu(publ, &tipc_nametbl->publ_list[scope], + list_for_each_entry_rcu(publ, &tn->nametbl->publ_list[scope], local_list) - publ->node = tipc_own_addr; + publ->node = tn->own_addr; - spin_unlock_bh(&tipc_nametbl_lock); + spin_unlock_bh(&tn->nametbl_lock); } diff --git a/net/tipc/name_distr.h b/net/tipc/name_distr.h index cef55cedcfb2..dd2d9fd80da2 100644 --- a/net/tipc/name_distr.h +++ b/net/tipc/name_distr.h @@ -67,13 +67,13 @@ struct distr_item { __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 dnode); -void tipc_named_rcv(struct sk_buff *buf); -void tipc_named_reinit(void); -void tipc_named_process_backlog(void); -void tipc_publ_notify(struct list_head *nsub_list, u32 addr); +struct sk_buff *tipc_named_publish(struct net *net, struct publication *publ); +struct sk_buff *tipc_named_withdraw(struct net *net, struct publication *publ); +void named_cluster_distribute(struct net *net, struct sk_buff *buf); +void tipc_named_node_up(struct net *net, u32 dnode); +void tipc_named_rcv(struct net *net, struct sk_buff_head *msg_queue); +void tipc_named_reinit(struct net *net); +void tipc_named_process_backlog(struct net *net); +void tipc_publ_notify(struct net *net, struct list_head *nsub_list, u32 addr); #endif diff --git a/net/tipc/name_table.c b/net/tipc/name_table.c index c8df0223371a..ab0ac62a1287 100644 --- a/net/tipc/name_table.c +++ b/net/tipc/name_table.c @@ -1,7 +1,7 @@ /* * net/tipc/name_table.c: TIPC name table code * - * Copyright (c) 2000-2006, 2014, Ericsson AB + * Copyright (c) 2000-2006, 2014-2015, Ericsson AB * Copyright (c) 2004-2008, 2010-2014, Wind River Systems * All rights reserved. * @@ -34,11 +34,15 @@ * POSSIBILITY OF SUCH DAMAGE. */ +#include <net/sock.h> #include "core.h" -#include "config.h" +#include "netlink.h" #include "name_table.h" #include "name_distr.h" #include "subscr.h" +#include "bcast.h" +#include "addr.h" +#include <net/genetlink.h> #define TIPC_NAMETBL_SIZE 1024 /* must be a power of 2 */ @@ -105,9 +109,6 @@ struct name_seq { struct rcu_head rcu; }; -struct name_table *tipc_nametbl; -DEFINE_SPINLOCK(tipc_nametbl_lock); - static int hash(int x) { return x & (TIPC_NAMETBL_SIZE - 1); @@ -228,9 +229,11 @@ static u32 nameseq_locate_subseq(struct name_seq *nseq, u32 instance) /** * tipc_nameseq_insert_publ */ -static struct publication *tipc_nameseq_insert_publ(struct name_seq *nseq, - u32 type, u32 lower, u32 upper, - u32 scope, u32 node, u32 port, u32 key) +static struct publication *tipc_nameseq_insert_publ(struct net *net, + struct name_seq *nseq, + u32 type, u32 lower, + u32 upper, u32 scope, + u32 node, u32 port, u32 key) { struct tipc_subscription *s; struct tipc_subscription *st; @@ -315,12 +318,12 @@ static struct publication *tipc_nameseq_insert_publ(struct name_seq *nseq, list_add(&publ->zone_list, &info->zone_list); info->zone_list_size++; - if (in_own_cluster(node)) { + if (in_own_cluster(net, node)) { list_add(&publ->cluster_list, &info->cluster_list); info->cluster_list_size++; } - if (in_own_node(node)) { + if (in_own_node(net, node)) { list_add(&publ->node_list, &info->node_list); info->node_list_size++; } @@ -349,8 +352,10 @@ static struct publication *tipc_nameseq_insert_publ(struct name_seq *nseq, * A failed withdraw request simply returns a failure indication and lets the * caller issue any error or warning messages associated with such a problem. */ -static struct publication *tipc_nameseq_remove_publ(struct name_seq *nseq, u32 inst, - u32 node, u32 ref, u32 key) +static struct publication *tipc_nameseq_remove_publ(struct net *net, + struct name_seq *nseq, + u32 inst, u32 node, + u32 ref, u32 key) { struct publication *publ; struct sub_seq *sseq = nameseq_find_subseq(nseq, inst); @@ -378,13 +383,13 @@ found: info->zone_list_size--; /* Remove publication from cluster scope list, if present */ - if (in_own_cluster(node)) { + if (in_own_cluster(net, node)) { list_del(&publ->cluster_list); info->cluster_list_size--; } /* Remove publication from node scope list, if present */ - if (in_own_node(node)) { + if (in_own_node(net, node)) { list_del(&publ->node_list); info->node_list_size--; } @@ -447,12 +452,13 @@ static void tipc_nameseq_subscribe(struct name_seq *nseq, } } -static struct name_seq *nametbl_find_seq(u32 type) +static struct name_seq *nametbl_find_seq(struct net *net, u32 type) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct hlist_head *seq_head; struct name_seq *ns; - seq_head = &tipc_nametbl->seq_hlist[hash(type)]; + seq_head = &tn->nametbl->seq_hlist[hash(type)]; hlist_for_each_entry_rcu(ns, seq_head, ns_list) { if (ns->type == type) return ns; @@ -461,11 +467,13 @@ static struct name_seq *nametbl_find_seq(u32 type) return NULL; }; -struct publication *tipc_nametbl_insert_publ(u32 type, u32 lower, u32 upper, - u32 scope, u32 node, u32 port, u32 key) +struct publication *tipc_nametbl_insert_publ(struct net *net, u32 type, + u32 lower, u32 upper, u32 scope, + u32 node, u32 port, u32 key) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct publication *publ; - struct name_seq *seq = nametbl_find_seq(type); + struct name_seq *seq = nametbl_find_seq(net, type); int index = hash(type); if ((scope < TIPC_ZONE_SCOPE) || (scope > TIPC_NODE_SCOPE) || @@ -476,29 +484,29 @@ struct publication *tipc_nametbl_insert_publ(u32 type, u32 lower, u32 upper, } if (!seq) - seq = tipc_nameseq_create(type, - &tipc_nametbl->seq_hlist[index]); + seq = tipc_nameseq_create(type, &tn->nametbl->seq_hlist[index]); if (!seq) return NULL; spin_lock_bh(&seq->lock); - publ = tipc_nameseq_insert_publ(seq, type, lower, upper, + publ = tipc_nameseq_insert_publ(net, seq, type, lower, upper, scope, node, port, key); spin_unlock_bh(&seq->lock); return publ; } -struct publication *tipc_nametbl_remove_publ(u32 type, u32 lower, - u32 node, u32 ref, u32 key) +struct publication *tipc_nametbl_remove_publ(struct net *net, u32 type, + u32 lower, u32 node, u32 ref, + u32 key) { struct publication *publ; - struct name_seq *seq = nametbl_find_seq(type); + struct name_seq *seq = nametbl_find_seq(net, type); if (!seq) return NULL; spin_lock_bh(&seq->lock); - publ = tipc_nameseq_remove_publ(seq, lower, node, ref, key); + publ = tipc_nameseq_remove_publ(net, seq, lower, node, ref, key); if (!seq->first_free && list_empty(&seq->subscriptions)) { hlist_del_init_rcu(&seq->ns_list); kfree(seq->sseqs); @@ -523,8 +531,10 @@ struct publication *tipc_nametbl_remove_publ(u32 type, u32 lower, * - if name translation is attempted and fails, sets 'destnode' to 0 * and returns 0 */ -u32 tipc_nametbl_translate(u32 type, u32 instance, u32 *destnode) +u32 tipc_nametbl_translate(struct net *net, u32 type, u32 instance, + u32 *destnode) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct sub_seq *sseq; struct name_info *info; struct publication *publ; @@ -532,11 +542,11 @@ u32 tipc_nametbl_translate(u32 type, u32 instance, u32 *destnode) u32 ref = 0; u32 node = 0; - if (!tipc_in_scope(*destnode, tipc_own_addr)) + if (!tipc_in_scope(*destnode, tn->own_addr)) return 0; rcu_read_lock(); - seq = nametbl_find_seq(type); + seq = nametbl_find_seq(net, type); if (unlikely(!seq)) goto not_found; spin_lock_bh(&seq->lock); @@ -569,13 +579,13 @@ u32 tipc_nametbl_translate(u32 type, u32 instance, u32 *destnode) } /* Round-Robin Algorithm */ - else if (*destnode == tipc_own_addr) { + else if (*destnode == tn->own_addr) { if (list_empty(&info->node_list)) goto no_match; publ = list_first_entry(&info->node_list, struct publication, node_list); list_move_tail(&publ->node_list, &info->node_list); - } else if (in_own_cluster_exact(*destnode)) { + } else if (in_own_cluster_exact(net, *destnode)) { if (list_empty(&info->cluster_list)) goto no_match; publ = list_first_entry(&info->cluster_list, struct publication, @@ -609,8 +619,8 @@ not_found: * * Returns non-zero if any off-node ports overlap */ -int tipc_nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, - struct tipc_port_list *dports) +int tipc_nametbl_mc_translate(struct net *net, u32 type, u32 lower, u32 upper, + u32 limit, struct tipc_plist *dports) { struct name_seq *seq; struct sub_seq *sseq; @@ -619,7 +629,7 @@ int tipc_nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, int res = 0; rcu_read_lock(); - seq = nametbl_find_seq(type); + seq = nametbl_find_seq(net, type); if (!seq) goto exit; @@ -635,7 +645,7 @@ int tipc_nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, info = sseq->info; list_for_each_entry(publ, &info->node_list, node_list) { if (publ->scope <= limit) - tipc_port_list_add(dports, publ->ref); + tipc_plist_push(dports, publ->ref); } if (info->cluster_list_size != info->node_list_size) @@ -650,50 +660,55 @@ exit: /* * tipc_nametbl_publish - add name publication to network name tables */ -struct publication *tipc_nametbl_publish(u32 type, u32 lower, u32 upper, - u32 scope, u32 port_ref, u32 key) +struct publication *tipc_nametbl_publish(struct net *net, u32 type, u32 lower, + u32 upper, u32 scope, u32 port_ref, + u32 key) { struct publication *publ; struct sk_buff *buf = NULL; + struct tipc_net *tn = net_generic(net, tipc_net_id); - spin_lock_bh(&tipc_nametbl_lock); - if (tipc_nametbl->local_publ_count >= TIPC_MAX_PUBLICATIONS) { + spin_lock_bh(&tn->nametbl_lock); + if (tn->nametbl->local_publ_count >= TIPC_MAX_PUBLICATIONS) { pr_warn("Publication failed, local publication limit reached (%u)\n", TIPC_MAX_PUBLICATIONS); - spin_unlock_bh(&tipc_nametbl_lock); + spin_unlock_bh(&tn->nametbl_lock); return NULL; } - publ = tipc_nametbl_insert_publ(type, lower, upper, scope, - tipc_own_addr, port_ref, key); + publ = tipc_nametbl_insert_publ(net, type, lower, upper, scope, + tn->own_addr, port_ref, key); if (likely(publ)) { - tipc_nametbl->local_publ_count++; - buf = tipc_named_publish(publ); + tn->nametbl->local_publ_count++; + buf = tipc_named_publish(net, publ); /* Any pending external events? */ - tipc_named_process_backlog(); + tipc_named_process_backlog(net); } - spin_unlock_bh(&tipc_nametbl_lock); + spin_unlock_bh(&tn->nametbl_lock); if (buf) - named_cluster_distribute(buf); + named_cluster_distribute(net, buf); return publ; } /** * tipc_nametbl_withdraw - withdraw name publication from network name tables */ -int tipc_nametbl_withdraw(u32 type, u32 lower, u32 ref, u32 key) +int tipc_nametbl_withdraw(struct net *net, u32 type, u32 lower, u32 ref, + u32 key) { struct publication *publ; struct sk_buff *skb = NULL; + struct tipc_net *tn = net_generic(net, tipc_net_id); - spin_lock_bh(&tipc_nametbl_lock); - publ = tipc_nametbl_remove_publ(type, lower, tipc_own_addr, ref, key); + spin_lock_bh(&tn->nametbl_lock); + publ = tipc_nametbl_remove_publ(net, type, lower, tn->own_addr, + ref, key); if (likely(publ)) { - tipc_nametbl->local_publ_count--; - skb = tipc_named_withdraw(publ); + tn->nametbl->local_publ_count--; + skb = tipc_named_withdraw(net, publ); /* Any pending external events? */ - tipc_named_process_backlog(); + tipc_named_process_backlog(net); list_del_init(&publ->pport_list); kfree_rcu(publ, rcu); } else { @@ -701,10 +716,10 @@ int tipc_nametbl_withdraw(u32 type, u32 lower, u32 ref, u32 key) "(type=%u, lower=%u, ref=%u, key=%u)\n", type, lower, ref, key); } - spin_unlock_bh(&tipc_nametbl_lock); + spin_unlock_bh(&tn->nametbl_lock); if (skb) { - named_cluster_distribute(skb); + named_cluster_distribute(net, skb); return 1; } return 0; @@ -715,15 +730,15 @@ int tipc_nametbl_withdraw(u32 type, u32 lower, u32 ref, u32 key) */ void tipc_nametbl_subscribe(struct tipc_subscription *s) { + struct tipc_net *tn = net_generic(s->net, tipc_net_id); u32 type = s->seq.type; int index = hash(type); struct name_seq *seq; - spin_lock_bh(&tipc_nametbl_lock); - seq = nametbl_find_seq(type); + spin_lock_bh(&tn->nametbl_lock); + seq = nametbl_find_seq(s->net, type); if (!seq) - seq = tipc_nameseq_create(type, - &tipc_nametbl->seq_hlist[index]); + seq = tipc_nameseq_create(type, &tn->nametbl->seq_hlist[index]); if (seq) { spin_lock_bh(&seq->lock); tipc_nameseq_subscribe(seq, s); @@ -732,7 +747,7 @@ void tipc_nametbl_subscribe(struct tipc_subscription *s) pr_warn("Failed to create subscription for {%u,%u,%u}\n", s->seq.type, s->seq.lower, s->seq.upper); } - spin_unlock_bh(&tipc_nametbl_lock); + spin_unlock_bh(&tn->nametbl_lock); } /** @@ -740,10 +755,11 @@ void tipc_nametbl_subscribe(struct tipc_subscription *s) */ void tipc_nametbl_unsubscribe(struct tipc_subscription *s) { + struct tipc_net *tn = net_generic(s->net, tipc_net_id); struct name_seq *seq; - spin_lock_bh(&tipc_nametbl_lock); - seq = nametbl_find_seq(s->seq.type); + spin_lock_bh(&tn->nametbl_lock); + seq = nametbl_find_seq(s->net, s->seq.type); if (seq != NULL) { spin_lock_bh(&seq->lock); list_del_init(&s->nameseq_list); @@ -756,193 +772,13 @@ void tipc_nametbl_unsubscribe(struct tipc_subscription *s) spin_unlock_bh(&seq->lock); } } - spin_unlock_bh(&tipc_nametbl_lock); -} - -/** - * subseq_list - print specified sub-sequence contents into the given buffer - */ -static int subseq_list(struct sub_seq *sseq, char *buf, int len, u32 depth, - u32 index) -{ - char portIdStr[27]; - const char *scope_str[] = {"", " zone", " cluster", " node"}; - struct publication *publ; - struct name_info *info; - int ret; - - ret = tipc_snprintf(buf, len, "%-10u %-10u ", sseq->lower, sseq->upper); - - if (depth == 2) { - ret += tipc_snprintf(buf - ret, len + ret, "\n"); - return ret; - } - - info = sseq->info; - - list_for_each_entry(publ, &info->zone_list, zone_list) { - sprintf(portIdStr, "<%u.%u.%u:%u>", - tipc_zone(publ->node), tipc_cluster(publ->node), - tipc_node(publ->node), publ->ref); - ret += tipc_snprintf(buf + ret, len - ret, "%-26s ", portIdStr); - if (depth > 3) { - ret += tipc_snprintf(buf + ret, len - ret, "%-10u %s", - publ->key, scope_str[publ->scope]); - } - if (!list_is_last(&publ->zone_list, &info->zone_list)) - ret += tipc_snprintf(buf + ret, len - ret, - "\n%33s", " "); - } - - ret += tipc_snprintf(buf + ret, len - ret, "\n"); - return ret; -} - -/** - * nameseq_list - print specified name sequence contents into the given buffer - */ -static int nameseq_list(struct name_seq *seq, char *buf, int len, u32 depth, - u32 type, u32 lowbound, u32 upbound, u32 index) -{ - struct sub_seq *sseq; - char typearea[11]; - int ret = 0; - - if (seq->first_free == 0) - return 0; - - sprintf(typearea, "%-10u", seq->type); - - if (depth == 1) { - ret += tipc_snprintf(buf, len, "%s\n", typearea); - return ret; - } - - for (sseq = seq->sseqs; sseq != &seq->sseqs[seq->first_free]; sseq++) { - if ((lowbound <= sseq->upper) && (upbound >= sseq->lower)) { - ret += tipc_snprintf(buf + ret, len - ret, "%s ", - typearea); - spin_lock_bh(&seq->lock); - ret += subseq_list(sseq, buf + ret, len - ret, - depth, index); - spin_unlock_bh(&seq->lock); - sprintf(typearea, "%10s", " "); - } - } - return ret; -} - -/** - * nametbl_header - print name table header into the given buffer - */ -static int nametbl_header(char *buf, int len, u32 depth) -{ - const char *header[] = { - "Type ", - "Lower Upper ", - "Port Identity ", - "Publication Scope" - }; - - int i; - int ret = 0; - - if (depth > 4) - depth = 4; - for (i = 0; i < depth; i++) - ret += tipc_snprintf(buf + ret, len - ret, header[i]); - ret += tipc_snprintf(buf + ret, len - ret, "\n"); - return ret; -} - -/** - * nametbl_list - print specified name table contents into the given buffer - */ -static int nametbl_list(char *buf, int len, u32 depth_info, - u32 type, u32 lowbound, u32 upbound) -{ - struct hlist_head *seq_head; - struct name_seq *seq; - int all_types; - int ret = 0; - u32 depth; - u32 i; - - all_types = (depth_info & TIPC_NTQ_ALLTYPES); - depth = (depth_info & ~TIPC_NTQ_ALLTYPES); - - if (depth == 0) - return 0; - - if (all_types) { - /* display all entries in name table to specified depth */ - ret += nametbl_header(buf, len, depth); - lowbound = 0; - upbound = ~0; - for (i = 0; i < TIPC_NAMETBL_SIZE; i++) { - seq_head = &tipc_nametbl->seq_hlist[i]; - hlist_for_each_entry_rcu(seq, seq_head, ns_list) { - ret += nameseq_list(seq, buf + ret, len - ret, - depth, seq->type, - lowbound, upbound, i); - } - } - } else { - /* display only the sequence that matches the specified type */ - if (upbound < lowbound) { - ret += tipc_snprintf(buf + ret, len - ret, - "invalid name sequence specified\n"); - return ret; - } - ret += nametbl_header(buf + ret, len - ret, depth); - i = hash(type); - seq_head = &tipc_nametbl->seq_hlist[i]; - hlist_for_each_entry_rcu(seq, seq_head, ns_list) { - if (seq->type == type) { - ret += nameseq_list(seq, buf + ret, len - ret, - depth, type, - lowbound, upbound, i); - break; - } - } - } - return ret; + spin_unlock_bh(&tn->nametbl_lock); } -struct sk_buff *tipc_nametbl_get(const void *req_tlv_area, int req_tlv_space) -{ - struct sk_buff *buf; - struct tipc_name_table_query *argv; - struct tlv_desc *rep_tlv; - char *pb; - int pb_len; - int str_len; - - if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NAME_TBL_QUERY)) - return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - - buf = tipc_cfg_reply_alloc(TLV_SPACE(ULTRA_STRING_MAX_LEN)); - if (!buf) - return NULL; - - rep_tlv = (struct tlv_desc *)buf->data; - pb = TLV_DATA(rep_tlv); - pb_len = ULTRA_STRING_MAX_LEN; - argv = (struct tipc_name_table_query *)TLV_DATA(req_tlv_area); - rcu_read_lock(); - str_len = nametbl_list(pb, pb_len, ntohl(argv->depth), - ntohl(argv->type), - ntohl(argv->lowbound), ntohl(argv->upbound)); - rcu_read_unlock(); - str_len += 1; /* for "\0" */ - skb_put(buf, TLV_SPACE(str_len)); - TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); - - return buf; -} - -int tipc_nametbl_init(void) +int tipc_nametbl_init(struct net *net) { + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct name_table *tipc_nametbl; int i; tipc_nametbl = kzalloc(sizeof(*tipc_nametbl), GFP_ATOMIC); @@ -955,6 +791,8 @@ int tipc_nametbl_init(void) INIT_LIST_HEAD(&tipc_nametbl->publ_list[TIPC_ZONE_SCOPE]); INIT_LIST_HEAD(&tipc_nametbl->publ_list[TIPC_CLUSTER_SCOPE]); INIT_LIST_HEAD(&tipc_nametbl->publ_list[TIPC_NODE_SCOPE]); + tn->nametbl = tipc_nametbl; + spin_lock_init(&tn->nametbl_lock); return 0; } @@ -963,7 +801,7 @@ int tipc_nametbl_init(void) * * tipc_nametbl_lock must be held when calling this function */ -static void tipc_purge_publications(struct name_seq *seq) +static void tipc_purge_publications(struct net *net, struct name_seq *seq) { struct publication *publ, *safe; struct sub_seq *sseq; @@ -973,7 +811,7 @@ static void tipc_purge_publications(struct name_seq *seq) sseq = seq->sseqs; info = sseq->info; list_for_each_entry_safe(publ, safe, &info->zone_list, zone_list) { - tipc_nametbl_remove_publ(publ->type, publ->lower, publ->node, + tipc_nameseq_remove_publ(net, seq, publ->lower, publ->node, publ->ref, publ->key); kfree_rcu(publ, rcu); } @@ -984,25 +822,27 @@ static void tipc_purge_publications(struct name_seq *seq) kfree_rcu(seq, rcu); } -void tipc_nametbl_stop(void) +void tipc_nametbl_stop(struct net *net) { u32 i; struct name_seq *seq; struct hlist_head *seq_head; + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct name_table *tipc_nametbl = tn->nametbl; /* Verify name table is empty and purge any lingering * publications, then release the name table */ - spin_lock_bh(&tipc_nametbl_lock); + spin_lock_bh(&tn->nametbl_lock); for (i = 0; i < TIPC_NAMETBL_SIZE; i++) { if (hlist_empty(&tipc_nametbl->seq_hlist[i])) continue; seq_head = &tipc_nametbl->seq_hlist[i]; hlist_for_each_entry_rcu(seq, seq_head, ns_list) { - tipc_purge_publications(seq); + tipc_purge_publications(net, seq); } } - spin_unlock_bh(&tipc_nametbl_lock); + spin_unlock_bh(&tn->nametbl_lock); synchronize_net(); kfree(tipc_nametbl); @@ -1033,7 +873,7 @@ static int __tipc_nl_add_nametable_publ(struct tipc_nl_msg *msg, *last_publ = p->key; hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, - &tipc_genl_v2_family, NLM_F_MULTI, + &tipc_genl_family, NLM_F_MULTI, TIPC_NL_NAME_TABLE_GET); if (!hdr) return -EMSGSIZE; @@ -1106,9 +946,10 @@ static int __tipc_nl_subseq_list(struct tipc_nl_msg *msg, struct name_seq *seq, return 0; } -static int __tipc_nl_seq_list(struct tipc_nl_msg *msg, u32 *last_type, - u32 *last_lower, u32 *last_publ) +static int tipc_nl_seq_list(struct net *net, struct tipc_nl_msg *msg, + u32 *last_type, u32 *last_lower, u32 *last_publ) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct hlist_head *seq_head; struct name_seq *seq = NULL; int err; @@ -1120,10 +961,10 @@ static int __tipc_nl_seq_list(struct tipc_nl_msg *msg, u32 *last_type, i = 0; for (; i < TIPC_NAMETBL_SIZE; i++) { - seq_head = &tipc_nametbl->seq_hlist[i]; + seq_head = &tn->nametbl->seq_hlist[i]; if (*last_type) { - seq = nametbl_find_seq(*last_type); + seq = nametbl_find_seq(net, *last_type); if (!seq) return -EPIPE; } else { @@ -1157,6 +998,7 @@ int tipc_nl_name_table_dump(struct sk_buff *skb, struct netlink_callback *cb) u32 last_type = cb->args[0]; u32 last_lower = cb->args[1]; u32 last_publ = cb->args[2]; + struct net *net = sock_net(skb->sk); struct tipc_nl_msg msg; if (done) @@ -1167,7 +1009,7 @@ int tipc_nl_name_table_dump(struct sk_buff *skb, struct netlink_callback *cb) msg.seq = cb->nlh->nlmsg_seq; rcu_read_lock(); - err = __tipc_nl_seq_list(&msg, &last_type, &last_lower, &last_publ); + err = tipc_nl_seq_list(net, &msg, &last_type, &last_lower, &last_publ); if (!err) { done = 1; } else if (err != -EMSGSIZE) { @@ -1188,3 +1030,41 @@ int tipc_nl_name_table_dump(struct sk_buff *skb, struct netlink_callback *cb) return skb->len; } + +void tipc_plist_push(struct tipc_plist *pl, u32 port) +{ + struct tipc_plist *nl; + + if (likely(!pl->port)) { + pl->port = port; + return; + } + if (pl->port == port) + return; + list_for_each_entry(nl, &pl->list, list) { + if (nl->port == port) + return; + } + nl = kmalloc(sizeof(*nl), GFP_ATOMIC); + if (nl) { + nl->port = port; + list_add(&nl->list, &pl->list); + } +} + +u32 tipc_plist_pop(struct tipc_plist *pl) +{ + struct tipc_plist *nl; + u32 port = 0; + + if (likely(list_empty(&pl->list))) { + port = pl->port; + pl->port = 0; + return port; + } + nl = list_first_entry(&pl->list, typeof(*nl), list); + port = nl->port; + list_del(&nl->list); + kfree(nl); + return port; +} diff --git a/net/tipc/name_table.h b/net/tipc/name_table.h index 5f0dee92010d..1524a73830f7 100644 --- a/net/tipc/name_table.h +++ b/net/tipc/name_table.h @@ -1,7 +1,7 @@ /* * net/tipc/name_table.h: Include file for TIPC name table code * - * Copyright (c) 2000-2006, 2014, Ericsson AB + * Copyright (c) 2000-2006, 2014-2015, Ericsson AB * Copyright (c) 2004-2005, 2010-2011, Wind River Systems * All rights reserved. * @@ -38,7 +38,7 @@ #define _TIPC_NAME_TABLE_H struct tipc_subscription; -struct tipc_port_list; +struct tipc_plist; /* * TIPC name types reserved for internal TIPC use (both current and planned) @@ -95,26 +95,39 @@ struct name_table { u32 local_publ_count; }; -extern spinlock_t tipc_nametbl_lock; -extern struct name_table *tipc_nametbl; - int tipc_nl_name_table_dump(struct sk_buff *skb, struct netlink_callback *cb); -struct sk_buff *tipc_nametbl_get(const void *req_tlv_area, int req_tlv_space); -u32 tipc_nametbl_translate(u32 type, u32 instance, u32 *node); -int tipc_nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, - struct tipc_port_list *dports); -struct publication *tipc_nametbl_publish(u32 type, u32 lower, u32 upper, - u32 scope, u32 port_ref, u32 key); -int tipc_nametbl_withdraw(u32 type, u32 lower, u32 ref, u32 key); -struct publication *tipc_nametbl_insert_publ(u32 type, u32 lower, u32 upper, - u32 scope, u32 node, u32 ref, +u32 tipc_nametbl_translate(struct net *net, u32 type, u32 instance, u32 *node); +int tipc_nametbl_mc_translate(struct net *net, u32 type, u32 lower, u32 upper, + u32 limit, struct tipc_plist *dports); +struct publication *tipc_nametbl_publish(struct net *net, u32 type, u32 lower, + u32 upper, u32 scope, u32 port_ref, + u32 key); +int tipc_nametbl_withdraw(struct net *net, u32 type, u32 lower, u32 ref, + u32 key); +struct publication *tipc_nametbl_insert_publ(struct net *net, u32 type, + u32 lower, u32 upper, u32 scope, + u32 node, u32 ref, u32 key); +struct publication *tipc_nametbl_remove_publ(struct net *net, u32 type, + u32 lower, u32 node, u32 ref, u32 key); -struct publication *tipc_nametbl_remove_publ(u32 type, u32 lower, u32 node, - u32 ref, u32 key); void tipc_nametbl_subscribe(struct tipc_subscription *s); void tipc_nametbl_unsubscribe(struct tipc_subscription *s); -int tipc_nametbl_init(void); -void tipc_nametbl_stop(void); +int tipc_nametbl_init(struct net *net); +void tipc_nametbl_stop(struct net *net); + +struct tipc_plist { + struct list_head list; + u32 port; +}; + +static inline void tipc_plist_init(struct tipc_plist *pl) +{ + INIT_LIST_HEAD(&pl->list); + pl->port = 0; +} + +void tipc_plist_push(struct tipc_plist *pl, u32 port); +u32 tipc_plist_pop(struct tipc_plist *pl); #endif diff --git a/net/tipc/net.c b/net/tipc/net.c index cf13df3cde8f..a54f3cbe2246 100644 --- a/net/tipc/net.c +++ b/net/tipc/net.c @@ -40,7 +40,6 @@ #include "subscr.h" #include "socket.h" #include "node.h" -#include "config.h" static const struct nla_policy tipc_nl_net_policy[TIPC_NLA_NET_MAX + 1] = { [TIPC_NLA_NET_UNSPEC] = { .type = NLA_UNSPEC }, @@ -108,48 +107,54 @@ static const struct nla_policy tipc_nl_net_policy[TIPC_NLA_NET_MAX + 1] = { * - A local spin_lock protecting the queue of subscriber events. */ -int tipc_net_start(u32 addr) +int tipc_net_start(struct net *net, u32 addr) { + struct tipc_net *tn = net_generic(net, tipc_net_id); char addr_string[16]; int res; - tipc_own_addr = addr; - tipc_named_reinit(); - tipc_sk_reinit(); - res = tipc_bclink_init(); + tn->own_addr = addr; + tipc_named_reinit(net); + tipc_sk_reinit(net); + res = tipc_bclink_init(net); if (res) return res; - tipc_nametbl_publish(TIPC_CFG_SRV, tipc_own_addr, tipc_own_addr, - TIPC_ZONE_SCOPE, 0, tipc_own_addr); + tipc_nametbl_publish(net, TIPC_CFG_SRV, tn->own_addr, tn->own_addr, + TIPC_ZONE_SCOPE, 0, tn->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); + tipc_addr_string_fill(addr_string, tn->own_addr), + tn->net_id); return 0; } -void tipc_net_stop(void) +void tipc_net_stop(struct net *net) { - if (!tipc_own_addr) + struct tipc_net *tn = net_generic(net, tipc_net_id); + + if (!tn->own_addr) return; - tipc_nametbl_withdraw(TIPC_CFG_SRV, tipc_own_addr, 0, tipc_own_addr); + tipc_nametbl_withdraw(net, TIPC_CFG_SRV, tn->own_addr, 0, + tn->own_addr); rtnl_lock(); - tipc_bearer_stop(); - tipc_bclink_stop(); - tipc_node_stop(); + tipc_bearer_stop(net); + tipc_bclink_stop(net); + tipc_node_stop(net); rtnl_unlock(); pr_info("Left network mode\n"); } -static int __tipc_nl_add_net(struct tipc_nl_msg *msg) +static int __tipc_nl_add_net(struct net *net, struct tipc_nl_msg *msg) { + struct tipc_net *tn = net_generic(net, tipc_net_id); void *hdr; struct nlattr *attrs; - hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_v2_family, + hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, NLM_F_MULTI, TIPC_NL_NET_GET); if (!hdr) return -EMSGSIZE; @@ -158,7 +163,7 @@ static int __tipc_nl_add_net(struct tipc_nl_msg *msg) if (!attrs) goto msg_full; - if (nla_put_u32(msg->skb, TIPC_NLA_NET_ID, tipc_net_id)) + if (nla_put_u32(msg->skb, TIPC_NLA_NET_ID, tn->net_id)) goto attr_msg_full; nla_nest_end(msg->skb, attrs); @@ -176,6 +181,7 @@ msg_full: int tipc_nl_net_dump(struct sk_buff *skb, struct netlink_callback *cb) { + struct net *net = sock_net(skb->sk); int err; int done = cb->args[0]; struct tipc_nl_msg msg; @@ -187,7 +193,7 @@ int tipc_nl_net_dump(struct sk_buff *skb, struct netlink_callback *cb) msg.portid = NETLINK_CB(cb->skb).portid; msg.seq = cb->nlh->nlmsg_seq; - err = __tipc_nl_add_net(&msg); + err = __tipc_nl_add_net(net, &msg); if (err) goto out; @@ -200,8 +206,10 @@ out: int tipc_nl_net_set(struct sk_buff *skb, struct genl_info *info) { - int err; + struct net *net = sock_net(skb->sk); + struct tipc_net *tn = net_generic(net, tipc_net_id); struct nlattr *attrs[TIPC_NLA_NET_MAX + 1]; + int err; if (!info->attrs[TIPC_NLA_NET]) return -EINVAL; @@ -216,21 +224,21 @@ int tipc_nl_net_set(struct sk_buff *skb, struct genl_info *info) u32 val; /* Can't change net id once TIPC has joined a network */ - if (tipc_own_addr) + if (tn->own_addr) return -EPERM; val = nla_get_u32(attrs[TIPC_NLA_NET_ID]); if (val < 1 || val > 9999) return -EINVAL; - tipc_net_id = val; + tn->net_id = val; } if (attrs[TIPC_NLA_NET_ADDR]) { u32 addr; /* Can't change net addr once TIPC has joined a network */ - if (tipc_own_addr) + if (tn->own_addr) return -EPERM; addr = nla_get_u32(attrs[TIPC_NLA_NET_ADDR]); @@ -238,7 +246,7 @@ int tipc_nl_net_set(struct sk_buff *skb, struct genl_info *info) return -EINVAL; rtnl_lock(); - tipc_net_start(addr); + tipc_net_start(net, addr); rtnl_unlock(); } diff --git a/net/tipc/net.h b/net/tipc/net.h index a81c1b9eb150..77a7a118911d 100644 --- a/net/tipc/net.h +++ b/net/tipc/net.h @@ -39,9 +39,9 @@ #include <net/genetlink.h> -int tipc_net_start(u32 addr); +int tipc_net_start(struct net *net, u32 addr); -void tipc_net_stop(void); +void tipc_net_stop(struct net *net); int tipc_nl_net_dump(struct sk_buff *skb, struct netlink_callback *cb); int tipc_nl_net_set(struct sk_buff *skb, struct genl_info *info); diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c index b891e3905bc4..7f6475efc984 100644 --- a/net/tipc/netlink.c +++ b/net/tipc/netlink.c @@ -35,7 +35,6 @@ */ #include "core.h" -#include "config.h" #include "socket.h" #include "name_table.h" #include "bearer.h" @@ -44,36 +43,6 @@ #include "net.h" #include <net/genetlink.h> -static int handle_cmd(struct sk_buff *skb, struct genl_info *info) -{ - struct sk_buff *rep_buf; - struct nlmsghdr *rep_nlh; - struct nlmsghdr *req_nlh = info->nlhdr; - struct tipc_genlmsghdr *req_userhdr = info->userhdr; - int hdr_space = nlmsg_total_size(GENL_HDRLEN + TIPC_GENL_HDRLEN); - u16 cmd; - - if ((req_userhdr->cmd & 0xC000) && (!netlink_capable(skb, CAP_NET_ADMIN))) - cmd = TIPC_CMD_NOT_NET_ADMIN; - else - cmd = req_userhdr->cmd; - - rep_buf = tipc_cfg_do_cmd(req_userhdr->dest, cmd, - nlmsg_data(req_nlh) + GENL_HDRLEN + TIPC_GENL_HDRLEN, - nlmsg_attrlen(req_nlh, GENL_HDRLEN + TIPC_GENL_HDRLEN), - hdr_space); - - if (rep_buf) { - skb_push(rep_buf, hdr_space); - rep_nlh = nlmsg_hdr(rep_buf); - memcpy(rep_nlh, req_nlh, hdr_space); - rep_nlh->nlmsg_len = rep_buf->len; - genlmsg_unicast(&init_net, rep_buf, NETLINK_CB(skb).portid); - } - - return 0; -} - static const struct nla_policy tipc_nl_policy[TIPC_NLA_MAX + 1] = { [TIPC_NLA_UNSPEC] = { .type = NLA_UNSPEC, }, [TIPC_NLA_BEARER] = { .type = NLA_NESTED, }, @@ -86,32 +55,16 @@ static const struct nla_policy tipc_nl_policy[TIPC_NLA_MAX + 1] = { [TIPC_NLA_NAME_TABLE] = { .type = NLA_NESTED, } }; -/* Legacy ASCII API */ -static struct genl_family tipc_genl_family = { - .id = GENL_ID_GENERATE, - .name = TIPC_GENL_NAME, - .version = TIPC_GENL_VERSION, - .hdrsize = TIPC_GENL_HDRLEN, - .maxattr = 0, -}; - -/* Legacy ASCII API */ -static struct genl_ops tipc_genl_ops[] = { - { - .cmd = TIPC_GENL_CMD, - .doit = handle_cmd, - }, -}; - /* Users of the legacy API (tipc-config) can't handle that we add operations, * so we have a separate genl handling for the new API. */ -struct genl_family tipc_genl_v2_family = { +struct genl_family tipc_genl_family = { .id = GENL_ID_GENERATE, .name = TIPC_GENL_V2_NAME, .version = TIPC_GENL_V2_VERSION, .hdrsize = 0, .maxattr = TIPC_NLA_MAX, + .netnsok = true, }; static const struct genl_ops tipc_genl_v2_ops[] = { @@ -197,9 +150,9 @@ static const struct genl_ops tipc_genl_v2_ops[] = { int tipc_nlmsg_parse(const struct nlmsghdr *nlh, struct nlattr ***attr) { - u32 maxattr = tipc_genl_v2_family.maxattr; + u32 maxattr = tipc_genl_family.maxattr; - *attr = tipc_genl_v2_family.attrbuf; + *attr = tipc_genl_family.attrbuf; if (!*attr) return -EOPNOTSUPP; @@ -210,13 +163,7 @@ int tipc_netlink_start(void) { int res; - res = genl_register_family_with_ops(&tipc_genl_family, tipc_genl_ops); - if (res) { - pr_err("Failed to register legacy interface\n"); - return res; - } - - res = genl_register_family_with_ops(&tipc_genl_v2_family, + res = genl_register_family_with_ops(&tipc_genl_family, tipc_genl_v2_ops); if (res) { pr_err("Failed to register netlink interface\n"); @@ -228,5 +175,4 @@ int tipc_netlink_start(void) void tipc_netlink_stop(void) { genl_unregister_family(&tipc_genl_family); - genl_unregister_family(&tipc_genl_v2_family); } diff --git a/net/tipc/netlink.h b/net/tipc/netlink.h index 1425c6869de0..08a1db67b927 100644 --- a/net/tipc/netlink.h +++ b/net/tipc/netlink.h @@ -36,7 +36,7 @@ #ifndef _TIPC_NETLINK_H #define _TIPC_NETLINK_H -extern struct genl_family tipc_genl_v2_family; +extern struct genl_family tipc_genl_family; int tipc_nlmsg_parse(const struct nlmsghdr *nlh, struct nlattr ***buf); struct tipc_nl_msg { @@ -45,4 +45,9 @@ struct tipc_nl_msg { u32 seq; }; +int tipc_netlink_start(void); +int tipc_netlink_compat_start(void); +void tipc_netlink_stop(void); +void tipc_netlink_compat_stop(void); + #endif diff --git a/net/tipc/netlink_compat.c b/net/tipc/netlink_compat.c new file mode 100644 index 000000000000..ce9121e8e990 --- /dev/null +++ b/net/tipc/netlink_compat.c @@ -0,0 +1,1084 @@ +/* + * Copyright (c) 2014, Ericsson AB + * 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" +#include "bearer.h" +#include "link.h" +#include "name_table.h" +#include "socket.h" +#include "node.h" +#include "net.h" +#include <net/genetlink.h> +#include <linux/tipc_config.h> + +/* The legacy API had an artificial message length limit called + * ULTRA_STRING_MAX_LEN. + */ +#define ULTRA_STRING_MAX_LEN 32768 + +#define TIPC_SKB_MAX TLV_SPACE(ULTRA_STRING_MAX_LEN) + +#define REPLY_TRUNCATED "<truncated>\n" + +struct tipc_nl_compat_msg { + u16 cmd; + int rep_type; + int rep_size; + int req_type; + struct sk_buff *rep; + struct tlv_desc *req; + struct sock *dst_sk; +}; + +struct tipc_nl_compat_cmd_dump { + int (*header)(struct tipc_nl_compat_msg *); + int (*dumpit)(struct sk_buff *, struct netlink_callback *); + int (*format)(struct tipc_nl_compat_msg *msg, struct nlattr **attrs); +}; + +struct tipc_nl_compat_cmd_doit { + int (*doit)(struct sk_buff *skb, struct genl_info *info); + int (*transcode)(struct sk_buff *skb, struct tipc_nl_compat_msg *msg); +}; + +static int tipc_skb_tailroom(struct sk_buff *skb) +{ + int tailroom; + int limit; + + tailroom = skb_tailroom(skb); + limit = TIPC_SKB_MAX - skb->len; + + if (tailroom < limit) + return tailroom; + + return limit; +} + +static int tipc_add_tlv(struct sk_buff *skb, u16 type, void *data, u16 len) +{ + struct tlv_desc *tlv = (struct tlv_desc *)skb_tail_pointer(skb); + + if (tipc_skb_tailroom(skb) < TLV_SPACE(len)) + return -EMSGSIZE; + + skb_put(skb, TLV_SPACE(len)); + tlv->tlv_type = htons(type); + tlv->tlv_len = htons(TLV_LENGTH(len)); + if (len && data) + memcpy(TLV_DATA(tlv), data, len); + + return 0; +} + +static void tipc_tlv_init(struct sk_buff *skb, u16 type) +{ + struct tlv_desc *tlv = (struct tlv_desc *)skb->data; + + TLV_SET_LEN(tlv, 0); + TLV_SET_TYPE(tlv, type); + skb_put(skb, sizeof(struct tlv_desc)); +} + +static int tipc_tlv_sprintf(struct sk_buff *skb, const char *fmt, ...) +{ + int n; + u16 len; + u32 rem; + char *buf; + struct tlv_desc *tlv; + va_list args; + + rem = tipc_skb_tailroom(skb); + + tlv = (struct tlv_desc *)skb->data; + len = TLV_GET_LEN(tlv); + buf = TLV_DATA(tlv) + len; + + va_start(args, fmt); + n = vscnprintf(buf, rem, fmt, args); + va_end(args); + + TLV_SET_LEN(tlv, n + len); + skb_put(skb, n); + + return n; +} + +static struct sk_buff *tipc_tlv_alloc(int size) +{ + int hdr_len; + struct sk_buff *buf; + + size = TLV_SPACE(size); + hdr_len = nlmsg_total_size(GENL_HDRLEN + TIPC_GENL_HDRLEN); + + buf = alloc_skb(hdr_len + size, GFP_KERNEL); + if (!buf) + return NULL; + + skb_reserve(buf, hdr_len); + + return buf; +} + +static struct sk_buff *tipc_get_err_tlv(char *str) +{ + int str_len = strlen(str) + 1; + struct sk_buff *buf; + + buf = tipc_tlv_alloc(TLV_SPACE(str_len)); + if (buf) + tipc_add_tlv(buf, TIPC_TLV_ERROR_STRING, str, str_len); + + return buf; +} + +static int __tipc_nl_compat_dumpit(struct tipc_nl_compat_cmd_dump *cmd, + struct tipc_nl_compat_msg *msg, + struct sk_buff *arg) +{ + int len = 0; + int err; + struct sk_buff *buf; + struct nlmsghdr *nlmsg; + struct netlink_callback cb; + + memset(&cb, 0, sizeof(cb)); + cb.nlh = (struct nlmsghdr *)arg->data; + cb.skb = arg; + + buf = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + if (!buf) + return -ENOMEM; + + buf->sk = msg->dst_sk; + + do { + int rem; + + len = (*cmd->dumpit)(buf, &cb); + + nlmsg_for_each_msg(nlmsg, nlmsg_hdr(buf), len, rem) { + struct nlattr **attrs; + + err = tipc_nlmsg_parse(nlmsg, &attrs); + if (err) + goto err_out; + + err = (*cmd->format)(msg, attrs); + if (err) + goto err_out; + + if (tipc_skb_tailroom(msg->rep) <= 1) { + err = -EMSGSIZE; + goto err_out; + } + } + + skb_reset_tail_pointer(buf); + buf->len = 0; + + } while (len); + + err = 0; + +err_out: + kfree_skb(buf); + + if (err == -EMSGSIZE) { + /* The legacy API only considered messages filling + * "ULTRA_STRING_MAX_LEN" to be truncated. + */ + if ((TIPC_SKB_MAX - msg->rep->len) <= 1) { + char *tail = skb_tail_pointer(msg->rep); + + if (*tail != '\0') + sprintf(tail - sizeof(REPLY_TRUNCATED) - 1, + REPLY_TRUNCATED); + } + + return 0; + } + + return err; +} + +static int tipc_nl_compat_dumpit(struct tipc_nl_compat_cmd_dump *cmd, + struct tipc_nl_compat_msg *msg) +{ + int err; + struct sk_buff *arg; + + if (msg->req_type && !TLV_CHECK_TYPE(msg->req, msg->req_type)) + return -EINVAL; + + msg->rep = tipc_tlv_alloc(msg->rep_size); + if (!msg->rep) + return -ENOMEM; + + if (msg->rep_type) + tipc_tlv_init(msg->rep, msg->rep_type); + + if (cmd->header) + (*cmd->header)(msg); + + arg = nlmsg_new(0, GFP_KERNEL); + if (!arg) { + kfree_skb(msg->rep); + return -ENOMEM; + } + + err = __tipc_nl_compat_dumpit(cmd, msg, arg); + if (err) + kfree_skb(msg->rep); + + kfree_skb(arg); + + return err; +} + +static int __tipc_nl_compat_doit(struct tipc_nl_compat_cmd_doit *cmd, + struct tipc_nl_compat_msg *msg) +{ + int err; + struct sk_buff *doit_buf; + struct sk_buff *trans_buf; + struct nlattr **attrbuf; + struct genl_info info; + + trans_buf = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!trans_buf) + return -ENOMEM; + + err = (*cmd->transcode)(trans_buf, msg); + if (err) + goto trans_out; + + attrbuf = kmalloc((tipc_genl_family.maxattr + 1) * + sizeof(struct nlattr *), GFP_KERNEL); + if (!attrbuf) { + err = -ENOMEM; + goto trans_out; + } + + err = nla_parse(attrbuf, tipc_genl_family.maxattr, + (const struct nlattr *)trans_buf->data, + trans_buf->len, NULL); + if (err) + goto parse_out; + + doit_buf = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); + if (!doit_buf) { + err = -ENOMEM; + goto parse_out; + } + + doit_buf->sk = msg->dst_sk; + + memset(&info, 0, sizeof(info)); + info.attrs = attrbuf; + + err = (*cmd->doit)(doit_buf, &info); + + kfree_skb(doit_buf); +parse_out: + kfree(attrbuf); +trans_out: + kfree_skb(trans_buf); + + return err; +} + +static int tipc_nl_compat_doit(struct tipc_nl_compat_cmd_doit *cmd, + struct tipc_nl_compat_msg *msg) +{ + int err; + + if (msg->req_type && !TLV_CHECK_TYPE(msg->req, msg->req_type)) + return -EINVAL; + + err = __tipc_nl_compat_doit(cmd, msg); + if (err) + return err; + + /* The legacy API considered an empty message a success message */ + msg->rep = tipc_tlv_alloc(0); + if (!msg->rep) + return -ENOMEM; + + return 0; +} + +static int tipc_nl_compat_bearer_dump(struct tipc_nl_compat_msg *msg, + struct nlattr **attrs) +{ + struct nlattr *bearer[TIPC_NLA_BEARER_MAX + 1]; + + nla_parse_nested(bearer, TIPC_NLA_BEARER_MAX, attrs[TIPC_NLA_BEARER], + NULL); + + return tipc_add_tlv(msg->rep, TIPC_TLV_BEARER_NAME, + nla_data(bearer[TIPC_NLA_BEARER_NAME]), + nla_len(bearer[TIPC_NLA_BEARER_NAME])); +} + +static int tipc_nl_compat_bearer_enable(struct sk_buff *skb, + struct tipc_nl_compat_msg *msg) +{ + struct nlattr *prop; + struct nlattr *bearer; + struct tipc_bearer_config *b; + + b = (struct tipc_bearer_config *)TLV_DATA(msg->req); + + bearer = nla_nest_start(skb, TIPC_NLA_BEARER); + if (!bearer) + return -EMSGSIZE; + + if (nla_put_string(skb, TIPC_NLA_BEARER_NAME, b->name)) + return -EMSGSIZE; + + if (nla_put_u32(skb, TIPC_NLA_BEARER_DOMAIN, ntohl(b->disc_domain))) + return -EMSGSIZE; + + if (ntohl(b->priority) <= TIPC_MAX_LINK_PRI) { + prop = nla_nest_start(skb, TIPC_NLA_BEARER_PROP); + if (!prop) + return -EMSGSIZE; + if (nla_put_u32(skb, TIPC_NLA_PROP_PRIO, ntohl(b->priority))) + return -EMSGSIZE; + nla_nest_end(skb, prop); + } + nla_nest_end(skb, bearer); + + return 0; +} + +static int tipc_nl_compat_bearer_disable(struct sk_buff *skb, + struct tipc_nl_compat_msg *msg) +{ + char *name; + struct nlattr *bearer; + + name = (char *)TLV_DATA(msg->req); + + bearer = nla_nest_start(skb, TIPC_NLA_BEARER); + if (!bearer) + return -EMSGSIZE; + + if (nla_put_string(skb, TIPC_NLA_BEARER_NAME, name)) + return -EMSGSIZE; + + nla_nest_end(skb, bearer); + + return 0; +} + +static inline u32 perc(u32 count, u32 total) +{ + return (count * 100 + (total / 2)) / total; +} + +static void __fill_bc_link_stat(struct tipc_nl_compat_msg *msg, + struct nlattr *prop[], struct nlattr *stats[]) +{ + tipc_tlv_sprintf(msg->rep, " Window:%u packets\n", + nla_get_u32(prop[TIPC_NLA_PROP_WIN])); + + tipc_tlv_sprintf(msg->rep, + " RX packets:%u fragments:%u/%u bundles:%u/%u\n", + nla_get_u32(stats[TIPC_NLA_STATS_RX_INFO]), + nla_get_u32(stats[TIPC_NLA_STATS_RX_FRAGMENTS]), + nla_get_u32(stats[TIPC_NLA_STATS_RX_FRAGMENTED]), + nla_get_u32(stats[TIPC_NLA_STATS_RX_BUNDLES]), + nla_get_u32(stats[TIPC_NLA_STATS_RX_BUNDLED])); + + tipc_tlv_sprintf(msg->rep, + " TX packets:%u fragments:%u/%u bundles:%u/%u\n", + nla_get_u32(stats[TIPC_NLA_STATS_TX_INFO]), + nla_get_u32(stats[TIPC_NLA_STATS_TX_FRAGMENTS]), + nla_get_u32(stats[TIPC_NLA_STATS_TX_FRAGMENTED]), + nla_get_u32(stats[TIPC_NLA_STATS_TX_BUNDLES]), + nla_get_u32(stats[TIPC_NLA_STATS_TX_BUNDLED])); + + tipc_tlv_sprintf(msg->rep, " RX naks:%u defs:%u dups:%u\n", + nla_get_u32(stats[TIPC_NLA_STATS_RX_NACKS]), + nla_get_u32(stats[TIPC_NLA_STATS_RX_DEFERRED]), + nla_get_u32(stats[TIPC_NLA_STATS_DUPLICATES])); + + tipc_tlv_sprintf(msg->rep, " TX naks:%u acks:%u dups:%u\n", + nla_get_u32(stats[TIPC_NLA_STATS_TX_NACKS]), + nla_get_u32(stats[TIPC_NLA_STATS_TX_ACKS]), + nla_get_u32(stats[TIPC_NLA_STATS_RETRANSMITTED])); + + tipc_tlv_sprintf(msg->rep, + " Congestion link:%u Send queue max:%u avg:%u", + nla_get_u32(stats[TIPC_NLA_STATS_LINK_CONGS]), + nla_get_u32(stats[TIPC_NLA_STATS_MAX_QUEUE]), + nla_get_u32(stats[TIPC_NLA_STATS_AVG_QUEUE])); +} + +static int tipc_nl_compat_link_stat_dump(struct tipc_nl_compat_msg *msg, + struct nlattr **attrs) +{ + char *name; + struct nlattr *link[TIPC_NLA_LINK_MAX + 1]; + struct nlattr *prop[TIPC_NLA_PROP_MAX + 1]; + struct nlattr *stats[TIPC_NLA_STATS_MAX + 1]; + + nla_parse_nested(link, TIPC_NLA_LINK_MAX, attrs[TIPC_NLA_LINK], NULL); + + nla_parse_nested(prop, TIPC_NLA_PROP_MAX, link[TIPC_NLA_LINK_PROP], + NULL); + + nla_parse_nested(stats, TIPC_NLA_STATS_MAX, link[TIPC_NLA_LINK_STATS], + NULL); + + name = (char *)TLV_DATA(msg->req); + if (strcmp(name, nla_data(link[TIPC_NLA_LINK_NAME])) != 0) + return 0; + + tipc_tlv_sprintf(msg->rep, "\nLink <%s>\n", + nla_data(link[TIPC_NLA_LINK_NAME])); + + if (link[TIPC_NLA_LINK_BROADCAST]) { + __fill_bc_link_stat(msg, prop, stats); + return 0; + } + + if (link[TIPC_NLA_LINK_ACTIVE]) + tipc_tlv_sprintf(msg->rep, " ACTIVE"); + else if (link[TIPC_NLA_LINK_UP]) + tipc_tlv_sprintf(msg->rep, " STANDBY"); + else + tipc_tlv_sprintf(msg->rep, " DEFUNCT"); + + tipc_tlv_sprintf(msg->rep, " MTU:%u Priority:%u", + nla_get_u32(link[TIPC_NLA_LINK_MTU]), + nla_get_u32(prop[TIPC_NLA_PROP_PRIO])); + + tipc_tlv_sprintf(msg->rep, " Tolerance:%u ms Window:%u packets\n", + nla_get_u32(prop[TIPC_NLA_PROP_TOL]), + nla_get_u32(prop[TIPC_NLA_PROP_WIN])); + + tipc_tlv_sprintf(msg->rep, + " RX packets:%u fragments:%u/%u bundles:%u/%u\n", + nla_get_u32(link[TIPC_NLA_LINK_RX]) - + nla_get_u32(stats[TIPC_NLA_STATS_RX_INFO]), + nla_get_u32(stats[TIPC_NLA_STATS_RX_FRAGMENTS]), + nla_get_u32(stats[TIPC_NLA_STATS_RX_FRAGMENTED]), + nla_get_u32(stats[TIPC_NLA_STATS_RX_BUNDLES]), + nla_get_u32(stats[TIPC_NLA_STATS_RX_BUNDLED])); + + tipc_tlv_sprintf(msg->rep, + " TX packets:%u fragments:%u/%u bundles:%u/%u\n", + nla_get_u32(link[TIPC_NLA_LINK_TX]) - + nla_get_u32(stats[TIPC_NLA_STATS_TX_INFO]), + nla_get_u32(stats[TIPC_NLA_STATS_TX_FRAGMENTS]), + nla_get_u32(stats[TIPC_NLA_STATS_TX_FRAGMENTED]), + nla_get_u32(stats[TIPC_NLA_STATS_TX_BUNDLES]), + nla_get_u32(stats[TIPC_NLA_STATS_TX_BUNDLED])); + + tipc_tlv_sprintf(msg->rep, + " TX profile sample:%u packets average:%u octets\n", + nla_get_u32(stats[TIPC_NLA_STATS_MSG_LEN_CNT]), + nla_get_u32(stats[TIPC_NLA_STATS_MSG_LEN_TOT]) / + nla_get_u32(stats[TIPC_NLA_STATS_MSG_PROF_TOT])); + + tipc_tlv_sprintf(msg->rep, + " 0-64:%u%% -256:%u%% -1024:%u%% -4096:%u%% ", + perc(nla_get_u32(stats[TIPC_NLA_STATS_MSG_LEN_P0]), + nla_get_u32(stats[TIPC_NLA_STATS_MSG_PROF_TOT])), + perc(nla_get_u32(stats[TIPC_NLA_STATS_MSG_LEN_P1]), + nla_get_u32(stats[TIPC_NLA_STATS_MSG_PROF_TOT])), + perc(nla_get_u32(stats[TIPC_NLA_STATS_MSG_LEN_P2]), + nla_get_u32(stats[TIPC_NLA_STATS_MSG_PROF_TOT])), + perc(nla_get_u32(stats[TIPC_NLA_STATS_MSG_LEN_P3]), + nla_get_u32(stats[TIPC_NLA_STATS_MSG_PROF_TOT]))); + + tipc_tlv_sprintf(msg->rep, "-16384:%u%% -32768:%u%% -66000:%u%%\n", + perc(nla_get_u32(stats[TIPC_NLA_STATS_MSG_LEN_P4]), + nla_get_u32(stats[TIPC_NLA_STATS_MSG_PROF_TOT])), + perc(nla_get_u32(stats[TIPC_NLA_STATS_MSG_LEN_P5]), + nla_get_u32(stats[TIPC_NLA_STATS_MSG_PROF_TOT])), + perc(nla_get_u32(stats[TIPC_NLA_STATS_MSG_LEN_P6]), + nla_get_u32(stats[TIPC_NLA_STATS_MSG_PROF_TOT]))); + + tipc_tlv_sprintf(msg->rep, + " RX states:%u probes:%u naks:%u defs:%u dups:%u\n", + nla_get_u32(stats[TIPC_NLA_STATS_RX_STATES]), + nla_get_u32(stats[TIPC_NLA_STATS_RX_PROBES]), + nla_get_u32(stats[TIPC_NLA_STATS_RX_NACKS]), + nla_get_u32(stats[TIPC_NLA_STATS_RX_DEFERRED]), + nla_get_u32(stats[TIPC_NLA_STATS_DUPLICATES])); + + tipc_tlv_sprintf(msg->rep, + " TX states:%u probes:%u naks:%u acks:%u dups:%u\n", + nla_get_u32(stats[TIPC_NLA_STATS_TX_STATES]), + nla_get_u32(stats[TIPC_NLA_STATS_TX_PROBES]), + nla_get_u32(stats[TIPC_NLA_STATS_TX_NACKS]), + nla_get_u32(stats[TIPC_NLA_STATS_TX_ACKS]), + nla_get_u32(stats[TIPC_NLA_STATS_RETRANSMITTED])); + + tipc_tlv_sprintf(msg->rep, + " Congestion link:%u Send queue max:%u avg:%u", + nla_get_u32(stats[TIPC_NLA_STATS_LINK_CONGS]), + nla_get_u32(stats[TIPC_NLA_STATS_MAX_QUEUE]), + nla_get_u32(stats[TIPC_NLA_STATS_AVG_QUEUE])); + + return 0; +} + +static int tipc_nl_compat_link_dump(struct tipc_nl_compat_msg *msg, + struct nlattr **attrs) +{ + struct nlattr *link[TIPC_NLA_LINK_MAX + 1]; + struct tipc_link_info link_info; + + nla_parse_nested(link, TIPC_NLA_LINK_MAX, attrs[TIPC_NLA_LINK], NULL); + + link_info.dest = nla_get_flag(link[TIPC_NLA_LINK_DEST]); + link_info.up = htonl(nla_get_flag(link[TIPC_NLA_LINK_UP])); + strcpy(link_info.str, nla_data(link[TIPC_NLA_LINK_NAME])); + + return tipc_add_tlv(msg->rep, TIPC_TLV_LINK_INFO, + &link_info, sizeof(link_info)); +} + +static int tipc_nl_compat_link_set(struct sk_buff *skb, + struct tipc_nl_compat_msg *msg) +{ + struct nlattr *link; + struct nlattr *prop; + struct tipc_link_config *lc; + + lc = (struct tipc_link_config *)TLV_DATA(msg->req); + + link = nla_nest_start(skb, TIPC_NLA_LINK); + if (!link) + return -EMSGSIZE; + + if (nla_put_string(skb, TIPC_NLA_LINK_NAME, lc->name)) + return -EMSGSIZE; + + prop = nla_nest_start(skb, TIPC_NLA_LINK_PROP); + if (!prop) + return -EMSGSIZE; + + if (msg->cmd == TIPC_CMD_SET_LINK_PRI) { + if (nla_put_u32(skb, TIPC_NLA_PROP_PRIO, ntohl(lc->value))) + return -EMSGSIZE; + } else if (msg->cmd == TIPC_CMD_SET_LINK_TOL) { + if (nla_put_u32(skb, TIPC_NLA_PROP_TOL, ntohl(lc->value))) + return -EMSGSIZE; + } else if (msg->cmd == TIPC_CMD_SET_LINK_WINDOW) { + if (nla_put_u32(skb, TIPC_NLA_PROP_WIN, ntohl(lc->value))) + return -EMSGSIZE; + } + + nla_nest_end(skb, prop); + nla_nest_end(skb, link); + + return 0; +} + +static int tipc_nl_compat_link_reset_stats(struct sk_buff *skb, + struct tipc_nl_compat_msg *msg) +{ + char *name; + struct nlattr *link; + + name = (char *)TLV_DATA(msg->req); + + link = nla_nest_start(skb, TIPC_NLA_LINK); + if (!link) + return -EMSGSIZE; + + if (nla_put_string(skb, TIPC_NLA_LINK_NAME, name)) + return -EMSGSIZE; + + nla_nest_end(skb, link); + + return 0; +} + +static int tipc_nl_compat_name_table_dump_header(struct tipc_nl_compat_msg *msg) +{ + int i; + u32 depth; + struct tipc_name_table_query *ntq; + static const char * const header[] = { + "Type ", + "Lower Upper ", + "Port Identity ", + "Publication Scope" + }; + + ntq = (struct tipc_name_table_query *)TLV_DATA(msg->req); + + depth = ntohl(ntq->depth); + + if (depth > 4) + depth = 4; + for (i = 0; i < depth; i++) + tipc_tlv_sprintf(msg->rep, header[i]); + tipc_tlv_sprintf(msg->rep, "\n"); + + return 0; +} + +static int tipc_nl_compat_name_table_dump(struct tipc_nl_compat_msg *msg, + struct nlattr **attrs) +{ + char port_str[27]; + struct tipc_name_table_query *ntq; + struct nlattr *nt[TIPC_NLA_NAME_TABLE_MAX + 1]; + struct nlattr *publ[TIPC_NLA_PUBL_MAX + 1]; + u32 node, depth, type, lowbound, upbound; + static const char * const scope_str[] = {"", " zone", " cluster", + " node"}; + + nla_parse_nested(nt, TIPC_NLA_NAME_TABLE_MAX, + attrs[TIPC_NLA_NAME_TABLE], NULL); + + nla_parse_nested(publ, TIPC_NLA_PUBL_MAX, nt[TIPC_NLA_NAME_TABLE_PUBL], + NULL); + + ntq = (struct tipc_name_table_query *)TLV_DATA(msg->req); + + depth = ntohl(ntq->depth); + type = ntohl(ntq->type); + lowbound = ntohl(ntq->lowbound); + upbound = ntohl(ntq->upbound); + + if (!(depth & TIPC_NTQ_ALLTYPES) && + (type != nla_get_u32(publ[TIPC_NLA_PUBL_TYPE]))) + return 0; + if (lowbound && (lowbound > nla_get_u32(publ[TIPC_NLA_PUBL_UPPER]))) + return 0; + if (upbound && (upbound < nla_get_u32(publ[TIPC_NLA_PUBL_LOWER]))) + return 0; + + tipc_tlv_sprintf(msg->rep, "%-10u ", + nla_get_u32(publ[TIPC_NLA_PUBL_TYPE])); + + if (depth == 1) + goto out; + + tipc_tlv_sprintf(msg->rep, "%-10u %-10u ", + nla_get_u32(publ[TIPC_NLA_PUBL_LOWER]), + nla_get_u32(publ[TIPC_NLA_PUBL_UPPER])); + + if (depth == 2) + goto out; + + node = nla_get_u32(publ[TIPC_NLA_PUBL_NODE]); + sprintf(port_str, "<%u.%u.%u:%u>", tipc_zone(node), tipc_cluster(node), + tipc_node(node), nla_get_u32(publ[TIPC_NLA_PUBL_REF])); + tipc_tlv_sprintf(msg->rep, "%-26s ", port_str); + + if (depth == 3) + goto out; + + tipc_tlv_sprintf(msg->rep, "%-10u %s", + nla_get_u32(publ[TIPC_NLA_PUBL_REF]), + scope_str[nla_get_u32(publ[TIPC_NLA_PUBL_SCOPE])]); +out: + tipc_tlv_sprintf(msg->rep, "\n"); + + return 0; +} + +static int __tipc_nl_compat_publ_dump(struct tipc_nl_compat_msg *msg, + struct nlattr **attrs) +{ + u32 type, lower, upper; + struct nlattr *publ[TIPC_NLA_PUBL_MAX + 1]; + + nla_parse_nested(publ, TIPC_NLA_PUBL_MAX, attrs[TIPC_NLA_PUBL], NULL); + + type = nla_get_u32(publ[TIPC_NLA_PUBL_TYPE]); + lower = nla_get_u32(publ[TIPC_NLA_PUBL_LOWER]); + upper = nla_get_u32(publ[TIPC_NLA_PUBL_UPPER]); + + if (lower == upper) + tipc_tlv_sprintf(msg->rep, " {%u,%u}", type, lower); + else + tipc_tlv_sprintf(msg->rep, " {%u,%u,%u}", type, lower, upper); + + return 0; +} + +static int tipc_nl_compat_publ_dump(struct tipc_nl_compat_msg *msg, u32 sock) +{ + int err; + void *hdr; + struct nlattr *nest; + struct sk_buff *args; + struct tipc_nl_compat_cmd_dump dump; + + args = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + if (!args) + return -ENOMEM; + + hdr = genlmsg_put(args, 0, 0, &tipc_genl_family, NLM_F_MULTI, + TIPC_NL_PUBL_GET); + + nest = nla_nest_start(args, TIPC_NLA_SOCK); + if (!nest) { + kfree_skb(args); + return -EMSGSIZE; + } + + if (nla_put_u32(args, TIPC_NLA_SOCK_REF, sock)) { + kfree_skb(args); + return -EMSGSIZE; + } + + nla_nest_end(args, nest); + genlmsg_end(args, hdr); + + dump.dumpit = tipc_nl_publ_dump; + dump.format = __tipc_nl_compat_publ_dump; + + err = __tipc_nl_compat_dumpit(&dump, msg, args); + + kfree_skb(args); + + return err; +} + +static int tipc_nl_compat_sk_dump(struct tipc_nl_compat_msg *msg, + struct nlattr **attrs) +{ + int err; + u32 sock_ref; + struct nlattr *sock[TIPC_NLA_SOCK_MAX + 1]; + + nla_parse_nested(sock, TIPC_NLA_SOCK_MAX, attrs[TIPC_NLA_SOCK], NULL); + + sock_ref = nla_get_u32(sock[TIPC_NLA_SOCK_REF]); + tipc_tlv_sprintf(msg->rep, "%u:", sock_ref); + + if (sock[TIPC_NLA_SOCK_CON]) { + u32 node; + struct nlattr *con[TIPC_NLA_CON_MAX + 1]; + + nla_parse_nested(con, TIPC_NLA_CON_MAX, sock[TIPC_NLA_SOCK_CON], + NULL); + + node = nla_get_u32(con[TIPC_NLA_CON_NODE]); + tipc_tlv_sprintf(msg->rep, " connected to <%u.%u.%u:%u>", + tipc_zone(node), + tipc_cluster(node), + tipc_node(node), + nla_get_u32(con[TIPC_NLA_CON_SOCK])); + + if (con[TIPC_NLA_CON_FLAG]) + tipc_tlv_sprintf(msg->rep, " via {%u,%u}\n", + nla_get_u32(con[TIPC_NLA_CON_TYPE]), + nla_get_u32(con[TIPC_NLA_CON_INST])); + else + tipc_tlv_sprintf(msg->rep, "\n"); + } else if (sock[TIPC_NLA_SOCK_HAS_PUBL]) { + tipc_tlv_sprintf(msg->rep, " bound to"); + + err = tipc_nl_compat_publ_dump(msg, sock_ref); + if (err) + return err; + } + tipc_tlv_sprintf(msg->rep, "\n"); + + return 0; +} + +static int tipc_nl_compat_media_dump(struct tipc_nl_compat_msg *msg, + struct nlattr **attrs) +{ + struct nlattr *media[TIPC_NLA_MEDIA_MAX + 1]; + + nla_parse_nested(media, TIPC_NLA_MEDIA_MAX, attrs[TIPC_NLA_MEDIA], + NULL); + + return tipc_add_tlv(msg->rep, TIPC_TLV_MEDIA_NAME, + nla_data(media[TIPC_NLA_MEDIA_NAME]), + nla_len(media[TIPC_NLA_MEDIA_NAME])); +} + +static int tipc_nl_compat_node_dump(struct tipc_nl_compat_msg *msg, + struct nlattr **attrs) +{ + struct tipc_node_info node_info; + struct nlattr *node[TIPC_NLA_NODE_MAX + 1]; + + nla_parse_nested(node, TIPC_NLA_NODE_MAX, attrs[TIPC_NLA_NODE], NULL); + + node_info.addr = htonl(nla_get_u32(node[TIPC_NLA_NODE_ADDR])); + node_info.up = htonl(nla_get_flag(node[TIPC_NLA_NODE_UP])); + + return tipc_add_tlv(msg->rep, TIPC_TLV_NODE_INFO, &node_info, + sizeof(node_info)); +} + +static int tipc_nl_compat_net_set(struct sk_buff *skb, + struct tipc_nl_compat_msg *msg) +{ + u32 val; + struct nlattr *net; + + val = ntohl(*(__be32 *)TLV_DATA(msg->req)); + + net = nla_nest_start(skb, TIPC_NLA_NET); + if (!net) + return -EMSGSIZE; + + if (msg->cmd == TIPC_CMD_SET_NODE_ADDR) { + if (nla_put_u32(skb, TIPC_NLA_NET_ADDR, val)) + return -EMSGSIZE; + } else if (msg->cmd == TIPC_CMD_SET_NETID) { + if (nla_put_u32(skb, TIPC_NLA_NET_ID, val)) + return -EMSGSIZE; + } + nla_nest_end(skb, net); + + return 0; +} + +static int tipc_nl_compat_net_dump(struct tipc_nl_compat_msg *msg, + struct nlattr **attrs) +{ + __be32 id; + struct nlattr *net[TIPC_NLA_NET_MAX + 1]; + + nla_parse_nested(net, TIPC_NLA_NET_MAX, attrs[TIPC_NLA_NET], NULL); + id = htonl(nla_get_u32(net[TIPC_NLA_NET_ID])); + + return tipc_add_tlv(msg->rep, TIPC_TLV_UNSIGNED, &id, sizeof(id)); +} + +static int tipc_cmd_show_stats_compat(struct tipc_nl_compat_msg *msg) +{ + msg->rep = tipc_tlv_alloc(ULTRA_STRING_MAX_LEN); + if (!msg->rep) + return -ENOMEM; + + tipc_tlv_init(msg->rep, TIPC_TLV_ULTRA_STRING); + tipc_tlv_sprintf(msg->rep, "TIPC version " TIPC_MOD_VER "\n"); + + return 0; +} + +static int tipc_nl_compat_handle(struct tipc_nl_compat_msg *msg) +{ + struct tipc_nl_compat_cmd_dump dump; + struct tipc_nl_compat_cmd_doit doit; + + memset(&dump, 0, sizeof(dump)); + memset(&doit, 0, sizeof(doit)); + + switch (msg->cmd) { + case TIPC_CMD_NOOP: + msg->rep = tipc_tlv_alloc(0); + if (!msg->rep) + return -ENOMEM; + return 0; + case TIPC_CMD_GET_BEARER_NAMES: + msg->rep_size = MAX_BEARERS * TLV_SPACE(TIPC_MAX_BEARER_NAME); + dump.dumpit = tipc_nl_bearer_dump; + dump.format = tipc_nl_compat_bearer_dump; + return tipc_nl_compat_dumpit(&dump, msg); + case TIPC_CMD_ENABLE_BEARER: + msg->req_type = TIPC_TLV_BEARER_CONFIG; + doit.doit = tipc_nl_bearer_enable; + doit.transcode = tipc_nl_compat_bearer_enable; + return tipc_nl_compat_doit(&doit, msg); + case TIPC_CMD_DISABLE_BEARER: + msg->req_type = TIPC_TLV_BEARER_NAME; + doit.doit = tipc_nl_bearer_disable; + doit.transcode = tipc_nl_compat_bearer_disable; + return tipc_nl_compat_doit(&doit, msg); + case TIPC_CMD_SHOW_LINK_STATS: + msg->req_type = TIPC_TLV_LINK_NAME; + msg->rep_size = ULTRA_STRING_MAX_LEN; + msg->rep_type = TIPC_TLV_ULTRA_STRING; + dump.dumpit = tipc_nl_link_dump; + dump.format = tipc_nl_compat_link_stat_dump; + return tipc_nl_compat_dumpit(&dump, msg); + case TIPC_CMD_GET_LINKS: + msg->req_type = TIPC_TLV_NET_ADDR; + msg->rep_size = ULTRA_STRING_MAX_LEN; + dump.dumpit = tipc_nl_link_dump; + dump.format = tipc_nl_compat_link_dump; + return tipc_nl_compat_dumpit(&dump, msg); + case TIPC_CMD_SET_LINK_TOL: + case TIPC_CMD_SET_LINK_PRI: + case TIPC_CMD_SET_LINK_WINDOW: + msg->req_type = TIPC_TLV_LINK_CONFIG; + doit.doit = tipc_nl_link_set; + doit.transcode = tipc_nl_compat_link_set; + return tipc_nl_compat_doit(&doit, msg); + case TIPC_CMD_RESET_LINK_STATS: + msg->req_type = TIPC_TLV_LINK_NAME; + doit.doit = tipc_nl_link_reset_stats; + doit.transcode = tipc_nl_compat_link_reset_stats; + return tipc_nl_compat_doit(&doit, msg); + case TIPC_CMD_SHOW_NAME_TABLE: + msg->req_type = TIPC_TLV_NAME_TBL_QUERY; + msg->rep_size = ULTRA_STRING_MAX_LEN; + msg->rep_type = TIPC_TLV_ULTRA_STRING; + dump.header = tipc_nl_compat_name_table_dump_header; + dump.dumpit = tipc_nl_name_table_dump; + dump.format = tipc_nl_compat_name_table_dump; + return tipc_nl_compat_dumpit(&dump, msg); + case TIPC_CMD_SHOW_PORTS: + msg->rep_size = ULTRA_STRING_MAX_LEN; + msg->rep_type = TIPC_TLV_ULTRA_STRING; + dump.dumpit = tipc_nl_sk_dump; + dump.format = tipc_nl_compat_sk_dump; + return tipc_nl_compat_dumpit(&dump, msg); + case TIPC_CMD_GET_MEDIA_NAMES: + msg->rep_size = MAX_MEDIA * TLV_SPACE(TIPC_MAX_MEDIA_NAME); + dump.dumpit = tipc_nl_media_dump; + dump.format = tipc_nl_compat_media_dump; + return tipc_nl_compat_dumpit(&dump, msg); + case TIPC_CMD_GET_NODES: + msg->rep_size = ULTRA_STRING_MAX_LEN; + dump.dumpit = tipc_nl_node_dump; + dump.format = tipc_nl_compat_node_dump; + return tipc_nl_compat_dumpit(&dump, msg); + case TIPC_CMD_SET_NODE_ADDR: + msg->req_type = TIPC_TLV_NET_ADDR; + doit.doit = tipc_nl_net_set; + doit.transcode = tipc_nl_compat_net_set; + return tipc_nl_compat_doit(&doit, msg); + case TIPC_CMD_SET_NETID: + msg->req_type = TIPC_TLV_UNSIGNED; + doit.doit = tipc_nl_net_set; + doit.transcode = tipc_nl_compat_net_set; + return tipc_nl_compat_doit(&doit, msg); + case TIPC_CMD_GET_NETID: + msg->rep_size = sizeof(u32); + dump.dumpit = tipc_nl_net_dump; + dump.format = tipc_nl_compat_net_dump; + return tipc_nl_compat_dumpit(&dump, msg); + case TIPC_CMD_SHOW_STATS: + return tipc_cmd_show_stats_compat(msg); + } + + return -EOPNOTSUPP; +} + +static int tipc_nl_compat_recv(struct sk_buff *skb, struct genl_info *info) +{ + int err; + int len; + struct tipc_nl_compat_msg msg; + struct nlmsghdr *req_nlh; + struct nlmsghdr *rep_nlh; + struct tipc_genlmsghdr *req_userhdr = info->userhdr; + struct net *net = genl_info_net(info); + + memset(&msg, 0, sizeof(msg)); + + req_nlh = (struct nlmsghdr *)skb->data; + msg.req = nlmsg_data(req_nlh) + GENL_HDRLEN + TIPC_GENL_HDRLEN; + msg.cmd = req_userhdr->cmd; + msg.dst_sk = info->dst_sk; + + if ((msg.cmd & 0xC000) && (!netlink_net_capable(skb, CAP_NET_ADMIN))) { + msg.rep = tipc_get_err_tlv(TIPC_CFG_NOT_NET_ADMIN); + err = -EACCES; + goto send; + } + + len = nlmsg_attrlen(req_nlh, GENL_HDRLEN + TIPC_GENL_HDRLEN); + if (TLV_GET_LEN(msg.req) && !TLV_OK(msg.req, len)) { + msg.rep = tipc_get_err_tlv(TIPC_CFG_NOT_SUPPORTED); + err = -EOPNOTSUPP; + goto send; + } + + err = tipc_nl_compat_handle(&msg); + if (err == -EOPNOTSUPP) + msg.rep = tipc_get_err_tlv(TIPC_CFG_NOT_SUPPORTED); + else if (err == -EINVAL) + msg.rep = tipc_get_err_tlv(TIPC_CFG_TLV_ERROR); +send: + if (!msg.rep) + return err; + + len = nlmsg_total_size(GENL_HDRLEN + TIPC_GENL_HDRLEN); + skb_push(msg.rep, len); + rep_nlh = nlmsg_hdr(msg.rep); + memcpy(rep_nlh, info->nlhdr, len); + rep_nlh->nlmsg_len = msg.rep->len; + genlmsg_unicast(net, msg.rep, NETLINK_CB(skb).portid); + + return err; +} + +static struct genl_family tipc_genl_compat_family = { + .id = GENL_ID_GENERATE, + .name = TIPC_GENL_NAME, + .version = TIPC_GENL_VERSION, + .hdrsize = TIPC_GENL_HDRLEN, + .maxattr = 0, + .netnsok = true, +}; + +static struct genl_ops tipc_genl_compat_ops[] = { + { + .cmd = TIPC_GENL_CMD, + .doit = tipc_nl_compat_recv, + }, +}; + +int tipc_netlink_compat_start(void) +{ + int res; + + res = genl_register_family_with_ops(&tipc_genl_compat_family, + tipc_genl_compat_ops); + if (res) { + pr_err("Failed to register legacy compat interface\n"); + return res; + } + + return 0; +} + +void tipc_netlink_compat_stop(void) +{ + genl_unregister_family(&tipc_genl_compat_family); +} diff --git a/net/tipc/node.c b/net/tipc/node.c index 8d353ec77a66..22c059ad2999 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -35,21 +35,14 @@ */ #include "core.h" -#include "config.h" +#include "link.h" #include "node.h" #include "name_distr.h" #include "socket.h" -#define NODE_HTABLE_SIZE 512 - static void node_lost_contact(struct tipc_node *n_ptr); static void node_established_contact(struct tipc_node *n_ptr); - -static struct hlist_head node_htable[NODE_HTABLE_SIZE]; -LIST_HEAD(tipc_node_list); -static u32 tipc_num_nodes; -static u32 tipc_num_links; -static DEFINE_SPINLOCK(node_list_lock); +static void tipc_node_delete(struct tipc_node *node); struct tipc_sock_conn { u32 port; @@ -75,19 +68,39 @@ 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 */ -struct tipc_node *tipc_node_find(u32 addr) +struct tipc_node *tipc_node_find(struct net *net, u32 addr) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_node *node; - if (unlikely(!in_own_cluster_exact(addr))) + if (unlikely(!in_own_cluster_exact(net, addr))) return NULL; rcu_read_lock(); - hlist_for_each_entry_rcu(node, &node_htable[tipc_hashfn(addr)], hash) { + 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; } @@ -96,79 +109,80 @@ struct tipc_node *tipc_node_find(u32 addr) return NULL; } -struct tipc_node *tipc_node_create(u32 addr) +struct tipc_node *tipc_node_create(struct net *net, u32 addr) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_node *n_ptr, *temp_node; - spin_lock_bh(&node_list_lock); - + spin_lock_bh(&tn->node_list_lock); + n_ptr = tipc_node_find(net, addr); + if (n_ptr) + goto exit; n_ptr = kzalloc(sizeof(*n_ptr), GFP_ATOMIC); if (!n_ptr) { - spin_unlock_bh(&node_list_lock); pr_warn("Node creation failed, no memory\n"); - return NULL; + goto exit; } - 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); INIT_LIST_HEAD(&n_ptr->publ_list); INIT_LIST_HEAD(&n_ptr->conn_sks); - skb_queue_head_init(&n_ptr->waiting_sks); - __skb_queue_head_init(&n_ptr->bclink.deferred_queue); - - hlist_add_head_rcu(&n_ptr->hash, &node_htable[tipc_hashfn(addr)]); - - list_for_each_entry_rcu(temp_node, &tipc_node_list, list) { + __skb_queue_head_init(&n_ptr->bclink.deferdq); + hlist_add_head_rcu(&n_ptr->hash, &tn->node_htable[tipc_hashfn(addr)]); + list_for_each_entry_rcu(temp_node, &tn->node_list, list) { if (n_ptr->addr < temp_node->addr) break; } 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_num_nodes++; - - spin_unlock_bh(&node_list_lock); + tipc_node_get(n_ptr); +exit: + spin_unlock_bh(&tn->node_list_lock); return n_ptr; } -static void tipc_node_delete(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); - - tipc_num_nodes--; + list_del_rcu(&node->list); + hlist_del_rcu(&node->hash); + kfree_rcu(node, rcu); } -void tipc_node_stop(void) +void tipc_node_stop(struct net *net) { + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_node *node, *t_node; - spin_lock_bh(&node_list_lock); - list_for_each_entry_safe(node, t_node, &tipc_node_list, list) - tipc_node_delete(node); - spin_unlock_bh(&node_list_lock); + spin_lock_bh(&tn->node_list_lock); + list_for_each_entry_safe(node, t_node, &tn->node_list, list) + tipc_node_put(node); + spin_unlock_bh(&tn->node_list_lock); } -int tipc_node_add_conn(u32 dnode, u32 port, u32 peer_port) +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(dnode)) + if (in_own_node(net, dnode)) return 0; - node = tipc_node_find(dnode); + node = tipc_node_find(net, dnode); if (!node) { pr_warn("Connecting sock to node 0x%x failed\n", dnode); 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; @@ -176,18 +190,20 @@ int tipc_node_add_conn(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(u32 dnode, u32 port) +void tipc_node_remove_conn(struct net *net, u32 dnode, u32 port) { struct tipc_node *node; struct tipc_sock_conn *conn, *safe; - if (in_own_node(dnode)) + if (in_own_node(net, dnode)) return; - node = tipc_node_find(dnode); + node = tipc_node_find(net, dnode); if (!node) return; @@ -199,23 +215,7 @@ void tipc_node_remove_conn(u32 dnode, u32 port) kfree(conn); } tipc_node_unlock(node); -} - -void tipc_node_abort_sock_conns(struct list_head *conns) -{ - struct tipc_sock_conn *conn, *safe; - struct sk_buff *buf; - - list_for_each_entry_safe(conn, safe, conns, list) { - buf = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, TIPC_CONN_MSG, - SHORT_H_SIZE, 0, tipc_own_addr, - conn->peer_node, conn->port, - conn->peer_port, TIPC_ERR_NO_NODE); - if (likely(buf)) - tipc_sk_rcv(buf); - list_del(&conn->list); - kfree(conn); - } + tipc_node_put(node); } /** @@ -231,8 +231,8 @@ void tipc_node_link_up(struct tipc_node *n_ptr, struct tipc_link *l_ptr) n_ptr->action_flags |= TIPC_NOTIFY_LINK_UP; n_ptr->link_id = l_ptr->peer_bearer_id << 16 | l_ptr->bearer_id; - pr_info("Established link <%s> on network plane %c\n", - l_ptr->name, l_ptr->net_plane); + pr_debug("Established link <%s> on network plane %c\n", + l_ptr->name, l_ptr->net_plane); if (!active[0]) { active[0] = active[1] = l_ptr; @@ -240,7 +240,7 @@ void tipc_node_link_up(struct tipc_node *n_ptr, struct tipc_link *l_ptr) goto exit; } if (l_ptr->priority < active[0]->priority) { - pr_info("New link <%s> becomes standby\n", l_ptr->name); + pr_debug("New link <%s> becomes standby\n", l_ptr->name); goto exit; } tipc_link_dup_queue_xmit(active[0], l_ptr); @@ -248,14 +248,14 @@ void tipc_node_link_up(struct tipc_node *n_ptr, struct tipc_link *l_ptr) active[0] = l_ptr; goto exit; } - pr_info("Old link <%s> becomes standby\n", active[0]->name); + pr_debug("Old link <%s> becomes standby\n", active[0]->name); if (active[1] != active[0]) - pr_info("Old link <%s> becomes standby\n", active[1]->name); + pr_debug("Old link <%s> becomes standby\n", active[1]->name); active[0] = active[1] = l_ptr; exit: /* Leave room for changeover header when returning 'mtu' to users: */ - n_ptr->act_mtus[0] = active[0]->max_pkt - INT_H_SIZE; - n_ptr->act_mtus[1] = active[1]->max_pkt - INT_H_SIZE; + n_ptr->act_mtus[0] = active[0]->mtu - INT_H_SIZE; + n_ptr->act_mtus[1] = active[1]->mtu - INT_H_SIZE; } /** @@ -290,6 +290,7 @@ 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_net *tn = net_generic(n_ptr->net, tipc_net_id); struct tipc_link **active; n_ptr->working_links--; @@ -297,12 +298,12 @@ void tipc_node_link_down(struct tipc_node *n_ptr, struct tipc_link *l_ptr) n_ptr->link_id = l_ptr->peer_bearer_id << 16 | l_ptr->bearer_id; if (!tipc_link_is_active(l_ptr)) { - pr_info("Lost standby link <%s> on network plane %c\n", - l_ptr->name, l_ptr->net_plane); + pr_debug("Lost standby link <%s> on network plane %c\n", + l_ptr->name, l_ptr->net_plane); return; } - pr_info("Lost link <%s> on network plane %c\n", - l_ptr->name, l_ptr->net_plane); + pr_debug("Lost link <%s> on network plane %c\n", + l_ptr->name, l_ptr->net_plane); active = &n_ptr->active_links[0]; if (active[0] == l_ptr) @@ -318,13 +319,12 @@ void tipc_node_link_down(struct tipc_node *n_ptr, struct tipc_link *l_ptr) /* Leave room for changeover header when returning 'mtu' to users: */ if (active[0]) { - n_ptr->act_mtus[0] = active[0]->max_pkt - INT_H_SIZE; - n_ptr->act_mtus[1] = active[1]->max_pkt - INT_H_SIZE; + n_ptr->act_mtus[0] = active[0]->mtu - INT_H_SIZE; + n_ptr->act_mtus[1] = active[1]->mtu - INT_H_SIZE; return; } - /* Loopback link went down? No fragmentation needed from now on. */ - if (n_ptr->addr == tipc_own_addr) { + if (n_ptr->addr == tn->own_addr) { n_ptr->act_mtus[0] = MAX_MSG_SIZE; n_ptr->act_mtus[1] = MAX_MSG_SIZE; } @@ -343,9 +343,6 @@ 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->bearer_id] = l_ptr; - spin_lock_bh(&node_list_lock); - tipc_num_links++; - spin_unlock_bh(&node_list_lock); n_ptr->link_cnt++; } @@ -357,9 +354,6 @@ void tipc_node_detach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr) if (l_ptr != n_ptr->links[i]) continue; n_ptr->links[i] = NULL; - spin_lock_bh(&node_list_lock); - tipc_num_links--; - spin_unlock_bh(&node_list_lock); n_ptr->link_cnt--; } } @@ -368,161 +362,71 @@ static void node_established_contact(struct tipc_node *n_ptr) { 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); + n_ptr->bclink.acked = tipc_bclink_get_last_sent(n_ptr->net); + tipc_bclink_add_node(n_ptr->net, n_ptr->addr); } static void node_lost_contact(struct tipc_node *n_ptr) { char addr_string[16]; - u32 i; + struct tipc_sock_conn *conn, *safe; + struct list_head *conns = &n_ptr->conn_sks; + struct sk_buff *skb; + struct tipc_net *tn = net_generic(n_ptr->net, tipc_net_id); + uint i; - pr_info("Lost contact with %s\n", - tipc_addr_string_fill(addr_string, n_ptr->addr)); + pr_debug("Lost contact with %s\n", + tipc_addr_string_fill(addr_string, n_ptr->addr)); /* Flush broadcast link info associated with lost node */ if (n_ptr->bclink.recv_permitted) { - __skb_queue_purge(&n_ptr->bclink.deferred_queue); + __skb_queue_purge(&n_ptr->bclink.deferdq); 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); + tipc_bclink_remove_node(n_ptr->net, n_ptr->addr); tipc_bclink_acknowledge(n_ptr, INVALID_LINK_SEQ); n_ptr->bclink.recv_permitted = false; } - /* Abort link changeover */ + /* Abort any ongoing link failover */ for (i = 0; i < MAX_BEARERS; i++) { struct tipc_link *l_ptr = n_ptr->links[i]; if (!l_ptr) continue; - l_ptr->reset_checkpoint = l_ptr->next_in_no; - l_ptr->exp_msg_count = 0; + l_ptr->flags &= ~LINK_FAILINGOVER; + l_ptr->failover_checkpt = 0; + l_ptr->failover_pkts = 0; + kfree_skb(l_ptr->failover_skb); + l_ptr->failover_skb = NULL; tipc_link_reset_fragments(l_ptr); } n_ptr->action_flags &= ~TIPC_WAIT_OWN_LINKS_DOWN; - /* 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; -} + /* Prevent re-contact with node until cleanup is done */ + n_ptr->action_flags |= TIPC_WAIT_PEER_LINKS_DOWN; -struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space) -{ - u32 domain; - struct sk_buff *buf; - struct tipc_node *n_ptr; - struct tipc_node_info node_info; - u32 payload_size; - - if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) - return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - - domain = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); - if (!tipc_addr_domain_valid(domain)) - return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (network address)"); - - spin_lock_bh(&node_list_lock); - if (!tipc_num_nodes) { - spin_unlock_bh(&node_list_lock); - return tipc_cfg_reply_none(); - } + /* Notify publications from this node */ + n_ptr->action_flags |= TIPC_NOTIFY_NODE_DOWN; - /* For now, get space for all other nodes */ - payload_size = TLV_SPACE(sizeof(node_info)) * tipc_num_nodes; - if (payload_size > 32768u) { - spin_unlock_bh(&node_list_lock); - return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (too many nodes)"); - } - spin_unlock_bh(&node_list_lock); - - buf = tipc_cfg_reply_alloc(payload_size); - if (!buf) - return NULL; - - /* Add TLVs for all nodes in scope */ - rcu_read_lock(); - list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { - if (!tipc_in_scope(domain, n_ptr->addr)) - continue; - node_info.addr = htonl(n_ptr->addr); - node_info.up = htonl(tipc_node_is_up(n_ptr)); - tipc_cfg_append_tlv(buf, TIPC_TLV_NODE_INFO, - &node_info, sizeof(node_info)); - } - rcu_read_unlock(); - return buf; -} - -struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) -{ - u32 domain; - struct sk_buff *buf; - struct tipc_node *n_ptr; - struct tipc_link_info link_info; - u32 payload_size; - - if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) - return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR); - - domain = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); - if (!tipc_addr_domain_valid(domain)) - return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE - " (network address)"); - - if (!tipc_own_addr) - return tipc_cfg_reply_none(); - - spin_lock_bh(&node_list_lock); - /* Get space for all unicast links + broadcast link */ - payload_size = TLV_SPACE((sizeof(link_info)) * (tipc_num_links + 1)); - if (payload_size > 32768u) { - spin_unlock_bh(&node_list_lock); - return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED - " (too many links)"); - } - spin_unlock_bh(&node_list_lock); - - buf = tipc_cfg_reply_alloc(payload_size); - if (!buf) - return NULL; - - /* Add TLV for broadcast link */ - link_info.dest = htonl(tipc_cluster_mask(tipc_own_addr)); - link_info.up = htonl(1); - strlcpy(link_info.str, tipc_bclink_name, TIPC_MAX_LINK_NAME); - tipc_cfg_append_tlv(buf, TIPC_TLV_LINK_INFO, &link_info, sizeof(link_info)); - - /* Add TLVs for any other links in scope */ - rcu_read_lock(); - list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { - u32 i; - - if (!tipc_in_scope(domain, n_ptr->addr)) - continue; - tipc_node_lock(n_ptr); - for (i = 0; i < MAX_BEARERS; i++) { - if (!n_ptr->links[i]) - continue; - link_info.dest = htonl(n_ptr->addr); - link_info.up = htonl(tipc_link_is_up(n_ptr->links[i])); - strcpy(link_info.str, n_ptr->links[i]->name); - tipc_cfg_append_tlv(buf, TIPC_TLV_LINK_INFO, - &link_info, sizeof(link_info)); + /* Notify sockets connected to node */ + list_for_each_entry_safe(conn, safe, conns, list) { + skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, TIPC_CONN_MSG, + SHORT_H_SIZE, 0, tn->own_addr, + conn->peer_node, conn->port, + conn->peer_port, TIPC_ERR_NO_NODE); + if (likely(skb)) { + skb_queue_tail(n_ptr->inputq, skb); + n_ptr->action_flags |= TIPC_MSG_EVT; } - tipc_node_unlock(n_ptr); + list_del(&conn->list); + kfree(conn); } - rcu_read_unlock(); - return buf; } /** @@ -534,78 +438,90 @@ struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) * * Returns 0 on success */ -int tipc_node_get_linkname(u32 bearer_id, u32 addr, char *linkname, size_t len) +int tipc_node_get_linkname(struct net *net, u32 bearer_id, u32 addr, + char *linkname, size_t len) { struct tipc_link *link; - struct tipc_node *node = tipc_node_find(addr); + int err = -EINVAL; + struct tipc_node *node = tipc_node_find(net, addr); + + if (!node) + return err; + + if (bearer_id >= MAX_BEARERS) + goto exit; - 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; + err = 0; } +exit: tipc_node_unlock(node); - return -EINVAL; + tipc_node_put(node); + return err; } void tipc_node_unlock(struct tipc_node *node) { - LIST_HEAD(nsub_list); - LIST_HEAD(conn_sks); - struct sk_buff_head waiting_sks; + struct net *net = node->net; u32 addr = 0; - int flags = node->action_flags; + u32 flags = node->action_flags; u32 link_id = 0; + struct list_head *publ_list; + struct sk_buff_head *inputq = node->inputq; + struct sk_buff_head *namedq; - if (likely(!flags)) { + if (likely(!flags || (flags == TIPC_MSG_EVT))) { + node->action_flags = 0; spin_unlock_bh(&node->lock); + if (flags == TIPC_MSG_EVT) + tipc_sk_rcv(net, inputq); return; } addr = node->addr; link_id = node->link_id; - __skb_queue_head_init(&waiting_sks); + namedq = node->namedq; + publ_list = &node->publ_list; - if (flags & TIPC_WAKEUP_USERS) - skb_queue_splice_init(&node->waiting_sks, &waiting_sks); - - if (flags & TIPC_NOTIFY_NODE_DOWN) { - list_replace_init(&node->publ_list, &nsub_list); - list_replace_init(&node->conn_sks, &conn_sks); - } - node->action_flags &= ~(TIPC_WAKEUP_USERS | TIPC_NOTIFY_NODE_DOWN | - TIPC_NOTIFY_NODE_UP | TIPC_NOTIFY_LINK_UP | - TIPC_NOTIFY_LINK_DOWN | - TIPC_WAKEUP_BCAST_USERS); + node->action_flags &= ~(TIPC_MSG_EVT | + 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_BCAST_RESET); spin_unlock_bh(&node->lock); - while (!skb_queue_empty(&waiting_sks)) - tipc_sk_rcv(__skb_dequeue(&waiting_sks)); - - if (!list_empty(&conn_sks)) - tipc_node_abort_sock_conns(&conn_sks); - - if (!list_empty(&nsub_list)) - tipc_publ_notify(&nsub_list, addr); + if (flags & TIPC_NOTIFY_NODE_DOWN) + tipc_publ_notify(net, publ_list, addr); if (flags & TIPC_WAKEUP_BCAST_USERS) - tipc_bclink_wakeup_users(); + tipc_bclink_wakeup_users(net); if (flags & TIPC_NOTIFY_NODE_UP) - tipc_named_node_up(addr); + tipc_named_node_up(net, addr); if (flags & TIPC_NOTIFY_LINK_UP) - tipc_nametbl_publish(TIPC_LINK_STATE, addr, addr, + tipc_nametbl_publish(net, TIPC_LINK_STATE, addr, addr, TIPC_NODE_SCOPE, link_id, addr); if (flags & TIPC_NOTIFY_LINK_DOWN) - tipc_nametbl_withdraw(TIPC_LINK_STATE, addr, + tipc_nametbl_withdraw(net, TIPC_LINK_STATE, addr, link_id, addr); + + if (flags & TIPC_MSG_EVT) + tipc_sk_rcv(net, inputq); + + if (flags & TIPC_NAMED_MSG_EVT) + tipc_named_rcv(net, namedq); + + 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 */ @@ -614,7 +530,7 @@ static int __tipc_nl_add_node(struct tipc_nl_msg *msg, struct tipc_node *node) void *hdr; struct nlattr *attrs; - hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_v2_family, + hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family, NLM_F_MULTI, TIPC_NL_NODE_GET); if (!hdr) return -EMSGSIZE; @@ -645,6 +561,8 @@ msg_full: int tipc_nl_node_dump(struct sk_buff *skb, struct netlink_callback *cb) { int err; + struct net *net = sock_net(skb->sk); + struct tipc_net *tn = net_generic(net, tipc_net_id); int done = cb->args[0]; int last_addr = cb->args[1]; struct tipc_node *node; @@ -658,20 +576,24 @@ 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(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, &tipc_node_list, list) { + list_for_each_entry_rcu(node, &tn->node_list, list) { if (last_addr) { if (node->addr == last_addr) last_addr = 0; diff --git a/net/tipc/node.h b/net/tipc/node.h index cbe0e950f1cc..02d5c20dc551 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -1,7 +1,7 @@ /* * net/tipc/node.h: Include file for TIPC node management routines * - * Copyright (c) 2000-2006, 2014, Ericsson AB + * Copyright (c) 2000-2006, 2014-2015, Ericsson AB * Copyright (c) 2005, 2010-2014, Wind River Systems * All rights reserved. * @@ -42,10 +42,10 @@ #include "bearer.h" #include "msg.h" -/* - * Out-of-range value for node signature - */ -#define INVALID_NODE_SIG 0x10000 +/* Out-of-range value for node signature */ +#define INVALID_NODE_SIG 0x10000 + +#define NODE_HTABLE_SIZE 512 /* Flags used to take different actions according to flag type * TIPC_WAIT_PEER_LINKS_DOWN: wait to see that peer's links are down @@ -55,14 +55,17 @@ * TIPC_DISTRIBUTE_NAME: publish or withdraw link state name type */ enum { + TIPC_MSG_EVT = 1, 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), - TIPC_WAKEUP_USERS = (1 << 5), - TIPC_WAKEUP_BCAST_USERS = (1 << 6), - TIPC_NOTIFY_LINK_UP = (1 << 7), - TIPC_NOTIFY_LINK_DOWN = (1 << 8) + TIPC_WAKEUP_BCAST_USERS = (1 << 5), + 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_RESET = (1 << 10) }; /** @@ -73,6 +76,7 @@ enum { * @oos_state: state tracker for handling OOS b'cast messages * @deferred_queue: deferred queue saved OOS b'cast message received from node * @reasm_buf: broadcast reassembly queue head from node + * @inputq_map: bitmap indicating which inqueues should be kicked * @recv_permitted: true if node is allowed to receive b'cast messages */ struct tipc_node_bclink { @@ -81,16 +85,22 @@ struct tipc_node_bclink { u32 last_sent; u32 oos_state; u32 deferred_size; - struct sk_buff_head deferred_queue; + struct sk_buff_head deferdq; struct sk_buff *reasm_buf; + int inputq_map; bool recv_permitted; }; /** * 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 + * @inputq: pointer to input queue containing messages for msg event + * @namedq: pointer to name table input queue with name table messages + * @curr_link: the link holding the node lock, if any * @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 @@ -98,6 +108,7 @@ struct tipc_node_bclink { * @list: links to adjacent nodes in sorted list of cluster's nodes * @working_links: number of working links to node (both active and standby) * @link_cnt: number of links to node + * @capabilities: bitmap, indicating peer node's functional capabilities * @signature: node instance identifier * @link_id: local and remote bearer ids of changing link, if any * @publ_list: list of publications @@ -105,41 +116,43 @@ struct tipc_node_bclink { */ struct tipc_node { u32 addr; + struct kref kref; spinlock_t lock; + struct net *net; struct hlist_node hash; + struct sk_buff_head *inputq; + struct sk_buff_head *namedq; struct tipc_link *active_links[2]; u32 act_mtus[2]; struct tipc_link *links[MAX_BEARERS]; - unsigned int action_flags; + int action_flags; struct tipc_node_bclink bclink; struct list_head list; int link_cnt; - int working_links; + u16 working_links; + u16 capabilities; u32 signature; u32 link_id; struct list_head publ_list; - struct sk_buff_head waiting_sks; struct list_head conn_sks; struct rcu_head rcu; }; -extern struct list_head tipc_node_list; - -struct tipc_node *tipc_node_find(u32 addr); -struct tipc_node *tipc_node_create(u32 addr); -void tipc_node_stop(void); +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); void tipc_node_detach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr); void tipc_node_link_down(struct tipc_node *n_ptr, struct tipc_link *l_ptr); void tipc_node_link_up(struct tipc_node *n_ptr, struct tipc_link *l_ptr); 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); +int tipc_node_get_linkname(struct net *net, u32 bearer_id, u32 node, + char *linkname, size_t len); void tipc_node_unlock(struct tipc_node *node); -int tipc_node_add_conn(u32 dnode, u32 port, u32 peer_port); -void tipc_node_remove_conn(u32 dnode, u32 port); +int tipc_node_add_conn(struct net *net, u32 dnode, u32 port, u32 peer_port); +void tipc_node_remove_conn(struct net *net, u32 dnode, u32 port); int tipc_nl_node_dump(struct sk_buff *skb, struct netlink_callback *cb); @@ -154,17 +167,19 @@ static inline bool tipc_node_blocked(struct tipc_node *node) TIPC_NOTIFY_NODE_DOWN | TIPC_WAIT_OWN_LINKS_DOWN)); } -static inline uint tipc_node_get_mtu(u32 addr, u32 selector) +static inline uint tipc_node_get_mtu(struct net *net, u32 addr, u32 selector) { struct tipc_node *node; u32 mtu; - node = tipc_node_find(addr); + 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; } diff --git a/net/tipc/server.c b/net/tipc/server.c index a538a02f869b..ab6183cdb121 100644 --- a/net/tipc/server.c +++ b/net/tipc/server.c @@ -35,12 +35,15 @@ #include "server.h" #include "core.h" +#include "socket.h" #include <net/sock.h> +#include <linux/module.h> /* Number of messages to send before rescheduling */ #define MAX_SEND_MSG_COUNT 25 #define MAX_RECV_MSG_COUNT 25 #define CF_CONNECTED 1 +#define CF_SERVER 2 #define sock2con(x) ((struct tipc_conn *)(x)->sk_user_data) @@ -87,9 +90,19 @@ static void tipc_clean_outqueues(struct tipc_conn *con); static void tipc_conn_kref_release(struct kref *kref) { struct tipc_conn *con = container_of(kref, struct tipc_conn, kref); + struct sockaddr_tipc *saddr = con->server->saddr; + struct socket *sock = con->sock; + struct sock *sk; - if (con->sock) { - tipc_sock_release_local(con->sock); + if (sock) { + sk = sock->sk; + if (test_bit(CF_SERVER, &con->flags)) { + __module_get(sock->ops->owner); + __module_get(sk->sk_prot_creator->owner); + } + saddr->scope = -TIPC_NODE_SCOPE; + kernel_bind(sock, (struct sockaddr *)saddr, sizeof(*saddr)); + sk_release_kernel(sk); con->sock = NULL; } @@ -255,7 +268,8 @@ static int tipc_receive_from_sock(struct tipc_conn *con) goto out_close; } - s->tipc_conn_recvmsg(con->conid, &addr, con->usr_data, buf, ret); + s->tipc_conn_recvmsg(sock_net(con->sock->sk), con->conid, &addr, + con->usr_data, buf, ret); kmem_cache_free(s->rcvbuf_cache, buf); @@ -279,7 +293,7 @@ static int tipc_accept_from_sock(struct tipc_conn *con) struct tipc_conn *newcon; int ret; - ret = tipc_sock_accept_local(sock, &newsock, O_NONBLOCK); + ret = kernel_accept(sock, &newsock, O_NONBLOCK); if (ret < 0) return ret; @@ -307,9 +321,12 @@ static struct socket *tipc_create_listen_sock(struct tipc_conn *con) struct socket *sock = NULL; int ret; - ret = tipc_sock_create_local(s->type, &sock); + ret = sock_create_kern(AF_TIPC, SOCK_SEQPACKET, 0, &sock); if (ret < 0) return NULL; + + sk_change_net(sock->sk, s->net); + ret = kernel_setsockopt(sock, SOL_TIPC, TIPC_IMPORTANCE, (char *)&s->imp, sizeof(s->imp)); if (ret < 0) @@ -335,11 +352,31 @@ static struct socket *tipc_create_listen_sock(struct tipc_conn *con) pr_err("Unknown socket type %d\n", s->type); goto create_err; } + + /* As server's listening socket owner and creator is the same module, + * we have to decrease TIPC module reference count to guarantee that + * it remains zero after the server socket is created, otherwise, + * executing "rmmod" command is unable to make TIPC module deleted + * after TIPC module is inserted successfully. + * + * However, the reference count is ever increased twice in + * sock_create_kern(): one is to increase the reference count of owner + * of TIPC socket's proto_ops struct; another is to increment the + * reference count of owner of TIPC proto struct. Therefore, we must + * decrement the module reference count twice to ensure that it keeps + * zero after server's listening socket is created. Of course, we + * must bump the module reference count twice as well before the socket + * is closed. + */ + module_put(sock->ops->owner); + module_put(sock->sk->sk_prot_creator->owner); + set_bit(CF_SERVER, &con->flags); + return sock; create_err: - sock_release(sock); - con->sock = NULL; + kernel_sock_shutdown(sock, SHUT_RDWR); + sk_release_kernel(sock->sk); return NULL; } diff --git a/net/tipc/server.h b/net/tipc/server.h index be817b0b547e..9015faedb1b0 100644 --- a/net/tipc/server.h +++ b/net/tipc/server.h @@ -36,7 +36,9 @@ #ifndef _TIPC_SERVER_H #define _TIPC_SERVER_H -#include "core.h" +#include <linux/idr.h> +#include <linux/tipc.h> +#include <net/net_namespace.h> #define TIPC_SERVER_NAME_LEN 32 @@ -45,6 +47,7 @@ * @conn_idr: identifier set of connection * @idr_lock: protect the connection identifier set * @idr_in_use: amount of allocated identifier entry + * @net: network namspace instance * @rcvbuf_cache: memory cache of server receive buffer * @rcv_wq: receive workqueue * @send_wq: send workqueue @@ -61,16 +64,18 @@ struct tipc_server { struct idr conn_idr; spinlock_t idr_lock; int idr_in_use; + struct net *net; struct kmem_cache *rcvbuf_cache; struct workqueue_struct *rcv_wq; struct workqueue_struct *send_wq; int max_rcvbuf_size; - void *(*tipc_conn_new) (int conid); - void (*tipc_conn_shutdown) (int conid, void *usr_data); - void (*tipc_conn_recvmsg) (int conid, struct sockaddr_tipc *addr, - void *usr_data, void *buf, size_t len); + void *(*tipc_conn_new)(int conid); + void (*tipc_conn_shutdown)(int conid, void *usr_data); + void (*tipc_conn_recvmsg)(struct net *net, int conid, + struct sockaddr_tipc *addr, void *usr_data, + void *buf, size_t len); struct sockaddr_tipc *saddr; - const char name[TIPC_SERVER_NAME_LEN]; + char name[TIPC_SERVER_NAME_LEN]; int imp; int type; }; diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 4731cad99d1c..ee90d74d7516 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -1,7 +1,7 @@ /* * net/tipc/socket.c: TIPC socket API * - * Copyright (c) 2001-2007, 2012-2014, Ericsson AB + * Copyright (c) 2001-2007, 2012-2015, Ericsson AB * Copyright (c) 2004-2008, 2010-2013, Wind River Systems * All rights reserved. * @@ -34,22 +34,24 @@ * POSSIBILITY OF SUCH DAMAGE. */ +#include <linux/rhashtable.h> #include "core.h" #include "name_table.h" #include "node.h" #include "link.h" -#include <linux/export.h> -#include "config.h" +#include "name_distr.h" #include "socket.h" -#define SS_LISTENING -1 /* socket is listening */ -#define SS_READY -2 /* socket is connectionless */ +#define SS_LISTENING -1 /* socket is listening */ +#define SS_READY -2 /* socket is connectionless */ -#define CONN_TIMEOUT_DEFAULT 8000 /* default connect timeout = 8s */ -#define CONN_PROBING_INTERVAL 3600000 /* [ms] => 1 h */ -#define TIPC_FWD_MSG 1 -#define TIPC_CONN_OK 0 -#define TIPC_CONN_PROBING 1 +#define CONN_TIMEOUT_DEFAULT 8000 /* default connect timeout = 8s */ +#define CONN_PROBING_INTERVAL msecs_to_jiffies(3600000) /* [ms] => 1 h */ +#define TIPC_FWD_MSG 1 +#define TIPC_CONN_OK 0 +#define TIPC_CONN_PROBING 1 +#define TIPC_MAX_PORT 0xffffffff +#define TIPC_MIN_PORT 1 /** * struct tipc_sock - TIPC socket structure @@ -59,21 +61,21 @@ * @conn_instance: TIPC instance used when connection was established * @published: non-zero if port has one or more associated names * @max_pkt: maximum packet size "hint" used when building messages sent by port - * @ref: unique reference to port in TIPC object registry + * @portid: unique port identity in TIPC socket hash table * @phdr: preformatted message header used when sending messages * @port_list: adjacent ports in TIPC's global list of ports * @publications: list of publications for port * @pub_count: total # of publications port has made during its lifetime * @probing_state: - * @probing_interval: - * @timer: - * @port: port - interacts with 'sk' and with the rest of the TIPC stack - * @peer_name: the peer of the connection, if any + * @probing_intv: * @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 * @link_cong: non-zero if owner must sleep because of link congestion * @sent_unacked: # messages sent by socket, and not yet acked by peer * @rcv_unacked: # messages read by user, but not yet acked back to peer + * @remote: 'connected' peer for dgram/rdm + * @node: hash table node + * @rcu: rcu struct for tipc_sock */ struct tipc_sock { struct sock sk; @@ -82,19 +84,21 @@ struct tipc_sock { u32 conn_instance; int published; u32 max_pkt; - u32 ref; + u32 portid; struct tipc_msg phdr; struct list_head sock_list; struct list_head publications; u32 pub_count; u32 probing_state; - u32 probing_interval; - struct timer_list timer; + unsigned long probing_intv; uint conn_timeout; atomic_t dupl_rcvcnt; bool link_cong; uint sent_unacked; uint rcv_unacked; + struct sockaddr_tipc remote; + struct rhash_head node; + struct rcu_head rcu; }; static int tipc_backlog_rcv(struct sock *sk, struct sk_buff *skb); @@ -103,23 +107,22 @@ static void tipc_write_space(struct sock *sk); static int tipc_release(struct socket *sock); static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags); static int tipc_wait_for_sndmsg(struct socket *sock, long *timeo_p); -static void tipc_sk_timeout(unsigned long ref); +static void tipc_sk_timeout(unsigned long data); static int tipc_sk_publish(struct tipc_sock *tsk, uint scope, struct tipc_name_seq const *seq); static int tipc_sk_withdraw(struct tipc_sock *tsk, uint scope, struct tipc_name_seq const *seq); -static u32 tipc_sk_ref_acquire(struct tipc_sock *tsk); -static void tipc_sk_ref_discard(u32 ref); -static struct tipc_sock *tipc_sk_get(u32 ref); -static struct tipc_sock *tipc_sk_get_next(u32 *ref); -static void tipc_sk_put(struct tipc_sock *tsk); +static struct tipc_sock *tipc_sk_lookup(struct net *net, u32 portid); +static int tipc_sk_insert(struct tipc_sock *tsk); +static void tipc_sk_remove(struct tipc_sock *tsk); +static int __tipc_send_stream(struct socket *sock, struct msghdr *m, + size_t dsz); +static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dsz); static const struct proto_ops packet_ops; static const struct proto_ops stream_ops; static const struct proto_ops msg_ops; - static struct proto tipc_proto; -static struct proto tipc_proto_kern; static const struct nla_policy tipc_nl_sock_policy[TIPC_NLA_SOCK_MAX + 1] = { [TIPC_NLA_SOCK_UNSPEC] = { .type = NLA_UNSPEC }, @@ -129,6 +132,8 @@ static const struct nla_policy tipc_nl_sock_policy[TIPC_NLA_SOCK_MAX + 1] = { [TIPC_NLA_SOCK_HAS_PUBL] = { .type = NLA_FLAG } }; +static const struct rhashtable_params tsk_rht_params; + /* * Revised TIPC socket locking policy: * @@ -174,6 +179,11 @@ static const struct nla_policy tipc_nl_sock_policy[TIPC_NLA_SOCK_MAX + 1] = { * - port reference */ +static u32 tsk_own_node(struct tipc_sock *tsk) +{ + return msg_prevnode(&tsk->phdr); +} + static u32 tsk_peer_node(struct tipc_sock *tsk) { return msg_destnode(&tsk->phdr); @@ -246,10 +256,11 @@ static void tsk_rej_rx_queue(struct sock *sk) { struct sk_buff *skb; u32 dnode; + u32 own_node = tsk_own_node(tipc_sk(sk)); while ((skb = __skb_dequeue(&sk->sk_receive_queue))) { - if (tipc_msg_reverse(skb, &dnode, TIPC_ERR_NO_PORT)) - tipc_link_xmit_skb(skb, dnode, 0); + if (tipc_msg_reverse(own_node, skb, &dnode, TIPC_ERR_NO_PORT)) + tipc_link_xmit_skb(sock_net(sk), skb, dnode, 0); } } @@ -260,6 +271,7 @@ static void tsk_rej_rx_queue(struct sock *sk) */ static bool tsk_peer_msg(struct tipc_sock *tsk, struct tipc_msg *msg) { + struct tipc_net *tn = net_generic(sock_net(&tsk->sk), tipc_net_id); u32 peer_port = tsk_peer_port(tsk); u32 orig_node; u32 peer_node; @@ -276,10 +288,10 @@ static bool tsk_peer_msg(struct tipc_sock *tsk, struct tipc_msg *msg) if (likely(orig_node == peer_node)) return true; - if (!orig_node && (peer_node == tipc_own_addr)) + if (!orig_node && (peer_node == tn->own_addr)) return true; - if (!peer_node && (orig_node == tipc_own_addr)) + if (!peer_node && (orig_node == tn->own_addr)) return true; return false; @@ -300,12 +312,12 @@ static bool tsk_peer_msg(struct tipc_sock *tsk, struct tipc_msg *msg) static int tipc_sk_create(struct net *net, struct socket *sock, int protocol, int kern) { + struct tipc_net *tn; const struct proto_ops *ops; socket_state state; struct sock *sk; struct tipc_sock *tsk; struct tipc_msg *msg; - u32 ref; /* Validate arguments */ if (unlikely(protocol != 0)) @@ -330,33 +342,28 @@ static int tipc_sk_create(struct net *net, struct socket *sock, } /* Allocate socket's protocol area */ - if (!kern) - sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto); - else - sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto_kern); - + sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto); if (sk == NULL) return -ENOMEM; tsk = tipc_sk(sk); - ref = tipc_sk_ref_acquire(tsk); - if (!ref) { - pr_warn("Socket create failed; reference table exhausted\n"); - return -ENOMEM; - } tsk->max_pkt = MAX_PKT_DEFAULT; - tsk->ref = ref; INIT_LIST_HEAD(&tsk->publications); msg = &tsk->phdr; - tipc_msg_init(msg, TIPC_LOW_IMPORTANCE, TIPC_NAMED_MSG, + tn = net_generic(sock_net(sk), tipc_net_id); + tipc_msg_init(tn->own_addr, msg, TIPC_LOW_IMPORTANCE, TIPC_NAMED_MSG, NAMED_H_SIZE, 0); - msg_set_origport(msg, ref); /* Finish initializing socket data structures */ sock->ops = ops; sock->state = state; sock_init_data(sock, sk); - k_init_timer(&tsk->timer, (Handler)tipc_sk_timeout, ref); + if (tipc_sk_insert(tsk)) { + pr_warn("Socket create failed; port numbrer exhausted\n"); + return -EINVAL; + } + msg_set_origport(msg, tsk->portid); + setup_timer(&sk->sk_timer, tipc_sk_timeout, (unsigned long)tsk); sk->sk_backlog_rcv = tipc_backlog_rcv; sk->sk_rcvbuf = sysctl_tipc_rmem[1]; sk->sk_data_ready = tipc_data_ready; @@ -373,73 +380,11 @@ static int tipc_sk_create(struct net *net, struct socket *sock, return 0; } -/** - * tipc_sock_create_local - create TIPC socket from inside TIPC module - * @type: socket type - SOCK_RDM or SOCK_SEQPACKET - * - * We cannot use sock_creat_kern here because it bumps module user count. - * Since socket owner and creator is the same module we must make sure - * that module count remains zero for module local sockets, otherwise - * we cannot do rmmod. - * - * Returns 0 on success, errno otherwise - */ -int tipc_sock_create_local(int type, struct socket **res) -{ - int rc; - - rc = sock_create_lite(AF_TIPC, type, 0, res); - if (rc < 0) { - pr_err("Failed to create kernel socket\n"); - return rc; - } - tipc_sk_create(&init_net, *res, 0, 1); - - return 0; -} - -/** - * tipc_sock_release_local - release socket created by tipc_sock_create_local - * @sock: the socket to be released. - * - * Module reference count is not incremented when such sockets are created, - * so we must keep it from being decremented when they are released. - */ -void tipc_sock_release_local(struct socket *sock) +static void tipc_sk_callback(struct rcu_head *head) { - tipc_release(sock); - sock->ops = NULL; - sock_release(sock); -} + struct tipc_sock *tsk = container_of(head, struct tipc_sock, rcu); -/** - * tipc_sock_accept_local - accept a connection on a socket created - * with tipc_sock_create_local. Use this function to avoid that - * module reference count is inadvertently incremented. - * - * @sock: the accepting socket - * @newsock: reference to the new socket to be created - * @flags: socket flags - */ - -int tipc_sock_accept_local(struct socket *sock, struct socket **newsock, - int flags) -{ - struct sock *sk = sock->sk; - int ret; - - ret = sock_create_lite(sk->sk_family, sk->sk_type, - sk->sk_protocol, newsock); - if (ret < 0) - return ret; - - ret = tipc_accept(sock, *newsock, flags); - if (ret < 0) { - sock_release(*newsock); - return ret; - } - (*newsock)->ops = sock->ops; - return ret; + sock_put(&tsk->sk); } /** @@ -461,9 +406,10 @@ int tipc_sock_accept_local(struct socket *sock, struct socket **newsock, static int tipc_release(struct socket *sock) { struct sock *sk = sock->sk; + struct net *net; struct tipc_sock *tsk; struct sk_buff *skb; - u32 dnode; + u32 dnode, probing_state; /* * Exit if socket isn't fully initialized (occurs when a failed accept() @@ -472,6 +418,7 @@ static int tipc_release(struct socket *sock) if (sk == NULL) return 0; + net = sock_net(sk); tsk = tipc_sk(sk); lock_sock(sk); @@ -491,26 +438,29 @@ static int tipc_release(struct socket *sock) (sock->state == SS_CONNECTED)) { sock->state = SS_DISCONNECTING; tsk->connected = 0; - tipc_node_remove_conn(dnode, tsk->ref); + tipc_node_remove_conn(net, dnode, tsk->portid); } - if (tipc_msg_reverse(skb, &dnode, TIPC_ERR_NO_PORT)) - tipc_link_xmit_skb(skb, dnode, 0); + if (tipc_msg_reverse(tsk_own_node(tsk), skb, &dnode, + TIPC_ERR_NO_PORT)) + tipc_link_xmit_skb(net, skb, dnode, 0); } } tipc_sk_withdraw(tsk, 0, NULL); - tipc_sk_ref_discard(tsk->ref); - k_cancel_timer(&tsk->timer); + probing_state = tsk->probing_state; + if (del_timer_sync(&sk->sk_timer) && + probing_state != TIPC_CONN_PROBING) + sock_put(sk); + tipc_sk_remove(tsk); if (tsk->connected) { - skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, TIPC_CONN_MSG, - SHORT_H_SIZE, 0, dnode, tipc_own_addr, - tsk_peer_port(tsk), - tsk->ref, TIPC_ERR_NO_PORT); + skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, + TIPC_CONN_MSG, SHORT_H_SIZE, 0, dnode, + tsk_own_node(tsk), tsk_peer_port(tsk), + tsk->portid, TIPC_ERR_NO_PORT); if (skb) - tipc_link_xmit_skb(skb, dnode, tsk->ref); - tipc_node_remove_conn(dnode, tsk->ref); + tipc_link_xmit_skb(net, skb, dnode, tsk->portid); + tipc_node_remove_conn(net, dnode, tsk->portid); } - k_term_timer(&tsk->timer); /* Discard any remaining (connection-based) messages in receive queue */ __skb_queue_purge(&sk->sk_receive_queue); @@ -518,7 +468,8 @@ static int tipc_release(struct socket *sock) /* Reject any messages that accumulated in backlog queue */ sock->state = SS_DISCONNECTING; release_sock(sk); - sock_put(sk); + + call_rcu(&tsk->rcu, tipc_sk_callback); sock->sk = NULL; return 0; @@ -602,6 +553,7 @@ static int tipc_getname(struct socket *sock, struct sockaddr *uaddr, { struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; struct tipc_sock *tsk = tipc_sk(sock->sk); + struct tipc_net *tn = net_generic(sock_net(sock->sk), tipc_net_id); memset(addr, 0, sizeof(*addr)); if (peer) { @@ -611,8 +563,8 @@ static int tipc_getname(struct socket *sock, struct sockaddr *uaddr, addr->addr.id.ref = tsk_peer_port(tsk); addr->addr.id.node = tsk_peer_node(tsk); } else { - addr->addr.id.ref = tsk->ref; - addr->addr.id.node = tipc_own_addr; + addr->addr.id.ref = tsk->portid; + addr->addr.id.node = tn->own_addr; } *uaddr_len = sizeof(*addr); @@ -711,8 +663,11 @@ static int tipc_sendmcast(struct socket *sock, struct tipc_name_seq *seq, struct msghdr *msg, size_t dsz, long timeo) { struct sock *sk = sock->sk; - struct tipc_msg *mhdr = &tipc_sk(sk)->phdr; - struct sk_buff_head head; + struct tipc_sock *tsk = tipc_sk(sk); + struct net *net = sock_net(sk); + struct tipc_msg *mhdr = &tsk->phdr; + struct sk_buff_head *pktchain = &sk->sk_write_queue; + struct iov_iter save = msg->msg_iter; uint mtu; int rc; @@ -727,83 +682,97 @@ static int tipc_sendmcast(struct socket *sock, struct tipc_name_seq *seq, new_mtu: mtu = tipc_bclink_get_mtu(); - __skb_queue_head_init(&head); - rc = tipc_msg_build(mhdr, msg, 0, dsz, mtu, &head); + rc = tipc_msg_build(mhdr, msg, 0, dsz, mtu, pktchain); if (unlikely(rc < 0)) return rc; do { - rc = tipc_bclink_xmit(&head); + rc = tipc_bclink_xmit(net, pktchain); if (likely(rc >= 0)) { rc = dsz; break; } - if (rc == -EMSGSIZE) + if (rc == -EMSGSIZE) { + msg->msg_iter = save; goto new_mtu; + } if (rc != -ELINKCONG) break; tipc_sk(sk)->link_cong = 1; rc = tipc_wait_for_sndmsg(sock, &timeo); if (rc) - __skb_queue_purge(&head); + __skb_queue_purge(pktchain); } while (!rc); return rc; } -/* tipc_sk_mcast_rcv - Deliver multicast message to all destination sockets +/** + * tipc_sk_mcast_rcv - Deliver multicast messages to all destination sockets + * @arrvq: queue with arriving messages, to be cloned after destination lookup + * @inputq: queue with cloned messages, delivered to socket after dest lookup + * + * Multi-threaded: parallel calls with reference to same queues may occur */ -void tipc_sk_mcast_rcv(struct sk_buff *buf) +void tipc_sk_mcast_rcv(struct net *net, struct sk_buff_head *arrvq, + struct sk_buff_head *inputq) { - struct tipc_msg *msg = buf_msg(buf); - struct tipc_port_list dports = {0, NULL, }; - struct tipc_port_list *item; - struct sk_buff *b; - uint i, last, dst = 0; + struct tipc_msg *msg; + struct tipc_plist dports; + u32 portid; u32 scope = TIPC_CLUSTER_SCOPE; - - if (in_own_node(msg_orignode(msg))) - scope = TIPC_NODE_SCOPE; - - /* Create destination port list: */ - tipc_nametbl_mc_translate(msg_nametype(msg), - msg_namelower(msg), - msg_nameupper(msg), - scope, - &dports); - last = dports.count; - if (!last) { - kfree_skb(buf); - return; - } - - for (item = &dports; item; item = item->next) { - for (i = 0; i < PLSIZE && ++dst <= last; i++) { - b = (dst != last) ? skb_clone(buf, GFP_ATOMIC) : buf; - if (!b) { - pr_warn("Failed do clone mcast rcv buffer\n"); + struct sk_buff_head tmpq; + uint hsz; + struct sk_buff *skb, *_skb; + + __skb_queue_head_init(&tmpq); + tipc_plist_init(&dports); + + skb = tipc_skb_peek(arrvq, &inputq->lock); + for (; skb; skb = tipc_skb_peek(arrvq, &inputq->lock)) { + msg = buf_msg(skb); + hsz = skb_headroom(skb) + msg_hdr_sz(msg); + + if (in_own_node(net, msg_orignode(msg))) + scope = TIPC_NODE_SCOPE; + + /* Create destination port list and message clones: */ + tipc_nametbl_mc_translate(net, + msg_nametype(msg), msg_namelower(msg), + msg_nameupper(msg), scope, &dports); + portid = tipc_plist_pop(&dports); + for (; portid; portid = tipc_plist_pop(&dports)) { + _skb = __pskb_copy(skb, hsz, GFP_ATOMIC); + if (_skb) { + msg_set_destport(buf_msg(_skb), portid); + __skb_queue_tail(&tmpq, _skb); continue; } - msg_set_destport(msg, item->ports[i]); - tipc_sk_rcv(b); + pr_warn("Failed to clone mcast rcv buffer\n"); + } + /* Append to inputq if not already done by other thread */ + spin_lock_bh(&inputq->lock); + if (skb_peek(arrvq) == skb) { + skb_queue_splice_tail_init(&tmpq, inputq); + kfree_skb(__skb_dequeue(arrvq)); } + spin_unlock_bh(&inputq->lock); + __skb_queue_purge(&tmpq); + kfree_skb(skb); } - tipc_port_list_free(&dports); + tipc_sk_rcv(net, inputq); } /** * tipc_sk_proto_rcv - receive a connection mng protocol message * @tsk: receiving socket - * @dnode: node to send response message to, if any - * @buf: buffer containing protocol message - * Returns 0 (TIPC_OK) if message was consumed, 1 (TIPC_FWD_MSG) if - * (CONN_PROBE_REPLY) message should be forwarded. + * @skb: pointer to message buffer. Set to NULL if buffer is consumed. */ -static int tipc_sk_proto_rcv(struct tipc_sock *tsk, u32 *dnode, - struct sk_buff *buf) +static void tipc_sk_proto_rcv(struct tipc_sock *tsk, struct sk_buff **skb) { - struct tipc_msg *msg = buf_msg(buf); + struct tipc_msg *msg = buf_msg(*skb); int conn_cong; - + u32 dnode; + u32 own_node = tsk_own_node(tsk); /* Ignore if connection cannot be validated: */ if (!tsk_peer_msg(tsk, msg)) goto exit; @@ -816,15 +785,15 @@ static int tipc_sk_proto_rcv(struct tipc_sock *tsk, u32 *dnode, if (conn_cong) tsk->sk.sk_write_space(&tsk->sk); } else if (msg_type(msg) == CONN_PROBE) { - if (!tipc_msg_reverse(buf, dnode, TIPC_OK)) - return TIPC_OK; - msg_set_type(msg, CONN_PROBE_REPLY); - return TIPC_FWD_MSG; + if (tipc_msg_reverse(own_node, *skb, &dnode, TIPC_OK)) { + msg_set_type(msg, CONN_PROBE_REPLY); + return; + } } /* Do nothing if msg_type() == CONN_PROBE_REPLY */ exit: - kfree_skb(buf); - return TIPC_OK; + kfree_skb(*skb); + *skb = NULL; } static int tipc_wait_for_sndmsg(struct socket *sock, long *timeo_p) @@ -854,7 +823,6 @@ static int tipc_wait_for_sndmsg(struct socket *sock, long *timeo_p) /** * tipc_sendmsg - send message in connectionless manner - * @iocb: if NULL, indicates that socket lock is already held * @sock: socket structure * @m: message to send * @dsz: amount of user data to be sent @@ -866,58 +834,63 @@ static int tipc_wait_for_sndmsg(struct socket *sock, long *timeo_p) * * Returns the number of bytes sent on success, or errno otherwise */ -static int tipc_sendmsg(struct kiocb *iocb, struct socket *sock, +static int tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dsz) { + struct sock *sk = sock->sk; + int ret; + + lock_sock(sk); + ret = __tipc_sendmsg(sock, m, dsz); + release_sock(sk); + + return ret; +} + +static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dsz) +{ DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name); struct sock *sk = sock->sk; struct tipc_sock *tsk = tipc_sk(sk); + struct net *net = sock_net(sk); struct tipc_msg *mhdr = &tsk->phdr; u32 dnode, dport; - struct sk_buff_head head; + struct sk_buff_head *pktchain = &sk->sk_write_queue; struct sk_buff *skb; - struct tipc_name_seq *seq = &dest->addr.nameseq; + struct tipc_name_seq *seq; + struct iov_iter save; u32 mtu; long timeo; int rc; - if (unlikely(!dest)) - return -EDESTADDRREQ; - - if (unlikely((m->msg_namelen < sizeof(*dest)) || - (dest->family != AF_TIPC))) - return -EINVAL; - if (dsz > TIPC_MAX_USER_MSG_SIZE) return -EMSGSIZE; - - if (iocb) - lock_sock(sk); - + if (unlikely(!dest)) { + if (tsk->connected && sock->state == SS_READY) + dest = &tsk->remote; + else + return -EDESTADDRREQ; + } else if (unlikely(m->msg_namelen < sizeof(*dest)) || + dest->family != AF_TIPC) { + return -EINVAL; + } if (unlikely(sock->state != SS_READY)) { - if (sock->state == SS_LISTENING) { - rc = -EPIPE; - goto exit; - } - if (sock->state != SS_UNCONNECTED) { - rc = -EISCONN; - goto exit; - } - if (tsk->published) { - rc = -EOPNOTSUPP; - goto exit; - } + if (sock->state == SS_LISTENING) + return -EPIPE; + if (sock->state != SS_UNCONNECTED) + return -EISCONN; + if (tsk->published) + return -EOPNOTSUPP; if (dest->addrtype == TIPC_ADDR_NAME) { tsk->conn_type = dest->addr.name.name.type; tsk->conn_instance = dest->addr.name.name.instance; } } - + seq = &dest->addr.nameseq; timeo = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); if (dest->addrtype == TIPC_ADDR_MCAST) { - rc = tipc_sendmcast(sock, seq, m, dsz, timeo); - goto exit; + return tipc_sendmcast(sock, seq, m, dsz, timeo); } else if (dest->addrtype == TIPC_ADDR_NAME) { u32 type = dest->addr.name.name.type; u32 inst = dest->addr.name.name.instance; @@ -929,13 +902,11 @@ static int tipc_sendmsg(struct kiocb *iocb, struct socket *sock, msg_set_nametype(mhdr, type); msg_set_nameinst(mhdr, inst); msg_set_lookup_scope(mhdr, tipc_addr_scope(domain)); - dport = tipc_nametbl_translate(type, inst, &dnode); + dport = tipc_nametbl_translate(net, type, inst, &dnode); msg_set_destnode(mhdr, dnode); msg_set_destport(mhdr, dport); - if (unlikely(!dport && !dnode)) { - rc = -EHOSTUNREACH; - goto exit; - } + if (unlikely(!dport && !dnode)) + return -EHOSTUNREACH; } else if (dest->addrtype == TIPC_ADDR_ID) { dnode = dest->addr.id.node; msg_set_type(mhdr, TIPC_DIRECT_MSG); @@ -945,35 +916,34 @@ static int tipc_sendmsg(struct kiocb *iocb, struct socket *sock, msg_set_hdr_sz(mhdr, BASIC_H_SIZE); } + save = m->msg_iter; new_mtu: - mtu = tipc_node_get_mtu(dnode, tsk->ref); - __skb_queue_head_init(&head); - rc = tipc_msg_build(mhdr, m, 0, dsz, mtu, &head); + mtu = tipc_node_get_mtu(net, dnode, tsk->portid); + rc = tipc_msg_build(mhdr, m, 0, dsz, mtu, pktchain); if (rc < 0) - goto exit; + return rc; do { - skb = skb_peek(&head); + skb = skb_peek(pktchain); TIPC_SKB_CB(skb)->wakeup_pending = tsk->link_cong; - rc = tipc_link_xmit(&head, dnode, tsk->ref); + rc = tipc_link_xmit(net, pktchain, dnode, tsk->portid); if (likely(rc >= 0)) { if (sock->state != SS_READY) sock->state = SS_CONNECTING; rc = dsz; break; } - if (rc == -EMSGSIZE) + if (rc == -EMSGSIZE) { + m->msg_iter = save; goto new_mtu; + } if (rc != -ELINKCONG) break; tsk->link_cong = 1; rc = tipc_wait_for_sndmsg(sock, &timeo); if (rc) - __skb_queue_purge(&head); + __skb_queue_purge(pktchain); } while (!rc); -exit: - if (iocb) - release_sock(sk); return rc; } @@ -1010,7 +980,6 @@ static int tipc_wait_for_sndpkt(struct socket *sock, long *timeo_p) /** * tipc_send_stream - send stream-oriented data - * @iocb: (unused) * @sock: socket structure * @m: data to send * @dsz: total length of data to be transmitted @@ -1020,23 +989,36 @@ static int tipc_wait_for_sndpkt(struct socket *sock, long *timeo_p) * Returns the number of bytes sent on success (or partial success), * or errno if no data sent */ -static int tipc_send_stream(struct kiocb *iocb, struct socket *sock, - struct msghdr *m, size_t dsz) +static int tipc_send_stream(struct socket *sock, struct msghdr *m, size_t dsz) +{ + struct sock *sk = sock->sk; + int ret; + + lock_sock(sk); + ret = __tipc_send_stream(sock, m, dsz); + release_sock(sk); + + return ret; +} + +static int __tipc_send_stream(struct socket *sock, struct msghdr *m, size_t dsz) { struct sock *sk = sock->sk; + struct net *net = sock_net(sk); struct tipc_sock *tsk = tipc_sk(sk); struct tipc_msg *mhdr = &tsk->phdr; - struct sk_buff_head head; + struct sk_buff_head *pktchain = &sk->sk_write_queue; DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name); - u32 ref = tsk->ref; + u32 portid = tsk->portid; int rc = -EINVAL; long timeo; u32 dnode; uint mtu, send, sent = 0; + struct iov_iter save; /* Handle implied connection establishment */ if (unlikely(dest)) { - rc = tipc_sendmsg(iocb, sock, m, dsz); + rc = __tipc_sendmsg(sock, m, dsz); if (dsz && (dsz == rc)) tsk->sent_unacked = 1; return rc; @@ -1044,30 +1026,26 @@ static int tipc_send_stream(struct kiocb *iocb, struct socket *sock, if (dsz > (uint)INT_MAX) return -EMSGSIZE; - if (iocb) - lock_sock(sk); - if (unlikely(sock->state != SS_CONNECTED)) { if (sock->state == SS_DISCONNECTING) - rc = -EPIPE; + return -EPIPE; else - rc = -ENOTCONN; - goto exit; + return -ENOTCONN; } timeo = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); dnode = tsk_peer_node(tsk); next: + save = m->msg_iter; mtu = tsk->max_pkt; send = min_t(uint, dsz - sent, TIPC_MAX_USER_MSG_SIZE); - __skb_queue_head_init(&head); - rc = tipc_msg_build(mhdr, m, sent, send, mtu, &head); + rc = tipc_msg_build(mhdr, m, sent, send, mtu, pktchain); if (unlikely(rc < 0)) - goto exit; + return rc; do { if (likely(!tsk_conn_cong(tsk))) { - rc = tipc_link_xmit(&head, dnode, ref); + rc = tipc_link_xmit(net, pktchain, dnode, portid); if (likely(!rc)) { tsk->sent_unacked++; sent += send; @@ -1076,7 +1054,9 @@ next: goto next; } if (rc == -EMSGSIZE) { - tsk->max_pkt = tipc_node_get_mtu(dnode, ref); + tsk->max_pkt = tipc_node_get_mtu(net, dnode, + portid); + m->msg_iter = save; goto next; } if (rc != -ELINKCONG) @@ -1085,17 +1065,14 @@ next: } rc = tipc_wait_for_sndpkt(sock, &timeo); if (rc) - __skb_queue_purge(&head); + __skb_queue_purge(pktchain); } while (!rc); -exit: - if (iocb) - release_sock(sk); + return sent ? sent : rc; } /** * tipc_send_packet - send a connection-oriented message - * @iocb: if NULL, indicates that socket lock is already held * @sock: socket structure * @m: message to send * @dsz: length of data to be transmitted @@ -1104,13 +1081,12 @@ exit: * * Returns the number of bytes sent on success, or errno otherwise */ -static int tipc_send_packet(struct kiocb *iocb, struct socket *sock, - struct msghdr *m, size_t dsz) +static int tipc_send_packet(struct socket *sock, struct msghdr *m, size_t dsz) { if (dsz > TIPC_MAX_USER_MSG_SIZE) return -EMSGSIZE; - return tipc_send_stream(iocb, sock, m, dsz); + return tipc_send_stream(sock, m, dsz); } /* tipc_sk_finish_conn - complete the setup of a connection @@ -1118,6 +1094,8 @@ static int tipc_send_packet(struct kiocb *iocb, struct socket *sock, static void tipc_sk_finish_conn(struct tipc_sock *tsk, u32 peer_port, u32 peer_node) { + struct sock *sk = &tsk->sk; + struct net *net = sock_net(sk); struct tipc_msg *msg = &tsk->phdr; msg_set_destnode(msg, peer_node); @@ -1126,12 +1104,12 @@ static void tipc_sk_finish_conn(struct tipc_sock *tsk, u32 peer_port, msg_set_lookup_scope(msg, 0); msg_set_hdr_sz(msg, SHORT_H_SIZE); - tsk->probing_interval = CONN_PROBING_INTERVAL; + tsk->probing_intv = CONN_PROBING_INTERVAL; tsk->probing_state = TIPC_CONN_OK; tsk->connected = 1; - k_start_timer(&tsk->timer, tsk->probing_interval); - tipc_node_add_conn(peer_node, tsk->ref, peer_port); - tsk->max_pkt = tipc_node_get_mtu(peer_node, tsk->ref); + sk_reset_timer(sk, &sk->sk_timer, jiffies + tsk->probing_intv); + tipc_node_add_conn(net, peer_node, tsk->portid, peer_port); + tsk->max_pkt = tipc_node_get_mtu(net, peer_node, tsk->portid); } /** @@ -1230,6 +1208,7 @@ static int tipc_sk_anc_data_recv(struct msghdr *m, struct tipc_msg *msg, static void tipc_sk_send_ack(struct tipc_sock *tsk, uint ack) { + struct net *net = sock_net(&tsk->sk); struct sk_buff *skb = NULL; struct tipc_msg *msg; u32 peer_port = tsk_peer_port(tsk); @@ -1237,13 +1216,14 @@ static void tipc_sk_send_ack(struct tipc_sock *tsk, uint ack) if (!tsk->connected) return; - skb = tipc_msg_create(CONN_MANAGER, CONN_ACK, INT_H_SIZE, 0, dnode, - tipc_own_addr, peer_port, tsk->ref, TIPC_OK); + skb = tipc_msg_create(CONN_MANAGER, CONN_ACK, INT_H_SIZE, 0, + dnode, tsk_own_node(tsk), peer_port, + tsk->portid, TIPC_OK); if (!skb) return; msg = buf_msg(skb); msg_set_msgcnt(msg, ack); - tipc_link_xmit_skb(skb, dnode, msg_link_selector(msg)); + tipc_link_xmit_skb(net, skb, dnode, msg_link_selector(msg)); } static int tipc_wait_for_rcvmsg(struct socket *sock, long *timeop) @@ -1267,12 +1247,12 @@ static int tipc_wait_for_rcvmsg(struct socket *sock, long *timeop) err = 0; if (!skb_queue_empty(&sk->sk_receive_queue)) break; - err = sock_intr_errno(timeo); - if (signal_pending(current)) - break; err = -EAGAIN; if (!timeo) break; + err = sock_intr_errno(timeo); + if (signal_pending(current)) + break; } finish_wait(sk_sleep(sk), &wait); *timeop = timeo; @@ -1281,7 +1261,6 @@ static int tipc_wait_for_rcvmsg(struct socket *sock, long *timeop) /** * tipc_recvmsg - receive packet-oriented message - * @iocb: (unused) * @m: descriptor for message info * @buf_len: total size of user buffer area * @flags: receive flags @@ -1291,8 +1270,8 @@ static int tipc_wait_for_rcvmsg(struct socket *sock, long *timeop) * * Returns size of returned message data, errno otherwise */ -static int tipc_recvmsg(struct kiocb *iocb, struct socket *sock, - struct msghdr *m, size_t buf_len, int flags) +static int tipc_recvmsg(struct socket *sock, struct msghdr *m, size_t buf_len, + int flags) { struct sock *sk = sock->sk; struct tipc_sock *tsk = tipc_sk(sk); @@ -1376,7 +1355,6 @@ exit: /** * tipc_recv_stream - receive stream-oriented data - * @iocb: (unused) * @m: descriptor for message info * @buf_len: total size of user buffer area * @flags: receive flags @@ -1386,8 +1364,8 @@ exit: * * Returns size of returned message data, errno otherwise */ -static int tipc_recv_stream(struct kiocb *iocb, struct socket *sock, - struct msghdr *m, size_t buf_len, int flags) +static int tipc_recv_stream(struct socket *sock, struct msghdr *m, + size_t buf_len, int flags) { struct sock *sk = sock->sk; struct tipc_sock *tsk = tipc_sk(sk); @@ -1529,15 +1507,16 @@ static void tipc_data_ready(struct sock *sk) /** * filter_connect - Handle all incoming messages for a connection-based socket * @tsk: TIPC socket - * @msg: message + * @skb: pointer to message buffer. Set to NULL if buffer is consumed * * Returns 0 (TIPC_OK) if everything ok, -TIPC_ERR_NO_PORT otherwise */ -static int filter_connect(struct tipc_sock *tsk, struct sk_buff **buf) +static int filter_connect(struct tipc_sock *tsk, struct sk_buff **skb) { struct sock *sk = &tsk->sk; + struct net *net = sock_net(sk); struct socket *sock = sk->sk_socket; - struct tipc_msg *msg = buf_msg(*buf); + struct tipc_msg *msg = buf_msg(*skb); int retval = -TIPC_ERR_NO_PORT; if (msg_mcast(msg)) @@ -1551,8 +1530,8 @@ static int filter_connect(struct tipc_sock *tsk, struct sk_buff **buf) sock->state = SS_DISCONNECTING; tsk->connected = 0; /* let timer expire on it's own */ - tipc_node_remove_conn(tsk_peer_node(tsk), - tsk->ref); + tipc_node_remove_conn(net, tsk_peer_node(tsk), + tsk->portid); } retval = TIPC_OK; } @@ -1587,8 +1566,8 @@ static int filter_connect(struct tipc_sock *tsk, struct sk_buff **buf) * connect() routine if sleeping. */ if (msg_data_sz(msg) == 0) { - kfree_skb(*buf); - *buf = NULL; + kfree_skb(*skb); + *skb = NULL; if (waitqueue_active(sk_sleep(sk))) wake_up_interruptible(sk_sleep(sk)); } @@ -1640,32 +1619,33 @@ static unsigned int rcvbuf_limit(struct sock *sk, struct sk_buff *buf) /** * filter_rcv - validate incoming message * @sk: socket - * @buf: message + * @skb: pointer to message. Set to NULL if buffer is consumed. * * Enqueues message on receive queue if acceptable; optionally handles * disconnect indication for a connected socket. * - * Called with socket lock already taken; port lock may also be taken. + * Called with socket lock already taken * - * Returns 0 (TIPC_OK) if message was consumed, -TIPC error code if message - * to be rejected, 1 (TIPC_FWD_MSG) if (CONN_MANAGER) message to be forwarded + * Returns 0 (TIPC_OK) if message was ok, -TIPC error code if rejected */ -static int filter_rcv(struct sock *sk, struct sk_buff *buf) +static int filter_rcv(struct sock *sk, struct sk_buff **skb) { struct socket *sock = sk->sk_socket; struct tipc_sock *tsk = tipc_sk(sk); - struct tipc_msg *msg = buf_msg(buf); - unsigned int limit = rcvbuf_limit(sk, buf); - u32 onode; + struct tipc_msg *msg = buf_msg(*skb); + unsigned int limit = rcvbuf_limit(sk, *skb); int rc = TIPC_OK; - if (unlikely(msg_user(msg) == CONN_MANAGER)) - return tipc_sk_proto_rcv(tsk, &onode, buf); + if (unlikely(msg_user(msg) == CONN_MANAGER)) { + tipc_sk_proto_rcv(tsk, skb); + return TIPC_OK; + } if (unlikely(msg_user(msg) == SOCK_WAKEUP)) { - kfree_skb(buf); + kfree_skb(*skb); tsk->link_cong = 0; sk->sk_write_space(sk); + *skb = NULL; return TIPC_OK; } @@ -1677,21 +1657,22 @@ static int filter_rcv(struct sock *sk, struct sk_buff *buf) if (msg_connected(msg)) return -TIPC_ERR_NO_PORT; } else { - rc = filter_connect(tsk, &buf); - if (rc != TIPC_OK || buf == NULL) + rc = filter_connect(tsk, skb); + if (rc != TIPC_OK || !*skb) return rc; } /* Reject message if there isn't room to queue it */ - if (sk_rmem_alloc_get(sk) + buf->truesize >= limit) + if (sk_rmem_alloc_get(sk) + (*skb)->truesize >= limit) return -TIPC_ERR_OVERLOAD; /* Enqueue message */ - TIPC_SKB_CB(buf)->handle = NULL; - __skb_queue_tail(&sk->sk_receive_queue, buf); - skb_set_owner_r(buf, sk); + TIPC_SKB_CB(*skb)->handle = NULL; + __skb_queue_tail(&sk->sk_receive_queue, *skb); + skb_set_owner_r(*skb, sk); sk->sk_data_ready(sk); + *skb = NULL; return TIPC_OK; } @@ -1700,78 +1681,125 @@ static int filter_rcv(struct sock *sk, struct sk_buff *buf) * @sk: socket * @skb: message * - * Caller must hold socket lock, but not port lock. + * Caller must hold socket lock * * Returns 0 */ static int tipc_backlog_rcv(struct sock *sk, struct sk_buff *skb) { - int rc; - u32 onode; + int err; + atomic_t *dcnt; + u32 dnode; struct tipc_sock *tsk = tipc_sk(sk); + struct net *net = sock_net(sk); uint truesize = skb->truesize; - rc = filter_rcv(sk, skb); - - if (likely(!rc)) { - if (atomic_read(&tsk->dupl_rcvcnt) < TIPC_CONN_OVERLOAD_LIMIT) - atomic_add(truesize, &tsk->dupl_rcvcnt); + err = filter_rcv(sk, &skb); + if (likely(!skb)) { + dcnt = &tsk->dupl_rcvcnt; + if (atomic_read(dcnt) < TIPC_CONN_OVERLOAD_LIMIT) + atomic_add(truesize, dcnt); return 0; } + if (!err || tipc_msg_reverse(tsk_own_node(tsk), skb, &dnode, -err)) + tipc_link_xmit_skb(net, skb, dnode, tsk->portid); + return 0; +} - if ((rc < 0) && !tipc_msg_reverse(skb, &onode, -rc)) - return 0; - - tipc_link_xmit_skb(skb, onode, 0); +/** + * tipc_sk_enqueue - extract all buffers with destination 'dport' from + * inputq and try adding them to socket or backlog queue + * @inputq: list of incoming buffers with potentially different destinations + * @sk: socket where the buffers should be enqueued + * @dport: port number for the socket + * @_skb: returned buffer to be forwarded or rejected, if applicable + * + * Caller must hold socket lock + * + * Returns TIPC_OK if all buffers enqueued, otherwise -TIPC_ERR_OVERLOAD + * or -TIPC_ERR_NO_PORT + */ +static int tipc_sk_enqueue(struct sk_buff_head *inputq, struct sock *sk, + u32 dport, struct sk_buff **_skb) +{ + unsigned int lim; + atomic_t *dcnt; + int err; + struct sk_buff *skb; + unsigned long time_limit = jiffies + 2; - return 0; + while (skb_queue_len(inputq)) { + if (unlikely(time_after_eq(jiffies, time_limit))) + return TIPC_OK; + skb = tipc_skb_dequeue(inputq, dport); + if (unlikely(!skb)) + return TIPC_OK; + if (!sock_owned_by_user(sk)) { + err = filter_rcv(sk, &skb); + if (likely(!skb)) + continue; + *_skb = skb; + return err; + } + dcnt = &tipc_sk(sk)->dupl_rcvcnt; + if (sk->sk_backlog.len) + atomic_set(dcnt, 0); + lim = rcvbuf_limit(sk, skb) + atomic_read(dcnt); + if (likely(!sk_add_backlog(sk, skb, lim))) + continue; + *_skb = skb; + return -TIPC_ERR_OVERLOAD; + } + return TIPC_OK; } /** - * tipc_sk_rcv - handle incoming message - * @skb: buffer containing arriving message - * Consumes buffer - * Returns 0 if success, or errno: -EHOSTUNREACH + * tipc_sk_rcv - handle a chain of incoming buffers + * @inputq: buffer list containing the buffers + * Consumes all buffers in list until inputq is empty + * Note: may be called in multiple threads referring to the same queue + * Returns 0 if last buffer was accepted, otherwise -EHOSTUNREACH + * Only node local calls check the return value, sending single-buffer queues */ -int tipc_sk_rcv(struct sk_buff *skb) +int tipc_sk_rcv(struct net *net, struct sk_buff_head *inputq) { + u32 dnode, dport = 0; + int err = -TIPC_ERR_NO_PORT; + struct sk_buff *skb; struct tipc_sock *tsk; + struct tipc_net *tn; struct sock *sk; - u32 dport = msg_destport(buf_msg(skb)); - int rc = TIPC_OK; - uint limit; - u32 dnode; - /* Validate destination and message */ - tsk = tipc_sk_get(dport); - if (unlikely(!tsk)) { - rc = tipc_msg_eval(skb, &dnode); - goto exit; + while (skb_queue_len(inputq)) { + skb = NULL; + dport = tipc_skb_peek_port(inputq, dport); + tsk = tipc_sk_lookup(net, dport); + if (likely(tsk)) { + sk = &tsk->sk; + if (likely(spin_trylock_bh(&sk->sk_lock.slock))) { + err = tipc_sk_enqueue(inputq, sk, dport, &skb); + spin_unlock_bh(&sk->sk_lock.slock); + dport = 0; + } + sock_put(sk); + } else { + skb = tipc_skb_dequeue(inputq, dport); + } + if (likely(!skb)) + continue; + if (tipc_msg_lookup_dest(net, skb, &dnode, &err)) + goto xmit; + if (!err) { + dnode = msg_destnode(buf_msg(skb)); + goto xmit; + } + tn = net_generic(net, tipc_net_id); + if (!tipc_msg_reverse(tn->own_addr, skb, &dnode, -err)) + continue; +xmit: + tipc_link_xmit_skb(net, skb, dnode, dport); } - sk = &tsk->sk; - - /* Queue message */ - spin_lock_bh(&sk->sk_lock.slock); - - if (!sock_owned_by_user(sk)) { - rc = filter_rcv(sk, skb); - } else { - if (sk->sk_backlog.len == 0) - atomic_set(&tsk->dupl_rcvcnt, 0); - limit = rcvbuf_limit(sk, skb) + atomic_read(&tsk->dupl_rcvcnt); - if (sk_add_backlog(sk, skb, limit)) - rc = -TIPC_ERR_OVERLOAD; - } - spin_unlock_bh(&sk->sk_lock.slock); - tipc_sk_put(tsk); - if (likely(!rc)) - return 0; -exit: - if ((rc < 0) && !tipc_msg_reverse(skb, &dnode, -rc)) - return -EHOSTUNREACH; - - tipc_link_xmit_skb(skb, dnode, 0); - return (rc < 0) ? -EHOSTUNREACH : 0; + return err ? -EHOSTUNREACH : 0; } static int tipc_wait_for_connect(struct socket *sock, long *timeo_p) @@ -1809,17 +1837,26 @@ static int tipc_connect(struct socket *sock, struct sockaddr *dest, int destlen, int flags) { struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); struct sockaddr_tipc *dst = (struct sockaddr_tipc *)dest; struct msghdr m = {NULL,}; - long timeout = (flags & O_NONBLOCK) ? 0 : tipc_sk(sk)->conn_timeout; + long timeout = (flags & O_NONBLOCK) ? 0 : tsk->conn_timeout; socket_state previous; - int res; + int res = 0; lock_sock(sk); - /* For now, TIPC does not allow use of connect() with DGRAM/RDM types */ + /* DGRAM/RDM connect(), just save the destaddr */ if (sock->state == SS_READY) { - res = -EOPNOTSUPP; + if (dst->family == AF_UNSPEC) { + memset(&tsk->remote, 0, sizeof(struct sockaddr_tipc)); + tsk->connected = 0; + } else if (destlen != sizeof(struct sockaddr_tipc)) { + res = -EINVAL; + } else { + memcpy(&tsk->remote, dest, destlen); + tsk->connected = 1; + } goto exit; } @@ -1847,7 +1884,7 @@ static int tipc_connect(struct socket *sock, struct sockaddr *dest, if (!timeout) m.msg_flags = MSG_DONTWAIT; - res = tipc_sendmsg(NULL, sock, &m, 0); + res = __tipc_sendmsg(sock, &m, 0); if ((res < 0) && (res != -EWOULDBLOCK)) goto exit; @@ -1927,12 +1964,12 @@ static int tipc_wait_for_accept(struct socket *sock, long timeo) err = -EINVAL; if (sock->state != SS_LISTENING) break; - err = sock_intr_errno(timeo); - if (signal_pending(current)) - break; err = -EAGAIN; if (!timeo) break; + err = sock_intr_errno(timeo); + if (signal_pending(current)) + break; } finish_wait(sk_sleep(sk), &wait); return err; @@ -2003,7 +2040,7 @@ static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags) struct msghdr m = {NULL,}; tsk_advance_rx_queue(sk); - tipc_send_packet(NULL, new_sock, &m, 0); + __tipc_send_stream(new_sock, &m, 0); } else { __skb_dequeue(&sk->sk_receive_queue); __skb_queue_head(&new_sk->sk_receive_queue, buf); @@ -2027,6 +2064,7 @@ exit: static int tipc_shutdown(struct socket *sock, int how) { struct sock *sk = sock->sk; + struct net *net = sock_net(sk); struct tipc_sock *tsk = tipc_sk(sk); struct sk_buff *skb; u32 dnode; @@ -2049,21 +2087,23 @@ restart: kfree_skb(skb); goto restart; } - if (tipc_msg_reverse(skb, &dnode, TIPC_CONN_SHUTDOWN)) - tipc_link_xmit_skb(skb, dnode, tsk->ref); - tipc_node_remove_conn(dnode, tsk->ref); + if (tipc_msg_reverse(tsk_own_node(tsk), skb, &dnode, + TIPC_CONN_SHUTDOWN)) + tipc_link_xmit_skb(net, skb, dnode, + tsk->portid); } else { dnode = tsk_peer_node(tsk); + skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, TIPC_CONN_MSG, SHORT_H_SIZE, - 0, dnode, tipc_own_addr, + 0, dnode, tsk_own_node(tsk), tsk_peer_port(tsk), - tsk->ref, TIPC_CONN_SHUTDOWN); - tipc_link_xmit_skb(skb, dnode, tsk->ref); + tsk->portid, TIPC_CONN_SHUTDOWN); + tipc_link_xmit_skb(net, skb, dnode, tsk->portid); } tsk->connected = 0; sock->state = SS_DISCONNECTING; - tipc_node_remove_conn(dnode, tsk->ref); + tipc_node_remove_conn(net, dnode, tsk->portid); /* fall through */ case SS_DISCONNECTING: @@ -2084,18 +2124,14 @@ restart: return res; } -static void tipc_sk_timeout(unsigned long ref) +static void tipc_sk_timeout(unsigned long data) { - struct tipc_sock *tsk; - struct sock *sk; + struct tipc_sock *tsk = (struct tipc_sock *)data; + struct sock *sk = &tsk->sk; struct sk_buff *skb = NULL; u32 peer_port, peer_node; + u32 own_node = tsk_own_node(tsk); - tsk = tipc_sk_get(ref); - if (!tsk) - return; - - sk = &tsk->sk; bh_lock_sock(sk); if (!tsk->connected) { bh_unlock_sock(sk); @@ -2106,38 +2142,39 @@ static void tipc_sk_timeout(unsigned long ref) if (tsk->probing_state == TIPC_CONN_PROBING) { /* Previous probe not answered -> self abort */ - skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, TIPC_CONN_MSG, - SHORT_H_SIZE, 0, tipc_own_addr, - peer_node, ref, peer_port, - TIPC_ERR_NO_PORT); + skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, + TIPC_CONN_MSG, SHORT_H_SIZE, 0, + own_node, peer_node, tsk->portid, + peer_port, TIPC_ERR_NO_PORT); } else { - skb = tipc_msg_create(CONN_MANAGER, CONN_PROBE, INT_H_SIZE, - 0, peer_node, tipc_own_addr, - peer_port, ref, TIPC_OK); + skb = tipc_msg_create(CONN_MANAGER, CONN_PROBE, + INT_H_SIZE, 0, peer_node, own_node, + peer_port, tsk->portid, TIPC_OK); tsk->probing_state = TIPC_CONN_PROBING; - k_start_timer(&tsk->timer, tsk->probing_interval); + sk_reset_timer(sk, &sk->sk_timer, jiffies + tsk->probing_intv); } bh_unlock_sock(sk); if (skb) - tipc_link_xmit_skb(skb, peer_node, ref); + tipc_link_xmit_skb(sock_net(sk), skb, peer_node, tsk->portid); exit: - tipc_sk_put(tsk); + sock_put(sk); } static int tipc_sk_publish(struct tipc_sock *tsk, uint scope, struct tipc_name_seq const *seq) { + struct net *net = sock_net(&tsk->sk); struct publication *publ; u32 key; if (tsk->connected) return -EINVAL; - key = tsk->ref + tsk->pub_count + 1; - if (key == tsk->ref) + key = tsk->portid + tsk->pub_count + 1; + if (key == tsk->portid) return -EADDRINUSE; - publ = tipc_nametbl_publish(seq->type, seq->lower, seq->upper, - scope, tsk->ref, key); + publ = tipc_nametbl_publish(net, seq->type, seq->lower, seq->upper, + scope, tsk->portid, key); if (unlikely(!publ)) return -EINVAL; @@ -2150,6 +2187,7 @@ static int tipc_sk_publish(struct tipc_sock *tsk, uint scope, static int tipc_sk_withdraw(struct tipc_sock *tsk, uint scope, struct tipc_name_seq const *seq) { + struct net *net = sock_net(&tsk->sk); struct publication *publ; struct publication *safe; int rc = -EINVAL; @@ -2164,12 +2202,12 @@ static int tipc_sk_withdraw(struct tipc_sock *tsk, uint scope, continue; if (publ->upper != seq->upper) break; - tipc_nametbl_withdraw(publ->type, publ->lower, + tipc_nametbl_withdraw(net, publ->type, publ->lower, publ->ref, publ->key); rc = 0; break; } - tipc_nametbl_withdraw(publ->type, publ->lower, + tipc_nametbl_withdraw(net, publ->type, publ->lower, publ->ref, publ->key); rc = 0; } @@ -2178,336 +2216,105 @@ static int tipc_sk_withdraw(struct tipc_sock *tsk, uint scope, return rc; } -static int tipc_sk_show(struct tipc_sock *tsk, char *buf, - int len, int full_id) -{ - struct publication *publ; - int ret; - - if (full_id) - ret = tipc_snprintf(buf, len, "<%u.%u.%u:%u>:", - tipc_zone(tipc_own_addr), - tipc_cluster(tipc_own_addr), - tipc_node(tipc_own_addr), tsk->ref); - else - ret = tipc_snprintf(buf, len, "%-10u:", tsk->ref); - - if (tsk->connected) { - u32 dport = tsk_peer_port(tsk); - u32 destnode = tsk_peer_node(tsk); - - ret += tipc_snprintf(buf + ret, len - ret, - " connected to <%u.%u.%u:%u>", - tipc_zone(destnode), - tipc_cluster(destnode), - tipc_node(destnode), dport); - if (tsk->conn_type != 0) - ret += tipc_snprintf(buf + ret, len - ret, - " via {%u,%u}", tsk->conn_type, - tsk->conn_instance); - } else if (tsk->published) { - ret += tipc_snprintf(buf + ret, len - ret, " bound to"); - list_for_each_entry(publ, &tsk->publications, pport_list) { - if (publ->lower == publ->upper) - ret += tipc_snprintf(buf + ret, len - ret, - " {%u,%u}", publ->type, - publ->lower); - else - ret += tipc_snprintf(buf + ret, len - ret, - " {%u,%u,%u}", publ->type, - publ->lower, publ->upper); - } - } - ret += tipc_snprintf(buf + ret, len - ret, "\n"); - return ret; -} - -struct sk_buff *tipc_sk_socks_show(void) -{ - struct sk_buff *buf; - struct tlv_desc *rep_tlv; - char *pb; - int pb_len; - struct tipc_sock *tsk; - int str_len = 0; - u32 ref = 0; - - buf = tipc_cfg_reply_alloc(TLV_SPACE(ULTRA_STRING_MAX_LEN)); - if (!buf) - return NULL; - rep_tlv = (struct tlv_desc *)buf->data; - pb = TLV_DATA(rep_tlv); - pb_len = ULTRA_STRING_MAX_LEN; - - tsk = tipc_sk_get_next(&ref); - for (; tsk; tsk = tipc_sk_get_next(&ref)) { - lock_sock(&tsk->sk); - str_len += tipc_sk_show(tsk, pb + str_len, - pb_len - str_len, 0); - release_sock(&tsk->sk); - tipc_sk_put(tsk); - } - str_len += 1; /* for "\0" */ - skb_put(buf, TLV_SPACE(str_len)); - TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); - - return buf; -} - /* tipc_sk_reinit: set non-zero address in all existing sockets * when we go from standalone to network mode. */ -void tipc_sk_reinit(void) +void tipc_sk_reinit(struct net *net) { + struct tipc_net *tn = net_generic(net, tipc_net_id); + const struct bucket_table *tbl; + struct rhash_head *pos; + struct tipc_sock *tsk; struct tipc_msg *msg; - u32 ref = 0; - struct tipc_sock *tsk = tipc_sk_get_next(&ref); + int i; - for (; tsk; tsk = tipc_sk_get_next(&ref)) { - lock_sock(&tsk->sk); - msg = &tsk->phdr; - msg_set_prevnode(msg, tipc_own_addr); - msg_set_orignode(msg, tipc_own_addr); - release_sock(&tsk->sk); - tipc_sk_put(tsk); + rcu_read_lock(); + tbl = rht_dereference_rcu((&tn->sk_rht)->tbl, &tn->sk_rht); + for (i = 0; i < tbl->size; i++) { + rht_for_each_entry_rcu(tsk, pos, tbl, i, node) { + spin_lock_bh(&tsk->sk.sk_lock.slock); + msg = &tsk->phdr; + msg_set_prevnode(msg, tn->own_addr); + msg_set_orignode(msg, tn->own_addr); + spin_unlock_bh(&tsk->sk.sk_lock.slock); + } } + rcu_read_unlock(); } -/** - * struct reference - TIPC socket reference entry - * @tsk: pointer to socket associated with reference entry - * @ref: reference value for socket (combines instance & array index info) - */ -struct reference { - struct tipc_sock *tsk; - u32 ref; -}; - -/** - * struct tipc_ref_table - table of TIPC socket reference entries - * @entries: pointer to array of reference entries - * @capacity: array index of first unusable entry - * @init_point: array index of first uninitialized entry - * @first_free: array index of first unused socket reference entry - * @last_free: array index of last unused socket reference entry - * @index_mask: bitmask for array index portion of reference values - * @start_mask: initial value for instance value portion of reference values - */ -struct ref_table { - struct reference *entries; - u32 capacity; - u32 init_point; - u32 first_free; - u32 last_free; - u32 index_mask; - u32 start_mask; -}; - -/* Socket reference table consists of 2**N entries. - * - * State Socket ptr Reference - * ----- ---------- --------- - * In use non-NULL XXXX|own index - * (XXXX changes each time entry is acquired) - * Free NULL YYYY|next free index - * (YYYY is one more than last used XXXX) - * Uninitialized NULL 0 - * - * Entry 0 is not used; this allows index 0 to denote the end of the free list. - * - * Note that a reference value of 0 does not necessarily indicate that an - * entry is uninitialized, since the last entry in the free list could also - * have a reference value of 0 (although this is unlikely). - */ - -static struct ref_table tipc_ref_table; - -static DEFINE_RWLOCK(ref_table_lock); - -/** - * tipc_ref_table_init - create reference table for sockets - */ -int tipc_sk_ref_table_init(u32 req_sz, u32 start) +static struct tipc_sock *tipc_sk_lookup(struct net *net, u32 portid) { - struct reference *table; - u32 actual_sz; - - /* account for unused entry, then round up size to a power of 2 */ - - req_sz++; - for (actual_sz = 16; actual_sz < req_sz; actual_sz <<= 1) { - /* do nothing */ - }; - - /* allocate table & mark all entries as uninitialized */ - table = vzalloc(actual_sz * sizeof(struct reference)); - if (table == NULL) - return -ENOMEM; - - tipc_ref_table.entries = table; - tipc_ref_table.capacity = req_sz; - tipc_ref_table.init_point = 1; - tipc_ref_table.first_free = 0; - tipc_ref_table.last_free = 0; - tipc_ref_table.index_mask = actual_sz - 1; - tipc_ref_table.start_mask = start & ~tipc_ref_table.index_mask; + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_sock *tsk; - return 0; -} + rcu_read_lock(); + tsk = rhashtable_lookup_fast(&tn->sk_rht, &portid, tsk_rht_params); + if (tsk) + sock_hold(&tsk->sk); + rcu_read_unlock(); -/** - * tipc_ref_table_stop - destroy reference table for sockets - */ -void tipc_sk_ref_table_stop(void) -{ - if (!tipc_ref_table.entries) - return; - vfree(tipc_ref_table.entries); - tipc_ref_table.entries = NULL; + return tsk; } -/* tipc_ref_acquire - create reference to a socket - * - * Register an socket pointer in the reference table. - * Returns a unique reference value that is used from then on to retrieve the - * socket pointer, or to determine if the socket has been deregistered. - */ -u32 tipc_sk_ref_acquire(struct tipc_sock *tsk) +static int tipc_sk_insert(struct tipc_sock *tsk) { - u32 index; - u32 index_mask; - u32 next_plus_upper; - u32 ref = 0; - struct reference *entry; - - if (unlikely(!tsk)) { - pr_err("Attempt to acquire ref. to non-existent obj\n"); - return 0; - } - if (unlikely(!tipc_ref_table.entries)) { - pr_err("Ref. table not found in acquisition attempt\n"); - return 0; - } - - /* Take a free entry, if available; otherwise initialize a new one */ - write_lock_bh(&ref_table_lock); - index = tipc_ref_table.first_free; - entry = &tipc_ref_table.entries[index]; - - if (likely(index)) { - index = tipc_ref_table.first_free; - entry = &tipc_ref_table.entries[index]; - index_mask = tipc_ref_table.index_mask; - next_plus_upper = entry->ref; - tipc_ref_table.first_free = next_plus_upper & index_mask; - ref = (next_plus_upper & ~index_mask) + index; - entry->tsk = tsk; - } else if (tipc_ref_table.init_point < tipc_ref_table.capacity) { - index = tipc_ref_table.init_point++; - entry = &tipc_ref_table.entries[index]; - ref = tipc_ref_table.start_mask + index; + struct sock *sk = &tsk->sk; + struct net *net = sock_net(sk); + struct tipc_net *tn = net_generic(net, tipc_net_id); + u32 remaining = (TIPC_MAX_PORT - TIPC_MIN_PORT) + 1; + u32 portid = prandom_u32() % remaining + TIPC_MIN_PORT; + + while (remaining--) { + portid++; + if ((portid < TIPC_MIN_PORT) || (portid > TIPC_MAX_PORT)) + portid = TIPC_MIN_PORT; + tsk->portid = portid; + sock_hold(&tsk->sk); + if (!rhashtable_lookup_insert_fast(&tn->sk_rht, &tsk->node, + tsk_rht_params)) + return 0; + sock_put(&tsk->sk); } - if (ref) { - entry->ref = ref; - entry->tsk = tsk; - } - write_unlock_bh(&ref_table_lock); - return ref; + return -1; } -/* tipc_sk_ref_discard - invalidate reference to an socket - * - * Disallow future references to an socket and free up the entry for re-use. - */ -void tipc_sk_ref_discard(u32 ref) +static void tipc_sk_remove(struct tipc_sock *tsk) { - struct reference *entry; - u32 index; - u32 index_mask; - - if (unlikely(!tipc_ref_table.entries)) { - pr_err("Ref. table not found during discard attempt\n"); - return; - } - - index_mask = tipc_ref_table.index_mask; - index = ref & index_mask; - entry = &tipc_ref_table.entries[index]; - - write_lock_bh(&ref_table_lock); + struct sock *sk = &tsk->sk; + struct tipc_net *tn = net_generic(sock_net(sk), tipc_net_id); - if (unlikely(!entry->tsk)) { - pr_err("Attempt to discard ref. to non-existent socket\n"); - goto exit; + if (!rhashtable_remove_fast(&tn->sk_rht, &tsk->node, tsk_rht_params)) { + WARN_ON(atomic_read(&sk->sk_refcnt) == 1); + __sock_put(sk); } - if (unlikely(entry->ref != ref)) { - pr_err("Attempt to discard non-existent reference\n"); - goto exit; - } - - /* Mark entry as unused; increment instance part of entry's - * reference to invalidate any subsequent references - */ - - entry->tsk = NULL; - entry->ref = (ref & ~index_mask) + (index_mask + 1); - - /* Append entry to free entry list */ - if (unlikely(tipc_ref_table.first_free == 0)) - tipc_ref_table.first_free = index; - else - tipc_ref_table.entries[tipc_ref_table.last_free].ref |= index; - tipc_ref_table.last_free = index; -exit: - write_unlock_bh(&ref_table_lock); } -/* tipc_sk_get - find referenced socket and return pointer to it - */ -struct tipc_sock *tipc_sk_get(u32 ref) +static const struct rhashtable_params tsk_rht_params = { + .nelem_hint = 192, + .head_offset = offsetof(struct tipc_sock, node), + .key_offset = offsetof(struct tipc_sock, portid), + .key_len = sizeof(u32), /* portid */ + .max_size = 1048576, + .min_size = 256, + .automatic_shrinking = true, +}; + +int tipc_sk_rht_init(struct net *net) { - struct reference *entry; - struct tipc_sock *tsk; + struct tipc_net *tn = net_generic(net, tipc_net_id); - if (unlikely(!tipc_ref_table.entries)) - return NULL; - read_lock_bh(&ref_table_lock); - entry = &tipc_ref_table.entries[ref & tipc_ref_table.index_mask]; - tsk = entry->tsk; - if (likely(tsk && (entry->ref == ref))) - sock_hold(&tsk->sk); - else - tsk = NULL; - read_unlock_bh(&ref_table_lock); - return tsk; + return rhashtable_init(&tn->sk_rht, &tsk_rht_params); } -/* tipc_sk_get_next - lock & return next socket after referenced one -*/ -struct tipc_sock *tipc_sk_get_next(u32 *ref) +void tipc_sk_rht_destroy(struct net *net) { - struct reference *entry; - struct tipc_sock *tsk = NULL; - uint index = *ref & tipc_ref_table.index_mask; + struct tipc_net *tn = net_generic(net, tipc_net_id); - read_lock_bh(&ref_table_lock); - while (++index < tipc_ref_table.capacity) { - entry = &tipc_ref_table.entries[index]; - if (!entry->tsk) - continue; - tsk = entry->tsk; - sock_hold(&tsk->sk); - *ref = entry->ref; - break; - } - read_unlock_bh(&ref_table_lock); - return tsk; -} + /* Wait for socket readers to complete */ + synchronize_net(); -static void tipc_sk_put(struct tipc_sock *tsk) -{ - sock_put(&tsk->sk); + rhashtable_destroy(&tn->sk_rht); } /** @@ -2639,8 +2446,9 @@ static int tipc_getsockopt(struct socket *sock, int lvl, int opt, return put_user(sizeof(value), ol); } -static int tipc_ioctl(struct socket *sk, unsigned int cmd, unsigned long arg) +static int tipc_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) { + struct sock *sk = sock->sk; struct tipc_sioc_ln_req lnr; void __user *argp = (void __user *)arg; @@ -2648,7 +2456,8 @@ static int tipc_ioctl(struct socket *sk, unsigned int cmd, unsigned long arg) case SIOCGETLINKNAME: if (copy_from_user(&lnr, argp, sizeof(lnr))) return -EFAULT; - if (!tipc_node_get_linkname(lnr.bearer_id & 0xffff, lnr.peer, + if (!tipc_node_get_linkname(sock_net(sk), + lnr.bearer_id & 0xffff, lnr.peer, lnr.linkname, TIPC_MAX_LINK_NAME)) { if (copy_to_user(argp, &lnr, sizeof(lnr))) return -EFAULT; @@ -2738,12 +2547,6 @@ static struct proto tipc_proto = { .sysctl_rmem = sysctl_tipc_rmem }; -static struct proto tipc_proto_kern = { - .name = "TIPC", - .obj_size = sizeof(struct tipc_sock), - .sysctl_rmem = sysctl_tipc_rmem -}; - /** * tipc_socket_init - initialize TIPC socket interface * @@ -2820,18 +2623,20 @@ static int __tipc_nl_add_sk(struct sk_buff *skb, struct netlink_callback *cb, int err; void *hdr; struct nlattr *attrs; + struct net *net = sock_net(skb->sk); + struct tipc_net *tn = net_generic(net, tipc_net_id); hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, - &tipc_genl_v2_family, NLM_F_MULTI, TIPC_NL_SOCK_GET); + &tipc_genl_family, NLM_F_MULTI, TIPC_NL_SOCK_GET); if (!hdr) goto msg_cancel; attrs = nla_nest_start(skb, TIPC_NLA_SOCK); if (!attrs) goto genlmsg_cancel; - if (nla_put_u32(skb, TIPC_NLA_SOCK_REF, tsk->ref)) + if (nla_put_u32(skb, TIPC_NLA_SOCK_REF, tsk->portid)) goto attr_msg_cancel; - if (nla_put_u32(skb, TIPC_NLA_SOCK_ADDR, tipc_own_addr)) + if (nla_put_u32(skb, TIPC_NLA_SOCK_ADDR, tn->own_addr)) goto attr_msg_cancel; if (tsk->connected) { @@ -2859,22 +2664,37 @@ int tipc_nl_sk_dump(struct sk_buff *skb, struct netlink_callback *cb) { int err; struct tipc_sock *tsk; - u32 prev_ref = cb->args[0]; - u32 ref = prev_ref; - - tsk = tipc_sk_get_next(&ref); - for (; tsk; tsk = tipc_sk_get_next(&ref)) { - lock_sock(&tsk->sk); - err = __tipc_nl_add_sk(skb, cb, tsk); - release_sock(&tsk->sk); - tipc_sk_put(tsk); - if (err) - break; + const struct bucket_table *tbl; + struct rhash_head *pos; + struct net *net = sock_net(skb->sk); + struct tipc_net *tn = net_generic(net, tipc_net_id); + u32 tbl_id = cb->args[0]; + u32 prev_portid = cb->args[1]; - prev_ref = ref; - } + rcu_read_lock(); + tbl = rht_dereference_rcu((&tn->sk_rht)->tbl, &tn->sk_rht); + for (; tbl_id < tbl->size; tbl_id++) { + rht_for_each_entry_rcu(tsk, pos, tbl, tbl_id, node) { + spin_lock_bh(&tsk->sk.sk_lock.slock); + if (prev_portid && prev_portid != tsk->portid) { + spin_unlock_bh(&tsk->sk.sk_lock.slock); + continue; + } - cb->args[0] = prev_ref; + err = __tipc_nl_add_sk(skb, cb, tsk); + if (err) { + prev_portid = tsk->portid; + spin_unlock_bh(&tsk->sk.sk_lock.slock); + goto out; + } + prev_portid = 0; + spin_unlock_bh(&tsk->sk.sk_lock.slock); + } + } +out: + rcu_read_unlock(); + cb->args[0] = tbl_id; + cb->args[1] = prev_portid; return skb->len; } @@ -2888,7 +2708,7 @@ static int __tipc_nl_add_sk_publ(struct sk_buff *skb, struct nlattr *attrs; hdr = genlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, - &tipc_genl_v2_family, NLM_F_MULTI, TIPC_NL_PUBL_GET); + &tipc_genl_family, NLM_F_MULTI, TIPC_NL_PUBL_GET); if (!hdr) goto msg_cancel; @@ -2962,12 +2782,13 @@ static int __tipc_nl_list_sk_publ(struct sk_buff *skb, int tipc_nl_publ_dump(struct sk_buff *skb, struct netlink_callback *cb) { int err; - u32 tsk_ref = cb->args[0]; + u32 tsk_portid = cb->args[0]; u32 last_publ = cb->args[1]; u32 done = cb->args[2]; + struct net *net = sock_net(skb->sk); struct tipc_sock *tsk; - if (!tsk_ref) { + if (!tsk_portid) { struct nlattr **attrs; struct nlattr *sock[TIPC_NLA_SOCK_MAX + 1]; @@ -2984,13 +2805,13 @@ int tipc_nl_publ_dump(struct sk_buff *skb, struct netlink_callback *cb) if (!sock[TIPC_NLA_SOCK_REF]) return -EINVAL; - tsk_ref = nla_get_u32(sock[TIPC_NLA_SOCK_REF]); + tsk_portid = nla_get_u32(sock[TIPC_NLA_SOCK_REF]); } if (done) return 0; - tsk = tipc_sk_get(tsk_ref); + tsk = tipc_sk_lookup(net, tsk_portid); if (!tsk) return -EINVAL; @@ -2999,9 +2820,9 @@ int tipc_nl_publ_dump(struct sk_buff *skb, struct netlink_callback *cb) if (!err) done = 1; release_sock(&tsk->sk); - tipc_sk_put(tsk); + sock_put(&tsk->sk); - cb->args[0] = tsk_ref; + cb->args[0] = tsk_portid; cb->args[1] = last_publ; cb->args[2] = done; diff --git a/net/tipc/socket.h b/net/tipc/socket.h index d34089387006..bf6551389522 100644 --- a/net/tipc/socket.h +++ b/net/tipc/socket.h @@ -1,6 +1,6 @@ /* net/tipc/socket.h: Include file for TIPC socket code * - * Copyright (c) 2014, Ericsson AB + * Copyright (c) 2014-2015, Ericsson AB * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -42,12 +42,14 @@ #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)) -int tipc_sk_rcv(struct sk_buff *buf); -struct sk_buff *tipc_sk_socks_show(void); -void tipc_sk_mcast_rcv(struct sk_buff *buf); -void tipc_sk_reinit(void); -int tipc_sk_ref_table_init(u32 requested_size, u32 start); -void tipc_sk_ref_table_stop(void); +int tipc_socket_init(void); +void tipc_socket_stop(void); +int tipc_sk_rcv(struct net *net, struct sk_buff_head *inputq); +void tipc_sk_mcast_rcv(struct net *net, struct sk_buff_head *arrvq, + struct sk_buff_head *inputq); +void tipc_sk_reinit(struct net *net); +int tipc_sk_rht_init(struct net *net); +void tipc_sk_rht_destroy(struct net *net); int tipc_nl_sk_dump(struct sk_buff *skb, struct netlink_callback *cb); int tipc_nl_publ_dump(struct sk_buff *skb, struct netlink_callback *cb); diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c index 0344206b984f..1c147c869c2e 100644 --- a/net/tipc/subscr.c +++ b/net/tipc/subscr.c @@ -50,33 +50,6 @@ struct tipc_subscriber { struct list_head subscription_list; }; -static void subscr_conn_msg_event(int conid, struct sockaddr_tipc *addr, - void *usr_data, void *buf, size_t len); -static void *subscr_named_msg_event(int conid); -static void subscr_conn_shutdown_event(int conid, void *usr_data); - -static atomic_t subscription_count = ATOMIC_INIT(0); - -static struct sockaddr_tipc topsrv_addr __read_mostly = { - .family = AF_TIPC, - .addrtype = TIPC_ADDR_NAMESEQ, - .addr.nameseq.type = TIPC_TOP_SRV, - .addr.nameseq.lower = TIPC_TOP_SRV, - .addr.nameseq.upper = TIPC_TOP_SRV, - .scope = TIPC_NODE_SCOPE -}; - -static struct tipc_server topsrv __read_mostly = { - .saddr = &topsrv_addr, - .imp = TIPC_CRITICAL_IMPORTANCE, - .type = SOCK_SEQPACKET, - .max_rcvbuf_size = sizeof(struct tipc_subscr), - .name = "topology_server", - .tipc_conn_recvmsg = subscr_conn_msg_event, - .tipc_conn_new = subscr_named_msg_event, - .tipc_conn_shutdown = subscr_conn_shutdown_event, -}; - /** * htohl - convert value to endianness used by destination * @in: value to convert @@ -93,6 +66,7 @@ static void subscr_send_event(struct tipc_subscription *sub, u32 found_lower, u32 found_upper, u32 event, u32 port_ref, u32 node) { + struct tipc_net *tn = net_generic(sub->net, tipc_net_id); struct tipc_subscriber *subscriber = sub->subscriber; struct kvec msg_sect; @@ -103,8 +77,8 @@ static void subscr_send_event(struct tipc_subscription *sub, u32 found_lower, sub->evt.found_upper = htohl(found_upper, sub->swap); sub->evt.port.ref = htohl(port_ref, sub->swap); sub->evt.port.node = htohl(node, sub->swap); - tipc_conn_sendmsg(&topsrv, subscriber->conid, NULL, msg_sect.iov_base, - msg_sect.iov_len); + tipc_conn_sendmsg(tn->topsrv, subscriber->conid, NULL, + msg_sect.iov_base, msg_sect.iov_len); } /** @@ -141,9 +115,11 @@ void tipc_subscr_report_overlap(struct tipc_subscription *sub, u32 found_lower, subscr_send_event(sub, found_lower, found_upper, event, port_ref, node); } -static void subscr_timeout(struct tipc_subscription *sub) +static void subscr_timeout(unsigned long data) { + struct tipc_subscription *sub = (struct tipc_subscription *)data; struct tipc_subscriber *subscriber = sub->subscriber; + struct tipc_net *tn = net_generic(sub->net, tipc_net_id); /* The spin lock per subscriber is used to protect its members */ spin_lock_bh(&subscriber->lock); @@ -167,9 +143,8 @@ static void subscr_timeout(struct tipc_subscription *sub) TIPC_SUBSCR_TIMEOUT, 0, 0); /* Now destroy subscription */ - k_term_timer(&sub->timer); kfree(sub); - atomic_dec(&subscription_count); + atomic_dec(&tn->subscription_count); } /** @@ -179,20 +154,12 @@ static void subscr_timeout(struct tipc_subscription *sub) */ static void subscr_del(struct tipc_subscription *sub) { + struct tipc_net *tn = net_generic(sub->net, tipc_net_id); + tipc_nametbl_unsubscribe(sub); list_del(&sub->subscription_list); kfree(sub); - atomic_dec(&subscription_count); -} - -/** - * subscr_terminate - terminate communication with a subscriber - * - * Note: Must call it in process context since it might sleep. - */ -static void subscr_terminate(struct tipc_subscriber *subscriber) -{ - tipc_conn_terminate(&topsrv, subscriber->conid); + atomic_dec(&tn->subscription_count); } static void subscr_release(struct tipc_subscriber *subscriber) @@ -207,8 +174,7 @@ static void subscr_release(struct tipc_subscriber *subscriber) subscription_list) { if (sub->timeout != TIPC_WAIT_FOREVER) { spin_unlock_bh(&subscriber->lock); - k_cancel_timer(&sub->timer); - k_term_timer(&sub->timer); + del_timer_sync(&sub->timer); spin_lock_bh(&subscriber->lock); } subscr_del(sub); @@ -250,8 +216,7 @@ static void subscr_cancel(struct tipc_subscr *s, if (sub->timeout != TIPC_WAIT_FOREVER) { sub->timeout = TIPC_WAIT_FOREVER; spin_unlock_bh(&subscriber->lock); - k_cancel_timer(&sub->timer); - k_term_timer(&sub->timer); + del_timer_sync(&sub->timer); spin_lock_bh(&subscriber->lock); } subscr_del(sub); @@ -262,9 +227,11 @@ static void subscr_cancel(struct tipc_subscr *s, * * Called with subscriber lock held. */ -static int subscr_subscribe(struct tipc_subscr *s, +static int subscr_subscribe(struct net *net, struct tipc_subscr *s, struct tipc_subscriber *subscriber, - struct tipc_subscription **sub_p) { + struct tipc_subscription **sub_p) +{ + struct tipc_net *tn = net_generic(net, tipc_net_id); struct tipc_subscription *sub; int swap; @@ -279,7 +246,7 @@ static int subscr_subscribe(struct tipc_subscr *s, } /* Refuse subscription if global limit exceeded */ - if (atomic_read(&subscription_count) >= TIPC_MAX_SUBSCRIPTIONS) { + if (atomic_read(&tn->subscription_count) >= TIPC_MAX_SUBSCRIPTIONS) { pr_warn("Subscription rejected, limit reached (%u)\n", TIPC_MAX_SUBSCRIPTIONS); return -EINVAL; @@ -293,10 +260,11 @@ static int subscr_subscribe(struct tipc_subscr *s, } /* Initialize subscription object */ + sub->net = net; sub->seq.type = htohl(s->seq.type, swap); sub->seq.lower = htohl(s->seq.lower, swap); sub->seq.upper = htohl(s->seq.upper, swap); - sub->timeout = htohl(s->timeout, swap); + sub->timeout = msecs_to_jiffies(htohl(s->timeout, swap)); sub->filter = htohl(s->filter, swap); if ((!(sub->filter & TIPC_SUB_PORTS) == !(sub->filter & TIPC_SUB_SERVICE)) || @@ -309,11 +277,10 @@ static int subscr_subscribe(struct tipc_subscr *s, sub->subscriber = subscriber; sub->swap = swap; memcpy(&sub->evt.s, s, sizeof(struct tipc_subscr)); - atomic_inc(&subscription_count); + atomic_inc(&tn->subscription_count); if (sub->timeout != TIPC_WAIT_FOREVER) { - k_init_timer(&sub->timer, - (Handler)subscr_timeout, (unsigned long)sub); - k_start_timer(&sub->timer, sub->timeout); + setup_timer(&sub->timer, subscr_timeout, (unsigned long)sub); + mod_timer(&sub->timer, jiffies + sub->timeout); } *sub_p = sub; return 0; @@ -326,24 +293,23 @@ static void subscr_conn_shutdown_event(int conid, void *usr_data) } /* Handle one request to create a new subscription for the subscriber */ -static void subscr_conn_msg_event(int conid, struct sockaddr_tipc *addr, - void *usr_data, void *buf, size_t len) +static void subscr_conn_msg_event(struct net *net, int conid, + struct sockaddr_tipc *addr, void *usr_data, + void *buf, size_t len) { struct tipc_subscriber *subscriber = usr_data; struct tipc_subscription *sub = NULL; + struct tipc_net *tn = net_generic(net, tipc_net_id); spin_lock_bh(&subscriber->lock); - if (subscr_subscribe((struct tipc_subscr *)buf, subscriber, &sub) < 0) { - spin_unlock_bh(&subscriber->lock); - subscr_terminate(subscriber); - return; - } + subscr_subscribe(net, (struct tipc_subscr *)buf, subscriber, &sub); if (sub) tipc_nametbl_subscribe(sub); + else + tipc_conn_terminate(tn->topsrv, subscriber->conid); spin_unlock_bh(&subscriber->lock); } - /* Handle one request to establish a new subscriber */ static void *subscr_named_msg_event(int conid) { @@ -362,12 +328,50 @@ static void *subscr_named_msg_event(int conid) return (void *)subscriber; } -int tipc_subscr_start(void) +int tipc_subscr_start(struct net *net) { - return tipc_server_start(&topsrv); + struct tipc_net *tn = net_generic(net, tipc_net_id); + const char name[] = "topology_server"; + struct tipc_server *topsrv; + struct sockaddr_tipc *saddr; + + saddr = kzalloc(sizeof(*saddr), GFP_ATOMIC); + if (!saddr) + return -ENOMEM; + saddr->family = AF_TIPC; + saddr->addrtype = TIPC_ADDR_NAMESEQ; + saddr->addr.nameseq.type = TIPC_TOP_SRV; + saddr->addr.nameseq.lower = TIPC_TOP_SRV; + saddr->addr.nameseq.upper = TIPC_TOP_SRV; + saddr->scope = TIPC_NODE_SCOPE; + + topsrv = kzalloc(sizeof(*topsrv), GFP_ATOMIC); + if (!topsrv) { + kfree(saddr); + return -ENOMEM; + } + topsrv->net = net; + topsrv->saddr = saddr; + topsrv->imp = TIPC_CRITICAL_IMPORTANCE; + topsrv->type = SOCK_SEQPACKET; + topsrv->max_rcvbuf_size = sizeof(struct tipc_subscr); + topsrv->tipc_conn_recvmsg = subscr_conn_msg_event; + topsrv->tipc_conn_new = subscr_named_msg_event; + topsrv->tipc_conn_shutdown = subscr_conn_shutdown_event; + + strncpy(topsrv->name, name, strlen(name) + 1); + tn->topsrv = topsrv; + atomic_set(&tn->subscription_count, 0); + + return tipc_server_start(topsrv); } -void tipc_subscr_stop(void) +void tipc_subscr_stop(struct net *net) { - tipc_server_stop(&topsrv); + struct tipc_net *tn = net_generic(net, tipc_net_id); + struct tipc_server *topsrv = tn->topsrv; + + tipc_server_stop(topsrv); + kfree(topsrv->saddr); + kfree(topsrv); } diff --git a/net/tipc/subscr.h b/net/tipc/subscr.h index 393e417bee3f..33488bd9fe3c 100644 --- a/net/tipc/subscr.h +++ b/net/tipc/subscr.h @@ -39,6 +39,9 @@ #include "server.h" +#define TIPC_MAX_SUBSCRIPTIONS 65535 +#define TIPC_MAX_PUBLICATIONS 65535 + struct tipc_subscription; struct tipc_subscriber; @@ -46,6 +49,7 @@ struct tipc_subscriber; * struct tipc_subscription - TIPC network topology subscription object * @subscriber: pointer to its subscriber * @seq: name sequence associated with subscription + * @net: point to network namespace * @timeout: duration of subscription (in ms) * @filter: event filtering to be done for subscription * @timer: timer governing subscription duration (optional) @@ -58,7 +62,8 @@ struct tipc_subscriber; struct tipc_subscription { struct tipc_subscriber *subscriber; struct tipc_name_seq seq; - u32 timeout; + struct net *net; + unsigned long timeout; u32 filter; struct timer_list timer; struct list_head nameseq_list; @@ -69,13 +74,10 @@ struct tipc_subscription { int tipc_subscr_overlap(struct tipc_subscription *sub, u32 found_lower, u32 found_upper); - void tipc_subscr_report_overlap(struct tipc_subscription *sub, u32 found_lower, u32 found_upper, u32 event, u32 port_ref, u32 node, int must); - -int tipc_subscr_start(void); - -void tipc_subscr_stop(void); +int tipc_subscr_start(struct net *net); +void tipc_subscr_stop(struct net *net); #endif diff --git a/net/tipc/udp_media.c b/net/tipc/udp_media.c new file mode 100644 index 000000000000..66deebc66aa1 --- /dev/null +++ b/net/tipc/udp_media.c @@ -0,0 +1,448 @@ +/* net/tipc/udp_media.c: IP bearer support for TIPC + * + * Copyright (c) 2015, Ericsson AB + * 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 <linux/socket.h> +#include <linux/ip.h> +#include <linux/udp.h> +#include <linux/inet.h> +#include <linux/inetdevice.h> +#include <linux/igmp.h> +#include <linux/kernel.h> +#include <linux/workqueue.h> +#include <linux/list.h> +#include <net/sock.h> +#include <net/ip.h> +#include <net/udp_tunnel.h> +#include <net/addrconf.h> +#include <linux/tipc_netlink.h> +#include "core.h" +#include "bearer.h" + +/* IANA assigned UDP port */ +#define UDP_PORT_DEFAULT 6118 + +static const struct nla_policy tipc_nl_udp_policy[TIPC_NLA_UDP_MAX + 1] = { + [TIPC_NLA_UDP_UNSPEC] = {.type = NLA_UNSPEC}, + [TIPC_NLA_UDP_LOCAL] = {.type = NLA_BINARY, + .len = sizeof(struct sockaddr_storage)}, + [TIPC_NLA_UDP_REMOTE] = {.type = NLA_BINARY, + .len = sizeof(struct sockaddr_storage)}, +}; + +/** + * struct udp_media_addr - IP/UDP addressing information + * + * This is the bearer level originating address used in neighbor discovery + * messages, and all fields should be in network byte order + */ +struct udp_media_addr { + __be16 proto; + __be16 udp_port; + union { + struct in_addr ipv4; + struct in6_addr ipv6; + }; +}; + +/** + * struct udp_bearer - ip/udp bearer data structure + * @bearer: associated generic tipc bearer + * @ubsock: bearer associated socket + * @ifindex: local address scope + * @work: used to schedule deferred work on a bearer + */ +struct udp_bearer { + struct tipc_bearer __rcu *bearer; + struct socket *ubsock; + u32 ifindex; + struct work_struct work; +}; + +/* udp_media_addr_set - convert a ip/udp address to a TIPC media address */ +static void tipc_udp_media_addr_set(struct tipc_media_addr *addr, + struct udp_media_addr *ua) +{ + memset(addr, 0, sizeof(struct tipc_media_addr)); + addr->media_id = TIPC_MEDIA_TYPE_UDP; + memcpy(addr->value, ua, sizeof(struct udp_media_addr)); + if (ntohs(ua->proto) == ETH_P_IP) { + if (ipv4_is_multicast(ua->ipv4.s_addr)) + addr->broadcast = 1; + } else if (ntohs(ua->proto) == ETH_P_IPV6) { + if (ipv6_addr_type(&ua->ipv6) & IPV6_ADDR_MULTICAST) + addr->broadcast = 1; + } else { + pr_err("Invalid UDP media address\n"); + } +} + +/* tipc_udp_addr2str - convert ip/udp address to string */ +static int tipc_udp_addr2str(struct tipc_media_addr *a, char *buf, int size) +{ + struct udp_media_addr *ua = (struct udp_media_addr *)&a->value; + + if (ntohs(ua->proto) == ETH_P_IP) + snprintf(buf, size, "%pI4:%u", &ua->ipv4, ntohs(ua->udp_port)); + else if (ntohs(ua->proto) == ETH_P_IPV6) + snprintf(buf, size, "%pI6:%u", &ua->ipv6, ntohs(ua->udp_port)); + else + pr_err("Invalid UDP media address\n"); + return 0; +} + +/* tipc_udp_msg2addr - extract an ip/udp address from a TIPC ndisc message */ +static int tipc_udp_msg2addr(struct tipc_bearer *b, struct tipc_media_addr *a, + char *msg) +{ + struct udp_media_addr *ua; + + ua = (struct udp_media_addr *) (msg + TIPC_MEDIA_ADDR_OFFSET); + if (msg[TIPC_MEDIA_TYPE_OFFSET] != TIPC_MEDIA_TYPE_UDP) + return -EINVAL; + tipc_udp_media_addr_set(a, ua); + return 0; +} + +/* tipc_udp_addr2msg - write an ip/udp address to a TIPC ndisc message */ +static int tipc_udp_addr2msg(char *msg, struct tipc_media_addr *a) +{ + memset(msg, 0, TIPC_MEDIA_INFO_SIZE); + msg[TIPC_MEDIA_TYPE_OFFSET] = TIPC_MEDIA_TYPE_UDP; + memcpy(msg + TIPC_MEDIA_ADDR_OFFSET, a->value, + sizeof(struct udp_media_addr)); + return 0; +} + +/* tipc_send_msg - enqueue a send request */ +static int tipc_udp_send_msg(struct net *net, struct sk_buff *skb, + struct tipc_bearer *b, + struct tipc_media_addr *dest) +{ + int ttl, err = 0; + struct udp_bearer *ub; + struct udp_media_addr *dst = (struct udp_media_addr *)&dest->value; + struct udp_media_addr *src = (struct udp_media_addr *)&b->addr.value; + struct sk_buff *clone; + struct rtable *rt; + + clone = skb_clone(skb, GFP_ATOMIC); + skb_set_inner_protocol(clone, htons(ETH_P_TIPC)); + ub = rcu_dereference_rtnl(b->media_ptr); + if (!ub) { + err = -ENODEV; + goto tx_error; + } + if (dst->proto == htons(ETH_P_IP)) { + struct flowi4 fl = { + .daddr = dst->ipv4.s_addr, + .saddr = src->ipv4.s_addr, + .flowi4_mark = clone->mark, + .flowi4_proto = IPPROTO_UDP + }; + rt = ip_route_output_key(net, &fl); + if (IS_ERR(rt)) { + err = PTR_ERR(rt); + goto tx_error; + } + ttl = ip4_dst_hoplimit(&rt->dst); + err = udp_tunnel_xmit_skb(rt, ub->ubsock->sk, clone, + src->ipv4.s_addr, + dst->ipv4.s_addr, 0, ttl, 0, + src->udp_port, dst->udp_port, + false, true); + if (err < 0) { + ip_rt_put(rt); + goto tx_error; + } +#if IS_ENABLED(CONFIG_IPV6) + } else { + struct dst_entry *ndst; + struct flowi6 fl6 = { + .flowi6_oif = ub->ifindex, + .daddr = dst->ipv6, + .saddr = src->ipv6, + .flowi6_proto = IPPROTO_UDP + }; + err = ipv6_stub->ipv6_dst_lookup(ub->ubsock->sk, &ndst, &fl6); + if (err) + goto tx_error; + ttl = ip6_dst_hoplimit(ndst); + err = udp_tunnel6_xmit_skb(ndst, ub->ubsock->sk, clone, + ndst->dev, &src->ipv6, + &dst->ipv6, 0, ttl, src->udp_port, + dst->udp_port, false); +#endif + } + return err; + +tx_error: + kfree_skb(clone); + return err; +} + +/* tipc_udp_recv - read data from bearer socket */ +static int tipc_udp_recv(struct sock *sk, struct sk_buff *skb) +{ + struct udp_bearer *ub; + struct tipc_bearer *b; + + ub = rcu_dereference_sk_user_data(sk); + if (!ub) { + pr_err_ratelimited("Failed to get UDP bearer reference"); + kfree_skb(skb); + return 0; + } + + skb_pull(skb, sizeof(struct udphdr)); + rcu_read_lock(); + b = rcu_dereference_rtnl(ub->bearer); + + if (b) { + tipc_rcv(sock_net(sk), skb, b); + rcu_read_unlock(); + return 0; + } + rcu_read_unlock(); + kfree_skb(skb); + return 0; +} + +static int enable_mcast(struct udp_bearer *ub, struct udp_media_addr *remote) +{ + int err = 0; + struct ip_mreqn mreqn; + struct sock *sk = ub->ubsock->sk; + + if (ntohs(remote->proto) == ETH_P_IP) { + if (!ipv4_is_multicast(remote->ipv4.s_addr)) + return 0; + mreqn.imr_multiaddr = remote->ipv4; + mreqn.imr_ifindex = ub->ifindex; + err = ip_mc_join_group(sk, &mreqn); +#if IS_ENABLED(CONFIG_IPV6) + } else { + if (!ipv6_addr_is_multicast(&remote->ipv6)) + return 0; + err = ipv6_stub->ipv6_sock_mc_join(sk, ub->ifindex, + &remote->ipv6); +#endif + } + return err; +} + +/** + * parse_options - build local/remote addresses from configuration + * @attrs: netlink config data + * @ub: UDP bearer instance + * @local: local bearer IP address/port + * @remote: peer or multicast IP/port + */ +static int parse_options(struct nlattr *attrs[], struct udp_bearer *ub, + struct udp_media_addr *local, + struct udp_media_addr *remote) +{ + struct nlattr *opts[TIPC_NLA_UDP_MAX + 1]; + struct sockaddr_storage *sa_local, *sa_remote; + + if (!attrs[TIPC_NLA_BEARER_UDP_OPTS]) + goto err; + if (nla_parse_nested(opts, TIPC_NLA_UDP_MAX, + attrs[TIPC_NLA_BEARER_UDP_OPTS], + tipc_nl_udp_policy)) + goto err; + if (opts[TIPC_NLA_UDP_LOCAL] && opts[TIPC_NLA_UDP_REMOTE]) { + sa_local = nla_data(opts[TIPC_NLA_UDP_LOCAL]); + sa_remote = nla_data(opts[TIPC_NLA_UDP_REMOTE]); + } else { +err: + pr_err("Invalid UDP bearer configuration"); + return -EINVAL; + } + if ((sa_local->ss_family & sa_remote->ss_family) == AF_INET) { + struct sockaddr_in *ip4; + + ip4 = (struct sockaddr_in *)sa_local; + local->proto = htons(ETH_P_IP); + local->udp_port = ip4->sin_port; + local->ipv4.s_addr = ip4->sin_addr.s_addr; + + ip4 = (struct sockaddr_in *)sa_remote; + remote->proto = htons(ETH_P_IP); + remote->udp_port = ip4->sin_port; + remote->ipv4.s_addr = ip4->sin_addr.s_addr; + return 0; + +#if IS_ENABLED(CONFIG_IPV6) + } else if ((sa_local->ss_family & sa_remote->ss_family) == AF_INET6) { + struct sockaddr_in6 *ip6; + + ip6 = (struct sockaddr_in6 *)sa_local; + local->proto = htons(ETH_P_IPV6); + local->udp_port = ip6->sin6_port; + local->ipv6 = ip6->sin6_addr; + ub->ifindex = ip6->sin6_scope_id; + + ip6 = (struct sockaddr_in6 *)sa_remote; + remote->proto = htons(ETH_P_IPV6); + remote->udp_port = ip6->sin6_port; + remote->ipv6 = ip6->sin6_addr; + return 0; +#endif + } + return -EADDRNOTAVAIL; +} + +/** + * tipc_udp_enable - callback to create a new udp bearer instance + * @net: network namespace + * @b: pointer to generic tipc_bearer + * @attrs: netlink bearer configuration + * + * validate the bearer parameters and initialize the udp bearer + * rtnl_lock should be held + */ +static int tipc_udp_enable(struct net *net, struct tipc_bearer *b, + struct nlattr *attrs[]) +{ + int err = -EINVAL; + struct udp_bearer *ub; + struct udp_media_addr *remote; + struct udp_media_addr local = {0}; + struct udp_port_cfg udp_conf = {0}; + struct udp_tunnel_sock_cfg tuncfg = {NULL}; + + ub = kzalloc(sizeof(*ub), GFP_ATOMIC); + if (!ub) + return -ENOMEM; + + remote = (struct udp_media_addr *)&b->bcast_addr.value; + memset(remote, 0, sizeof(struct udp_media_addr)); + err = parse_options(attrs, ub, &local, remote); + if (err) + goto err; + + b->bcast_addr.media_id = TIPC_MEDIA_TYPE_UDP; + b->bcast_addr.broadcast = 1; + rcu_assign_pointer(b->media_ptr, ub); + rcu_assign_pointer(ub->bearer, b); + tipc_udp_media_addr_set(&b->addr, &local); + if (local.proto == htons(ETH_P_IP)) { + struct net_device *dev; + + dev = __ip_dev_find(net, local.ipv4.s_addr, false); + if (!dev) { + err = -ENODEV; + goto err; + } + udp_conf.family = AF_INET; + udp_conf.local_ip.s_addr = htonl(INADDR_ANY); + udp_conf.use_udp_checksums = false; + ub->ifindex = dev->ifindex; + b->mtu = dev->mtu - sizeof(struct iphdr) + - sizeof(struct udphdr); +#if IS_ENABLED(CONFIG_IPV6) + } else if (local.proto == htons(ETH_P_IPV6)) { + udp_conf.family = AF_INET6; + udp_conf.use_udp6_tx_checksums = true; + udp_conf.use_udp6_rx_checksums = true; + udp_conf.local_ip6 = in6addr_any; + b->mtu = 1280; +#endif + } else { + err = -EAFNOSUPPORT; + goto err; + } + udp_conf.local_udp_port = local.udp_port; + err = udp_sock_create(net, &udp_conf, &ub->ubsock); + if (err) + goto err; + tuncfg.sk_user_data = ub; + tuncfg.encap_type = 1; + tuncfg.encap_rcv = tipc_udp_recv; + tuncfg.encap_destroy = NULL; + setup_udp_tunnel_sock(net, ub->ubsock, &tuncfg); + + if (enable_mcast(ub, remote)) + goto err; + return 0; +err: + kfree(ub); + return err; +} + +/* cleanup_bearer - break the socket/bearer association */ +static void cleanup_bearer(struct work_struct *work) +{ + struct udp_bearer *ub = container_of(work, struct udp_bearer, work); + + if (ub->ubsock) + udp_tunnel_sock_release(ub->ubsock); + synchronize_net(); + kfree(ub); +} + +/* tipc_udp_disable - detach bearer from socket */ +static void tipc_udp_disable(struct tipc_bearer *b) +{ + struct udp_bearer *ub; + + ub = rcu_dereference_rtnl(b->media_ptr); + if (!ub) { + pr_err("UDP bearer instance not found\n"); + return; + } + if (ub->ubsock) + sock_set_flag(ub->ubsock->sk, SOCK_DEAD); + RCU_INIT_POINTER(b->media_ptr, NULL); + RCU_INIT_POINTER(ub->bearer, NULL); + + /* sock_release need to be done outside of rtnl lock */ + INIT_WORK(&ub->work, cleanup_bearer); + schedule_work(&ub->work); +} + +struct tipc_media udp_media_info = { + .send_msg = tipc_udp_send_msg, + .enable_media = tipc_udp_enable, + .disable_media = tipc_udp_disable, + .addr2str = tipc_udp_addr2str, + .addr2msg = tipc_udp_addr2msg, + .msg2addr = tipc_udp_msg2addr, + .priority = TIPC_DEF_LINK_PRI, + .tolerance = TIPC_DEF_LINK_TOL, + .window = TIPC_DEF_LINK_WIN, + .type_id = TIPC_MEDIA_TYPE_UDP, + .hwaddr_len = 0, + .name = "udp" +}; |