diff options
Diffstat (limited to 'net/tipc')
38 files changed, 4056 insertions, 5216 deletions
diff --git a/net/tipc/Makefile b/net/tipc/Makefile index b282f7130d2b..b8a13caad59a 100644 --- a/net/tipc/Makefile +++ b/net/tipc/Makefile @@ -5,9 +5,9 @@ obj-$(CONFIG_TIPC) := tipc.o tipc-y += addr.o bcast.o bearer.o config.o \ - core.o handler.o link.o discover.o msg.o \ + core.o link.o discover.o msg.o \ name_distr.o subscr.o name_table.o net.o \ - netlink.o node.o node_subscr.o port.o ref.o \ + netlink.o node.o node_subscr.o \ socket.o log.o eth_media.o server.o tipc-$(CONFIG_TIPC_MEDIA_IB) += ib_media.o diff --git a/net/tipc/addr.h b/net/tipc/addr.h index 60b00ab93d74..a74acf9ee804 100644 --- a/net/tipc/addr.h +++ b/net/tipc/addr.h @@ -37,6 +37,8 @@ #ifndef _TIPC_ADDR_H #define _TIPC_ADDR_H +#include "core.h" + #define TIPC_ZONE_MASK 0xff000000u #define TIPC_CLUSTER_MASK 0xfffff000u diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c index 0d4402587fdf..b8670bf262e2 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, Ericsson AB + * Copyright (c) 2004-2006, 2014, Ericsson AB * Copyright (c) 2004, Intel Corporation. * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. @@ -37,13 +37,14 @@ #include "core.h" #include "link.h" -#include "port.h" +#include "socket.h" +#include "msg.h" #include "bcast.h" #include "name_distr.h" -#define MAX_PKT_DEFAULT_MCAST 1500 /* bcast link max packet size (fixed) */ - -#define BCLINK_WIN_DEFAULT 20 /* bcast link window size (default) */ +#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 @@ -71,7 +72,7 @@ struct tipc_bcbearer_pair { * Note: The fields labelled "temporary" are incorporated into the bearer * to avoid consuming potentially limited stack space through the use of * large local variables within multicast routines. Concurrent access is - * prevented through use of the spinlock "bc_lock". + * prevented through use of the spinlock "bclink_lock". */ struct tipc_bcbearer { struct tipc_bearer bearer; @@ -84,34 +85,69 @@ struct tipc_bcbearer { /** * struct tipc_bclink - link used for broadcast messages + * @lock: spinlock governing access to structure * @link: (non-standard) broadcast link structure * @node: (non-standard) node structure representing b'cast link's peer node + * @flags: represent bclink states * @bcast_nodes: map of broadcast-capable nodes * @retransmit_to: node that most recently requested a retransmit * * Handles sequence numbering, fragmentation, bundling, etc. */ struct tipc_bclink { + spinlock_t lock; struct tipc_link link; struct tipc_node node; + unsigned int flags; struct tipc_node_map bcast_nodes; struct tipc_node *retransmit_to; }; -static struct tipc_bcbearer bcast_bearer; -static struct tipc_bclink bcast_link; - -static struct tipc_bcbearer *bcbearer = &bcast_bearer; -static struct tipc_bclink *bclink = &bcast_link; -static struct tipc_link *bcl = &bcast_link.link; - -static DEFINE_SPINLOCK(bc_lock); +static struct tipc_bcbearer *bcbearer; +static struct tipc_bclink *bclink; +static struct tipc_link *bcl; const char tipc_bclink_name[] = "broadcast-link"; static void tipc_nmap_diff(struct tipc_node_map *nm_a, struct tipc_node_map *nm_b, struct tipc_node_map *nm_diff); +static void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node); +static void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node); + +static void tipc_bclink_lock(void) +{ + spin_lock_bh(&bclink->lock); +} + +static void tipc_bclink_unlock(void) +{ + struct tipc_node *node = NULL; + + if (likely(!bclink->flags)) { + spin_unlock_bh(&bclink->lock); + return; + } + + if (bclink->flags & TIPC_BCLINK_RESET) { + bclink->flags &= ~TIPC_BCLINK_RESET; + node = tipc_bclink_retransmit_to(); + } + spin_unlock_bh(&bclink->lock); + + if (node) + tipc_link_reset_all(node); +} + +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) { @@ -130,16 +166,16 @@ static void bcbuf_decr_acks(struct sk_buff *buf) void tipc_bclink_add_node(u32 addr) { - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); tipc_nmap_add(&bclink->bcast_nodes, addr); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); } void tipc_bclink_remove_node(u32 addr) { - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); tipc_nmap_remove(&bclink->bcast_nodes, addr); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); } static void bclink_set_last_sent(void) @@ -165,7 +201,7 @@ static void bclink_update_last_sent(struct tipc_node *node, u32 seqno) /** * tipc_bclink_retransmit_to - get most recent node to request retransmission * - * Called with bc_lock locked + * Called with bclink_lock locked */ struct tipc_node *tipc_bclink_retransmit_to(void) { @@ -177,7 +213,7 @@ struct tipc_node *tipc_bclink_retransmit_to(void) * @after: sequence number of last packet to *not* retransmit * @to: sequence number of last packet to retransmit * - * Called with bc_lock locked + * Called with bclink_lock locked */ static void bclink_retransmit_pkt(u32 after, u32 to) { @@ -190,11 +226,22 @@ static void bclink_retransmit_pkt(u32 after, u32 to) } /** + * tipc_bclink_wakeup_users - wake up pending users + * + * Called with no locks taken + */ +void tipc_bclink_wakeup_users(void) +{ + while (skb_queue_len(&bclink->link.waiting_sks)) + tipc_sk_rcv(skb_dequeue(&bclink->link.waiting_sks)); +} + +/** * tipc_bclink_acknowledge - handle acknowledgement of broadcast packets * @n_ptr: node that sent acknowledgement info * @acked: broadcast sequence # that has been acknowledged * - * Node is locked, bc_lock unlocked. + * Node is locked, bclink_lock unlocked. */ void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked) { @@ -202,8 +249,7 @@ void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked) struct sk_buff *next; unsigned int released = 0; - spin_lock_bh(&bc_lock); - + tipc_bclink_lock(); /* Bail out if tx queue is empty (no clean up is required) */ crs = bcl->first_out; if (!crs) @@ -264,16 +310,17 @@ void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked) tipc_link_push_queue(bcl); bclink_set_last_sent(); } - if (unlikely(released && !list_empty(&bcl->waiting_ports))) - tipc_link_wakeup_ports(bcl, 0); + if (unlikely(released && !skb_queue_empty(&bcl->waiting_sks))) + n_ptr->action_flags |= TIPC_WAKEUP_BCAST_USERS; + exit: - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); } /** * tipc_bclink_update_link_state - update broadcast link state * - * tipc_net_lock and node lock set + * RCU and node lock set */ void tipc_bclink_update_link_state(struct tipc_node *n_ptr, u32 last_sent) { @@ -320,10 +367,10 @@ void tipc_bclink_update_link_state(struct tipc_node *n_ptr, u32 last_sent) ? buf_seqno(n_ptr->bclink.deferred_head) - 1 : n_ptr->bclink.last_sent); - spin_lock_bh(&bc_lock); - tipc_bearer_send(&bcbearer->bearer, buf, NULL); + tipc_bclink_lock(); + tipc_bearer_send(MAX_BEARERS, buf, NULL); bcl->stats.sent_nacks++; - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); kfree_skb(buf); n_ptr->bclink.oos_state++; @@ -335,8 +382,6 @@ void tipc_bclink_update_link_state(struct tipc_node *n_ptr, u32 last_sent) * * Delay any upcoming NACK by this node if another node has already * requested the first message this node is going to ask for. - * - * Only tipc_net_lock set. */ static void bclink_peek_nack(struct tipc_msg *msg) { @@ -355,36 +400,56 @@ static void bclink_peek_nack(struct tipc_msg *msg) tipc_node_unlock(n_ptr); } -/* - * tipc_bclink_send_msg - broadcast a packet to all nodes in cluster +/* tipc_bclink_xmit - broadcast buffer chain to all nodes in cluster + * and to identified node local sockets + * @buf: 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_send_msg(struct sk_buff *buf) +int tipc_bclink_xmit(struct sk_buff *buf) { - int res; - - spin_lock_bh(&bc_lock); - - if (!bclink->bcast_nodes.count) { - res = msg_data_sz(buf_msg(buf)); - kfree_skb(buf); - goto exit; + int rc = 0; + int bc = 0; + struct sk_buff *clbuf; + + /* Prepare clone of message for local node */ + clbuf = tipc_msg_reassemble(buf); + if (unlikely(!clbuf)) { + kfree_skb_list(buf); + return -EHOSTUNREACH; } - res = tipc_link_send_buf(bcl, buf); - if (likely(res >= 0)) { - bclink_set_last_sent(); - bcl->stats.queue_sz_counts++; - bcl->stats.accu_queue_sz += bcl->out_queue_size; + /* Broadcast to all other nodes */ + if (likely(bclink)) { + tipc_bclink_lock(); + if (likely(bclink->bcast_nodes.count)) { + rc = __tipc_link_xmit(bcl, buf); + if (likely(!rc)) { + bclink_set_last_sent(); + bcl->stats.queue_sz_counts++; + bcl->stats.accu_queue_sz += bcl->out_queue_size; + } + bc = 1; + } + tipc_bclink_unlock(); } -exit: - spin_unlock_bh(&bc_lock); - return res; + + if (unlikely(!bc)) + kfree_skb_list(buf); + + /* Deliver message clone */ + if (likely(!rc)) + tipc_sk_mcast_rcv(clbuf); + else + kfree_skb(clbuf); + + return rc; } /** * bclink_accept_pkt - accept an incoming, in-sequence broadcast packet * - * Called with both sending node's lock and bc_lock taken. + * Called with both sending node's lock and bclink_lock taken. */ static void bclink_accept_pkt(struct tipc_node *node, u32 seqno) { @@ -399,25 +464,24 @@ static void bclink_accept_pkt(struct tipc_node *node, u32 seqno) */ if (((seqno - tipc_own_addr) % TIPC_MIN_LINK_WIN) == 0) { - tipc_link_send_proto_msg( - node->active_links[node->addr & 1], - STATE_MSG, 0, 0, 0, 0, 0); + tipc_link_proto_xmit(node->active_links[node->addr & 1], + STATE_MSG, 0, 0, 0, 0, 0); bcl->stats.sent_acks++; } } /** - * tipc_bclink_recv_pkt - receive a broadcast packet, and deliver upwards + * tipc_bclink_rcv - receive a broadcast packet, and deliver upwards * - * tipc_net_lock is read_locked, no other locks set + * RCU is locked, no other locks set */ -void tipc_bclink_recv_pkt(struct sk_buff *buf) +void tipc_bclink_rcv(struct sk_buff *buf) { struct tipc_msg *msg = buf_msg(buf); struct tipc_node *node; u32 next_in; u32 seqno; - int deferred; + int deferred = 0; /* Screen out unwanted broadcast messages */ @@ -440,12 +504,12 @@ void tipc_bclink_recv_pkt(struct sk_buff *buf) if (msg_destnode(msg) == tipc_own_addr) { tipc_bclink_acknowledge(node, msg_bcast_ack(msg)); tipc_node_unlock(node); - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); bcl->stats.recv_nacks++; bclink->retransmit_to = node; bclink_retransmit_pkt(msg_bcgap_after(msg), msg_bcgap_to(msg)); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); } else { tipc_node_unlock(node); bclink_peek_nack(msg); @@ -463,51 +527,47 @@ receive: /* Deliver message to destination */ if (likely(msg_isdata(msg))) { - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); bclink_accept_pkt(node, seqno); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); tipc_node_unlock(node); if (likely(msg_mcast(msg))) - tipc_port_recv_mcast(buf, NULL); + tipc_sk_mcast_rcv(buf); else kfree_skb(buf); } else if (msg_user(msg) == MSG_BUNDLER) { - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); bclink_accept_pkt(node, seqno); bcl->stats.recv_bundles++; bcl->stats.recv_bundled += msg_msgcnt(msg); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); tipc_node_unlock(node); - tipc_link_recv_bundle(buf); + tipc_link_bundle_rcv(buf); } else if (msg_user(msg) == MSG_FRAGMENTER) { - int ret; - ret = tipc_link_recv_fragment(&node->bclink.reasm_head, - &node->bclink.reasm_tail, - &buf); - if (ret == LINK_REASM_ERROR) + tipc_buf_append(&node->bclink.reasm_buf, &buf); + if (unlikely(!buf && !node->bclink.reasm_buf)) goto unlock; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); bclink_accept_pkt(node, seqno); bcl->stats.recv_fragments++; - if (ret == LINK_REASM_COMPLETE) { + if (buf) { bcl->stats.recv_fragmented++; - /* Point msg to inner header */ msg = buf_msg(buf); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); goto receive; } - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); tipc_node_unlock(node); } else if (msg_user(msg) == NAME_DISTRIBUTOR) { - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); bclink_accept_pkt(node, seqno); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); tipc_node_unlock(node); - tipc_named_recv(buf); + tipc_named_rcv(buf); } else { - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); bclink_accept_pkt(node, seqno); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); tipc_node_unlock(node); kfree_skb(buf); } @@ -537,6 +597,7 @@ receive: buf = node->bclink.deferred_head; node->bclink.deferred_head = buf->next; + buf->next = NULL; node->bclink.deferred_size--; goto receive; } @@ -550,17 +611,16 @@ receive: node->bclink.deferred_size += deferred; bclink_update_last_sent(node, seqno); buf = NULL; - } else - deferred = 0; + } - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); if (deferred) bcl->stats.deferred_recv++; else bcl->stats.duplicates++; - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); unlock: tipc_node_unlock(node); @@ -588,6 +648,7 @@ static int tipc_bcbearer_send(struct sk_buff *buf, struct tipc_bearer *unused1, struct tipc_media_addr *unused2) { int bp_index; + struct tipc_msg *msg = buf_msg(buf); /* Prepare broadcast link message for reliable transmission, * if first time trying to send it; @@ -595,10 +656,7 @@ static int tipc_bcbearer_send(struct sk_buff *buf, struct tipc_bearer *unused1, * since they are sent in an unreliable manner and don't need it */ if (likely(!msg_non_seq(buf_msg(buf)))) { - struct tipc_msg *msg; - bcbuf_set_acks(buf, bclink->bcast_nodes.count); - msg = buf_msg(buf); msg_set_non_seq(msg, 1); msg_set_mc_netid(msg, tipc_net_id); bcl->stats.sent_info++; @@ -615,18 +673,14 @@ static int tipc_bcbearer_send(struct sk_buff *buf, struct tipc_bearer *unused1, for (bp_index = 0; bp_index < MAX_BEARERS; bp_index++) { struct tipc_bearer *p = bcbearer->bpairs[bp_index].primary; struct tipc_bearer *s = bcbearer->bpairs[bp_index].secondary; - struct tipc_bearer *b = p; + struct tipc_bearer *bp[2] = {p, s}; + struct tipc_bearer *b = bp[msg_link_selector(msg)]; struct sk_buff *tbuf; if (!p) break; /* No more bearers to try */ - - if (tipc_bearer_blocked(p)) { - if (!s || tipc_bearer_blocked(s)) - continue; /* Can't use either bearer */ - b = s; - } - + if (!b) + b = p; tipc_nmap_diff(&bcbearer->remains, &b->nodes, &bcbearer->remains_new); if (bcbearer->remains_new.count == bcbearer->remains.count) @@ -634,22 +688,15 @@ static int tipc_bcbearer_send(struct sk_buff *buf, struct tipc_bearer *unused1, if (bp_index == 0) { /* Use original buffer for first bearer */ - tipc_bearer_send(b, buf, &b->bcast_addr); + tipc_bearer_send(b->identity, buf, &b->bcast_addr); } else { /* Avoid concurrent buffer access */ - tbuf = pskb_copy(buf, GFP_ATOMIC); + tbuf = pskb_copy_for_clone(buf, GFP_ATOMIC); if (!tbuf) break; - tipc_bearer_send(b, tbuf, &b->bcast_addr); + tipc_bearer_send(b->identity, tbuf, &b->bcast_addr); kfree_skb(tbuf); /* Bearer keeps a clone */ } - - /* Swap bearers for next packet */ - if (s) { - bcbearer->bpairs[bp_index].primary = s; - bcbearer->bpairs[bp_index].secondary = p; - } - if (bcbearer->remains_new.count == 0) break; /* All targets reached */ @@ -662,22 +709,28 @@ static int tipc_bcbearer_send(struct sk_buff *buf, struct tipc_bearer *unused1, /** * tipc_bcbearer_sort - create sets of bearer pairs used by broadcast bearer */ -void tipc_bcbearer_sort(void) +void tipc_bcbearer_sort(struct tipc_node_map *nm_ptr, u32 node, bool action) { struct tipc_bcbearer_pair *bp_temp = bcbearer->bpairs_temp; struct tipc_bcbearer_pair *bp_curr; + struct tipc_bearer *b; int b_index; int pri; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); + + if (action) + tipc_nmap_add(nm_ptr, node); + else + tipc_nmap_remove(nm_ptr, node); /* Group bearers by priority (can assume max of two per priority) */ memset(bp_temp, 0, sizeof(bcbearer->bpairs_temp)); + rcu_read_lock(); for (b_index = 0; b_index < MAX_BEARERS; b_index++) { - struct tipc_bearer *b = &tipc_bearers[b_index]; - - if (!b->active || !b->nodes.count) + b = rcu_dereference_rtnl(bearer_list[b_index]); + if (!b || !b->nodes.count) continue; if (!bp_temp[b->priority].primary) @@ -685,6 +738,7 @@ void tipc_bcbearer_sort(void) else bp_temp[b->priority].secondary = b; } + rcu_read_unlock(); /* Create array of bearer pairs for broadcasting */ bp_curr = bcbearer->bpairs; @@ -710,7 +764,7 @@ void tipc_bcbearer_sort(void) bp_curr++; } - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); } @@ -722,7 +776,7 @@ int tipc_bclink_stats(char *buf, const u32 buf_size) if (!bcl) return 0; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); s = &bcl->stats; @@ -751,7 +805,7 @@ int tipc_bclink_stats(char *buf, const u32 buf_size) s->queue_sz_counts ? (s->accu_queue_sz / s->queue_sz_counts) : 0); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); return ret; } @@ -760,9 +814,9 @@ int tipc_bclink_reset_stats(void) if (!bcl) return -ENOPROTOOPT; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); memset(&bcl->stats, 0, sizeof(bcl->stats)); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); return 0; } @@ -773,45 +827,60 @@ int tipc_bclink_set_queue_limits(u32 limit) if ((limit < TIPC_MIN_LINK_WIN) || (limit > TIPC_MAX_LINK_WIN)) return -EINVAL; - spin_lock_bh(&bc_lock); + tipc_bclink_lock(); tipc_link_set_queue_limits(bcl, limit); - spin_unlock_bh(&bc_lock); + tipc_bclink_unlock(); return 0; } -void tipc_bclink_init(void) +int tipc_bclink_init(void) { + bcbearer = kzalloc(sizeof(*bcbearer), GFP_ATOMIC); + if (!bcbearer) + return -ENOMEM; + + bclink = kzalloc(sizeof(*bclink), GFP_ATOMIC); + if (!bclink) { + kfree(bcbearer); + return -ENOMEM; + } + + bcl = &bclink->link; bcbearer->bearer.media = &bcbearer->media; bcbearer->media.send_msg = tipc_bcbearer_send; sprintf(bcbearer->media.name, "tipc-broadcast"); - INIT_LIST_HEAD(&bcl->waiting_ports); + spin_lock_init(&bclink->lock); + __skb_queue_head_init(&bcl->waiting_sks); bcl->next_out_no = 1; spin_lock_init(&bclink->node.lock); + __skb_queue_head_init(&bclink->node.waiting_sks); bcl->owner = &bclink->node; bcl->max_pkt = MAX_PKT_DEFAULT_MCAST; tipc_link_set_queue_limits(bcl, BCLINK_WIN_DEFAULT); - spin_lock_init(&bcbearer->bearer.lock); - bcl->b_ptr = &bcbearer->bearer; + bcl->bearer_id = MAX_BEARERS; + rcu_assign_pointer(bearer_list[MAX_BEARERS], &bcbearer->bearer); bcl->state = WORKING_WORKING; strlcpy(bcl->name, tipc_bclink_name, TIPC_MAX_LINK_NAME); + return 0; } void tipc_bclink_stop(void) { - spin_lock_bh(&bc_lock); - tipc_link_stop(bcl); - spin_unlock_bh(&bc_lock); - - memset(bclink, 0, sizeof(*bclink)); - memset(bcbearer, 0, sizeof(*bcbearer)); + tipc_bclink_lock(); + tipc_link_purge_queues(bcl); + tipc_bclink_unlock(); + + RCU_INIT_POINTER(bearer_list[BCBEARER], NULL); + synchronize_net(); + kfree(bcbearer); + kfree(bclink); } - /** * tipc_nmap_add - add a node to a node map */ -void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node) +static void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node) { int n = tipc_node(node); int w = n / WSIZE; @@ -826,7 +895,7 @@ void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node) /** * tipc_nmap_remove - remove a node from a node map */ -void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node) +static void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node) { int n = tipc_node(node); int w = n / WSIZE; diff --git a/net/tipc/bcast.h b/net/tipc/bcast.h index 6ee587b469fd..e7b0f85a82bc 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, Ericsson AB + * Copyright (c) 2003-2006, 2014, Ericsson AB * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * @@ -39,6 +39,7 @@ #define MAX_NODES 4096 #define WSIZE 32 +#define TIPC_BCLINK_RESET 1 /** * struct tipc_node_map - set of node identifiers @@ -69,9 +70,6 @@ struct tipc_node; extern const char tipc_bclink_name[]; -void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node); -void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node); - /** * tipc_nmap_equal - test for equality of node maps */ @@ -84,20 +82,22 @@ static inline int tipc_nmap_equal(struct tipc_node_map *nm_a, void tipc_port_list_add(struct tipc_port_list *pl_ptr, u32 port); void tipc_port_list_free(struct tipc_port_list *pl_ptr); -void tipc_bclink_init(void); +int tipc_bclink_init(void); void tipc_bclink_stop(void); +void tipc_bclink_set_flags(unsigned int flags); void tipc_bclink_add_node(u32 addr); void tipc_bclink_remove_node(u32 addr); struct tipc_node *tipc_bclink_retransmit_to(void); void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked); -int tipc_bclink_send_msg(struct sk_buff *buf); -void tipc_bclink_recv_pkt(struct sk_buff *buf); +void tipc_bclink_rcv(struct sk_buff *buf); u32 tipc_bclink_get_last_sent(void); 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(void); - +void tipc_bcbearer_sort(struct tipc_node_map *nm_ptr, u32 node, bool action); +uint tipc_bclink_get_mtu(void); +int tipc_bclink_xmit(struct sk_buff *buf); +void tipc_bclink_wakeup_users(void); #endif diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c index 3f9707a16d06..264474394f9f 100644 --- a/net/tipc/bearer.c +++ b/net/tipc/bearer.c @@ -1,8 +1,8 @@ /* * net/tipc/bearer.c: TIPC bearer code * - * Copyright (c) 1996-2006, Ericsson AB - * Copyright (c) 2004-2006, 2010-2011, Wind River Systems + * Copyright (c) 1996-2006, 2013, Ericsson AB + * Copyright (c) 2004-2006, 2010-2013, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -41,12 +41,17 @@ #define MAX_ADDR_STR 60 -static struct tipc_media *media_list[MAX_MEDIA]; -static u32 media_count; +static struct tipc_media * const media_info_array[] = { + ð_media_info, +#ifdef CONFIG_TIPC_MEDIA_IB + &ib_media_info, +#endif + NULL +}; -struct tipc_bearer tipc_bearers[MAX_BEARERS]; +struct tipc_bearer __rcu *bearer_list[MAX_BEARERS + 1]; -static void bearer_disable(struct tipc_bearer *b_ptr); +static void bearer_disable(struct tipc_bearer *b_ptr, bool shutting_down); /** * tipc_media_find - locates specified media object by name @@ -55,11 +60,11 @@ struct tipc_media *tipc_media_find(const char *name) { u32 i; - for (i = 0; i < media_count; i++) { - if (!strcmp(media_list[i]->name, name)) - return media_list[i]; + for (i = 0; media_info_array[i] != NULL; i++) { + if (!strcmp(media_info_array[i]->name, name)) + break; } - return NULL; + return media_info_array[i]; } /** @@ -69,44 +74,11 @@ static struct tipc_media *media_find_id(u8 type) { u32 i; - for (i = 0; i < media_count; i++) { - if (media_list[i]->type_id == type) - return media_list[i]; + for (i = 0; media_info_array[i] != NULL; i++) { + if (media_info_array[i]->type_id == type) + break; } - return NULL; -} - -/** - * tipc_register_media - register a media type - * - * Bearers for this media type must be activated separately at a later stage. - */ -int tipc_register_media(struct tipc_media *m_ptr) -{ - int res = -EINVAL; - - write_lock_bh(&tipc_net_lock); - - if ((strlen(m_ptr->name) + 1) > TIPC_MAX_MEDIA_NAME) - goto exit; - if (m_ptr->priority > TIPC_MAX_LINK_PRI) - goto exit; - if ((m_ptr->tolerance < TIPC_MIN_LINK_TOL) || - (m_ptr->tolerance > TIPC_MAX_LINK_TOL)) - goto exit; - if (media_count >= MAX_MEDIA) - goto exit; - if (tipc_media_find(m_ptr->name) || media_find_id(m_ptr->type_id)) - goto exit; - - media_list[media_count] = m_ptr; - media_count++; - res = 0; -exit: - write_unlock_bh(&tipc_net_lock); - if (res) - pr_warn("Media <%s> registration error\n", m_ptr->name); - return res; + return media_info_array[i]; } /** @@ -144,13 +116,11 @@ struct sk_buff *tipc_media_get_names(void) if (!buf) return NULL; - read_lock_bh(&tipc_net_lock); - for (i = 0; i < media_count; i++) { + for (i = 0; media_info_array[i] != NULL; i++) { tipc_cfg_append_tlv(buf, TIPC_TLV_MEDIA_NAME, - media_list[i]->name, - strlen(media_list[i]->name) + 1); + media_info_array[i]->name, + strlen(media_info_array[i]->name) + 1); } - read_unlock_bh(&tipc_net_lock); return buf; } @@ -207,27 +177,9 @@ struct tipc_bearer *tipc_bearer_find(const char *name) struct tipc_bearer *b_ptr; u32 i; - for (i = 0, b_ptr = tipc_bearers; i < MAX_BEARERS; i++, b_ptr++) { - if (b_ptr->active && (!strcmp(b_ptr->name, name))) - return b_ptr; - } - return NULL; -} - -/** - * tipc_bearer_find_interface - locates bearer object with matching interface name - */ -struct tipc_bearer *tipc_bearer_find_interface(const char *if_name) -{ - struct tipc_bearer *b_ptr; - char *b_if_name; - u32 i; - - for (i = 0, b_ptr = tipc_bearers; i < MAX_BEARERS; i++, b_ptr++) { - if (!b_ptr->active) - continue; - b_if_name = strchr(b_ptr->name, ':') + 1; - if (!strcmp(b_if_name, if_name)) + for (i = 0; i < MAX_BEARERS; i++) { + b_ptr = rtnl_dereference(bearer_list[i]); + if (b_ptr && (!strcmp(b_ptr->name, name))) return b_ptr; } return NULL; @@ -239,65 +191,52 @@ struct tipc_bearer *tipc_bearer_find_interface(const char *if_name) struct sk_buff *tipc_bearer_get_names(void) { struct sk_buff *buf; - struct tipc_bearer *b_ptr; + struct tipc_bearer *b; int i, j; buf = tipc_cfg_reply_alloc(MAX_BEARERS * TLV_SPACE(TIPC_MAX_BEARER_NAME)); if (!buf) return NULL; - read_lock_bh(&tipc_net_lock); - for (i = 0; i < media_count; i++) { + for (i = 0; media_info_array[i] != NULL; i++) { for (j = 0; j < MAX_BEARERS; j++) { - b_ptr = &tipc_bearers[j]; - if (b_ptr->active && (b_ptr->media == media_list[i])) { + 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_ptr->name, - strlen(b_ptr->name) + 1); + b->name, + strlen(b->name) + 1); } } } - read_unlock_bh(&tipc_net_lock); return buf; } -void tipc_bearer_add_dest(struct tipc_bearer *b_ptr, u32 dest) +void tipc_bearer_add_dest(u32 bearer_id, u32 dest) { - tipc_nmap_add(&b_ptr->nodes, dest); - tipc_bcbearer_sort(); - tipc_disc_add_dest(b_ptr->link_req); -} - -void tipc_bearer_remove_dest(struct tipc_bearer *b_ptr, u32 dest) -{ - tipc_nmap_remove(&b_ptr->nodes, dest); - tipc_bcbearer_sort(); - tipc_disc_remove_dest(b_ptr->link_req); -} + struct tipc_bearer *b_ptr; -/* - * Interrupt enabling new requests after bearer blocking: - * See bearer_send(). - */ -void tipc_continue(struct tipc_bearer *b) -{ - spin_lock_bh(&b->lock); - b->blocked = 0; - spin_unlock_bh(&b->lock); + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[bearer_id]); + if (b_ptr) { + tipc_bcbearer_sort(&b_ptr->nodes, dest, true); + tipc_disc_add_dest(b_ptr->link_req); + } + rcu_read_unlock(); } -/* - * tipc_bearer_blocked - determines if bearer is currently blocked - */ -int tipc_bearer_blocked(struct tipc_bearer *b) +void tipc_bearer_remove_dest(u32 bearer_id, u32 dest) { - int res; - - spin_lock_bh(&b->lock); - res = b->blocked; - spin_unlock_bh(&b->lock); + struct tipc_bearer *b_ptr; - return res; + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[bearer_id]); + if (b_ptr) { + tipc_bcbearer_sort(&b_ptr->nodes, dest, false); + tipc_disc_remove_dest(b_ptr->link_req); + } + rcu_read_unlock(); } /** @@ -342,13 +281,11 @@ int tipc_enable_bearer(const char *name, u32 disc_domain, u32 priority) return -EINVAL; } - write_lock_bh(&tipc_net_lock); - m_ptr = tipc_media_find(b_names.media_name); if (!m_ptr) { pr_warn("Bearer <%s> rejected, media <%s> not registered\n", name, b_names.media_name); - goto exit; + return -EINVAL; } if (priority == TIPC_MEDIA_LINK_PRI) @@ -358,21 +295,22 @@ restart: bearer_id = MAX_BEARERS; with_this_prio = 1; for (i = MAX_BEARERS; i-- != 0; ) { - if (!tipc_bearers[i].active) { + b_ptr = rtnl_dereference(bearer_list[i]); + if (!b_ptr) { bearer_id = i; continue; } - if (!strcmp(name, tipc_bearers[i].name)) { + if (!strcmp(name, b_ptr->name)) { pr_warn("Bearer <%s> rejected, already enabled\n", name); - goto exit; + return -EINVAL; } - if ((tipc_bearers[i].priority == priority) && + if ((b_ptr->priority == priority) && (++with_this_prio > 2)) { if (priority-- == 0) { pr_warn("Bearer <%s> rejected, duplicate priority\n", name); - goto exit; + return -EINVAL; } pr_warn("Bearer <%s> priority adjustment required %u->%u\n", name, priority + 1, priority); @@ -382,93 +320,79 @@ restart: if (bearer_id >= MAX_BEARERS) { pr_warn("Bearer <%s> rejected, bearer limit reached (%u)\n", name, MAX_BEARERS); - goto exit; + return -EINVAL; } - b_ptr = &tipc_bearers[bearer_id]; + b_ptr = kzalloc(sizeof(*b_ptr), GFP_ATOMIC); + if (!b_ptr) + return -ENOMEM; + strcpy(b_ptr->name, name); + b_ptr->media = m_ptr; res = m_ptr->enable_media(b_ptr); if (res) { pr_warn("Bearer <%s> rejected, enable failure (%d)\n", name, -res); - goto exit; + return -EINVAL; } b_ptr->identity = bearer_id; - b_ptr->media = m_ptr; b_ptr->tolerance = m_ptr->tolerance; b_ptr->window = m_ptr->window; + b_ptr->domain = disc_domain; b_ptr->net_plane = bearer_id + 'A'; - b_ptr->active = 1; b_ptr->priority = priority; - INIT_LIST_HEAD(&b_ptr->links); - spin_lock_init(&b_ptr->lock); - res = tipc_disc_create(b_ptr, &b_ptr->bcast_addr, disc_domain); + res = tipc_disc_create(b_ptr, &b_ptr->bcast_addr); if (res) { - bearer_disable(b_ptr); + bearer_disable(b_ptr, false); pr_warn("Bearer <%s> rejected, discovery object creation failed\n", name); - goto exit; + return -EINVAL; } + + rcu_assign_pointer(bearer_list[bearer_id], b_ptr); + pr_info("Enabled bearer <%s>, discovery domain %s, priority %u\n", name, tipc_addr_string_fill(addr_string, disc_domain), priority); -exit: - write_unlock_bh(&tipc_net_lock); return res; } /** - * tipc_block_bearer - Block the bearer, and reset all its links + * tipc_reset_bearer - Reset all links established over this bearer */ -int tipc_block_bearer(struct tipc_bearer *b_ptr) +static int tipc_reset_bearer(struct tipc_bearer *b_ptr) { - struct tipc_link *l_ptr; - struct tipc_link *temp_l_ptr; - - read_lock_bh(&tipc_net_lock); - pr_info("Blocking bearer <%s>\n", b_ptr->name); - spin_lock_bh(&b_ptr->lock); - b_ptr->blocked = 1; - list_for_each_entry_safe(l_ptr, temp_l_ptr, &b_ptr->links, link_list) { - struct tipc_node *n_ptr = l_ptr->owner; - - spin_lock_bh(&n_ptr->lock); - tipc_link_reset(l_ptr); - spin_unlock_bh(&n_ptr->lock); - } - spin_unlock_bh(&b_ptr->lock); - read_unlock_bh(&tipc_net_lock); + pr_info("Resetting bearer <%s>\n", b_ptr->name); + tipc_link_reset_list(b_ptr->identity); + tipc_disc_reset(b_ptr); return 0; } /** * bearer_disable * - * Note: This routine assumes caller holds tipc_net_lock. + * Note: This routine assumes caller holds RTNL lock. */ -static void bearer_disable(struct tipc_bearer *b_ptr) +static void bearer_disable(struct tipc_bearer *b_ptr, bool shutting_down) { - struct tipc_link *l_ptr; - struct tipc_link *temp_l_ptr; - struct tipc_link_req *temp_req; + u32 i; pr_info("Disabling bearer <%s>\n", b_ptr->name); - spin_lock_bh(&b_ptr->lock); - b_ptr->blocked = 1; b_ptr->media->disable_media(b_ptr); - list_for_each_entry_safe(l_ptr, temp_l_ptr, &b_ptr->links, link_list) { - tipc_link_delete(l_ptr); - } - temp_req = b_ptr->link_req; - b_ptr->link_req = NULL; - spin_unlock_bh(&b_ptr->lock); - if (temp_req) - tipc_disc_delete(temp_req); + tipc_link_delete_list(b_ptr->identity, shutting_down); + if (b_ptr->link_req) + tipc_disc_delete(b_ptr->link_req); - memset(b_ptr, 0, sizeof(struct tipc_bearer)); + for (i = 0; i < MAX_BEARERS; i++) { + if (b_ptr == rtnl_dereference(bearer_list[i])) { + RCU_INIT_POINTER(bearer_list[i], NULL); + break; + } + } + kfree_rcu(b_ptr, rcu); } int tipc_disable_bearer(const char *name) @@ -476,28 +400,230 @@ int tipc_disable_bearer(const char *name) struct tipc_bearer *b_ptr; int res; - write_lock_bh(&tipc_net_lock); b_ptr = tipc_bearer_find(name); if (b_ptr == NULL) { pr_warn("Attempt to disable unknown bearer <%s>\n", name); res = -EINVAL; } else { - bearer_disable(b_ptr); + bearer_disable(b_ptr, false); res = 0; } - write_unlock_bh(&tipc_net_lock); return res; } +int tipc_enable_l2_media(struct tipc_bearer *b) +{ + 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); + if (!dev) + return -ENODEV; + + /* Associate TIPC bearer with L2 bearer */ + rcu_assign_pointer(b->media_ptr, dev); + memset(&b->bcast_addr, 0, sizeof(b->bcast_addr)); + memcpy(b->bcast_addr.value, dev->broadcast, b->media->hwaddr_len); + b->bcast_addr.media_id = b->media->type_id; + b->bcast_addr.broadcast = 1; + b->mtu = dev->mtu; + b->media->raw2addr(b, &b->addr, (char *)dev->dev_addr); + rcu_assign_pointer(dev->tipc_ptr, b); + return 0; +} +/* tipc_disable_l2_media - detach TIPC bearer from an L2 interface + * + * Mark L2 bearer as inactive so that incoming buffers are thrown away, + * then get worker thread to complete bearer cleanup. (Can't do cleanup + * here because cleanup code needs to sleep and caller holds spinlocks.) + */ +void tipc_disable_l2_media(struct tipc_bearer *b) +{ + struct net_device *dev; + + dev = (struct net_device *)rtnl_dereference(b->media_ptr); + RCU_INIT_POINTER(b->media_ptr, NULL); + RCU_INIT_POINTER(dev->tipc_ptr, NULL); + synchronize_net(); + dev_put(dev); +} + +/** + * tipc_l2_send_msg - send a TIPC packet out over an L2 interface + * @buf: the packet to be sent + * @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) +{ + struct sk_buff *clone; + struct net_device *dev; + int delta; + + dev = (struct net_device *)rcu_dereference_rtnl(b->media_ptr); + if (!dev) + return 0; + + clone = skb_clone(buf, GFP_ATOMIC); + if (!clone) + return 0; + + delta = dev->hard_header_len - skb_headroom(buf); + if ((delta > 0) && + pskb_expand_head(clone, SKB_DATA_ALIGN(delta), 0, GFP_ATOMIC)) { + kfree_skb(clone); + return 0; + } + + skb_reset_network_header(clone); + clone->dev = dev; + clone->protocol = htons(ETH_P_TIPC); + dev_hard_header(clone, dev, ETH_P_TIPC, dest->value, + dev->dev_addr, clone->len); + dev_queue_xmit(clone); + return 0; +} + +/* tipc_bearer_send- sends buffer to destination over bearer + * + * IMPORTANT: + * 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, + struct tipc_media_addr *dest) +{ + struct tipc_bearer *b_ptr; + + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[bearer_id]); + if (likely(b_ptr)) + b_ptr->media->send_msg(buf, b_ptr, dest); + rcu_read_unlock(); +} + +/** + * tipc_l2_rcv_msg - handle incoming TIPC message from an interface + * @buf: the received packet + * @dev: the net device that the packet was received on + * @pt: the packet_type structure which was used to register this handler + * @orig_dev: the original receive net device in case the device is a bond + * + * Accept only packets explicitly sent to this node, or broadcast packets; + * ignores packets sent using interface multicast, and traffic sent to other + * nodes (which can happen if interface is running in promiscuous mode). + */ +static int tipc_l2_rcv_msg(struct sk_buff *buf, struct net_device *dev, + struct packet_type *pt, struct net_device *orig_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); + rcu_read_unlock(); + return NET_RX_SUCCESS; + } + } + rcu_read_unlock(); + + kfree_skb(buf); + return NET_RX_DROP; +} + +/** + * tipc_l2_device_event - handle device events from network device + * @nb: the context of the notification + * @evt: the type of event + * @ptr: the net device that the event was on + * + * This function is called by the Ethernet driver in case of link + * change event. + */ +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; + + b_ptr = rtnl_dereference(dev->tipc_ptr); + if (!b_ptr) + return NOTIFY_DONE; + + b_ptr->mtu = dev->mtu; + + switch (evt) { + case NETDEV_CHANGE: + if (netif_carrier_ok(dev)) + break; + case NETDEV_DOWN: + case NETDEV_CHANGEMTU: + tipc_reset_bearer(b_ptr); + break; + case NETDEV_CHANGEADDR: + b_ptr->media->raw2addr(b_ptr, &b_ptr->addr, + (char *)dev->dev_addr); + tipc_reset_bearer(b_ptr); + break; + case NETDEV_UNREGISTER: + case NETDEV_CHANGENAME: + bearer_disable(b_ptr, false); + break; + } + return NOTIFY_OK; +} + +static struct packet_type tipc_packet_type __read_mostly = { + .type = htons(ETH_P_TIPC), + .func = tipc_l2_rcv_msg, +}; + +static struct notifier_block notifier = { + .notifier_call = tipc_l2_device_event, + .priority = 0, +}; + +int tipc_bearer_setup(void) +{ + int err; + + err = register_netdevice_notifier(¬ifier); + if (err) + return err; + dev_add_pack(&tipc_packet_type); + return 0; +} + +void tipc_bearer_cleanup(void) +{ + unregister_netdevice_notifier(¬ifier); + dev_remove_pack(&tipc_packet_type); +} void tipc_bearer_stop(void) { + struct tipc_bearer *b_ptr; u32 i; for (i = 0; i < MAX_BEARERS; i++) { - if (tipc_bearers[i].active) - bearer_disable(&tipc_bearers[i]); + b_ptr = rtnl_dereference(bearer_list[i]); + if (b_ptr) { + bearer_disable(b_ptr, true); + bearer_list[i] = NULL; + } } - media_count = 0; } diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h index e5e04be6fffa..78fccc49de23 100644 --- a/net/tipc/bearer.h +++ b/net/tipc/bearer.h @@ -1,7 +1,7 @@ /* * net/tipc/bearer.h: Include file for TIPC bearer code * - * Copyright (c) 1996-2006, Ericsson AB + * Copyright (c) 1996-2006, 2013, Ericsson AB * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * @@ -42,14 +42,12 @@ #define MAX_BEARERS 2 #define MAX_MEDIA 2 -/* - * Identifiers associated with TIPC message header media address info - * - * - address info field is 20 bytes long - * - media type identifier located at offset 3 - * - remaining bytes vary according to media type +/* Identifiers associated with TIPC message header media address info + * - address info field is 32 bytes long + * - the field's actual content and length is defined per media + * - remaining unused bytes in the field are set to zero */ -#define TIPC_MEDIA_ADDR_SIZE 20 +#define TIPC_MEDIA_ADDR_SIZE 32 #define TIPC_MEDIA_TYPE_OFFSET 3 /* @@ -73,18 +71,19 @@ struct tipc_media_addr { struct tipc_bearer; /** - * struct tipc_media - TIPC media information available to internal users + * struct tipc_media - Media specific info exposed to generic bearer layer * @send_msg: routine which handles buffer transmission * @enable_media: routine which enables a media * @disable_media: routine which disables a media - * @addr2str: routine which converts media address to string - * @addr2msg: routine which converts media address to protocol message area - * @msg2addr: routine which converts media address from protocol message area - * @bcast_addr: media address used in broadcasting + * @addr2str: convert media address format to string + * @addr2msg: convert from media addr format to discovery msg addr format + * @msg2addr: convert from discovery msg addr format to media addr format + * @raw2addr: convert from raw addr format to media addr format * @priority: default link (and bearer) priority * @tolerance: default time (in ms) before declaring link failure * @window: default window (in packets) before declaring link congestion * @type_id: TIPC media identifier + * @hwaddr_len: TIPC media address len * @name: media name */ struct tipc_media { @@ -93,33 +92,39 @@ struct tipc_media { struct tipc_media_addr *dest); int (*enable_media)(struct tipc_bearer *b_ptr); void (*disable_media)(struct tipc_bearer *b_ptr); - int (*addr2str)(struct tipc_media_addr *a, char *str_buf, int str_size); - int (*addr2msg)(struct tipc_media_addr *a, char *msg_area); - int (*msg2addr)(const struct tipc_bearer *b_ptr, - struct tipc_media_addr *a, char *msg_area); + int (*addr2str)(struct tipc_media_addr *addr, + char *strbuf, + int bufsz); + int (*addr2msg)(char *msg, struct tipc_media_addr *addr); + int (*msg2addr)(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg); + int (*raw2addr)(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *raw); u32 priority; u32 tolerance; u32 window; u32 type_id; + u32 hwaddr_len; char name[TIPC_MAX_MEDIA_NAME]; }; /** - * struct tipc_bearer - TIPC bearer structure - * @usr_handle: pointer to additional media-specific information about bearer + * struct tipc_bearer - Generic TIPC bearer structure + * @media_ptr: pointer to additional media-specific information about bearer * @mtu: max packet size bearer can support - * @blocked: non-zero if bearer is blocked - * @lock: spinlock for controlling access to bearer * @addr: media-specific address associated with bearer * @name: bearer name (format = media:interface) * @media: ptr to media structure associated with bearer + * @bcast_addr: media address used in broadcasting + * @rcu: rcu struct for tipc_bearer * @priority: default link priority for bearer * @window: default window size for bearer * @tolerance: default link tolerance for bearer + * @domain: network domain to which links can be established * @identity: array index of this bearer within TIPC bearer array * @link_req: ptr to (optional) structure making periodic link setup requests - * @links: list of non-congested links associated with bearer - * @active: non-zero if bearer structure is represents a bearer * @net_plane: network plane ('A' through 'H') currently associated with bearer * @nodes: indicates which nodes in cluster can be reached through bearer * @@ -128,21 +133,19 @@ struct tipc_media { * care of initializing all other fields. */ struct tipc_bearer { - void *usr_handle; /* initalized by media */ + void __rcu *media_ptr; /* initalized by media */ u32 mtu; /* initalized by media */ - int blocked; /* initalized by media */ struct tipc_media_addr addr; /* initalized by media */ char name[TIPC_MAX_BEARER_NAME]; - spinlock_t lock; struct tipc_media *media; struct tipc_media_addr bcast_addr; + struct rcu_head rcu; u32 priority; u32 window; u32 tolerance; + u32 domain; u32 identity; struct tipc_link_req *link_req; - struct list_head links; - int active; char net_plane; struct tipc_node_map nodes; }; @@ -154,60 +157,43 @@ struct tipc_bearer_names { struct tipc_link; -extern struct tipc_bearer tipc_bearers[]; +extern struct tipc_bearer __rcu *bearer_list[]; /* * TIPC routines available to supported media types */ -int tipc_register_media(struct tipc_media *m_ptr); - -void tipc_recv_msg(struct sk_buff *buf, struct tipc_bearer *tb_ptr); - -int tipc_block_bearer(struct tipc_bearer *b_ptr); -void tipc_continue(struct tipc_bearer *tb_ptr); +void tipc_rcv(struct sk_buff *buf, 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); /* * Routines made available to TIPC by supported media types */ -int tipc_eth_media_start(void); -void tipc_eth_media_stop(void); +extern struct tipc_media eth_media_info; #ifdef CONFIG_TIPC_MEDIA_IB -int tipc_ib_media_start(void); -void tipc_ib_media_stop(void); -#else -static inline int tipc_ib_media_start(void) { return 0; } -static inline void tipc_ib_media_stop(void) { return; } +extern struct tipc_media ib_media_info; #endif 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); +void tipc_disable_l2_media(struct tipc_bearer *b); +int tipc_l2_send_msg(struct sk_buff *buf, struct tipc_bearer *b, + struct tipc_media_addr *dest); struct sk_buff *tipc_bearer_get_names(void); -void tipc_bearer_add_dest(struct tipc_bearer *b_ptr, u32 dest); -void tipc_bearer_remove_dest(struct tipc_bearer *b_ptr, u32 dest); +void tipc_bearer_add_dest(u32 bearer_id, u32 dest); +void tipc_bearer_remove_dest(u32 bearer_id, u32 dest); struct tipc_bearer *tipc_bearer_find(const char *name); -struct tipc_bearer *tipc_bearer_find_interface(const char *if_name); struct tipc_media *tipc_media_find(const char *name); -int tipc_bearer_blocked(struct tipc_bearer *b_ptr); +int tipc_bearer_setup(void); +void tipc_bearer_cleanup(void); void tipc_bearer_stop(void); - -/** - * tipc_bearer_send- sends buffer to destination over bearer - * - * IMPORTANT: - * The media send routine must not alter the buffer being passed in - * as it may be needed for later retransmission! - */ -static inline void tipc_bearer_send(struct tipc_bearer *b, struct sk_buff *buf, - struct tipc_media_addr *dest) -{ - b->media->send_msg(buf, b, dest); -} +void tipc_bearer_send(u32 bearer_id, struct sk_buff *buf, + struct tipc_media_addr *dest); #endif /* _TIPC_BEARER_H */ diff --git a/net/tipc/config.c b/net/tipc/config.c index c301a9a592d8..876f4c6a2631 100644 --- a/net/tipc/config.c +++ b/net/tipc/config.c @@ -35,21 +35,17 @@ */ #include "core.h" -#include "port.h" +#include "socket.h" #include "name_table.h" #include "config.h" #include "server.h" #define REPLY_TRUNCATED "<truncated>\n" -static DEFINE_MUTEX(config_mutex); -static struct tipc_server cfgsrv; - 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; @@ -181,20 +177,10 @@ static struct sk_buff *cfg_set_own_addr(void) if (tipc_own_addr) return tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED " (cannot change node address once assigned)"); - tipc_core_start_net(addr); - return tipc_cfg_reply_none(); -} - -static struct sk_buff *cfg_set_remote_mng(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); + if (!tipc_net_start(addr)) + return tipc_cfg_reply_none(); - value = ntohl(*(__be32 *)TLV_DATA(req_tlv_area)); - tipc_remote_management = (value != 0); - 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) @@ -237,7 +223,7 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area { struct sk_buff *rep_tlv_buf; - mutex_lock(&config_mutex); + rtnl_lock(); /* Save request and reply details in a well-known location */ req_tlv_area = request_area; @@ -247,21 +233,10 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area /* Check command authorization */ if (likely(in_own_node(orig_node))) { /* command is permitted */ - } else if (cmd >= 0x8000) { + } else { rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED " (cannot be done remotely)"); goto exit; - } else if (!tipc_remote_management) { - rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NO_REMOTE); - goto exit; - } else if (cmd >= 0x4000) { - u32 domain = 0; - - if ((tipc_nametbl_translate(TIPC_ZM_SRV, 0, &domain) == 0) || - (domain != orig_node)) { - rep_tlv_buf = tipc_cfg_reply_error_string(TIPC_CFG_NOT_ZONE_MSTR); - goto exit; - } } /* Call appropriate processing routine */ @@ -291,7 +266,7 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area rep_tlv_buf = tipc_media_get_names(); break; case TIPC_CMD_SHOW_PORTS: - rep_tlv_buf = tipc_port_get_ports(); + rep_tlv_buf = tipc_sk_socks_show(); break; case TIPC_CMD_SHOW_STATS: rep_tlv_buf = tipc_show_stats(); @@ -310,18 +285,12 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area case TIPC_CMD_SET_NODE_ADDR: rep_tlv_buf = cfg_set_own_addr(); break; - case TIPC_CMD_SET_REMOTE_MNG: - rep_tlv_buf = cfg_set_remote_mng(); - 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_REMOTE_MNG: - rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_remote_management); - break; case TIPC_CMD_GET_MAX_PORTS: rep_tlv_buf = tipc_cfg_reply_unsigned(tipc_max_ports); break; @@ -345,6 +314,8 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area 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)"); @@ -366,83 +337,6 @@ struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area /* Return reply buffer */ exit: - mutex_unlock(&config_mutex); + rtnl_unlock(); return rep_tlv_buf; } - -static void cfg_conn_msg_event(int conid, struct sockaddr_tipc *addr, - void *usr_data, void *buf, size_t len) -{ - struct tipc_cfg_msg_hdr *req_hdr; - struct tipc_cfg_msg_hdr *rep_hdr; - struct sk_buff *rep_buf; - int ret; - - /* Validate configuration message header (ignore invalid message) */ - req_hdr = (struct tipc_cfg_msg_hdr *)buf; - if ((len < sizeof(*req_hdr)) || - (len != TCM_ALIGN(ntohl(req_hdr->tcm_len))) || - (ntohs(req_hdr->tcm_flags) != TCM_F_REQUEST)) { - pr_warn("Invalid configuration message discarded\n"); - return; - } - - /* Generate reply for request (if can't, return request) */ - rep_buf = tipc_cfg_do_cmd(addr->addr.id.node, ntohs(req_hdr->tcm_type), - buf + sizeof(*req_hdr), - len - sizeof(*req_hdr), - BUF_HEADROOM + MAX_H_SIZE + sizeof(*rep_hdr)); - if (rep_buf) { - skb_push(rep_buf, sizeof(*rep_hdr)); - rep_hdr = (struct tipc_cfg_msg_hdr *)rep_buf->data; - memcpy(rep_hdr, req_hdr, sizeof(*rep_hdr)); - rep_hdr->tcm_len = htonl(rep_buf->len); - rep_hdr->tcm_flags &= htons(~TCM_F_REQUEST); - - ret = tipc_conn_sendmsg(&cfgsrv, conid, addr, rep_buf->data, - rep_buf->len); - if (ret < 0) - pr_err("Sending cfg reply message failed, no memory\n"); - - kfree_skb(rep_buf); - } -} - -static struct sockaddr_tipc cfgsrv_addr __read_mostly = { - .family = AF_TIPC, - .addrtype = TIPC_ADDR_NAMESEQ, - .addr.nameseq.type = TIPC_CFG_SRV, - .addr.nameseq.lower = 0, - .addr.nameseq.upper = 0, - .scope = TIPC_ZONE_SCOPE -}; - -static struct tipc_server cfgsrv __read_mostly = { - .saddr = &cfgsrv_addr, - .imp = TIPC_CRITICAL_IMPORTANCE, - .type = SOCK_RDM, - .max_rcvbuf_size = 64 * 1024, - .name = "cfg_server", - .tipc_conn_recvmsg = cfg_conn_msg_event, - .tipc_conn_new = NULL, - .tipc_conn_shutdown = NULL -}; - -int tipc_cfg_init(void) -{ - return tipc_server_start(&cfgsrv); -} - -void tipc_cfg_reinit(void) -{ - tipc_server_stop(&cfgsrv); - - cfgsrv_addr.addr.nameseq.lower = tipc_own_addr; - cfgsrv_addr.addr.nameseq.upper = tipc_own_addr; - tipc_server_start(&cfgsrv); -} - -void tipc_cfg_stop(void) -{ - tipc_server_stop(&cfgsrv); -} diff --git a/net/tipc/config.h b/net/tipc/config.h index 1f252f3fa058..47b1bf181612 100644 --- a/net/tipc/config.h +++ b/net/tipc/config.h @@ -64,9 +64,4 @@ static inline struct sk_buff *tipc_cfg_reply_ultra_string(char *string) struct sk_buff *tipc_cfg_do_cmd(u32 orig_node, u16 cmd, const void *req_tlv_area, int req_tlv_space, int headroom); - -int tipc_cfg_init(void); -void tipc_cfg_reinit(void); -void tipc_cfg_stop(void); - #endif diff --git a/net/tipc/core.c b/net/tipc/core.c index c6d3f75a9e1b..a5737b8407dd 100644 --- a/net/tipc/core.c +++ b/net/tipc/core.c @@ -1,7 +1,7 @@ /* * net/tipc/core.c: TIPC module code * - * Copyright (c) 2003-2006, Ericsson AB + * Copyright (c) 2003-2006, 2013, Ericsson AB * Copyright (c) 2005-2006, 2010-2013, Wind River Systems * All rights reserved. * @@ -35,11 +35,10 @@ */ #include "core.h" -#include "ref.h" #include "name_table.h" #include "subscr.h" #include "config.h" -#include "port.h" +#include "socket.h" #include <linux/module.h> @@ -50,7 +49,6 @@ int tipc_random __read_mostly; u32 tipc_own_addr __read_mostly; int tipc_max_ports __read_mostly; int tipc_net_id __read_mostly; -int tipc_remote_management __read_mostly; int sysctl_tipc_rmem[3] __read_mostly; /* min/default/max */ /** @@ -77,46 +75,16 @@ struct sk_buff *tipc_buf_acquire(u32 size) } /** - * tipc_core_stop_net - shut down TIPC networking sub-systems - */ -static void tipc_core_stop_net(void) -{ - tipc_net_stop(); - tipc_eth_media_stop(); - tipc_ib_media_stop(); -} - -/** - * start_net - start TIPC networking sub-systems - */ -int tipc_core_start_net(unsigned long addr) -{ - int res; - - tipc_net_start(addr); - res = tipc_eth_media_start(); - if (res < 0) - goto err; - res = tipc_ib_media_start(); - if (res < 0) - goto err; - return res; - -err: - tipc_core_stop_net(); - return res; -} - -/** * tipc_core_stop - switch TIPC from SINGLE NODE to NOT RUNNING mode */ static void tipc_core_stop(void) { + tipc_net_stop(); + tipc_bearer_cleanup(); tipc_netlink_stop(); - tipc_cfg_stop(); tipc_subscr_stop(); tipc_nametbl_stop(); - tipc_ref_table_stop(); + tipc_sk_ref_table_stop(); tipc_socket_stop(); tipc_unregister_sysctl(); } @@ -126,30 +94,53 @@ static void tipc_core_stop(void) */ static int tipc_core_start(void) { - int res; + int err; get_random_bytes(&tipc_random, sizeof(tipc_random)); - res = tipc_handler_start(); - if (!res) - res = tipc_ref_table_init(tipc_max_ports, tipc_random); - if (!res) - res = tipc_nametbl_init(); - if (!res) - res = tipc_netlink_start(); - if (!res) - res = tipc_socket_init(); - if (!res) - res = tipc_register_sysctl(); - if (!res) - res = tipc_subscr_start(); - if (!res) - res = tipc_cfg_init(); - if (res) { - tipc_handler_stop(); - tipc_core_stop(); - } - return res; + err = tipc_sk_ref_table_init(tipc_max_ports, tipc_random); + if (err) + goto out_reftbl; + + err = tipc_nametbl_init(); + if (err) + goto out_nametbl; + + err = tipc_netlink_start(); + if (err) + goto out_netlink; + + err = tipc_socket_init(); + if (err) + goto out_socket; + + err = tipc_register_sysctl(); + if (err) + goto out_sysctl; + + err = tipc_subscr_start(); + if (err) + goto out_subscr; + + err = tipc_bearer_setup(); + if (err) + goto out_bearer; + + return 0; +out_bearer: + tipc_subscr_stop(); +out_subscr: + tipc_unregister_sysctl(); +out_sysctl: + tipc_socket_stop(); +out_socket: + tipc_netlink_stop(); +out_netlink: + tipc_nametbl_stop(); +out_nametbl: + tipc_sk_ref_table_stop(); +out_reftbl: + return err; } static int __init tipc_init(void) @@ -159,14 +150,14 @@ static int __init tipc_init(void) pr_info("Activated (version " TIPC_MOD_VER ")\n"); tipc_own_addr = 0; - tipc_remote_management = 1; tipc_max_ports = CONFIG_TIPC_PORTS; tipc_net_id = 4711; - sysctl_tipc_rmem[0] = CONN_OVERLOAD_LIMIT >> 4 << TIPC_LOW_IMPORTANCE; - sysctl_tipc_rmem[1] = CONN_OVERLOAD_LIMIT >> 4 << + sysctl_tipc_rmem[0] = TIPC_CONN_OVERLOAD_LIMIT >> 4 << + TIPC_LOW_IMPORTANCE; + sysctl_tipc_rmem[1] = TIPC_CONN_OVERLOAD_LIMIT >> 4 << TIPC_CRITICAL_IMPORTANCE; - sysctl_tipc_rmem[2] = CONN_OVERLOAD_LIMIT; + sysctl_tipc_rmem[2] = TIPC_CONN_OVERLOAD_LIMIT; res = tipc_core_start(); if (res) @@ -178,8 +169,6 @@ static int __init tipc_init(void) static void __exit tipc_exit(void) { - tipc_handler_stop(); - tipc_core_stop_net(); tipc_core_stop(); pr_info("Deactivated\n"); } diff --git a/net/tipc/core.h b/net/tipc/core.h index 94895d4e86ab..f773b148722f 100644 --- a/net/tipc/core.h +++ b/net/tipc/core.h @@ -47,7 +47,7 @@ #include <linux/mm.h> #include <linux/timer.h> #include <linux/string.h> -#include <asm/uaccess.h> +#include <linux/uaccess.h> #include <linux/interrupt.h> #include <linux/atomic.h> #include <asm/hardirq.h> @@ -56,7 +56,8 @@ #include <linux/list.h> #include <linux/slab.h> #include <linux/vmalloc.h> - +#include <linux/rtnetlink.h> +#include <linux/etherdevice.h> #define TIPC_MOD_VER "2.0.0" @@ -79,8 +80,8 @@ int tipc_snprintf(char *buf, int len, const char *fmt, ...); extern u32 tipc_own_addr __read_mostly; extern int tipc_max_ports __read_mostly; extern int tipc_net_id __read_mostly; -extern int tipc_remote_management __read_mostly; extern int sysctl_tipc_rmem[3] __read_mostly; +extern int sysctl_tipc_named_timeout __read_mostly; /* * Other global variables @@ -90,9 +91,6 @@ extern int tipc_random __read_mostly; /* * Routines available to privileged subsystems */ -int tipc_core_start_net(unsigned long); -int tipc_handler_start(void); -void tipc_handler_stop(void); int tipc_netlink_start(void); void tipc_netlink_stop(void); int tipc_socket_init(void); @@ -111,12 +109,10 @@ void tipc_unregister_sysctl(void); #endif /* - * TIPC timer and signal code + * TIPC timer code */ typedef void (*Handler) (unsigned long); -u32 tipc_k_signal(Handler routine, unsigned long argument); - /** * k_init_timer - initialize a timer * @timer: pointer to timer structure @@ -192,6 +188,11 @@ static inline void k_term_timer(struct timer_list *timer) struct tipc_skb_cb { void *handle; + struct sk_buff *tail; + bool deferred; + bool wakeup_pending; + u16 chain_sz; + u16 chain_imp; }; #define TIPC_SKB_CB(__skb) ((struct tipc_skb_cb *)&((__skb)->cb[0])) diff --git a/net/tipc/discover.c b/net/tipc/discover.c index ecc758c6eacf..aa722a42ef8b 100644 --- a/net/tipc/discover.c +++ b/net/tipc/discover.c @@ -1,7 +1,7 @@ /* * net/tipc/discover.c * - * Copyright (c) 2003-2006, Ericsson AB + * Copyright (c) 2003-2006, 2014, Ericsson AB * Copyright (c) 2005-2006, 2010-2011, Wind River Systems * All rights reserved. * @@ -46,19 +46,21 @@ /** * struct tipc_link_req - information about an ongoing link setup request - * @bearer: bearer issuing requests + * @bearer_id: identity of bearer issuing requests * @dest: destination address for request messages * @domain: network domain to which links can be established * @num_nodes: number of nodes currently discovered (i.e. with an active link) + * @lock: spinlock for controlling access to requests * @buf: request message to be (repeatedly) sent * @timer: timer governing period between requests * @timer_intv: current interval between requests (in ms) */ struct tipc_link_req { - struct tipc_bearer *bearer; + u32 bearer_id; struct tipc_media_addr dest; u32 domain; int num_nodes; + spinlock_t lock; struct sk_buff *buf; struct timer_list timer; unsigned int timer_intv; @@ -67,25 +69,21 @@ struct tipc_link_req { /** * tipc_disc_init_msg - initialize a link setup message * @type: message type (request or response) - * @dest_domain: network domain of node(s) which should respond to message * @b_ptr: ptr to bearer issuing message */ -static struct sk_buff *tipc_disc_init_msg(u32 type, u32 dest_domain, - struct tipc_bearer *b_ptr) +static void tipc_disc_init_msg(struct sk_buff *buf, u32 type, + struct tipc_bearer *b_ptr) { - struct sk_buff *buf = tipc_buf_acquire(INT_H_SIZE); struct tipc_msg *msg; - - if (buf) { - msg = buf_msg(buf); - tipc_msg_init(msg, LINK_CONFIG, type, INT_H_SIZE, dest_domain); - msg_set_non_seq(msg, 1); - msg_set_node_sig(msg, tipc_random); - msg_set_dest_domain(msg, dest_domain); - msg_set_bc_netid(msg, tipc_net_id); - b_ptr->media->addr2msg(&b_ptr->addr, msg_media_addr(msg)); - } - return buf; + u32 dest_domain = b_ptr->domain; + + msg = buf_msg(buf); + tipc_msg_init(msg, LINK_CONFIG, type, INT_H_SIZE, dest_domain); + msg_set_non_seq(msg, 1); + msg_set_node_sig(msg, tipc_random); + msg_set_dest_domain(msg, dest_domain); + msg_set_bc_netid(msg, tipc_net_id); + b_ptr->media->addr2msg(msg_media_addr(msg), &b_ptr->addr); } /** @@ -108,146 +106,150 @@ static void disc_dupl_alert(struct tipc_bearer *b_ptr, u32 node_addr, } /** - * tipc_disc_recv_msg - handle incoming link setup message (request or response) + * tipc_disc_rcv - handle incoming discovery message (request or response) * @buf: buffer containing message - * @b_ptr: bearer that message arrived on + * @bearer: bearer that message arrived on */ -void tipc_disc_recv_msg(struct sk_buff *buf, struct tipc_bearer *b_ptr) +void tipc_disc_rcv(struct sk_buff *buf, struct tipc_bearer *bearer) { - struct tipc_node *n_ptr; + struct tipc_node *node; struct tipc_link *link; - struct tipc_media_addr media_addr; + struct tipc_media_addr maddr; struct sk_buff *rbuf; struct tipc_msg *msg = buf_msg(buf); - u32 dest = msg_dest_domain(msg); - u32 orig = msg_prevnode(msg); + u32 ddom = msg_dest_domain(msg); + u32 onode = msg_prevnode(msg); u32 net_id = msg_bc_netid(msg); - u32 type = msg_type(msg); + u32 mtyp = msg_type(msg); u32 signature = msg_node_sig(msg); - int addr_mismatch; - int link_fully_up; - - media_addr.broadcast = 1; - b_ptr->media->msg2addr(b_ptr, &media_addr, msg_media_addr(msg)); + bool addr_match = false; + bool sign_match = false; + bool link_up = false; + bool accept_addr = false; + bool accept_sign = false; + bool respond = false; + + bearer->media->msg2addr(bearer, &maddr, msg_media_addr(msg)); kfree_skb(buf); /* Ensure message from node is valid and communication is permitted */ if (net_id != tipc_net_id) return; - if (media_addr.broadcast) + if (maddr.broadcast) return; - if (!tipc_addr_domain_valid(dest)) + if (!tipc_addr_domain_valid(ddom)) return; - if (!tipc_addr_node_valid(orig)) + if (!tipc_addr_node_valid(onode)) return; - if (orig == tipc_own_addr) { - if (memcmp(&media_addr, &b_ptr->addr, sizeof(media_addr))) - disc_dupl_alert(b_ptr, tipc_own_addr, &media_addr); + + if (in_own_node(onode)) { + if (memcmp(&maddr, &bearer->addr, sizeof(maddr))) + disc_dupl_alert(bearer, tipc_own_addr, &maddr); return; } - if (!tipc_in_scope(dest, tipc_own_addr)) + if (!tipc_in_scope(ddom, tipc_own_addr)) return; - if (!tipc_in_scope(b_ptr->link_req->domain, orig)) + if (!tipc_in_scope(bearer->domain, onode)) return; - /* Locate structure corresponding to requesting node */ - n_ptr = tipc_node_find(orig); - if (!n_ptr) { - n_ptr = tipc_node_create(orig); - if (!n_ptr) - return; - } - tipc_node_lock(n_ptr); + /* Locate, or if necessary, create, node: */ + node = tipc_node_find(onode); + if (!node) + node = tipc_node_create(onode); + if (!node) + return; - /* Prepare to validate requesting node's signature and media address */ - link = n_ptr->links[b_ptr->identity]; - addr_mismatch = (link != NULL) && - memcmp(&link->media_addr, &media_addr, sizeof(media_addr)); + tipc_node_lock(node); + link = node->links[bearer->identity]; - /* - * Ensure discovery message's signature is correct - * - * If signature is incorrect and there is no working link to the node, - * accept the new signature but invalidate all existing links to the - * node so they won't re-activate without a new discovery message. - * - * If signature is incorrect and the requested link to the node is - * working, accept the new signature. (This is an instance of delayed - * rediscovery, where a link endpoint was able to re-establish contact - * with its peer endpoint on a node that rebooted before receiving a - * discovery message from that node.) - * - * If signature is incorrect and there is a working link to the node - * that is not the requested link, reject the request (must be from - * a duplicate node). - */ - if (signature != n_ptr->signature) { - if (n_ptr->working_links == 0) { - struct tipc_link *curr_link; - int i; - - for (i = 0; i < MAX_BEARERS; i++) { - curr_link = n_ptr->links[i]; - if (curr_link) { - memset(&curr_link->media_addr, 0, - sizeof(media_addr)); - tipc_link_reset(curr_link); - } - } - addr_mismatch = (link != NULL); - } else if (tipc_link_is_up(link) && !addr_mismatch) { - /* delayed rediscovery */ - } else { - disc_dupl_alert(b_ptr, orig, &media_addr); - tipc_node_unlock(n_ptr); - return; - } - n_ptr->signature = signature; + /* Prepare to validate requesting node's signature and media address */ + sign_match = (signature == node->signature); + addr_match = link && !memcmp(&link->media_addr, &maddr, sizeof(maddr)); + link_up = link && tipc_link_is_up(link); + + + /* These three flags give us eight permutations: */ + + if (sign_match && addr_match && link_up) { + /* All is fine. Do nothing. */ + } else if (sign_match && addr_match && !link_up) { + /* Respond. The link will come up in due time */ + respond = true; + } else if (sign_match && !addr_match && link_up) { + /* Peer has changed i/f address without rebooting. + * If so, the link will reset soon, and the next + * discovery will be accepted. So we can ignore it. + * It may also be an cloned or malicious peer having + * chosen the same node address and signature as an + * existing one. + * Ignore requests until the link goes down, if ever. + */ + disc_dupl_alert(bearer, onode, &maddr); + } else if (sign_match && !addr_match && !link_up) { + /* Peer link has changed i/f address without rebooting. + * It may also be a cloned or malicious peer; we can't + * distinguish between the two. + * The signature is correct, so we must accept. + */ + accept_addr = true; + respond = true; + } else if (!sign_match && addr_match && link_up) { + /* Peer node rebooted. Two possibilities: + * - Delayed re-discovery; this link endpoint has already + * reset and re-established contact with the peer, before + * receiving a discovery message from that node. + * (The peer happened to receive one from this node first). + * - The peer came back so fast that our side has not + * discovered it yet. Probing from this side will soon + * reset the link, since there can be no working link + * endpoint at the peer end, and the link will re-establish. + * Accept the signature, since it comes from a known peer. + */ + accept_sign = true; + } else if (!sign_match && addr_match && !link_up) { + /* The peer node has rebooted. + * Accept signature, since it is a known peer. + */ + accept_sign = true; + respond = true; + } else if (!sign_match && !addr_match && link_up) { + /* Peer rebooted with new address, or a new/duplicate peer. + * Ignore until the link goes down, if ever. + */ + disc_dupl_alert(bearer, onode, &maddr); + } else if (!sign_match && !addr_match && !link_up) { + /* Peer rebooted with new address, or it is a new peer. + * Accept signature and address. + */ + accept_sign = true; + accept_addr = true; + respond = true; } - /* - * Ensure requesting node's media address is correct - * - * If media address doesn't match and the link is working, reject the - * request (must be from a duplicate node). - * - * If media address doesn't match and the link is not working, accept - * the new media address and reset the link to ensure it starts up - * cleanly. - */ - if (addr_mismatch) { - if (tipc_link_is_up(link)) { - disc_dupl_alert(b_ptr, orig, &media_addr); - tipc_node_unlock(n_ptr); - return; - } else { - memcpy(&link->media_addr, &media_addr, - sizeof(media_addr)); - tipc_link_reset(link); - } - } + if (accept_sign) + node->signature = signature; - /* Create a link endpoint for this bearer, if necessary */ - if (!link) { - link = tipc_link_create(n_ptr, b_ptr, &media_addr); - if (!link) { - tipc_node_unlock(n_ptr); - return; + if (accept_addr) { + if (!link) + link = tipc_link_create(node, bearer, &maddr); + if (link) { + memcpy(&link->media_addr, &maddr, sizeof(maddr)); + tipc_link_reset(link); + } else { + respond = false; } } - /* Accept discovery message & send response, if necessary */ - link_fully_up = link_working_working(link); - - if ((type == DSC_REQ_MSG) && !link_fully_up && !b_ptr->blocked) { - rbuf = tipc_disc_init_msg(DSC_RESP_MSG, orig, b_ptr); + /* Send response, if necessary */ + if (respond && (mtyp == DSC_REQ_MSG)) { + rbuf = tipc_buf_acquire(INT_H_SIZE); if (rbuf) { - tipc_bearer_send(b_ptr, rbuf, &media_addr); + tipc_disc_init_msg(rbuf, DSC_RESP_MSG, bearer); + tipc_bearer_send(bearer->identity, rbuf, &maddr); kfree_skb(rbuf); } } - - tipc_node_unlock(n_ptr); + tipc_node_unlock(node); } /** @@ -274,7 +276,9 @@ static void disc_update(struct tipc_link_req *req) */ void tipc_disc_add_dest(struct tipc_link_req *req) { + spin_lock_bh(&req->lock); req->num_nodes++; + spin_unlock_bh(&req->lock); } /** @@ -283,18 +287,10 @@ void tipc_disc_add_dest(struct tipc_link_req *req) */ void tipc_disc_remove_dest(struct tipc_link_req *req) { + spin_lock_bh(&req->lock); req->num_nodes--; disc_update(req); -} - -/** - * disc_send_msg - send link setup request message - * @req: ptr to link request structure - */ -static void disc_send_msg(struct tipc_link_req *req) -{ - if (!req->bearer->blocked) - tipc_bearer_send(req->bearer, req->buf, &req->dest); + spin_unlock_bh(&req->lock); } /** @@ -307,7 +303,7 @@ static void disc_timeout(struct tipc_link_req *req) { int max_delay; - spin_lock_bh(&req->bearer->lock); + spin_lock_bh(&req->lock); /* Stop searching if only desired node has been found */ if (tipc_node(req->domain) && req->num_nodes) { @@ -322,7 +318,8 @@ 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 */ - disc_send_msg(req); + tipc_bearer_send(req->bearer_id, req->buf, &req->dest); + req->timer_intv *= 2; if (req->num_nodes) @@ -334,7 +331,7 @@ static void disc_timeout(struct tipc_link_req *req) k_start_timer(&req->timer, req->timer_intv); exit: - spin_unlock_bh(&req->bearer->lock); + spin_unlock_bh(&req->lock); } /** @@ -345,8 +342,7 @@ exit: * * Returns 0 if successful, otherwise -errno. */ -int tipc_disc_create(struct tipc_bearer *b_ptr, struct tipc_media_addr *dest, - u32 dest_domain) +int tipc_disc_create(struct tipc_bearer *b_ptr, struct tipc_media_addr *dest) { struct tipc_link_req *req; @@ -354,21 +350,23 @@ int tipc_disc_create(struct tipc_bearer *b_ptr, struct tipc_media_addr *dest, if (!req) return -ENOMEM; - req->buf = tipc_disc_init_msg(DSC_REQ_MSG, dest_domain, b_ptr); + req->buf = tipc_buf_acquire(INT_H_SIZE); if (!req->buf) { kfree(req); - return -ENOMSG; + return -ENOMEM; } + tipc_disc_init_msg(req->buf, DSC_REQ_MSG, b_ptr); memcpy(&req->dest, dest, sizeof(*dest)); - req->bearer = b_ptr; - req->domain = dest_domain; + req->bearer_id = b_ptr->identity; + req->domain = b_ptr->domain; req->num_nodes = 0; req->timer_intv = TIPC_LINK_REQ_INIT; + spin_lock_init(&req->lock); k_init_timer(&req->timer, (Handler)disc_timeout, (unsigned long)req); k_start_timer(&req->timer, req->timer_intv); b_ptr->link_req = req; - disc_send_msg(req); + tipc_bearer_send(req->bearer_id, req->buf, &req->dest); return 0; } @@ -383,3 +381,23 @@ void tipc_disc_delete(struct tipc_link_req *req) kfree_skb(req->buf); kfree(req); } + +/** + * tipc_disc_reset - reset object to send periodic link setup requests + * @b_ptr: ptr to bearer issuing requests + * @dest_domain: network domain to which links can be established + */ +void tipc_disc_reset(struct tipc_bearer *b_ptr) +{ + struct tipc_link_req *req = b_ptr->link_req; + + spin_lock_bh(&req->lock); + tipc_disc_init_msg(req->buf, DSC_REQ_MSG, b_ptr); + req->bearer_id = b_ptr->identity; + req->domain = b_ptr->domain; + req->num_nodes = 0; + req->timer_intv = TIPC_LINK_REQ_INIT; + k_start_timer(&req->timer, req->timer_intv); + tipc_bearer_send(req->bearer_id, req->buf, &req->dest); + spin_unlock_bh(&req->lock); +} diff --git a/net/tipc/discover.h b/net/tipc/discover.h index 75b67c403aa3..515b57392f4d 100644 --- a/net/tipc/discover.h +++ b/net/tipc/discover.h @@ -39,11 +39,11 @@ struct tipc_link_req; -int tipc_disc_create(struct tipc_bearer *b_ptr, struct tipc_media_addr *dest, - u32 dest_domain); +int tipc_disc_create(struct tipc_bearer *b_ptr, struct tipc_media_addr *dest); void tipc_disc_delete(struct tipc_link_req *req); +void tipc_disc_reset(struct tipc_bearer *b_ptr); void tipc_disc_add_dest(struct tipc_link_req *req); void tipc_disc_remove_dest(struct tipc_link_req *req); -void tipc_disc_recv_msg(struct sk_buff *buf, struct tipc_bearer *b_ptr); +void tipc_disc_rcv(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 f80d59f5a161..5e1426f1751f 100644 --- a/net/tipc/eth_media.c +++ b/net/tipc/eth_media.c @@ -1,7 +1,7 @@ /* * net/tipc/eth_media.c: Ethernet bearer support for TIPC * - * Copyright (c) 2001-2007, Ericsson AB + * Copyright (c) 2001-2007, 2013-2014, Ericsson AB * Copyright (c) 2005-2008, 2011-2013, Wind River Systems * All rights reserved. * @@ -37,340 +37,65 @@ #include "core.h" #include "bearer.h" -#define MAX_ETH_MEDIA MAX_BEARERS +#define ETH_ADDR_OFFSET 4 /* MAC addr position inside address field */ -#define ETH_ADDR_OFFSET 4 /* message header offset of MAC address */ - -/** - * struct eth_media - Ethernet bearer data structure - * @bearer: ptr to associated "generic" bearer structure - * @dev: ptr to associated Ethernet network device - * @tipc_packet_type: used in binding TIPC to Ethernet driver - * @setup: work item used when enabling bearer - * @cleanup: work item used when disabling bearer - */ -struct eth_media { - struct tipc_bearer *bearer; - struct net_device *dev; - struct packet_type tipc_packet_type; - struct work_struct setup; - struct work_struct cleanup; -}; - -static struct tipc_media eth_media_info; -static struct eth_media eth_media_array[MAX_ETH_MEDIA]; -static int eth_started; - -static int recv_notification(struct notifier_block *nb, unsigned long evt, - void *dv); -/* - * Network device notifier info - */ -static struct notifier_block notifier = { - .notifier_call = recv_notification, - .priority = 0 -}; - -/** - * eth_media_addr_set - initialize Ethernet media address structure - * - * Media-dependent "value" field stores MAC address in first 6 bytes - * and zeroes out the remaining bytes. - */ -static void eth_media_addr_set(const struct tipc_bearer *tb_ptr, - struct tipc_media_addr *a, char *mac) +/* Convert Ethernet address (media address format) to string */ +static int tipc_eth_addr2str(struct tipc_media_addr *addr, + char *strbuf, int bufsz) { - memcpy(a->value, mac, ETH_ALEN); - memset(a->value + ETH_ALEN, 0, sizeof(a->value) - ETH_ALEN); - a->media_id = TIPC_MEDIA_TYPE_ETH; - a->broadcast = !memcmp(mac, tb_ptr->bcast_addr.value, ETH_ALEN); -} - -/** - * send_msg - send a TIPC message out over an Ethernet interface - */ -static int send_msg(struct sk_buff *buf, struct tipc_bearer *tb_ptr, - struct tipc_media_addr *dest) -{ - struct sk_buff *clone; - struct net_device *dev; - int delta; - - clone = skb_clone(buf, GFP_ATOMIC); - if (!clone) - return 0; - - dev = ((struct eth_media *)(tb_ptr->usr_handle))->dev; - delta = dev->hard_header_len - skb_headroom(buf); - - if ((delta > 0) && - pskb_expand_head(clone, SKB_DATA_ALIGN(delta), 0, GFP_ATOMIC)) { - kfree_skb(clone); - return 0; - } + if (bufsz < 18) /* 18 = strlen("aa:bb:cc:dd:ee:ff\0") */ + return 1; - skb_reset_network_header(clone); - clone->dev = dev; - clone->protocol = htons(ETH_P_TIPC); - dev_hard_header(clone, dev, ETH_P_TIPC, dest->value, - dev->dev_addr, clone->len); - dev_queue_xmit(clone); + sprintf(strbuf, "%pM", addr->value); return 0; } -/** - * recv_msg - handle incoming TIPC message from an Ethernet interface - * - * Accept only packets explicitly sent to this node, or broadcast packets; - * ignores packets sent using Ethernet multicast, and traffic sent to other - * nodes (which can happen if interface is running in promiscuous mode). - */ -static int recv_msg(struct sk_buff *buf, struct net_device *dev, - struct packet_type *pt, struct net_device *orig_dev) -{ - struct eth_media *eb_ptr = (struct eth_media *)pt->af_packet_priv; - - if (!net_eq(dev_net(dev), &init_net)) { - kfree_skb(buf); - return NET_RX_DROP; - } - - if (likely(eb_ptr->bearer)) { - if (likely(buf->pkt_type <= PACKET_BROADCAST)) { - buf->next = NULL; - tipc_recv_msg(buf, eb_ptr->bearer); - return NET_RX_SUCCESS; - } - } - kfree_skb(buf); - return NET_RX_DROP; -} - -/** - * setup_media - setup association between Ethernet bearer and interface - */ -static void setup_media(struct work_struct *work) -{ - struct eth_media *eb_ptr = - container_of(work, struct eth_media, setup); - - dev_add_pack(&eb_ptr->tipc_packet_type); -} - -/** - * enable_media - attach TIPC bearer to an Ethernet interface - */ -static int enable_media(struct tipc_bearer *tb_ptr) +/* Convert from media address format to discovery message addr format */ +static int tipc_eth_addr2msg(char *msg, struct tipc_media_addr *addr) { - struct net_device *dev; - struct eth_media *eb_ptr = ð_media_array[0]; - struct eth_media *stop = ð_media_array[MAX_ETH_MEDIA]; - char *driver_name = strchr((const char *)tb_ptr->name, ':') + 1; - int pending_dev = 0; - - /* Find unused Ethernet bearer structure */ - while (eb_ptr->dev) { - if (!eb_ptr->bearer) - pending_dev++; - if (++eb_ptr == stop) - return pending_dev ? -EAGAIN : -EDQUOT; - } - - /* Find device with specified name */ - dev = dev_get_by_name(&init_net, driver_name); - if (!dev) - return -ENODEV; - - /* Create Ethernet bearer for device */ - eb_ptr->dev = dev; - eb_ptr->tipc_packet_type.type = htons(ETH_P_TIPC); - eb_ptr->tipc_packet_type.dev = dev; - eb_ptr->tipc_packet_type.func = recv_msg; - eb_ptr->tipc_packet_type.af_packet_priv = eb_ptr; - INIT_LIST_HEAD(&(eb_ptr->tipc_packet_type.list)); - INIT_WORK(&eb_ptr->setup, setup_media); - schedule_work(&eb_ptr->setup); - - /* Associate TIPC bearer with Ethernet bearer */ - eb_ptr->bearer = tb_ptr; - tb_ptr->usr_handle = (void *)eb_ptr; - memset(tb_ptr->bcast_addr.value, 0, sizeof(tb_ptr->bcast_addr.value)); - memcpy(tb_ptr->bcast_addr.value, dev->broadcast, ETH_ALEN); - tb_ptr->bcast_addr.media_id = TIPC_MEDIA_TYPE_ETH; - tb_ptr->bcast_addr.broadcast = 1; - tb_ptr->mtu = dev->mtu; - tb_ptr->blocked = 0; - eth_media_addr_set(tb_ptr, &tb_ptr->addr, (char *)dev->dev_addr); + memset(msg, 0, TIPC_MEDIA_ADDR_SIZE); + msg[TIPC_MEDIA_TYPE_OFFSET] = TIPC_MEDIA_TYPE_ETH; + memcpy(msg + ETH_ADDR_OFFSET, addr->value, ETH_ALEN); return 0; } -/** - * cleanup_media - break association between Ethernet bearer and interface - * - * This routine must be invoked from a work queue because it can sleep. - */ -static void cleanup_media(struct work_struct *work) +/* Convert raw mac address format to media addr format */ +static int tipc_eth_raw2addr(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg) { - struct eth_media *eb_ptr = - container_of(work, struct eth_media, cleanup); - - dev_remove_pack(&eb_ptr->tipc_packet_type); - dev_put(eb_ptr->dev); - eb_ptr->dev = NULL; -} + char bcast_mac[ETH_ALEN] = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff}; -/** - * disable_media - detach TIPC bearer from an Ethernet interface - * - * Mark Ethernet bearer as inactive so that incoming buffers are thrown away, - * then get worker thread to complete bearer cleanup. (Can't do cleanup - * here because cleanup code needs to sleep and caller holds spinlocks.) - */ -static void disable_media(struct tipc_bearer *tb_ptr) -{ - struct eth_media *eb_ptr = (struct eth_media *)tb_ptr->usr_handle; - - eb_ptr->bearer = NULL; - INIT_WORK(&eb_ptr->cleanup, cleanup_media); - schedule_work(&eb_ptr->cleanup); -} - -/** - * recv_notification - handle device updates from OS - * - * Change the state of the Ethernet bearer (if any) associated with the - * specified device. - */ -static int recv_notification(struct notifier_block *nb, unsigned long evt, - void *ptr) -{ - struct net_device *dev = netdev_notifier_info_to_dev(ptr); - struct eth_media *eb_ptr = ð_media_array[0]; - struct eth_media *stop = ð_media_array[MAX_ETH_MEDIA]; - - if (!net_eq(dev_net(dev), &init_net)) - return NOTIFY_DONE; - - while ((eb_ptr->dev != dev)) { - if (++eb_ptr == stop) - return NOTIFY_DONE; /* couldn't find device */ - } - if (!eb_ptr->bearer) - return NOTIFY_DONE; /* bearer had been disabled */ - - eb_ptr->bearer->mtu = dev->mtu; - - switch (evt) { - case NETDEV_CHANGE: - if (netif_carrier_ok(dev)) - tipc_continue(eb_ptr->bearer); - else - tipc_block_bearer(eb_ptr->bearer); - break; - case NETDEV_UP: - tipc_continue(eb_ptr->bearer); - break; - case NETDEV_DOWN: - tipc_block_bearer(eb_ptr->bearer); - break; - case NETDEV_CHANGEMTU: - case NETDEV_CHANGEADDR: - tipc_block_bearer(eb_ptr->bearer); - tipc_continue(eb_ptr->bearer); - break; - case NETDEV_UNREGISTER: - case NETDEV_CHANGENAME: - tipc_disable_bearer(eb_ptr->bearer->name); - break; - } - return NOTIFY_OK; -} - -/** - * eth_addr2str - convert Ethernet address to string - */ -static int eth_addr2str(struct tipc_media_addr *a, char *str_buf, int str_size) -{ - if (str_size < 18) /* 18 = strlen("aa:bb:cc:dd:ee:ff\0") */ - return 1; - - sprintf(str_buf, "%pM", a->value); + memset(addr, 0, sizeof(*addr)); + ether_addr_copy(addr->value, msg); + addr->media_id = TIPC_MEDIA_TYPE_ETH; + addr->broadcast = !memcmp(addr->value, bcast_mac, ETH_ALEN); return 0; } -/** - * eth_str2addr - convert Ethernet address format to message header format - */ -static int eth_addr2msg(struct tipc_media_addr *a, char *msg_area) +/* Convert discovery msg addr format to Ethernet media addr format */ +static int tipc_eth_msg2addr(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg) { - memset(msg_area, 0, TIPC_MEDIA_ADDR_SIZE); - msg_area[TIPC_MEDIA_TYPE_OFFSET] = TIPC_MEDIA_TYPE_ETH; - memcpy(msg_area + ETH_ADDR_OFFSET, a->value, ETH_ALEN); - return 0; + /* Skip past preamble: */ + msg += ETH_ADDR_OFFSET; + return tipc_eth_raw2addr(b, addr, msg); } -/** - * eth_str2addr - convert message header address format to Ethernet format - */ -static int eth_msg2addr(const struct tipc_bearer *tb_ptr, - struct tipc_media_addr *a, char *msg_area) -{ - if (msg_area[TIPC_MEDIA_TYPE_OFFSET] != TIPC_MEDIA_TYPE_ETH) - return 1; - - eth_media_addr_set(tb_ptr, a, msg_area + ETH_ADDR_OFFSET); - return 0; -} - -/* - * Ethernet media registration info - */ -static struct tipc_media eth_media_info = { - .send_msg = send_msg, - .enable_media = enable_media, - .disable_media = disable_media, - .addr2str = eth_addr2str, - .addr2msg = eth_addr2msg, - .msg2addr = eth_msg2addr, +/* Ethernet media registration info */ +struct tipc_media eth_media_info = { + .send_msg = tipc_l2_send_msg, + .enable_media = tipc_enable_l2_media, + .disable_media = tipc_disable_l2_media, + .addr2str = tipc_eth_addr2str, + .addr2msg = tipc_eth_addr2msg, + .msg2addr = tipc_eth_msg2addr, + .raw2addr = tipc_eth_raw2addr, .priority = TIPC_DEF_LINK_PRI, .tolerance = TIPC_DEF_LINK_TOL, .window = TIPC_DEF_LINK_WIN, .type_id = TIPC_MEDIA_TYPE_ETH, + .hwaddr_len = ETH_ALEN, .name = "eth" }; - -/** - * tipc_eth_media_start - activate Ethernet bearer support - * - * Register Ethernet media type with TIPC bearer code. Also register - * with OS for notifications about device state changes. - */ -int tipc_eth_media_start(void) -{ - int res; - - if (eth_started) - return -EINVAL; - - res = tipc_register_media(ð_media_info); - if (res) - return res; - - res = register_netdevice_notifier(¬ifier); - if (!res) - eth_started = 1; - return res; -} - -/** - * tipc_eth_media_stop - deactivate Ethernet bearer support - */ -void tipc_eth_media_stop(void) -{ - if (!eth_started) - return; - - flush_scheduled_work(); - unregister_netdevice_notifier(¬ifier); - eth_started = 0; -} diff --git a/net/tipc/handler.c b/net/tipc/handler.c deleted file mode 100644 index e4bc8a296744..000000000000 --- a/net/tipc/handler.c +++ /dev/null @@ -1,135 +0,0 @@ -/* - * net/tipc/handler.c: TIPC signal handling - * - * Copyright (c) 2000-2006, Ericsson AB - * Copyright (c) 2005, Wind River Systems - * All rights reserved. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions are met: - * - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * 3. Neither the names of the copyright holders nor the names of its - * contributors may be used to endorse or promote products derived from - * this software without specific prior written permission. - * - * Alternatively, this software may be distributed under the terms of the - * GNU General Public License ("GPL") version 2 as published by the Free - * Software Foundation. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" - * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE - * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR - * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF - * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS - * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN - * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) - * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE - * POSSIBILITY OF SUCH DAMAGE. - */ - -#include "core.h" - -struct queue_item { - struct list_head next_signal; - void (*handler) (unsigned long); - unsigned long data; -}; - -static struct kmem_cache *tipc_queue_item_cache; -static struct list_head signal_queue_head; -static DEFINE_SPINLOCK(qitem_lock); -static int handler_enabled __read_mostly; - -static void process_signal_queue(unsigned long dummy); - -static DECLARE_TASKLET_DISABLED(tipc_tasklet, process_signal_queue, 0); - - -unsigned int tipc_k_signal(Handler routine, unsigned long argument) -{ - struct queue_item *item; - - spin_lock_bh(&qitem_lock); - if (!handler_enabled) { - pr_err("Signal request ignored by handler\n"); - spin_unlock_bh(&qitem_lock); - return -ENOPROTOOPT; - } - - item = kmem_cache_alloc(tipc_queue_item_cache, GFP_ATOMIC); - if (!item) { - pr_err("Signal queue out of memory\n"); - spin_unlock_bh(&qitem_lock); - return -ENOMEM; - } - item->handler = routine; - item->data = argument; - list_add_tail(&item->next_signal, &signal_queue_head); - spin_unlock_bh(&qitem_lock); - tasklet_schedule(&tipc_tasklet); - return 0; -} - -static void process_signal_queue(unsigned long dummy) -{ - struct queue_item *__volatile__ item; - struct list_head *l, *n; - - spin_lock_bh(&qitem_lock); - list_for_each_safe(l, n, &signal_queue_head) { - item = list_entry(l, struct queue_item, next_signal); - list_del(&item->next_signal); - spin_unlock_bh(&qitem_lock); - item->handler(item->data); - spin_lock_bh(&qitem_lock); - kmem_cache_free(tipc_queue_item_cache, item); - } - spin_unlock_bh(&qitem_lock); -} - -int tipc_handler_start(void) -{ - tipc_queue_item_cache = - kmem_cache_create("tipc_queue_items", sizeof(struct queue_item), - 0, SLAB_HWCACHE_ALIGN, NULL); - if (!tipc_queue_item_cache) - return -ENOMEM; - - INIT_LIST_HEAD(&signal_queue_head); - tasklet_enable(&tipc_tasklet); - handler_enabled = 1; - return 0; -} - -void tipc_handler_stop(void) -{ - struct list_head *l, *n; - struct queue_item *item; - - spin_lock_bh(&qitem_lock); - if (!handler_enabled) { - spin_unlock_bh(&qitem_lock); - return; - } - handler_enabled = 0; - spin_unlock_bh(&qitem_lock); - - tasklet_kill(&tipc_tasklet); - - spin_lock_bh(&qitem_lock); - list_for_each_safe(l, n, &signal_queue_head) { - item = list_entry(l, struct queue_item, next_signal); - list_del(&item->next_signal); - kmem_cache_free(tipc_queue_item_cache, item); - } - spin_unlock_bh(&qitem_lock); - - kmem_cache_destroy(tipc_queue_item_cache); -} diff --git a/net/tipc/ib_media.c b/net/tipc/ib_media.c index c13989297464..8522eef9c136 100644 --- a/net/tipc/ib_media.c +++ b/net/tipc/ib_media.c @@ -42,252 +42,9 @@ #include "core.h" #include "bearer.h" -#define MAX_IB_MEDIA MAX_BEARERS - -/** - * struct ib_media - Infiniband media data structure - * @bearer: ptr to associated "generic" bearer structure - * @dev: ptr to associated Infiniband network device - * @tipc_packet_type: used in binding TIPC to Infiniband driver - * @cleanup: work item used when disabling bearer - */ - -struct ib_media { - struct tipc_bearer *bearer; - struct net_device *dev; - struct packet_type tipc_packet_type; - struct work_struct setup; - struct work_struct cleanup; -}; - -static struct tipc_media ib_media_info; -static struct ib_media ib_media_array[MAX_IB_MEDIA]; -static int ib_started; - -/** - * ib_media_addr_set - initialize Infiniband media address structure - * - * Media-dependent "value" field stores MAC address in first 6 bytes - * and zeroes out the remaining bytes. - */ -static void ib_media_addr_set(const struct tipc_bearer *tb_ptr, - struct tipc_media_addr *a, char *mac) -{ - BUILD_BUG_ON(sizeof(a->value) < INFINIBAND_ALEN); - memcpy(a->value, mac, INFINIBAND_ALEN); - a->media_id = TIPC_MEDIA_TYPE_IB; - a->broadcast = !memcmp(mac, tb_ptr->bcast_addr.value, INFINIBAND_ALEN); -} - -/** - * send_msg - send a TIPC message out over an InfiniBand interface - */ -static int send_msg(struct sk_buff *buf, struct tipc_bearer *tb_ptr, - struct tipc_media_addr *dest) -{ - struct sk_buff *clone; - struct net_device *dev; - int delta; - - clone = skb_clone(buf, GFP_ATOMIC); - if (!clone) - return 0; - - dev = ((struct ib_media *)(tb_ptr->usr_handle))->dev; - delta = dev->hard_header_len - skb_headroom(buf); - - if ((delta > 0) && - pskb_expand_head(clone, SKB_DATA_ALIGN(delta), 0, GFP_ATOMIC)) { - kfree_skb(clone); - return 0; - } - - skb_reset_network_header(clone); - clone->dev = dev; - clone->protocol = htons(ETH_P_TIPC); - dev_hard_header(clone, dev, ETH_P_TIPC, dest->value, - dev->dev_addr, clone->len); - dev_queue_xmit(clone); - return 0; -} - -/** - * recv_msg - handle incoming TIPC message from an InfiniBand interface - * - * Accept only packets explicitly sent to this node, or broadcast packets; - * ignores packets sent using InfiniBand multicast, and traffic sent to other - * nodes (which can happen if interface is running in promiscuous mode). - */ -static int recv_msg(struct sk_buff *buf, struct net_device *dev, - struct packet_type *pt, struct net_device *orig_dev) -{ - struct ib_media *ib_ptr = (struct ib_media *)pt->af_packet_priv; - - if (!net_eq(dev_net(dev), &init_net)) { - kfree_skb(buf); - return NET_RX_DROP; - } - - if (likely(ib_ptr->bearer)) { - if (likely(buf->pkt_type <= PACKET_BROADCAST)) { - buf->next = NULL; - tipc_recv_msg(buf, ib_ptr->bearer); - return NET_RX_SUCCESS; - } - } - kfree_skb(buf); - return NET_RX_DROP; -} - -/** - * setup_bearer - setup association between InfiniBand bearer and interface - */ -static void setup_media(struct work_struct *work) -{ - struct ib_media *ib_ptr = - container_of(work, struct ib_media, setup); - - dev_add_pack(&ib_ptr->tipc_packet_type); -} - -/** - * enable_media - attach TIPC bearer to an InfiniBand interface - */ -static int enable_media(struct tipc_bearer *tb_ptr) -{ - struct net_device *dev; - struct ib_media *ib_ptr = &ib_media_array[0]; - struct ib_media *stop = &ib_media_array[MAX_IB_MEDIA]; - char *driver_name = strchr((const char *)tb_ptr->name, ':') + 1; - int pending_dev = 0; - - /* Find unused InfiniBand bearer structure */ - while (ib_ptr->dev) { - if (!ib_ptr->bearer) - pending_dev++; - if (++ib_ptr == stop) - return pending_dev ? -EAGAIN : -EDQUOT; - } - - /* Find device with specified name */ - dev = dev_get_by_name(&init_net, driver_name); - if (!dev) - return -ENODEV; - - /* Create InfiniBand bearer for device */ - ib_ptr->dev = dev; - ib_ptr->tipc_packet_type.type = htons(ETH_P_TIPC); - ib_ptr->tipc_packet_type.dev = dev; - ib_ptr->tipc_packet_type.func = recv_msg; - ib_ptr->tipc_packet_type.af_packet_priv = ib_ptr; - INIT_LIST_HEAD(&(ib_ptr->tipc_packet_type.list)); - INIT_WORK(&ib_ptr->setup, setup_media); - schedule_work(&ib_ptr->setup); - - /* Associate TIPC bearer with InfiniBand bearer */ - ib_ptr->bearer = tb_ptr; - tb_ptr->usr_handle = (void *)ib_ptr; - memset(tb_ptr->bcast_addr.value, 0, sizeof(tb_ptr->bcast_addr.value)); - memcpy(tb_ptr->bcast_addr.value, dev->broadcast, INFINIBAND_ALEN); - tb_ptr->bcast_addr.media_id = TIPC_MEDIA_TYPE_IB; - tb_ptr->bcast_addr.broadcast = 1; - tb_ptr->mtu = dev->mtu; - tb_ptr->blocked = 0; - ib_media_addr_set(tb_ptr, &tb_ptr->addr, (char *)dev->dev_addr); - return 0; -} - -/** - * cleanup_bearer - break association between InfiniBand bearer and interface - * - * This routine must be invoked from a work queue because it can sleep. - */ -static void cleanup_bearer(struct work_struct *work) -{ - struct ib_media *ib_ptr = - container_of(work, struct ib_media, cleanup); - - dev_remove_pack(&ib_ptr->tipc_packet_type); - dev_put(ib_ptr->dev); - ib_ptr->dev = NULL; -} - -/** - * disable_media - detach TIPC bearer from an InfiniBand interface - * - * Mark InfiniBand bearer as inactive so that incoming buffers are thrown away, - * then get worker thread to complete bearer cleanup. (Can't do cleanup - * here because cleanup code needs to sleep and caller holds spinlocks.) - */ -static void disable_media(struct tipc_bearer *tb_ptr) -{ - struct ib_media *ib_ptr = (struct ib_media *)tb_ptr->usr_handle; - - ib_ptr->bearer = NULL; - INIT_WORK(&ib_ptr->cleanup, cleanup_bearer); - schedule_work(&ib_ptr->cleanup); -} - -/** - * recv_notification - handle device updates from OS - * - * Change the state of the InfiniBand bearer (if any) associated with the - * specified device. - */ -static int recv_notification(struct notifier_block *nb, unsigned long evt, - void *ptr) -{ - struct net_device *dev = netdev_notifier_info_to_dev(ptr); - struct ib_media *ib_ptr = &ib_media_array[0]; - struct ib_media *stop = &ib_media_array[MAX_IB_MEDIA]; - - if (!net_eq(dev_net(dev), &init_net)) - return NOTIFY_DONE; - - while ((ib_ptr->dev != dev)) { - if (++ib_ptr == stop) - return NOTIFY_DONE; /* couldn't find device */ - } - if (!ib_ptr->bearer) - return NOTIFY_DONE; /* bearer had been disabled */ - - ib_ptr->bearer->mtu = dev->mtu; - - switch (evt) { - case NETDEV_CHANGE: - if (netif_carrier_ok(dev)) - tipc_continue(ib_ptr->bearer); - else - tipc_block_bearer(ib_ptr->bearer); - break; - case NETDEV_UP: - tipc_continue(ib_ptr->bearer); - break; - case NETDEV_DOWN: - tipc_block_bearer(ib_ptr->bearer); - break; - case NETDEV_CHANGEMTU: - case NETDEV_CHANGEADDR: - tipc_block_bearer(ib_ptr->bearer); - tipc_continue(ib_ptr->bearer); - break; - case NETDEV_UNREGISTER: - case NETDEV_CHANGENAME: - tipc_disable_bearer(ib_ptr->bearer->name); - break; - } - return NOTIFY_OK; -} - -static struct notifier_block notifier = { - .notifier_call = recv_notification, - .priority = 0, -}; - -/** - * ib_addr2str - convert InfiniBand address to string - */ -static int ib_addr2str(struct tipc_media_addr *a, char *str_buf, int str_size) +/* convert InfiniBand address (media address format) media address to string */ +static int tipc_ib_addr2str(struct tipc_media_addr *a, char *str_buf, + int str_size) { if (str_size < 60) /* 60 = 19 * strlen("xx:") + strlen("xx\0") */ return 1; @@ -297,76 +54,48 @@ static int ib_addr2str(struct tipc_media_addr *a, char *str_buf, int str_size) return 0; } -/** - * ib_addr2msg - convert InfiniBand address format to message header format - */ -static int ib_addr2msg(struct tipc_media_addr *a, char *msg_area) +/* Convert from media address format to discovery message addr format */ +static int tipc_ib_addr2msg(char *msg, struct tipc_media_addr *addr) { - memset(msg_area, 0, TIPC_MEDIA_ADDR_SIZE); - msg_area[TIPC_MEDIA_TYPE_OFFSET] = TIPC_MEDIA_TYPE_IB; - memcpy(msg_area, a->value, INFINIBAND_ALEN); + memset(msg, 0, TIPC_MEDIA_ADDR_SIZE); + memcpy(msg, addr->value, INFINIBAND_ALEN); return 0; } -/** - * ib_msg2addr - convert message header address format to InfiniBand format - */ -static int ib_msg2addr(const struct tipc_bearer *tb_ptr, - struct tipc_media_addr *a, char *msg_area) +/* Convert raw InfiniBand address format to media addr format */ +static int tipc_ib_raw2addr(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg) { - ib_media_addr_set(tb_ptr, a, msg_area); + memset(addr, 0, sizeof(*addr)); + memcpy(addr->value, msg, INFINIBAND_ALEN); + addr->media_id = TIPC_MEDIA_TYPE_IB; + addr->broadcast = !memcmp(msg, b->bcast_addr.value, + INFINIBAND_ALEN); return 0; } -/* - * InfiniBand media registration info - */ -static struct tipc_media ib_media_info = { - .send_msg = send_msg, - .enable_media = enable_media, - .disable_media = disable_media, - .addr2str = ib_addr2str, - .addr2msg = ib_addr2msg, - .msg2addr = ib_msg2addr, +/* Convert discovery msg addr format to InfiniBand media addr format */ +static int tipc_ib_msg2addr(struct tipc_bearer *b, + struct tipc_media_addr *addr, + char *msg) +{ + return tipc_ib_raw2addr(b, addr, msg); +} + +/* InfiniBand media registration info */ +struct tipc_media ib_media_info = { + .send_msg = tipc_l2_send_msg, + .enable_media = tipc_enable_l2_media, + .disable_media = tipc_disable_l2_media, + .addr2str = tipc_ib_addr2str, + .addr2msg = tipc_ib_addr2msg, + .msg2addr = tipc_ib_msg2addr, + .raw2addr = tipc_ib_raw2addr, .priority = TIPC_DEF_LINK_PRI, .tolerance = TIPC_DEF_LINK_TOL, .window = TIPC_DEF_LINK_WIN, .type_id = TIPC_MEDIA_TYPE_IB, + .hwaddr_len = INFINIBAND_ALEN, .name = "ib" }; - -/** - * tipc_ib_media_start - activate InfiniBand bearer support - * - * Register InfiniBand media type with TIPC bearer code. Also register - * with OS for notifications about device state changes. - */ -int tipc_ib_media_start(void) -{ - int res; - - if (ib_started) - return -EINVAL; - - res = tipc_register_media(&ib_media_info); - if (res) - return res; - - res = register_netdevice_notifier(¬ifier); - if (!res) - ib_started = 1; - return res; -} - -/** - * tipc_ib_media_stop - deactivate InfiniBand bearer support - */ -void tipc_ib_media_stop(void) -{ - if (!ib_started) - return; - - flush_scheduled_work(); - unregister_netdevice_notifier(¬ifier); - ib_started = 0; -} diff --git a/net/tipc/link.c b/net/tipc/link.c index 69cd9bf3f561..65410e18b8a6 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, Ericsson AB + * Copyright (c) 1996-2007, 2012-2014, Ericsson AB * Copyright (c) 2004-2007, 2010-2013, Wind River Systems * All rights reserved. * @@ -36,7 +36,7 @@ #include "core.h" #include "link.h" -#include "port.h" +#include "socket.h" #include "name_distr.h" #include "discover.h" #include "config.h" @@ -77,20 +77,17 @@ static const char *link_unk_evt = "Unknown link event "; static void link_handle_out_of_seq_msg(struct tipc_link *l_ptr, struct sk_buff *buf); -static void link_recv_proto_msg(struct tipc_link *l_ptr, struct sk_buff *buf); -static int link_recv_changeover_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 int link_send_sections_long(struct tipc_port *sender, - struct iovec const *msg_sect, - unsigned int len, u32 destnode); 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 link_start(struct tipc_link *l_ptr); -static int link_send_long_buf(struct tipc_link *l_ptr, struct sk_buff *buf); -static void tipc_link_send_sync(struct tipc_link *l); -static void tipc_link_recv_sync(struct tipc_node *n, struct sk_buff *buf); +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); /* * Simple link routines @@ -102,9 +99,18 @@ static unsigned int align(unsigned int i) static void link_init_max_pkt(struct tipc_link *l_ptr) { + struct tipc_bearer *b_ptr; u32 max_pkt; - max_pkt = (l_ptr->b_ptr->mtu & ~3); + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[l_ptr->bearer_id]); + if (!b_ptr) { + rcu_read_unlock(); + return; + } + max_pkt = (b_ptr->mtu & ~3); + rcu_read_unlock(); + if (max_pkt > MAX_MSG_SIZE) max_pkt = MAX_MSG_SIZE; @@ -148,11 +154,6 @@ int tipc_link_is_active(struct tipc_link *l_ptr) /** * link_timeout - handle expiration of link timer * @l_ptr: pointer to link - * - * This routine must not grab "tipc_net_lock" to avoid a potential deadlock conflict - * with tipc_link_delete(). (There is no risk that the node will be deleted by - * another thread because tipc_link_delete() always cancels the link timer before - * tipc_node_delete() is called.) */ static void link_timeout(struct tipc_link *l_ptr) { @@ -214,8 +215,8 @@ static void link_set_timer(struct tipc_link *l_ptr, u32 time) * Returns pointer to link. */ 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_bearer *b_ptr, + const struct tipc_media_addr *media_addr) { struct tipc_link *l_ptr; struct tipc_msg *msg; @@ -254,7 +255,7 @@ struct tipc_link *tipc_link_create(struct tipc_node *n_ptr, l_ptr->owner = n_ptr; l_ptr->checkpoint = 1; l_ptr->peer_session = INVALID_SESSION; - l_ptr->b_ptr = b_ptr; + l_ptr->bearer_id = b_ptr->identity; link_set_supervision_props(l_ptr, b_ptr->tolerance); l_ptr->state = RESET_UNKNOWN; @@ -269,115 +270,98 @@ struct tipc_link *tipc_link_create(struct tipc_node *n_ptr, l_ptr->priority = b_ptr->priority; tipc_link_set_queue_limits(l_ptr, b_ptr->window); + l_ptr->net_plane = b_ptr->net_plane; link_init_max_pkt(l_ptr); l_ptr->next_out_no = 1; - INIT_LIST_HEAD(&l_ptr->waiting_ports); + __skb_queue_head_init(&l_ptr->waiting_sks); 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); - list_add_tail(&l_ptr->link_list, &b_ptr->links); - tipc_k_signal((Handler)link_start, (unsigned long)l_ptr); + k_init_timer(&l_ptr->timer, (Handler)link_timeout, + (unsigned long)l_ptr); + + link_state_event(l_ptr, STARTING_EVT); return l_ptr; } -/** - * tipc_link_delete - delete a link - * @l_ptr: pointer to link - * - * Note: 'tipc_net_lock' is write_locked, bearer is locked. - * This routine must not grab the node lock until after link timer cancellation - * to avoid a potential deadlock situation. - */ -void tipc_link_delete(struct tipc_link *l_ptr) +void tipc_link_delete_list(unsigned int bearer_id, bool shutting_down) { - if (!l_ptr) { - pr_err("Attempt to delete non-existent link\n"); - return; - } - - k_cancel_timer(&l_ptr->timer); + struct tipc_link *l_ptr; + struct tipc_node *n_ptr; - tipc_node_lock(l_ptr->owner); - tipc_link_reset(l_ptr); - tipc_node_detach_link(l_ptr->owner, l_ptr); - tipc_link_stop(l_ptr); - list_del_init(&l_ptr->link_list); - tipc_node_unlock(l_ptr->owner); - k_term_timer(&l_ptr->timer); - kfree(l_ptr); -} + 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); -static void link_start(struct tipc_link *l_ptr) -{ - tipc_node_lock(l_ptr->owner); - link_state_event(l_ptr, STARTING_EVT); - tipc_node_unlock(l_ptr->owner); + /* 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); + } + rcu_read_unlock(); } /** - * link_schedule_port - schedule port for deferred sending - * @l_ptr: pointer to link - * @origport: reference to sending port - * @sz: amount of data to be sent - * - * Schedules port for renewed sending of messages after link congestion - * has abated. + * link_schedule_user - schedule user 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 + * Create pseudo msg to send back to user when congestion abates */ -static int link_schedule_port(struct tipc_link *l_ptr, u32 origport, u32 sz) +static bool link_schedule_user(struct tipc_link *link, u32 oport, + uint chain_sz, uint imp) { - struct tipc_port *p_ptr; + struct sk_buff *buf; - spin_lock_bh(&tipc_port_list_lock); - p_ptr = tipc_port_lock(origport); - if (p_ptr) { - if (!p_ptr->wakeup) - goto exit; - if (!list_empty(&p_ptr->wait_list)) - goto exit; - p_ptr->congested = 1; - p_ptr->waiting_pkts = 1 + ((sz - 1) / l_ptr->max_pkt); - list_add_tail(&p_ptr->wait_list, &l_ptr->waiting_ports); - l_ptr->stats.link_congs++; -exit: - tipc_port_unlock(p_ptr); - } - spin_unlock_bh(&tipc_port_list_lock); - return -ELINKCONG; + 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); + link->stats.link_congs++; + return true; } -void tipc_link_wakeup_ports(struct tipc_link *l_ptr, int all) +/** + * link_prepare_wakeup - prepare users for wakeup after congestion + * @link: congested link + * 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) { - struct tipc_port *p_ptr; - struct tipc_port *temp_p_ptr; - int win = l_ptr->queue_limit[0] - l_ptr->out_queue_size; + struct sk_buff_head *wq = &link->waiting_sks; + struct sk_buff *buf; + uint pend_qsz = link->out_queue_size; - if (all) - win = 100000; - if (win <= 0) - return; - if (!spin_trylock_bh(&tipc_port_list_lock)) - return; - if (link_congested(l_ptr)) - goto exit; - list_for_each_entry_safe(p_ptr, temp_p_ptr, &l_ptr->waiting_ports, - wait_list) { - if (win <= 0) + for (buf = skb_peek(wq); buf; buf = skb_peek(wq)) { + if (pend_qsz >= link->queue_limit[TIPC_SKB_CB(buf)->chain_imp]) break; - list_del_init(&p_ptr->wait_list); - spin_lock_bh(p_ptr->lock); - p_ptr->congested = 0; - p_ptr->wakeup(p_ptr); - win -= p_ptr->waiting_pkts; - spin_unlock_bh(p_ptr->lock); + pend_qsz += TIPC_SKB_CB(buf)->chain_sz; + __skb_queue_tail(&link->owner->waiting_sks, __skb_dequeue(wq)); } - -exit: - spin_unlock_bh(&tipc_port_list_lock); } /** @@ -386,14 +370,7 @@ exit: */ static void link_release_outqueue(struct tipc_link *l_ptr) { - struct sk_buff *buf = l_ptr->first_out; - struct sk_buff *next; - - while (buf) { - next = buf->next; - kfree_skb(buf); - buf = next; - } + kfree_skb_list(l_ptr->first_out); l_ptr->first_out = NULL; l_ptr->out_queue_size = 0; } @@ -404,46 +381,29 @@ static void link_release_outqueue(struct tipc_link *l_ptr) */ void tipc_link_reset_fragments(struct tipc_link *l_ptr) { - kfree_skb(l_ptr->reasm_head); - l_ptr->reasm_head = NULL; - l_ptr->reasm_tail = NULL; + kfree_skb(l_ptr->reasm_buf); + l_ptr->reasm_buf = NULL; } /** - * tipc_link_stop - purge all inbound and outbound messages associated with link + * tipc_link_purge_queues - purge all pkt queues associated with link * @l_ptr: pointer to link */ -void tipc_link_stop(struct tipc_link *l_ptr) +void tipc_link_purge_queues(struct tipc_link *l_ptr) { - struct sk_buff *buf; - struct sk_buff *next; - - buf = l_ptr->oldest_deferred_in; - while (buf) { - next = buf->next; - kfree_skb(buf); - buf = next; - } - - buf = l_ptr->first_out; - while (buf) { - next = buf->next; - kfree_skb(buf); - buf = next; - } - + kfree_skb_list(l_ptr->oldest_deferred_in); + kfree_skb_list(l_ptr->first_out); tipc_link_reset_fragments(l_ptr); - kfree_skb(l_ptr->proto_msg_queue); l_ptr->proto_msg_queue = NULL; } void tipc_link_reset(struct tipc_link *l_ptr) { - struct sk_buff *buf; 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; msg_set_session(l_ptr->pmsg, ((msg_session(l_ptr->pmsg) + 1) & 0xffff)); @@ -459,10 +419,9 @@ void tipc_link_reset(struct tipc_link *l_ptr) return; tipc_node_link_down(l_ptr->owner, l_ptr); - tipc_bearer_remove_dest(l_ptr->b_ptr, l_ptr->addr); + tipc_bearer_remove_dest(l_ptr->bearer_id, l_ptr->addr); - if (was_active_link && tipc_node_active_links(l_ptr->owner) && - l_ptr->owner->permit_changeover) { + if (was_active_link && tipc_node_active_links(l_ptr->owner)) { l_ptr->reset_checkpoint = checkpoint; l_ptr->exp_msg_count = START_CHANGEOVER; } @@ -471,15 +430,11 @@ void tipc_link_reset(struct tipc_link *l_ptr) link_release_outqueue(l_ptr); kfree_skb(l_ptr->proto_msg_queue); l_ptr->proto_msg_queue = NULL; - buf = l_ptr->oldest_deferred_in; - while (buf) { - struct sk_buff *next = buf->next; - kfree_skb(buf); - buf = next; + kfree_skb_list(l_ptr->oldest_deferred_in); + 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; } - if (!list_empty(&l_ptr->waiting_ports)) - tipc_link_wakeup_ports(l_ptr, 1); - l_ptr->retransm_queue_head = 0; l_ptr->retransm_queue_size = 0; l_ptr->last_out = NULL; @@ -496,12 +451,27 @@ void tipc_link_reset(struct tipc_link *l_ptr) link_reset_statistics(l_ptr); } +void tipc_link_reset_list(unsigned int bearer_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) { + tipc_node_lock(n_ptr); + l_ptr = n_ptr->links[bearer_id]; + if (l_ptr) + tipc_link_reset(l_ptr); + tipc_node_unlock(n_ptr); + } + rcu_read_unlock(); +} static void link_activate(struct tipc_link *l_ptr) { l_ptr->next_in_no = l_ptr->stats.recv_info = 1; tipc_node_link_up(l_ptr->owner, l_ptr); - tipc_bearer_add_dest(l_ptr->b_ptr, l_ptr->addr); + tipc_bearer_add_dest(l_ptr->bearer_id, l_ptr->addr); } /** @@ -514,13 +484,17 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) struct tipc_link *other; u32 cont_intv = l_ptr->continuity_interval; - if (!l_ptr->started && (event != STARTING_EVT)) + if (l_ptr->flags & LINK_STOPPED) + return; + + if (!(l_ptr->flags & LINK_STARTED) && (event != STARTING_EVT)) return; /* Not yet. */ - if (link_blocked(l_ptr)) { + /* Check whether changeover is going on */ + if (l_ptr->exp_msg_count) { if (event == TIMEOUT_EVT) link_set_timer(l_ptr, cont_intv); - return; /* Changeover going on */ + return; } switch (l_ptr->state) { @@ -533,12 +507,12 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) if (l_ptr->next_in_no != l_ptr->checkpoint) { l_ptr->checkpoint = l_ptr->next_in_no; if (tipc_bclink_acks_missing(l_ptr->owner)) { - tipc_link_send_proto_msg(l_ptr, STATE_MSG, - 0, 0, 0, 0, 0); + 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_send_proto_msg(l_ptr, STATE_MSG, - 1, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, STATE_MSG, + 1, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; } link_set_timer(l_ptr, cont_intv); @@ -546,7 +520,7 @@ 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_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv / 4); break; @@ -556,7 +530,8 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) tipc_link_reset(l_ptr); l_ptr->state = RESET_RESET; l_ptr->fsm_msg_cnt = 0; - tipc_link_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, ACTIVATE_MSG, + 0, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; @@ -578,7 +553,8 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) tipc_link_reset(l_ptr); l_ptr->state = RESET_RESET; l_ptr->fsm_msg_cnt = 0; - tipc_link_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, ACTIVATE_MSG, + 0, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; @@ -588,14 +564,14 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) l_ptr->fsm_msg_cnt = 0; l_ptr->checkpoint = l_ptr->next_in_no; if (tipc_bclink_acks_missing(l_ptr->owner)) { - tipc_link_send_proto_msg(l_ptr, STATE_MSG, - 0, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, STATE_MSG, + 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_send_proto_msg(l_ptr, STATE_MSG, - 1, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, STATE_MSG, + 1, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv / 4); } else { /* Link has failed */ @@ -604,8 +580,8 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) tipc_link_reset(l_ptr); l_ptr->state = RESET_UNKNOWN; l_ptr->fsm_msg_cnt = 0; - tipc_link_send_proto_msg(l_ptr, RESET_MSG, - 0, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, RESET_MSG, + 0, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); } @@ -625,24 +601,25 @@ 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_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; if (l_ptr->owner->working_links == 1) - tipc_link_send_sync(l_ptr); + tipc_link_sync_xmit(l_ptr); link_set_timer(l_ptr, cont_intv); break; case RESET_MSG: l_ptr->state = RESET_RESET; l_ptr->fsm_msg_cnt = 0; - tipc_link_send_proto_msg(l_ptr, ACTIVATE_MSG, 1, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, ACTIVATE_MSG, + 1, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; case STARTING_EVT: - l_ptr->started = 1; + l_ptr->flags |= LINK_STARTED; /* fall through */ case TIMEOUT_EVT: - tipc_link_send_proto_msg(l_ptr, RESET_MSG, 0, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, RESET_MSG, 0, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; @@ -660,16 +637,17 @@ 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_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; if (l_ptr->owner->working_links == 1) - tipc_link_send_sync(l_ptr); + tipc_link_sync_xmit(l_ptr); link_set_timer(l_ptr, cont_intv); break; case RESET_MSG: break; case TIMEOUT_EVT: - tipc_link_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, ACTIVATE_MSG, + 0, 0, 0, 0, 0); l_ptr->fsm_msg_cnt++; link_set_timer(l_ptr, cont_intv); break; @@ -682,192 +660,157 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event) } } -/* - * link_bundle_buf(): Append contents of a buffer to - * the tail of an existing one. +/* 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 link_bundle_buf(struct tipc_link *l_ptr, struct sk_buff *bundler, - struct sk_buff *buf) +static int tipc_link_cong(struct tipc_link *link, struct sk_buff *buf) { - struct tipc_msg *bundler_msg = buf_msg(bundler); struct tipc_msg *msg = buf_msg(buf); - u32 size = msg_size(msg); - u32 bundle_size = msg_size(bundler_msg); - u32 to_pos = align(bundle_size); - u32 pad = to_pos - bundle_size; - - if (msg_user(bundler_msg) != MSG_BUNDLER) - return 0; - if (msg_type(bundler_msg) != OPEN_MSG) - return 0; - if (skb_tailroom(bundler) < (pad + size)) - return 0; - if (l_ptr->max_pkt < (to_pos + size)) - return 0; - - skb_put(bundler, pad + size); - skb_copy_to_linear_data_offset(bundler, to_pos, buf->data, size); - msg_set_size(bundler_msg, to_pos + size); - msg_set_msgcnt(bundler_msg, msg_msgcnt(bundler_msg) + 1); - kfree_skb(buf); - l_ptr->stats.sent_bundled++; - return 1; -} - -static void link_add_to_outqueue(struct tipc_link *l_ptr, - struct sk_buff *buf, - struct tipc_msg *msg) -{ - u32 ack = mod(l_ptr->next_in_no - 1); - u32 seqno = mod(l_ptr->next_out_no++); - - msg_set_word(msg, 2, ((ack << 16) | seqno)); - msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); - buf->next = NULL; - if (l_ptr->first_out) { - l_ptr->last_out->next = buf; - l_ptr->last_out = buf; - } else - l_ptr->first_out = l_ptr->last_out = buf; - - l_ptr->out_queue_size++; - if (l_ptr->out_queue_size > l_ptr->stats.max_queue_sz) - l_ptr->stats.max_queue_sz = l_ptr->out_queue_size; -} - -static void link_add_chain_to_outqueue(struct tipc_link *l_ptr, - struct sk_buff *buf_chain, - u32 long_msgno) -{ - struct sk_buff *buf; - struct tipc_msg *msg; - - if (!l_ptr->next_out) - l_ptr->next_out = buf_chain; - while (buf_chain) { - buf = buf_chain; - buf_chain = buf_chain->next; - - msg = buf_msg(buf); - msg_set_long_msgno(msg, long_msgno); - link_add_to_outqueue(l_ptr, buf, msg); - } + 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(buf)->wakeup_pending) + return -ELINKCONG; + if (link_schedule_user(link, oport, TIPC_SKB_CB(buf)->chain_sz, imp)) + return -ELINKCONG; +drop: + kfree_skb_list(buf); + return -EHOSTUNREACH; } -/* - * tipc_link_send_buf() is the 'full path' for messages, called from - * inside TIPC when the 'fast path' in tipc_send_buf - * has failed, and from link_send() +/** + * __tipc_link_xmit(): same as tipc_link_xmit, but destlink is known & locked + * @link: link to use + * @buf: 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. */ -int tipc_link_send_buf(struct tipc_link *l_ptr, struct sk_buff *buf) +int __tipc_link_xmit(struct tipc_link *link, struct sk_buff *buf) { struct tipc_msg *msg = buf_msg(buf); - u32 size = msg_size(msg); - u32 dsz = msg_data_sz(msg); - u32 queue_size = l_ptr->out_queue_size; - u32 imp = tipc_msg_tot_importance(msg); - u32 queue_limit = l_ptr->queue_limit[imp]; - u32 max_packet = l_ptr->max_pkt; - - /* Match msg importance against queue limits: */ - if (unlikely(queue_size >= queue_limit)) { - if (imp <= TIPC_CRITICAL_IMPORTANCE) { - link_schedule_port(l_ptr, msg_origport(msg), size); - kfree_skb(buf); - return -ELINKCONG; - } - kfree_skb(buf); - if (imp > CONN_MANAGER) { - pr_warn("%s<%s>, send queue full", link_rst_msg, - l_ptr->name); - tipc_link_reset(l_ptr); + uint psz = msg_size(msg); + uint qsz = link->out_queue_size; + uint sndlim = link->queue_limit[0]; + uint imp = tipc_msg_tot_importance(msg); + uint mtu = link->max_pkt; + 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 *next = buf->next; + + /* Match queue limits against msg importance: */ + if (unlikely(qsz >= link->queue_limit[imp])) + return tipc_link_cong(link, buf); + + /* Has valid packet limit been used ? */ + if (unlikely(psz > mtu)) { + kfree_skb_list(buf); + return -EMSGSIZE; + } + + /* Prepare each packet for sending, and add to outqueue: */ + while (buf) { + next = buf->next; + msg = buf_msg(buf); + msg_set_word(msg, 2, ((ack << 16) | mod(seqno))); + msg_set_bcast_ack(msg, bc_last_in); + + if (!link->first_out) { + link->first_out = buf; + } else if (qsz < sndlim) { + link->last_out->next = buf; + } else if (tipc_msg_bundle(link->last_out, buf, mtu)) { + link->stats.sent_bundled++; + buf = next; + next = buf->next; + continue; + } else if (tipc_msg_make_bundle(&buf, mtu, link->addr)) { + link->stats.sent_bundled++; + link->stats.sent_bundles++; + link->last_out->next = buf; + if (!link->next_out) + link->next_out = buf; + } else { + link->last_out->next = buf; + if (!link->next_out) + link->next_out = buf; } - return dsz; - } - - /* Fragmentation needed ? */ - if (size > max_packet) - return link_send_long_buf(l_ptr, buf); - /* Packet can be queued or sent. */ - if (likely(!tipc_bearer_blocked(l_ptr->b_ptr) && - !link_congested(l_ptr))) { - link_add_to_outqueue(l_ptr, buf, msg); - - tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr); - l_ptr->unacked_window = 0; - return dsz; - } - /* Congestion: can message be bundled ? */ - if ((msg_user(msg) != CHANGEOVER_PROTOCOL) && - (msg_user(msg) != MSG_FRAGMENTER)) { - - /* Try adding message to an existing bundle */ - if (l_ptr->next_out && - link_bundle_buf(l_ptr, l_ptr->last_out, buf)) - return dsz; - - /* Try creating a new bundle */ - if (size <= max_packet * 2 / 3) { - struct sk_buff *bundler = tipc_buf_acquire(max_packet); - struct tipc_msg bundler_hdr; - - if (bundler) { - tipc_msg_init(&bundler_hdr, MSG_BUNDLER, OPEN_MSG, - INT_H_SIZE, l_ptr->addr); - skb_copy_to_linear_data(bundler, &bundler_hdr, - INT_H_SIZE); - skb_trim(bundler, INT_H_SIZE); - link_bundle_buf(l_ptr, bundler, buf); - buf = bundler; - msg = buf_msg(buf); - l_ptr->stats.sent_bundles++; - } + /* Send packet if possible: */ + if (likely(++qsz <= sndlim)) { + tipc_bearer_send(link->bearer_id, buf, addr); + link->next_out = next; + link->unacked_window = 0; } + seqno++; + link->last_out = buf; + buf = next; } - if (!l_ptr->next_out) - l_ptr->next_out = buf; - link_add_to_outqueue(l_ptr, buf, msg); - return dsz; + link->next_out_no = seqno; + link->out_queue_size = qsz; + return 0; } -/* - * tipc_link_send(): same as tipc_link_send_buf(), but the link to use has - * not been selected yet, and the the owner node is not locked - * Called by TIPC internal users, e.g. the name distributor +/** + * tipc_link_xmit() is the general link level function for message sending + * @buf: chain of buffers containing message + * @dsz: amount of user data to be sent + * @dnode: address of destination node + * @selector: a number used for deterministic link selection + * Consumes the buffer chain, except when returning -ELINKCONG + * Returns 0 if success, otherwise errno: -ELINKCONG,-EHOSTUNREACH,-EMSGSIZE */ -int tipc_link_send(struct sk_buff *buf, u32 dest, u32 selector) +int tipc_link_xmit(struct sk_buff *buf, u32 dnode, u32 selector) { - struct tipc_link *l_ptr; - struct tipc_node *n_ptr; - int res = -ELINKCONG; + struct tipc_link *link = NULL; + struct tipc_node *node; + int rc = -EHOSTUNREACH; - read_lock_bh(&tipc_net_lock); - 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 = tipc_link_send_buf(l_ptr, buf); - else - kfree_skb(buf); - tipc_node_unlock(n_ptr); - } else { - kfree_skb(buf); + node = tipc_node_find(dnode); + if (node) { + tipc_node_lock(node); + link = node->active_links[selector & 1]; + if (link) + rc = __tipc_link_xmit(link, buf); + tipc_node_unlock(node); } - read_unlock_bh(&tipc_net_lock); - return res; + + if (link) + return rc; + + if (likely(in_own_node(dnode))) + return tipc_sk_rcv(buf); + + kfree_skb_list(buf); + return rc; } /* - * tipc_link_send_sync - synchronize broadcast link endpoints. + * tipc_link_sync_xmit - synchronize broadcast link endpoints. * * Give a newly added peer node the sequence number where it should * start receiving and acking broadcast packets. * * Called with node locked */ -static void tipc_link_send_sync(struct tipc_link *l) +static void tipc_link_sync_xmit(struct tipc_link *link) { struct sk_buff *buf; struct tipc_msg *msg; @@ -877,21 +820,20 @@ static void tipc_link_send_sync(struct tipc_link *l) return; msg = buf_msg(buf); - tipc_msg_init(msg, BCAST_PROTOCOL, STATE_MSG, INT_H_SIZE, l->addr); - msg_set_last_bcast(msg, l->owner->bclink.acked); - link_add_chain_to_outqueue(l, buf, 0); - tipc_link_push_queue(l); + tipc_msg_init(msg, BCAST_PROTOCOL, STATE_MSG, INT_H_SIZE, link->addr); + msg_set_last_bcast(msg, link->owner->bclink.acked); + __tipc_link_xmit(link, buf); } /* - * tipc_link_recv_sync - synchronize broadcast link endpoints. + * tipc_link_sync_rcv - synchronize broadcast link endpoints. * Receive the sequence number where we should start receiving and * acking broadcast packets from a newly added peer node, and open * up for reception of such packets. * * Called with node locked */ -static void tipc_link_recv_sync(struct tipc_node *n, struct sk_buff *buf) +static void tipc_link_sync_rcv(struct tipc_node *n, struct sk_buff *buf) { struct tipc_msg *msg = buf_msg(buf); @@ -901,315 +843,9 @@ static void tipc_link_recv_sync(struct tipc_node *n, struct sk_buff *buf) } /* - * tipc_link_send_names - send name table entries to new neighbor - * - * Send routine for bulk delivery of name table messages when contact - * with a new neighbor occurs. No link congestion checking is performed - * because name table messages *must* be delivered. The messages must be - * small enough not to require fragmentation. - * Called without any locks held. - */ -void tipc_link_send_names(struct list_head *message_list, u32 dest) -{ - struct tipc_node *n_ptr; - struct tipc_link *l_ptr; - struct sk_buff *buf; - struct sk_buff *temp_buf; - - if (list_empty(message_list)) - return; - - read_lock_bh(&tipc_net_lock); - n_ptr = tipc_node_find(dest); - if (n_ptr) { - tipc_node_lock(n_ptr); - l_ptr = n_ptr->active_links[0]; - if (l_ptr) { - /* convert circular list to linear list */ - ((struct sk_buff *)message_list->prev)->next = NULL; - link_add_chain_to_outqueue(l_ptr, - (struct sk_buff *)message_list->next, 0); - tipc_link_push_queue(l_ptr); - INIT_LIST_HEAD(message_list); - } - tipc_node_unlock(n_ptr); - } - read_unlock_bh(&tipc_net_lock); - - /* discard the messages if they couldn't be sent */ - list_for_each_safe(buf, temp_buf, ((struct sk_buff *)message_list)) { - list_del((struct list_head *)buf); - kfree_skb(buf); - } -} - -/* - * link_send_buf_fast: Entry for data messages where the - * destination link is known and the header is complete, - * inclusive total message length. Very time critical. - * Link is locked. Returns user data length. - */ -static int link_send_buf_fast(struct tipc_link *l_ptr, struct sk_buff *buf, - u32 *used_max_pkt) -{ - struct tipc_msg *msg = buf_msg(buf); - int res = msg_data_sz(msg); - - if (likely(!link_congested(l_ptr))) { - if (likely(msg_size(msg) <= l_ptr->max_pkt)) { - if (likely(!tipc_bearer_blocked(l_ptr->b_ptr))) { - link_add_to_outqueue(l_ptr, buf, msg); - tipc_bearer_send(l_ptr->b_ptr, buf, - &l_ptr->media_addr); - l_ptr->unacked_window = 0; - return res; - } - } else - *used_max_pkt = l_ptr->max_pkt; - } - return tipc_link_send_buf(l_ptr, buf); /* All other cases */ -} - -/* - * tipc_link_send_sections_fast: Entry for messages where the - * destination processor is known and the header is complete, - * except for total message length. - * Returns user data length or errno. - */ -int tipc_link_send_sections_fast(struct tipc_port *sender, - struct iovec const *msg_sect, - unsigned int len, u32 destaddr) -{ - struct tipc_msg *hdr = &sender->phdr; - struct tipc_link *l_ptr; - struct sk_buff *buf; - struct tipc_node *node; - int res; - u32 selector = msg_origport(hdr) & 1; - -again: - /* - * Try building message using port's max_pkt hint. - * (Must not hold any locks while building message.) - */ - res = tipc_msg_build(hdr, msg_sect, len, sender->max_pkt, &buf); - /* Exit if build request was invalid */ - if (unlikely(res < 0)) - return res; - - read_lock_bh(&tipc_net_lock); - node = tipc_node_find(destaddr); - if (likely(node)) { - tipc_node_lock(node); - l_ptr = node->active_links[selector]; - if (likely(l_ptr)) { - if (likely(buf)) { - res = link_send_buf_fast(l_ptr, buf, - &sender->max_pkt); -exit: - tipc_node_unlock(node); - read_unlock_bh(&tipc_net_lock); - return res; - } - - /* Exit if link (or bearer) is congested */ - if (link_congested(l_ptr) || - tipc_bearer_blocked(l_ptr->b_ptr)) { - res = link_schedule_port(l_ptr, - sender->ref, res); - goto exit; - } - - /* - * Message size exceeds max_pkt hint; update hint, - * then re-try fast path or fragment the message - */ - sender->max_pkt = l_ptr->max_pkt; - tipc_node_unlock(node); - read_unlock_bh(&tipc_net_lock); - - - if ((msg_hdr_sz(hdr) + res) <= sender->max_pkt) - goto again; - - return link_send_sections_long(sender, msg_sect, len, - destaddr); - } - tipc_node_unlock(node); - } - read_unlock_bh(&tipc_net_lock); - - /* Couldn't find a link to the destination node */ - if (buf) - return tipc_reject_msg(buf, TIPC_ERR_NO_NODE); - if (res >= 0) - return tipc_port_reject_sections(sender, hdr, msg_sect, - len, TIPC_ERR_NO_NODE); - return res; -} - -/* - * link_send_sections_long(): Entry for long messages where the - * destination node is known and the header is complete, - * inclusive total message length. - * Link and bearer congestion status have been checked to be ok, - * and are ignored if they change. - * - * Note that fragments do not use the full link MTU so that they won't have - * to undergo refragmentation if link changeover causes them to be sent - * over another link with an additional tunnel header added as prefix. - * (Refragmentation will still occur if the other link has a smaller MTU.) - * - * Returns user data length or errno. - */ -static int link_send_sections_long(struct tipc_port *sender, - struct iovec const *msg_sect, - unsigned int len, u32 destaddr) -{ - struct tipc_link *l_ptr; - struct tipc_node *node; - struct tipc_msg *hdr = &sender->phdr; - u32 dsz = len; - u32 max_pkt, fragm_sz, rest; - struct tipc_msg fragm_hdr; - struct sk_buff *buf, *buf_chain, *prev; - u32 fragm_crs, fragm_rest, hsz, sect_rest; - const unchar __user *sect_crs; - int curr_sect; - u32 fragm_no; - int res = 0; - -again: - fragm_no = 1; - max_pkt = sender->max_pkt - INT_H_SIZE; - /* leave room for tunnel header in case of link changeover */ - fragm_sz = max_pkt - INT_H_SIZE; - /* leave room for fragmentation header in each fragment */ - rest = dsz; - fragm_crs = 0; - fragm_rest = 0; - sect_rest = 0; - sect_crs = NULL; - curr_sect = -1; - - /* Prepare reusable fragment header */ - tipc_msg_init(&fragm_hdr, MSG_FRAGMENTER, FIRST_FRAGMENT, - INT_H_SIZE, msg_destnode(hdr)); - msg_set_size(&fragm_hdr, max_pkt); - msg_set_fragm_no(&fragm_hdr, 1); - - /* Prepare header of first fragment */ - buf_chain = buf = tipc_buf_acquire(max_pkt); - if (!buf) - return -ENOMEM; - buf->next = NULL; - skb_copy_to_linear_data(buf, &fragm_hdr, INT_H_SIZE); - hsz = msg_hdr_sz(hdr); - skb_copy_to_linear_data_offset(buf, INT_H_SIZE, hdr, hsz); - - /* Chop up message */ - fragm_crs = INT_H_SIZE + hsz; - fragm_rest = fragm_sz - hsz; - - do { /* For all sections */ - u32 sz; - - if (!sect_rest) { - sect_rest = msg_sect[++curr_sect].iov_len; - sect_crs = msg_sect[curr_sect].iov_base; - } - - if (sect_rest < fragm_rest) - sz = sect_rest; - else - sz = fragm_rest; - - if (copy_from_user(buf->data + fragm_crs, sect_crs, sz)) { - res = -EFAULT; -error: - for (; buf_chain; buf_chain = buf) { - buf = buf_chain->next; - kfree_skb(buf_chain); - } - return res; - } - sect_crs += sz; - sect_rest -= sz; - fragm_crs += sz; - fragm_rest -= sz; - rest -= sz; - - if (!fragm_rest && rest) { - - /* Initiate new fragment: */ - if (rest <= fragm_sz) { - fragm_sz = rest; - msg_set_type(&fragm_hdr, LAST_FRAGMENT); - } else { - msg_set_type(&fragm_hdr, FRAGMENT); - } - msg_set_size(&fragm_hdr, fragm_sz + INT_H_SIZE); - msg_set_fragm_no(&fragm_hdr, ++fragm_no); - prev = buf; - buf = tipc_buf_acquire(fragm_sz + INT_H_SIZE); - if (!buf) { - res = -ENOMEM; - goto error; - } - - buf->next = NULL; - prev->next = buf; - skb_copy_to_linear_data(buf, &fragm_hdr, INT_H_SIZE); - fragm_crs = INT_H_SIZE; - fragm_rest = fragm_sz; - } - } while (rest > 0); - - /* - * Now we have a buffer chain. Select a link and check - * that packet size is still OK - */ - node = tipc_node_find(destaddr); - if (likely(node)) { - tipc_node_lock(node); - l_ptr = node->active_links[sender->ref & 1]; - if (!l_ptr) { - tipc_node_unlock(node); - goto reject; - } - if (l_ptr->max_pkt < max_pkt) { - sender->max_pkt = l_ptr->max_pkt; - tipc_node_unlock(node); - for (; buf_chain; buf_chain = buf) { - buf = buf_chain->next; - kfree_skb(buf_chain); - } - goto again; - } - } else { -reject: - for (; buf_chain; buf_chain = buf) { - buf = buf_chain->next; - kfree_skb(buf_chain); - } - return tipc_port_reject_sections(sender, hdr, msg_sect, - len, TIPC_ERR_NO_NODE); - } - - /* Append chain of fragments to send queue & send them */ - l_ptr->long_msg_seq_no++; - link_add_chain_to_outqueue(l_ptr, buf_chain, l_ptr->long_msg_seq_no); - l_ptr->stats.sent_fragments += fragm_no; - l_ptr->stats.sent_fragmented++; - tipc_link_push_queue(l_ptr); - tipc_node_unlock(node); - return dsz; -} - -/* * tipc_link_push_packet: Push one unsent packet to the media */ -u32 tipc_link_push_packet(struct tipc_link *l_ptr) +static u32 tipc_link_push_packet(struct tipc_link *l_ptr) { struct sk_buff *buf = l_ptr->first_out; u32 r_q_size = l_ptr->retransm_queue_size; @@ -1234,7 +870,7 @@ u32 tipc_link_push_packet(struct tipc_link *l_ptr) if (r_q_size && buf) { msg_set_ack(buf_msg(buf), mod(l_ptr->next_in_no - 1)); msg_set_bcast_ack(buf_msg(buf), l_ptr->owner->bclink.last_in); - tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr); + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); l_ptr->retransm_queue_head = mod(++r_q_head); l_ptr->retransm_queue_size = --r_q_size; l_ptr->stats.retransmitted++; @@ -1246,7 +882,7 @@ u32 tipc_link_push_packet(struct tipc_link *l_ptr) if (buf) { msg_set_ack(buf_msg(buf), mod(l_ptr->next_in_no - 1)); msg_set_bcast_ack(buf_msg(buf), l_ptr->owner->bclink.last_in); - tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr); + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); l_ptr->unacked_window = 0; kfree_skb(buf); l_ptr->proto_msg_queue = NULL; @@ -1263,9 +899,10 @@ u32 tipc_link_push_packet(struct tipc_link *l_ptr) if (mod(next - first) < l_ptr->queue_limit[0]) { msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); - tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr); + tipc_bearer_send(l_ptr->bearer_id, buf, + &l_ptr->media_addr); if (msg_user(msg) == MSG_BUNDLER) - msg_set_type(msg, CLOSED_MSG); + msg_set_type(msg, BUNDLE_CLOSED); l_ptr->next_out = buf->next; return 0; } @@ -1281,41 +918,29 @@ void tipc_link_push_queue(struct tipc_link *l_ptr) { u32 res; - if (tipc_bearer_blocked(l_ptr->b_ptr)) - return; - do { res = tipc_link_push_packet(l_ptr); } while (!res); } -static void link_reset_all(unsigned long addr) +void tipc_link_reset_all(struct tipc_node *node) { - struct tipc_node *n_ptr; char addr_string[16]; u32 i; - read_lock_bh(&tipc_net_lock); - n_ptr = tipc_node_find((u32)addr); - if (!n_ptr) { - read_unlock_bh(&tipc_net_lock); - return; /* node no longer exists */ - } - - tipc_node_lock(n_ptr); + tipc_node_lock(node); pr_warn("Resetting all links to %s\n", - tipc_addr_string_fill(addr_string, n_ptr->addr)); + tipc_addr_string_fill(addr_string, node->addr)); for (i = 0; i < MAX_BEARERS; i++) { - if (n_ptr->links[i]) { - link_print(n_ptr->links[i], "Resetting link\n"); - tipc_link_reset(n_ptr->links[i]); + if (node->links[i]) { + link_print(node->links[i], "Resetting link\n"); + tipc_link_reset(node->links[i]); } } - tipc_node_unlock(n_ptr); - read_unlock_bh(&tipc_net_lock); + tipc_node_unlock(node); } static void link_retransmit_failure(struct tipc_link *l_ptr, @@ -1352,10 +977,9 @@ static void link_retransmit_failure(struct tipc_link *l_ptr, n_ptr->bclink.oos_state, n_ptr->bclink.last_sent); - tipc_k_signal((Handler)link_reset_all, (unsigned long)n_ptr->addr); - tipc_node_unlock(n_ptr); + tipc_bclink_set_flags(TIPC_BCLINK_RESET); l_ptr->stale_count = 0; } } @@ -1370,33 +994,22 @@ void tipc_link_retransmit(struct tipc_link *l_ptr, struct sk_buff *buf, msg = buf_msg(buf); - if (tipc_bearer_blocked(l_ptr->b_ptr)) { - if (l_ptr->retransm_queue_size == 0) { - l_ptr->retransm_queue_head = msg_seqno(msg); - l_ptr->retransm_queue_size = retransmits; - } else { - pr_err("Unexpected retransmit on link %s (qsize=%d)\n", - l_ptr->name, l_ptr->retransm_queue_size); + /* Detect repeated retransmit failures */ + if (l_ptr->last_retransmitted == msg_seqno(msg)) { + if (++l_ptr->stale_count > 100) { + link_retransmit_failure(l_ptr, buf); + return; } - return; } else { - /* Detect repeated retransmit failures on unblocked bearer */ - if (l_ptr->last_retransmitted == msg_seqno(msg)) { - if (++l_ptr->stale_count > 100) { - link_retransmit_failure(l_ptr, buf); - return; - } - } else { - l_ptr->last_retransmitted = msg_seqno(msg); - l_ptr->stale_count = 1; - } + l_ptr->last_retransmitted = msg_seqno(msg); + l_ptr->stale_count = 1; } while (retransmits && (buf != l_ptr->next_out) && buf) { msg = buf_msg(buf); msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); - tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr); + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); buf = buf->next; retransmits--; l_ptr->stats.retransmitted++; @@ -1451,6 +1064,12 @@ static int link_recv_buf_validate(struct sk_buff *buf) u32 hdr_size; u32 min_hdr_size; + /* 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; @@ -1476,16 +1095,15 @@ static int link_recv_buf_validate(struct sk_buff *buf) } /** - * tipc_recv_msg - process TIPC messages arriving from off-node + * tipc_rcv - process TIPC packets/messages arriving from off-node * @head: pointer to message buffer chain - * @tb_ptr: pointer to bearer message arrived on + * @b_ptr: pointer to bearer message arrived on * * Invoked with no locks held. Bearer pointer must point to a valid bearer * structure (i.e. cannot be NULL), but bearer can be inactive. */ -void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *b_ptr) +void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr) { - read_lock_bh(&tipc_net_lock); while (head) { struct tipc_node *n_ptr; struct tipc_link *l_ptr; @@ -1495,13 +1113,9 @@ void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *b_ptr) u32 seq_no; u32 ackd; u32 released = 0; - int type; head = head->next; - - /* Ensure bearer is still enabled */ - if (unlikely(!b_ptr->active)) - goto discard; + buf->next = NULL; /* Ensure message is well-formed */ if (unlikely(!link_recv_buf_validate(buf))) @@ -1516,9 +1130,9 @@ void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *b_ptr) if (unlikely(msg_non_seq(msg))) { if (msg_user(msg) == LINK_CONFIG) - tipc_disc_recv_msg(buf, b_ptr); + tipc_disc_rcv(buf, b_ptr); else - tipc_bclink_recv_pkt(buf); + tipc_bclink_rcv(buf); continue; } @@ -1539,14 +1153,14 @@ void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *b_ptr) goto unlock_discard; /* Verify that communication with node is currently allowed */ - if ((n_ptr->block_setup & WAIT_PEER_DOWN) && - msg_user(msg) == LINK_PROTOCOL && - (msg_type(msg) == RESET_MSG || - msg_type(msg) == ACTIVATE_MSG) && - !msg_redundant_link(msg)) - n_ptr->block_setup &= ~WAIT_PEER_DOWN; - - if (n_ptr->block_setup) + if ((n_ptr->action_flags & TIPC_WAIT_PEER_LINKS_DOWN) && + msg_user(msg) == LINK_PROTOCOL && + (msg_type(msg) == RESET_MSG || + msg_type(msg) == ACTIVATE_MSG) && + !msg_redundant_link(msg)) + n_ptr->action_flags &= ~TIPC_WAIT_PEER_LINKS_DOWN; + + if (tipc_node_blocked(n_ptr)) goto unlock_discard; /* Validate message sequence number info */ @@ -1561,7 +1175,6 @@ void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *b_ptr) while ((crs != l_ptr->next_out) && less_eq(buf_seqno(crs), ackd)) { struct sk_buff *next = crs->next; - kfree_skb(crs); crs = next; released++; @@ -1574,18 +1187,16 @@ void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *b_ptr) /* Try sending any messages link endpoint has pending */ if (unlikely(l_ptr->next_out)) tipc_link_push_queue(l_ptr); - if (unlikely(!list_empty(&l_ptr->waiting_ports))) - tipc_link_wakeup_ports(l_ptr, 0); - if (unlikely(++l_ptr->unacked_window >= TIPC_MIN_LINK_WIN)) { - l_ptr->stats.sent_acks++; - tipc_link_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); + + if (released && !skb_queue_empty(&l_ptr->waiting_sks)) { + link_prepare_wakeup(l_ptr); + l_ptr->owner->action_flags |= TIPC_WAKEUP_USERS; } - /* Now (finally!) process the incoming message */ -protocol_check: + /* Process the incoming packet */ if (unlikely(!link_working_working(l_ptr))) { if (msg_user(msg) == LINK_PROTOCOL) { - link_recv_proto_msg(l_ptr, buf); + tipc_link_proto_rcv(l_ptr, buf); head = link_insert_deferred_queue(l_ptr, head); tipc_node_unlock(n_ptr); continue; @@ -1614,72 +1225,100 @@ protocol_check: l_ptr->next_in_no++; if (unlikely(l_ptr->oldest_deferred_in)) head = link_insert_deferred_queue(l_ptr, head); -deliver: - if (likely(msg_isdata(msg))) { - tipc_node_unlock(n_ptr); - tipc_port_recv_msg(buf); - continue; + + if (unlikely(++l_ptr->unacked_window >= TIPC_MIN_LINK_WIN)) { + l_ptr->stats.sent_acks++; + tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); } - switch (msg_user(msg)) { - int ret; - case MSG_BUNDLER: - l_ptr->stats.recv_bundles++; - l_ptr->stats.recv_bundled += msg_msgcnt(msg); - tipc_node_unlock(n_ptr); - tipc_link_recv_bundle(buf); - continue; - case NAME_DISTRIBUTOR: - n_ptr->bclink.recv_permitted = true; - tipc_node_unlock(n_ptr); - tipc_named_recv(buf); - continue; - case BCAST_PROTOCOL: - tipc_link_recv_sync(n_ptr, buf); - tipc_node_unlock(n_ptr); - continue; - case CONN_MANAGER: - tipc_node_unlock(n_ptr); - tipc_port_recv_proto_msg(buf); - continue; - case MSG_FRAGMENTER: - l_ptr->stats.recv_fragments++; - ret = tipc_link_recv_fragment(&l_ptr->reasm_head, - &l_ptr->reasm_tail, - &buf); - if (ret == LINK_REASM_COMPLETE) { - l_ptr->stats.recv_fragmented++; - msg = buf_msg(buf); - goto deliver; - } - if (ret == LINK_REASM_ERROR) - tipc_link_reset(l_ptr); + + if (tipc_link_prepare_input(l_ptr, &buf)) { tipc_node_unlock(n_ptr); continue; - case CHANGEOVER_PROTOCOL: - type = msg_type(msg); - if (link_recv_changeover_msg(&l_ptr, &buf)) { - msg = buf_msg(buf); - seq_no = msg_seqno(msg); - if (type == ORIGINAL_MSG) - goto deliver; - goto protocol_check; - } - break; - default: - kfree_skb(buf); - buf = NULL; - break; } tipc_node_unlock(n_ptr); - tipc_net_route_msg(buf); + msg = buf_msg(buf); + if (tipc_link_input(l_ptr, buf) != 0) + goto discard; continue; unlock_discard: - tipc_node_unlock(n_ptr); discard: kfree_skb(buf); } - read_unlock_bh(&tipc_net_lock); +} + +/** + * tipc_link_prepare_input - process TIPC link messages + * + * returns nonzero if the message was consumed + * + * Node lock must be held + */ +static int tipc_link_prepare_input(struct tipc_link *l, struct sk_buff **buf) +{ + struct tipc_node *n; + struct tipc_msg *msg; + int res = -EINVAL; + + 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); + } + break; + case MSG_BUNDLER: + l->stats.recv_bundles++; + l->stats.recv_bundled += msg_msgcnt(msg); + res = 0; + break; + case NAME_DISTRIBUTOR: + n->bclink.recv_permitted = true; + res = 0; + break; + case BCAST_PROTOCOL: + tipc_link_sync_rcv(n, *buf); + break; + default: + res = 0; + } + return res; +} +/** + * tipc_link_input - Deliver message too higher layers + */ +static int tipc_link_input(struct tipc_link *l, struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + int res = 0; + + 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); + break; + case NAME_DISTRIBUTOR: + tipc_named_rcv(buf); + break; + case MSG_BUNDLER: + tipc_link_bundle_rcv(buf); + break; + default: + res = -EINVAL; + } + return res; } /** @@ -1741,7 +1380,7 @@ static void link_handle_out_of_seq_msg(struct tipc_link *l_ptr, u32 seq_no = buf_seqno(buf); if (likely(msg_user(buf_msg(buf)) == LINK_PROTOCOL)) { - link_recv_proto_msg(l_ptr, buf); + tipc_link_proto_rcv(l_ptr, buf); return; } @@ -1762,8 +1401,9 @@ static void link_handle_out_of_seq_msg(struct tipc_link *l_ptr, &l_ptr->newest_deferred_in, buf)) { l_ptr->deferred_inqueue_sz++; l_ptr->stats.deferred_recv++; + TIPC_SKB_CB(buf)->deferred = true; if ((l_ptr->deferred_inqueue_sz % 16) == 1) - tipc_link_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); + tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); } else l_ptr->stats.duplicates++; } @@ -1771,9 +1411,8 @@ static void link_handle_out_of_seq_msg(struct tipc_link *l_ptr, /* * Send protocol message to the other endpoint. */ -void tipc_link_send_proto_msg(struct tipc_link *l_ptr, u32 msg_typ, - int probe_msg, u32 gap, u32 tolerance, - u32 priority, u32 ack_mtu) +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) { struct sk_buff *buf = NULL; struct tipc_msg *msg = l_ptr->pmsg; @@ -1786,16 +1425,17 @@ void tipc_link_send_proto_msg(struct tipc_link *l_ptr, u32 msg_typ, l_ptr->proto_msg_queue = NULL; } - if (link_blocked(l_ptr)) + /* Don't send protocol message during link changeover */ + if (l_ptr->exp_msg_count) return; /* Abort non-RESET send if communication with node is prohibited */ - if ((l_ptr->owner->block_setup) && (msg_typ != RESET_MSG)) + if ((tipc_node_blocked(l_ptr->owner)) && (msg_typ != RESET_MSG)) return; /* Create protocol message with "out-of-sequence" sequence number */ msg_set_type(msg, msg_typ); - msg_set_net_plane(msg, l_ptr->b_ptr->net_plane); + msg_set_net_plane(msg, l_ptr->net_plane); msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); msg_set_last_bcast(msg, tipc_bclink_get_last_sent()); @@ -1861,13 +1501,7 @@ void tipc_link_send_proto_msg(struct tipc_link *l_ptr, u32 msg_typ, skb_copy_to_linear_data(buf, msg, sizeof(l_ptr->proto_msg)); buf->priority = TC_PRIO_CONTROL; - /* Defer message if bearer is already blocked */ - if (tipc_bearer_blocked(l_ptr->b_ptr)) { - l_ptr->proto_msg_queue = buf; - return; - } - - tipc_bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr); + tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr); l_ptr->unacked_window = 0; kfree_skb(buf); } @@ -1877,7 +1511,7 @@ void tipc_link_send_proto_msg(struct tipc_link *l_ptr, u32 msg_typ, * Note that network plane id propagates through the network, and may * change at any time. The node with lowest address rules */ -static void link_recv_proto_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) { u32 rec_gap = 0; u32 max_pkt_info; @@ -1885,17 +1519,13 @@ static void link_recv_proto_msg(struct tipc_link *l_ptr, struct sk_buff *buf) u32 msg_tol; struct tipc_msg *msg = buf_msg(buf); - if (link_blocked(l_ptr)) + /* Discard protocol message during link changeover */ + if (l_ptr->exp_msg_count) goto exit; - /* record unnumbered packet arrival (force mismatch on next timeout) */ - l_ptr->checkpoint--; - - if (l_ptr->b_ptr->net_plane != msg_net_plane(msg)) + if (l_ptr->net_plane != msg_net_plane(msg)) if (tipc_own_addr > msg_prevnode(msg)) - l_ptr->b_ptr->net_plane = msg_net_plane(msg); - - l_ptr->owner->permit_changeover = msg_redundant_link(msg); + l_ptr->net_plane = msg_net_plane(msg); switch (msg_type(msg)) { @@ -1912,7 +1542,7 @@ static void link_recv_proto_msg(struct tipc_link *l_ptr, struct sk_buff *buf) * peer has lost contact -- don't allow peer's links * to reactivate before we recognize loss & clean up */ - l_ptr->owner->block_setup = WAIT_NODE_DOWN; + l_ptr->owner->action_flags |= TIPC_WAIT_OWN_LINKS_DOWN; } link_state_event(l_ptr, RESET_MSG); @@ -1968,6 +1598,10 @@ static void link_recv_proto_msg(struct tipc_link *l_ptr, struct sk_buff *buf) tipc_link_reset(l_ptr); /* Enforce change to take effect */ break; } + + /* Record reception; force mismatch at next timeout: */ + l_ptr->checkpoint--; + link_state_event(l_ptr, TRAFFIC_MSG_EVT); l_ptr->stats.recv_states++; if (link_reset_unknown(l_ptr)) @@ -1997,8 +1631,8 @@ static void link_recv_proto_msg(struct tipc_link *l_ptr, struct sk_buff *buf) msg_last_bcast(msg)); if (rec_gap || (msg_probe(msg))) { - tipc_link_send_proto_msg(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, max_pkt_ack); } if (msg_seq_gap(msg)) { l_ptr->stats.recv_nacks++; @@ -2012,13 +1646,13 @@ exit: } -/* - * tipc_link_tunnel(): Send one message via a link belonging to - * another bearer. Owner node is locked. +/* tipc_link_tunnel_xmit(): Tunnel one packet via a link belonging to + * a different bearer. Owner node is locked. */ -static void tipc_link_tunnel(struct tipc_link *l_ptr, - struct tipc_msg *tunnel_hdr, struct tipc_msg *msg, - u32 selector) +static void tipc_link_tunnel_xmit(struct tipc_link *l_ptr, + struct tipc_msg *tunnel_hdr, + struct tipc_msg *msg, + u32 selector) { struct tipc_link *tunnel; struct sk_buff *buf; @@ -2037,16 +1671,17 @@ static void tipc_link_tunnel(struct tipc_link *l_ptr, } skb_copy_to_linear_data(buf, tunnel_hdr, INT_H_SIZE); skb_copy_to_linear_data_offset(buf, INT_H_SIZE, msg, length); - tipc_link_send_buf(tunnel, buf); + __tipc_link_xmit(tunnel, buf); } - -/* - * changeover(): Send whole message queue via the remaining link - * Owner node is locked. +/* tipc_link_failover_send_queue(): A link has gone down, but a second + * link is still active. We can do failover. Tunnel the failing link's + * whole send queue via the remaining link. This way, we don't lose + * any packets, and sequence order is preserved for subsequent traffic + * sent over the remaining link. Owner node is locked. */ -void tipc_link_changeover(struct tipc_link *l_ptr) +void tipc_link_failover_send_queue(struct tipc_link *l_ptr) { u32 msgcount = l_ptr->out_queue_size; struct sk_buff *crs = l_ptr->first_out; @@ -2057,11 +1692,6 @@ void tipc_link_changeover(struct tipc_link *l_ptr) if (!tunnel) return; - if (!l_ptr->owner->permit_changeover) { - pr_warn("%speer did not permit changeover\n", link_co_err); - return; - } - tipc_msg_init(&tunnel_hdr, CHANGEOVER_PROTOCOL, ORIGINAL_MSG, INT_H_SIZE, l_ptr->addr); msg_set_bearer_id(&tunnel_hdr, l_ptr->peer_bearer_id); @@ -2074,7 +1704,7 @@ void tipc_link_changeover(struct tipc_link *l_ptr) if (buf) { skb_copy_to_linear_data(buf, &tunnel_hdr, INT_H_SIZE); msg_set_size(&tunnel_hdr, INT_H_SIZE); - tipc_link_send_buf(tunnel, buf); + __tipc_link_xmit(tunnel, buf); } else { pr_warn("%sunable to send changeover msg\n", link_co_err); @@ -2095,20 +1725,30 @@ void tipc_link_changeover(struct tipc_link *l_ptr) msgcount = msg_msgcnt(msg); while (msgcount--) { msg_set_seqno(m, msg_seqno(msg)); - tipc_link_tunnel(l_ptr, &tunnel_hdr, m, - msg_link_selector(m)); + tipc_link_tunnel_xmit(l_ptr, &tunnel_hdr, m, + msg_link_selector(m)); pos += align(msg_size(m)); m = (struct tipc_msg *)pos; } } else { - tipc_link_tunnel(l_ptr, &tunnel_hdr, msg, - msg_link_selector(msg)); + tipc_link_tunnel_xmit(l_ptr, &tunnel_hdr, msg, + msg_link_selector(msg)); } crs = crs->next; } } -void tipc_link_send_duplicate(struct tipc_link *l_ptr, struct tipc_link *tunnel) +/* tipc_link_dup_queue_xmit(): A second link has become active. Tunnel a + * duplicate of the first link's send queue via the new link. This way, we + * are guaranteed that currently queued packets from a socket are delivered + * before future traffic from the same socket, even if this is using the + * new link. The last arriving copy of each duplicate packet is dropped at + * the receiving end by the regular protocol check, so packet cardinality + * 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) { struct sk_buff *iter; struct tipc_msg tunnel_hdr; @@ -2137,7 +1777,7 @@ void tipc_link_send_duplicate(struct tipc_link *l_ptr, struct tipc_link *tunnel) skb_copy_to_linear_data(outbuf, &tunnel_hdr, INT_H_SIZE); skb_copy_to_linear_data_offset(outbuf, INT_H_SIZE, iter->data, length); - tipc_link_send_buf(tunnel, outbuf); + __tipc_link_xmit(tunnel, outbuf); if (!tipc_link_is_up(l_ptr)) return; iter = iter->next; @@ -2164,91 +1804,117 @@ static struct sk_buff *buf_extract(struct sk_buff *skb, u32 from_pos) return eb; } -/* - * link_recv_changeover_msg(): Receive tunneled packet sent - * via other link. Node is locked. Return extracted buffer. + + +/* tipc_link_dup_rcv(): Receive a tunnelled DUPLICATE_MSG packet. + * Owner node is locked. */ -static int link_recv_changeover_msg(struct tipc_link **l_ptr, - struct sk_buff **buf) +static void tipc_link_dup_rcv(struct tipc_link *l_ptr, + struct sk_buff *t_buf) { - struct sk_buff *tunnel_buf = *buf; - struct tipc_link *dest_link; - struct tipc_msg *msg; - struct tipc_msg *tunnel_msg = buf_msg(tunnel_buf); - u32 msg_typ = msg_type(tunnel_msg); - u32 msg_count = msg_msgcnt(tunnel_msg); - u32 bearer_id = msg_bearer_id(tunnel_msg); + struct sk_buff *buf; - if (bearer_id >= MAX_BEARERS) - goto exit; - dest_link = (*l_ptr)->owner->links[bearer_id]; - if (!dest_link) - goto exit; - if (dest_link == *l_ptr) { - pr_err("Unexpected changeover message on link <%s>\n", - (*l_ptr)->name); - goto exit; + 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); + return; } - *l_ptr = dest_link; - msg = msg_get_wrapped(tunnel_msg); - if (msg_typ == DUPLICATE_MSG) { - if (less(msg_seqno(msg), mod(dest_link->next_in_no))) - goto exit; - *buf = buf_extract(tunnel_buf, INT_H_SIZE); - if (*buf == NULL) { - pr_warn("%sduplicate msg dropped\n", link_co_err); + /* Add buffer to deferred queue, if applicable: */ + link_handle_out_of_seq_msg(l_ptr, buf); +} + +/* tipc_link_failover_rcv(): Receive a tunnelled ORIGINAL_MSG packet + * Owner node is locked. + */ +static struct sk_buff *tipc_link_failover_rcv(struct tipc_link *l_ptr, + struct sk_buff *t_buf) +{ + struct tipc_msg *t_msg = buf_msg(t_buf); + struct sk_buff *buf = NULL; + struct tipc_msg *msg; + + 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; } - kfree_skb(tunnel_buf); - return 1; - } + msg = buf_msg(buf); - /* First original message ?: */ - if (tipc_link_is_up(dest_link)) { - pr_info("%s<%s>, changeover initiated by peer\n", link_rst_msg, - dest_link->name); - tipc_link_reset(dest_link); - dest_link->exp_msg_count = msg_count; - if (!msg_count) - goto exit; - } else if (dest_link->exp_msg_count == START_CHANGEOVER) { - dest_link->exp_msg_count = msg_count; - if (!msg_count) + 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); + } + return buf; +} + +/* 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); + + *buf = NULL; - /* Receive original message */ - if (dest_link->exp_msg_count == 0) { - pr_warn("%sgot too many tunnelled messages\n", link_co_err); + if (bearer_id >= MAX_BEARERS) goto exit; - } - dest_link->exp_msg_count--; - if (less(msg_seqno(msg), dest_link->reset_checkpoint)) { + + l_ptr = n_ptr->links[bearer_id]; + if (!l_ptr) goto exit; - } else { - *buf = buf_extract(tunnel_buf, INT_H_SIZE); - if (*buf != NULL) { - kfree_skb(tunnel_buf); - return 1; - } else { - pr_warn("%soriginal msg dropped\n", link_co_err); - } - } + + 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); exit: - *buf = NULL; - kfree_skb(tunnel_buf); - return 0; + kfree_skb(t_buf); + return *buf != NULL; } /* * Bundler functionality: */ -void tipc_link_recv_bundle(struct sk_buff *buf) +void tipc_link_bundle_rcv(struct sk_buff *buf) { 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); @@ -2256,131 +1922,18 @@ void tipc_link_recv_bundle(struct sk_buff *buf) pr_warn("Link unable to unbundle message(s)\n"); break; } - pos += align(msg_size(buf_msg(obuf))); - tipc_net_route_msg(obuf); - } - kfree_skb(buf); -} - -/* - * Fragmentation/defragmentation: - */ - -/* - * link_send_long_buf: Entry for buffers needing fragmentation. - * The buffer is complete, inclusive total message length. - * Returns user data length. - */ -static int link_send_long_buf(struct tipc_link *l_ptr, struct sk_buff *buf) -{ - struct sk_buff *buf_chain = NULL; - struct sk_buff *buf_chain_tail = (struct sk_buff *)&buf_chain; - struct tipc_msg *inmsg = buf_msg(buf); - struct tipc_msg fragm_hdr; - u32 insize = msg_size(inmsg); - u32 dsz = msg_data_sz(inmsg); - unchar *crs = buf->data; - u32 rest = insize; - u32 pack_sz = l_ptr->max_pkt; - u32 fragm_sz = pack_sz - INT_H_SIZE; - u32 fragm_no = 0; - u32 destaddr; - - if (msg_short(inmsg)) - destaddr = l_ptr->addr; - else - destaddr = msg_destnode(inmsg); - - /* Prepare reusable fragment header: */ - tipc_msg_init(&fragm_hdr, MSG_FRAGMENTER, FIRST_FRAGMENT, - INT_H_SIZE, destaddr); - - /* Chop up message: */ - while (rest > 0) { - struct sk_buff *fragm; - - if (rest <= fragm_sz) { - fragm_sz = rest; - msg_set_type(&fragm_hdr, LAST_FRAGMENT); - } - fragm = tipc_buf_acquire(fragm_sz + INT_H_SIZE); - if (fragm == NULL) { - kfree_skb(buf); - while (buf_chain) { - buf = buf_chain; - buf_chain = buf_chain->next; - kfree_skb(buf); - } - return -ENOMEM; + omsg = buf_msg(obuf); + pos += align(msg_size(omsg)); + if (msg_isdata(omsg) || (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); } - msg_set_size(&fragm_hdr, fragm_sz + INT_H_SIZE); - fragm_no++; - msg_set_fragm_no(&fragm_hdr, fragm_no); - skb_copy_to_linear_data(fragm, &fragm_hdr, INT_H_SIZE); - skb_copy_to_linear_data_offset(fragm, INT_H_SIZE, crs, - fragm_sz); - buf_chain_tail->next = fragm; - buf_chain_tail = fragm; - - rest -= fragm_sz; - crs += fragm_sz; - msg_set_type(&fragm_hdr, FRAGMENT); } kfree_skb(buf); - - /* Append chain of fragments to send queue & send them */ - l_ptr->long_msg_seq_no++; - link_add_chain_to_outqueue(l_ptr, buf_chain, l_ptr->long_msg_seq_no); - l_ptr->stats.sent_fragments += fragm_no; - l_ptr->stats.sent_fragmented++; - tipc_link_push_queue(l_ptr); - - return dsz; -} - -/* - * tipc_link_recv_fragment(): Called with node lock on. Returns - * the reassembled buffer if message is complete. - */ -int tipc_link_recv_fragment(struct sk_buff **head, struct sk_buff **tail, - struct sk_buff **fbuf) -{ - struct sk_buff *frag = *fbuf; - struct tipc_msg *msg = buf_msg(frag); - u32 fragid = msg_type(msg); - bool headstolen; - int delta; - - skb_pull(frag, msg_hdr_sz(msg)); - if (fragid == FIRST_FRAGMENT) { - if (*head || skb_unclone(frag, GFP_ATOMIC)) - goto out_free; - *head = frag; - skb_frag_list_init(*head); - return 0; - } else if (*head && - skb_try_coalesce(*head, frag, &headstolen, &delta)) { - kfree_skb_partial(frag, headstolen); - } else { - if (!*head) - goto out_free; - if (!skb_has_frag_list(*head)) - skb_shinfo(*head)->frag_list = frag; - else - (*tail)->next = frag; - *tail = frag; - (*head)->truesize += frag->truesize; - } - if (fragid == LAST_FRAGMENT) { - *fbuf = *head; - *tail = *head = NULL; - return LINK_REASM_COMPLETE; - } - return 0; -out_free: - pr_warn_ratelimited("Link unable to reassemble fragmented message\n"); - kfree_skb(*fbuf); - return LINK_REASM_ERROR; } static void link_set_supervision_props(struct tipc_link *l_ptr, u32 tolerance) @@ -2413,35 +1966,39 @@ void tipc_link_set_queue_limits(struct tipc_link *l_ptr, u32 window) l_ptr->queue_limit[MSG_FRAGMENTER] = 4000; } -/** - * link_find_link - locate link by name - * @name: ptr to link name string - * @node: ptr to area to be filled with ptr to associated node - * - * Caller must hold 'tipc_net_lock' to ensure node and bearer are not deleted; - * this also prevents link deletion. +/* tipc_link_find_owner - locate owner node of link by link's name + * @name: pointer to link name string + * @bearer_id: pointer to index in 'node->links' array where the link was found. * - * Returns pointer to link (or 0 if invalid link name). + * Returns pointer to node owning the link, or 0 if no matching link is found. */ -static struct tipc_link *link_find_link(const char *name, - struct tipc_node **node) +static struct tipc_node *tipc_link_find_owner(const char *link_name, + unsigned int *bearer_id) { struct tipc_link *l_ptr; struct tipc_node *n_ptr; + struct tipc_node *found_node = 0; int i; - list_for_each_entry(n_ptr, &tipc_node_list, list) { + *bearer_id = 0; + rcu_read_lock(); + list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { + tipc_node_lock(n_ptr); for (i = 0; i < MAX_BEARERS; i++) { l_ptr = n_ptr->links[i]; - if (l_ptr && !strcmp(l_ptr->name, name)) - goto found; + if (l_ptr && !strcmp(l_ptr->name, link_name)) { + *bearer_id = i; + found_node = n_ptr; + break; + } } + tipc_node_unlock(n_ptr); + if (found_node) + break; } - l_ptr = NULL; - n_ptr = NULL; -found: - *node = n_ptr; - return l_ptr; + rcu_read_unlock(); + + return found_node; } /** @@ -2473,7 +2030,7 @@ static int link_value_is_valid(u16 cmd, u32 new_value) * @new_value: new value of link, bearer, or media setting * @cmd: which link, bearer, or media attribute to set (TIPC_CMD_SET_LINK_*) * - * Caller must hold 'tipc_net_lock' to ensure link/bearer/media is not deleted. + * Caller must hold RTNL lock to ensure link/bearer/media is not deleted. * * Returns 0 if value updated and negative value on error. */ @@ -2483,32 +2040,33 @@ static int link_cmd_set_value(const char *name, u32 new_value, u16 cmd) struct tipc_link *l_ptr; struct tipc_bearer *b_ptr; struct tipc_media *m_ptr; + int bearer_id; int res = 0; - l_ptr = link_find_link(name, &node); - if (l_ptr) { - /* - * acquire node lock for tipc_link_send_proto_msg(). - * see "TIPC locking policy" in net.c. - */ + node = tipc_link_find_owner(name, &bearer_id); + if (node) { tipc_node_lock(node); - switch (cmd) { - case TIPC_CMD_SET_LINK_TOL: - link_set_supervision_props(l_ptr, new_value); - tipc_link_send_proto_msg(l_ptr, - STATE_MSG, 0, 0, new_value, 0, 0); - break; - case TIPC_CMD_SET_LINK_PRI: - l_ptr->priority = new_value; - tipc_link_send_proto_msg(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; + 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; @@ -2578,9 +2136,7 @@ struct sk_buff *tipc_link_cmd_config(const void *req_tlv_area, int req_tlv_space " (cannot change setting on broadcast link)"); } - read_lock_bh(&tipc_net_lock); res = link_cmd_set_value(args->name, new_value, cmd); - read_unlock_bh(&tipc_net_lock); if (res) return tipc_cfg_reply_error_string("cannot change link setting"); @@ -2603,6 +2159,7 @@ struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_ 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); @@ -2613,18 +2170,18 @@ struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_ return tipc_cfg_reply_error_string("link not found"); return tipc_cfg_reply_none(); } + node = tipc_link_find_owner(link_name, &bearer_id); + if (!node) + return tipc_cfg_reply_error_string("link not found"); - read_lock_bh(&tipc_net_lock); - l_ptr = link_find_link(link_name, &node); + tipc_node_lock(node); + l_ptr = node->links[bearer_id]; if (!l_ptr) { - read_unlock_bh(&tipc_net_lock); + tipc_node_unlock(node); return tipc_cfg_reply_error_string("link not found"); } - - tipc_node_lock(node); link_reset_statistics(l_ptr); tipc_node_unlock(node); - read_unlock_bh(&tipc_net_lock); return tipc_cfg_reply_none(); } @@ -2651,18 +2208,24 @@ static int tipc_link_stats(const char *name, char *buf, const u32 buf_size) 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); - read_lock_bh(&tipc_net_lock); - l = link_find_link(name, &node); + node = tipc_link_find_owner(name, &bearer_id); + if (!node) + return 0; + + tipc_node_lock(node); + + l = node->links[bearer_id]; if (!l) { - read_unlock_bh(&tipc_net_lock); + tipc_node_unlock(node); return 0; } - tipc_node_lock(node); + s = &l->stats; if (tipc_link_is_active(l)) @@ -2725,7 +2288,6 @@ static int tipc_link_stats(const char *name, char *buf, const u32 buf_size) (s->accu_queue_sz / s->queue_sz_counts) : 0); tipc_node_unlock(node); - read_unlock_bh(&tipc_net_lock); return ret; } @@ -2776,7 +2338,6 @@ u32 tipc_link_get_max_pkt(u32 dest, u32 selector) if (dest == tipc_own_addr) return MAX_MSG_SIZE; - read_lock_bh(&tipc_net_lock); n_ptr = tipc_node_find(dest); if (n_ptr) { tipc_node_lock(n_ptr); @@ -2785,13 +2346,18 @@ u32 tipc_link_get_max_pkt(u32 dest, u32 selector) res = l_ptr->max_pkt; tipc_node_unlock(n_ptr); } - read_unlock_bh(&tipc_net_lock); return res; } static void link_print(struct tipc_link *l_ptr, const char *str) { - pr_info("%s Link %x<%s>:", str, l_ptr->addr, l_ptr->b_ptr->name); + struct tipc_bearer *b_ptr; + + rcu_read_lock(); + b_ptr = rcu_dereference_rtnl(bearer_list[l_ptr->bearer_id]); + if (b_ptr) + pr_info("%s Link %x<%s>:", str, l_ptr->addr, b_ptr->name); + rcu_read_unlock(); if (link_working_unknown(l_ptr)) pr_cont(":WU\n"); diff --git a/net/tipc/link.h b/net/tipc/link.h index 8a6c1026644d..b567a3427fda 100644 --- a/net/tipc/link.h +++ b/net/tipc/link.h @@ -1,7 +1,7 @@ /* * net/tipc/link.h: Include file for TIPC link code * - * Copyright (c) 1995-2006, Ericsson AB + * Copyright (c) 1995-2006, 2013-2014, Ericsson AB * Copyright (c) 2004-2005, 2010-2011, Wind River Systems * All rights reserved. * @@ -40,27 +40,23 @@ #include "msg.h" #include "node.h" -/* - * Link reassembly status codes - */ -#define LINK_REASM_ERROR -1 -#define LINK_REASM_COMPLETE 1 - -/* - * Out-of-range value for link sequence numbers +/* Out-of-range value for link sequence numbers */ #define INVALID_LINK_SEQ 0x10000 -/* - * Link states +/* Link working states */ #define WORKING_WORKING 560810u #define WORKING_UNKNOWN 560811u #define RESET_UNKNOWN 560812u #define RESET_RESET 560813u -/* - * Starting value for maximum packet size negotiation on unicast links +/* Link endpoint execution states + */ +#define LINK_STARTED 0x0001 +#define LINK_STOPPED 0x0002 + +/* Starting value for maximum packet size negotiation on unicast links * (unless bearer MTU is less) */ #define MAX_PKT_DEFAULT 1500 @@ -102,21 +98,20 @@ struct tipc_stats { * @media_addr: media address to use when sending messages over link * @timer: link timer * @owner: pointer to peer node - * @link_list: adjacent links in bearer's list of links - * @started: indicates if link has been started + * @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 - * @b_ptr: pointer to bearer used by link + * @bearer_id: local bearer id used by link * @tolerance: minimum link continuity loss needed to reset link [in ms] * @continuity_interval: link continuity testing interval [in ms] * @abort_limit: # of unacknowledged continuity probes needed to reset link * @state: current state of link FSM - * @blocked: indicates if link has been administratively blocked * @fsm_msg_cnt: # of protocol messages link FSM has sent in current state * @proto_msg: template for control messages generated by link * @pmsg: convenience pointer to "proto_msg" field * @priority: current link priority + * @net_plane: current link network plane ('A' through 'H') * @queue_limit: outbound message queue congestion thresholds (indexed by user) * @exp_msg_count: # of tunnelled messages expected during link changeover * @reset_checkpoint: seq # of last acknowledged message at time of link reset @@ -138,10 +133,9 @@ struct tipc_stats { * @retransm_queue_size: number of messages to retransmit * @retransm_queue_head: sequence number of first message to retransmit * @next_out: ptr to first unsent outbound message in queue - * @waiting_ports: linked list of ports waiting for link congestion to abate + * @waiting_sks: linked list of sockets waiting for link congestion to abate * @long_msg_seq_no: next identifier to use for outbound fragmented messages - * @reasm_head: list head of partially reassembled inbound message fragments - * @reasm_tail: last fragment received + * @reasm_buf: head of partially reassembled inbound message fragments * @stats: collects statistics regarding link activity */ struct tipc_link { @@ -150,19 +144,17 @@ struct tipc_link { struct tipc_media_addr media_addr; struct timer_list timer; struct tipc_node *owner; - struct list_head link_list; /* Management and link supervision data */ - int started; + unsigned int flags; u32 checkpoint; u32 peer_session; u32 peer_bearer_id; - struct tipc_bearer *b_ptr; + u32 bearer_id; u32 tolerance; u32 continuity_interval; u32 abort_limit; int state; - int blocked; u32 fsm_msg_cnt; struct { unchar hdr[INT_H_SIZE]; @@ -170,6 +162,7 @@ struct tipc_link { } proto_msg; struct tipc_msg *pmsg; u32 priority; + char net_plane; u32 queue_limit[15]; /* queue_limit[0]==window limit */ /* Changeover */ @@ -201,12 +194,11 @@ struct tipc_link { u32 retransm_queue_size; u32 retransm_queue_head; struct sk_buff *next_out; - struct list_head waiting_ports; + struct sk_buff_head waiting_sks; /* Fragmentation/reassembly */ u32 long_msg_seq_no; - struct sk_buff *reasm_head; - struct sk_buff *reasm_tail; + struct sk_buff *reasm_buf; /* Statistics */ struct tipc_stats stats; @@ -217,36 +209,32 @@ 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(struct tipc_link *l_ptr); -void tipc_link_changeover(struct tipc_link *l_ptr); -void tipc_link_send_duplicate(struct tipc_link *l_ptr, struct tipc_link *dest); +void tipc_link_delete_list(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); -u32 tipc_link_push_packet(struct tipc_link *l_ptr); -void tipc_link_stop(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_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); -int tipc_link_send(struct sk_buff *buf, u32 dest, u32 selector); -void tipc_link_send_names(struct list_head *message_list, u32 dest); -int tipc_link_send_buf(struct tipc_link *l_ptr, struct sk_buff *buf); +void tipc_link_reset_list(unsigned int bearer_id); +int tipc_link_xmit(struct sk_buff *buf, u32 dest, u32 selector); +int __tipc_link_xmit(struct tipc_link *link, struct sk_buff *buf); u32 tipc_link_get_max_pkt(u32 dest, u32 selector); -int tipc_link_send_sections_fast(struct tipc_port *sender, - struct iovec const *msg_sect, - unsigned int len, u32 destnode); -void tipc_link_recv_bundle(struct sk_buff *buf); -int tipc_link_recv_fragment(struct sk_buff **reasm_head, - struct sk_buff **reasm_tail, - struct sk_buff **fbuf); -void tipc_link_send_proto_msg(struct tipc_link *l_ptr, u32 msg_typ, int prob, - u32 gap, u32 tolerance, u32 priority, - u32 acked_mtu); +void tipc_link_bundle_rcv(struct sk_buff *buf); +void tipc_link_proto_xmit(struct tipc_link *l_ptr, u32 msg_typ, int prob, + u32 gap, u32 tolerance, u32 priority, u32 acked_mtu); void tipc_link_push_queue(struct tipc_link *l_ptr); u32 tipc_link_defer_pkt(struct sk_buff **head, struct sk_buff **tail, - struct sk_buff *buf); -void tipc_link_wakeup_ports(struct tipc_link *l_ptr, int all); + struct sk_buff *buf); void tipc_link_set_queue_limits(struct tipc_link *l_ptr, u32 window); void tipc_link_retransmit(struct tipc_link *l_ptr, struct sk_buff *start, u32 retransmits); @@ -312,11 +300,6 @@ static inline int link_reset_reset(struct tipc_link *l_ptr) return l_ptr->state == RESET_RESET; } -static inline int link_blocked(struct tipc_link *l_ptr) -{ - return l_ptr->exp_msg_count || l_ptr->blocked; -} - static inline int link_congested(struct tipc_link *l_ptr) { return l_ptr->out_queue_size >= l_ptr->queue_limit[0]; diff --git a/net/tipc/msg.c b/net/tipc/msg.c index e525f8ce1dee..74745a47d72a 100644 --- a/net/tipc/msg.c +++ b/net/tipc/msg.c @@ -1,7 +1,7 @@ /* * net/tipc/msg.c: TIPC message header routines * - * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2000-2006, 2014, Ericsson AB * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * @@ -36,21 +36,16 @@ #include "core.h" #include "msg.h" +#include "addr.h" +#include "name_table.h" -u32 tipc_msg_tot_importance(struct tipc_msg *m) +#define MAX_FORWARD_SIZE 1024 + +static unsigned int align(unsigned int i) { - if (likely(msg_isdata(m))) { - if (likely(msg_orignode(m) == tipc_own_addr)) - return msg_importance(m); - return msg_importance(m) + 4; - } - if ((msg_user(m) == MSG_FRAGMENTER) && - (msg_type(m) == FIRST_FRAGMENT)) - return msg_importance(msg_get_wrapped(m)); - return msg_importance(m); + return (i + 3) & ~3u; } - void tipc_msg_init(struct tipc_msg *m, u32 user, u32 type, u32 hsize, u32 destnode) { @@ -61,41 +56,407 @@ void tipc_msg_init(struct tipc_msg *m, u32 user, u32 type, u32 hsize, msg_set_size(m, hsize); msg_set_prevnode(m, tipc_own_addr); msg_set_type(m, type); - msg_set_orignode(m, tipc_own_addr); - msg_set_destnode(m, destnode); + if (hsize > SHORT_H_SIZE) { + msg_set_orignode(m, tipc_own_addr); + msg_set_destnode(m, destnode); + } } -/** - * tipc_msg_build - create message using specified header and data - * - * Note: Caller must not hold any locks in case copy_from_user() is interrupted! - * - * Returns message data size or errno +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; + + buf = tipc_buf_acquire(hdr_sz + data_sz); + if (unlikely(!buf)) + return NULL; + + msg = buf_msg(buf); + tipc_msg_init(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); + if (hdr_sz > SHORT_H_SIZE) { + msg_set_orignode(msg, onode); + msg_set_destnode(msg, dnode); + } + return buf; +} + +/* tipc_buf_append(): Append a buffer to the fragment list of another buffer + * @*headbuf: in: NULL for first frag, otherwise value returned from prev call + * out: set when successful non-complete reassembly, otherwise NULL + * @*buf: in: the buffer to append. Always defined + * out: head buf after sucessful complete reassembly, otherwise NULL + * Returns 1 when reassembly complete, otherwise 0 */ -int tipc_msg_build(struct tipc_msg *hdr, struct iovec const *msg_sect, - unsigned int len, int max_size, struct sk_buff **buf) +int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf) { - int dsz, sz, hsz; - unsigned char *to; - - dsz = len; - hsz = msg_hdr_sz(hdr); - sz = hsz + dsz; - msg_set_size(hdr, sz); - if (unlikely(sz > max_size)) { + struct sk_buff *head = *headbuf; + struct sk_buff *frag = *buf; + struct sk_buff *tail; + struct tipc_msg *msg; + u32 fragid; + int delta; + bool headstolen; + + if (!frag) + goto err; + + msg = buf_msg(frag); + fragid = msg_type(msg); + frag->next = NULL; + skb_pull(frag, msg_hdr_sz(msg)); + + if (fragid == FIRST_FRAGMENT) { + if (unlikely(head)) + goto err; + if (unlikely(skb_unclone(frag, GFP_ATOMIC))) + goto err; + head = *headbuf = frag; + skb_frag_list_init(head); + TIPC_SKB_CB(head)->tail = NULL; *buf = NULL; - return dsz; + return 0; } - *buf = tipc_buf_acquire(sz); - if (!(*buf)) - return -ENOMEM; - skb_copy_to_linear_data(*buf, hdr, hsz); - to = (*buf)->data + hsz; - if (len && memcpy_fromiovecend(to, msg_sect, 0, dsz)) { - kfree_skb(*buf); - *buf = NULL; - return -EFAULT; + if (!head) + goto err; + + if (skb_try_coalesce(head, frag, &headstolen, &delta)) { + kfree_skb_partial(frag, headstolen); + } else { + tail = TIPC_SKB_CB(head)->tail; + if (!skb_has_frag_list(head)) + skb_shinfo(head)->frag_list = frag; + else + tail->next = frag; + head->truesize += frag->truesize; + head->data_len += frag->len; + head->len += frag->len; + TIPC_SKB_CB(head)->tail = frag; } + + if (fragid == LAST_FRAGMENT) { + *buf = head; + TIPC_SKB_CB(head)->tail = NULL; + *headbuf = NULL; + return 1; + } + *buf = NULL; + return 0; + +err: + pr_warn_ratelimited("Unable to build fragment list\n"); + kfree_skb(*buf); + kfree_skb(*headbuf); + *buf = *headbuf = NULL; + return 0; +} + + +/** + * tipc_msg_build - create buffer chain containing specified header and data + * @mhdr: Message header, to be prepended to data + * @iov: User data + * @offset: Posision in iov to start copying from + * @dsz: Total length of user data + * @pktmax: Max packet size that can be used + * @chain: 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 iovec const *iov, + int offset, int dsz, int pktmax , struct sk_buff **chain) +{ + int mhsz = msg_hdr_sz(mhdr); + int msz = mhsz + dsz; + int pktno = 1; + int pktsz; + int pktrem = pktmax; + int drem = dsz; + struct tipc_msg pkthdr; + struct sk_buff *buf, *prev; + char *pktpos; + int rc; + uint chain_sz = 0; + msg_set_size(mhdr, msz); + + /* No fragmentation needed? */ + if (likely(msz <= pktmax)) { + buf = tipc_buf_acquire(msz); + *chain = buf; + if (unlikely(!buf)) + return -ENOMEM; + skb_copy_to_linear_data(buf, mhdr, mhsz); + pktpos = buf->data + mhsz; + TIPC_SKB_CB(buf)->chain_sz = 1; + if (!dsz || !memcpy_fromiovecend(pktpos, iov, offset, 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)); + msg_set_size(&pkthdr, pktmax); + msg_set_fragm_no(&pkthdr, pktno); + + /* Prepare first fragment */ + *chain = buf = tipc_buf_acquire(pktmax); + if (!buf) + return -ENOMEM; + chain_sz = 1; + pktpos = buf->data; + skb_copy_to_linear_data(buf, &pkthdr, INT_H_SIZE); + pktpos += INT_H_SIZE; + pktrem -= INT_H_SIZE; + skb_copy_to_linear_data_offset(buf, INT_H_SIZE, mhdr, mhsz); + pktpos += mhsz; + pktrem -= mhsz; + + do { + if (drem < pktrem) + pktrem = drem; + + if (memcpy_fromiovecend(pktpos, iov, offset, pktrem)) { + rc = -EFAULT; + goto error; + } + drem -= pktrem; + offset += pktrem; + + if (!drem) + break; + + /* Prepare new fragment: */ + if (drem < (pktmax - INT_H_SIZE)) + pktsz = drem + INT_H_SIZE; + else + pktsz = pktmax; + prev = buf; + buf = tipc_buf_acquire(pktsz); + if (!buf) { + rc = -ENOMEM; + goto error; + } + chain_sz++; + prev->next = buf; + msg_set_type(&pkthdr, FRAGMENT); + msg_set_size(&pkthdr, pktsz); + msg_set_fragm_no(&pkthdr, ++pktno); + skb_copy_to_linear_data(buf, &pkthdr, INT_H_SIZE); + pktpos = buf->data + INT_H_SIZE; + pktrem = pktsz - INT_H_SIZE; + + } while (1); + TIPC_SKB_CB(*chain)->chain_sz = chain_sz; + msg_set_type(buf_msg(buf), LAST_FRAGMENT); return dsz; +error: + kfree_skb_list(*chain); + *chain = NULL; + return rc; +} + +/** + * tipc_msg_bundle(): Append contents of a buffer to tail of an existing one + * @bbuf: the existing buffer ("bundle") + * @buf: 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 *bbuf, struct sk_buff *buf, u32 mtu) +{ + struct tipc_msg *bmsg = buf_msg(bbuf); + struct tipc_msg *msg = buf_msg(buf); + unsigned int bsz = msg_size(bmsg); + unsigned int msz = msg_size(msg); + u32 start = align(bsz); + 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)) + return false; + if (unlikely(msg_user(msg) == BCAST_PROTOCOL)) + return false; + if (likely(msg_user(bmsg) != MSG_BUNDLER)) + return false; + if (likely(msg_type(bmsg) != BUNDLE_OPEN)) + return false; + if (unlikely(skb_tailroom(bbuf) < (pad + msz))) + return false; + if (unlikely(max < (start + msz))) + return false; + + skb_put(bbuf, pad + msz); + skb_copy_to_linear_data_offset(bbuf, start, buf->data, msz); + msg_set_size(bmsg, start + msz); + msg_set_msgcnt(bmsg, msg_msgcnt(bmsg) + 1); + bbuf->next = buf->next; + kfree_skb(buf); + return true; +} + +/** + * tipc_msg_make_bundle(): Create bundle buf and append message to its tail + * @buf: buffer to be appended and replaced + * @mtu: max allowable size for the bundle buffer, inclusive header + * @dnode: destination node for message. (Not always present in header) + * Replaces buffer if successful + * Returns true if sucess, otherwise false + */ +bool tipc_msg_make_bundle(struct sk_buff **buf, u32 mtu, u32 dnode) +{ + struct sk_buff *bbuf; + struct tipc_msg *bmsg; + struct tipc_msg *msg = buf_msg(*buf); + 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) + return false; + if (msg_user(msg) == BCAST_PROTOCOL) + return false; + if (msz > (max / 2)) + return false; + + bbuf = tipc_buf_acquire(max); + if (!bbuf) + return false; + + skb_trim(bbuf, INT_H_SIZE); + bmsg = buf_msg(bbuf); + tipc_msg_init(bmsg, MSG_BUNDLER, BUNDLE_OPEN, 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)); + bbuf->next = (*buf)->next; + tipc_msg_bundle(bbuf, *buf, mtu); + *buf = bbuf; + return true; +} + +/** + * tipc_msg_reverse(): swap source and destination addresses and add error code + * @buf: buffer containing message to be reversed + * @dnode: return value: node where to send message after reversal + * @err: error code to be set in message + * Consumes buffer if failure + * Returns true if success, otherwise false + */ +bool tipc_msg_reverse(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; + 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); + if (!msg_short(msg)) { + msg_set_orignode(msg, msg_destnode(&ohdr)); + msg_set_destnode(msg, msg_orignode(&ohdr)); + } + msg_set_size(msg, msg_hdr_sz(msg) + rdsz); + skb_trim(buf, msg_size(msg)); + skb_orphan(buf); + *dnode = msg_orignode(&ohdr); + return true; +exit: + kfree_skb(buf); + 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 + * + * Does not consume buffer + * Returns 0 (TIPC_OK) if message ok and we can try again, -TIPC error + * code if message to be rejected + */ +int tipc_msg_eval(struct sk_buff *buf, u32 *dnode) +{ + struct tipc_msg *msg = buf_msg(buf); + u32 dport; + + 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 (!dport) + return -TIPC_ERR_NO_NAME; + msg_incr_reroute_cnt(msg); + msg_set_destnode(msg, *dnode); + msg_set_destport(msg, dport); + return TIPC_OK; +} + +/* tipc_msg_reassemble() - clone a buffer chain of fragments and + * reassemble the clones into one message + */ +struct sk_buff *tipc_msg_reassemble(struct sk_buff *chain) +{ + struct sk_buff *buf = chain; + struct sk_buff *frag = buf; + struct sk_buff *head = NULL; + int hdr_sz; + + /* Copy header if single buffer */ + if (!buf->next) { + hdr_sz = skb_headroom(buf) + msg_hdr_sz(buf_msg(buf)); + return __pskb_copy(buf, hdr_sz, GFP_ATOMIC); + } + + /* Clone all fragments and reassemble */ + while (buf) { + frag = skb_clone(buf, GFP_ATOMIC); + if (!frag) + goto error; + frag->next = NULL; + if (tipc_buf_append(&head, &frag)) + break; + if (!head) + goto error; + buf = buf->next; + } + return frag; +error: + pr_warn("Failed do clone local mcast rcv buffer\n"); + kfree_skb(head); + return NULL; } diff --git a/net/tipc/msg.h b/net/tipc/msg.h index 76d1269b9443..0ea7b695ac4d 100644 --- a/net/tipc/msg.h +++ b/net/tipc/msg.h @@ -1,7 +1,7 @@ /* * net/tipc/msg.h: Include file for TIPC message header routines * - * Copyright (c) 2000-2007, Ericsson AB + * Copyright (c) 2000-2007, 2014, Ericsson AB * Copyright (c) 2005-2008, 2010-2011, Wind River Systems * All rights reserved. * @@ -442,6 +442,7 @@ static inline struct tipc_msg *msg_get_wrapped(struct tipc_msg *m) #define NAME_DISTRIBUTOR 11 #define MSG_FRAGMENTER 12 #define LINK_CONFIG 13 +#define SOCK_WAKEUP 14 /* pseudo user */ /* * Connection management protocol message types @@ -463,6 +464,11 @@ static inline struct tipc_msg *msg_get_wrapped(struct tipc_msg *m) #define FRAGMENT 1 #define LAST_FRAGMENT 2 +/* Bundling protocol message types + */ +#define BUNDLE_OPEN 0 +#define BUNDLE_CLOSED 1 + /* * Link management protocol message types */ @@ -706,9 +712,40 @@ static inline void msg_set_link_tolerance(struct tipc_msg *m, u32 n) msg_set_bits(m, 9, 0, 0xffff, n); } -u32 tipc_msg_tot_importance(struct tipc_msg *m); +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); -int tipc_msg_build(struct tipc_msg *hdr, struct iovec const *msg_sect, - unsigned int len, int max_size, struct sk_buff **buf); + +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 *bbuf, struct sk_buff *buf, u32 mtu); + +bool tipc_msg_make_bundle(struct sk_buff **buf, u32 mtu, u32 dnode); + +int tipc_msg_build(struct tipc_msg *mhdr, struct iovec const *iov, + int offset, int dsz, int mtu , struct sk_buff **chain); + +struct sk_buff *tipc_msg_reassemble(struct sk_buff *chain); + #endif diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c index e0d08055754e..376d2bb51d8d 100644 --- a/net/tipc/name_distr.c +++ b/net/tipc/name_distr.c @@ -1,7 +1,7 @@ /* * net/tipc/name_distr.c: TIPC name distribution code * - * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2000-2006, 2014, Ericsson AB * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * @@ -38,34 +38,6 @@ #include "link.h" #include "name_distr.h" -#define ITEM_SIZE sizeof(struct distr_item) - -/** - * struct distr_item - publication info distributed to other nodes - * @type: name sequence type - * @lower: name sequence lower bound - * @upper: name sequence upper bound - * @ref: publishing port reference - * @key: publication key - * - * ===> All fields are stored in network byte order. <=== - * - * First 3 fields identify (name or) name sequence being published. - * Reference field uniquely identifies port that published name sequence. - * Key field uniquely identifies publication, in the event a port has - * multiple publications of the same name sequence. - * - * Note: There is no field that identifies the publishing node because it is - * the same for all items contained within a publication message. - */ -struct distr_item { - __be32 type; - __be32 lower; - __be32 upper; - __be32 ref; - __be32 key; -}; - /** * struct publ_list - list of publications made by this node * @list: circular list of publications @@ -99,6 +71,21 @@ static struct publ_list *publ_lists[] = { }; +int sysctl_tipc_named_timeout __read_mostly = 2000; + +/** + * struct tipc_dist_queue - queue holding deferred name table updates + */ +static struct list_head tipc_dist_queue = LIST_HEAD_INIT(tipc_dist_queue); + +struct distr_queue_item { + struct distr_item i; + u32 dtype; + u32 node; + unsigned long expires; + struct list_head next; +}; + /** * publ_to_item - add publication info to a publication message */ @@ -127,20 +114,26 @@ static struct sk_buff *named_prepare_buf(u32 type, u32 size, u32 dest) return buf; } -static void named_cluster_distribute(struct sk_buff *buf) +void named_cluster_distribute(struct sk_buff *buf) { - struct sk_buff *buf_copy; - struct tipc_node *n_ptr; - - list_for_each_entry(n_ptr, &tipc_node_list, list) { - if (tipc_node_active_links(n_ptr)) { - buf_copy = skb_copy(buf, GFP_ATOMIC); - if (!buf_copy) - break; - msg_set_destnode(buf_msg(buf_copy), n_ptr->addr); - tipc_link_send(buf_copy, n_ptr->addr, n_ptr->addr); - } + struct sk_buff *obuf; + struct tipc_node *node; + u32 dnode; + + rcu_read_lock(); + list_for_each_entry_rcu(node, &tipc_node_list, list) { + dnode = node->addr; + if (in_own_node(dnode)) + continue; + if (!tipc_node_active_links(node)) + continue; + obuf = skb_copy(buf, GFP_ATOMIC); + if (!obuf) + break; + msg_set_destnode(buf_msg(obuf), dnode); + tipc_link_xmit(obuf, dnode, dnode); } + rcu_read_unlock(); kfree_skb(buf); } @@ -148,7 +141,7 @@ static void named_cluster_distribute(struct sk_buff *buf) /** * tipc_named_publish - tell other nodes about a new publication by this node */ -void tipc_named_publish(struct publication *publ) +struct sk_buff *tipc_named_publish(struct publication *publ) { struct sk_buff *buf; struct distr_item *item; @@ -157,23 +150,23 @@ void tipc_named_publish(struct publication *publ) publ_lists[publ->scope]->size++; if (publ->scope == TIPC_NODE_SCOPE) - return; + return NULL; buf = named_prepare_buf(PUBLICATION, ITEM_SIZE, 0); if (!buf) { pr_warn("Publication distribution failure\n"); - return; + return NULL; } item = (struct distr_item *)msg_data(buf_msg(buf)); publ_to_item(item, publ); - named_cluster_distribute(buf); + return buf; } /** * tipc_named_withdraw - tell other nodes about a withdrawn publication by this node */ -void tipc_named_withdraw(struct publication *publ) +struct sk_buff *tipc_named_withdraw(struct publication *publ) { struct sk_buff *buf; struct distr_item *item; @@ -182,47 +175,57 @@ void tipc_named_withdraw(struct publication *publ) publ_lists[publ->scope]->size--; if (publ->scope == TIPC_NODE_SCOPE) - return; + return NULL; buf = named_prepare_buf(WITHDRAWAL, ITEM_SIZE, 0); if (!buf) { pr_warn("Withdrawal distribution failure\n"); - return; + return NULL; } item = (struct distr_item *)msg_data(buf_msg(buf)); publ_to_item(item, publ); - named_cluster_distribute(buf); + return buf; } -/* +/** * named_distribute - prepare name info for bulk distribution to another node + * @msg_list: list of messages (buffers) to be returned from this function + * @dnode: node to be updated + * @pls: linked list of publication items to be packed into buffer chain */ -static void named_distribute(struct list_head *message_list, u32 node, - struct publ_list *pls, u32 max_item_buf) +static void named_distribute(struct list_head *msg_list, u32 dnode, + struct publ_list *pls) { struct publication *publ; struct sk_buff *buf = NULL; struct distr_item *item = NULL; - u32 left = 0; - u32 rest = pls->size * ITEM_SIZE; + uint dsz = pls->size * ITEM_SIZE; + uint msg_dsz = (tipc_node_get_mtu(dnode, 0) / ITEM_SIZE) * ITEM_SIZE; + uint rem = dsz; + uint msg_rem = 0; list_for_each_entry(publ, &pls->list, local_list) { + /* Prepare next buffer: */ if (!buf) { - left = (rest <= max_item_buf) ? rest : max_item_buf; - rest -= left; - buf = named_prepare_buf(PUBLICATION, left, node); + msg_rem = min_t(uint, rem, msg_dsz); + rem -= msg_rem; + buf = named_prepare_buf(PUBLICATION, msg_rem, dnode); if (!buf) { pr_warn("Bulk publication failure\n"); return; } item = (struct distr_item *)msg_data(buf_msg(buf)); } + + /* Pack publication into message: */ publ_to_item(item, publ); item++; - left -= ITEM_SIZE; - if (!left) { - list_add_tail((struct list_head *)buf, message_list); + msg_rem -= ITEM_SIZE; + + /* Append full buffer to list: */ + if (!msg_rem) { + list_add_tail((struct list_head *)buf, msg_list); buf = NULL; } } @@ -231,38 +234,20 @@ static void named_distribute(struct list_head *message_list, u32 node, /** * tipc_named_node_up - tell specified node about all publications by this node */ -void tipc_named_node_up(unsigned long nodearg) +void tipc_named_node_up(u32 dnode) { - struct tipc_node *n_ptr; - struct tipc_link *l_ptr; - struct list_head message_list; - u32 node = (u32)nodearg; - u32 max_item_buf = 0; - - /* compute maximum amount of publication data to send per message */ - read_lock_bh(&tipc_net_lock); - n_ptr = tipc_node_find(node); - if (n_ptr) { - tipc_node_lock(n_ptr); - l_ptr = n_ptr->active_links[0]; - if (l_ptr) - max_item_buf = ((l_ptr->max_pkt - INT_H_SIZE) / - ITEM_SIZE) * ITEM_SIZE; - tipc_node_unlock(n_ptr); - } - read_unlock_bh(&tipc_net_lock); - if (!max_item_buf) - return; - - /* create list of publication messages, then send them as a unit */ - INIT_LIST_HEAD(&message_list); + LIST_HEAD(msg_list); + struct sk_buff *buf_chain; read_lock_bh(&tipc_nametbl_lock); - named_distribute(&message_list, node, &publ_cluster, max_item_buf); - named_distribute(&message_list, node, &publ_zone, max_item_buf); + named_distribute(&msg_list, dnode, &publ_cluster); + named_distribute(&msg_list, dnode, &publ_zone); read_unlock_bh(&tipc_nametbl_lock); - tipc_link_send_names(&message_list, node); + /* Convert circular list to linear list and send: */ + buf_chain = (struct sk_buff *)msg_list.next; + ((struct sk_buff *)msg_list.prev)->next = NULL; + tipc_link_xmit(buf_chain, dnode, dnode); } /** @@ -293,54 +278,105 @@ static void named_purge_publ(struct publication *publ) } /** - * tipc_named_recv - process name table update message sent by another node + * tipc_update_nametbl - try to process a nametable update and notify + * subscribers + * + * tipc_nametbl_lock must be held. + * Returns the publication item if successful, otherwise NULL. */ -void tipc_named_recv(struct sk_buff *buf) +static bool tipc_update_nametbl(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), + ntohl(i->upper), + TIPC_CLUSTER_SCOPE, node, + ntohl(i->ref), ntohl(i->key)); + if (publ) { + tipc_nodesub_subscribe(&publ->subscr, node, publ, + (net_ev_handler) + named_purge_publ); + return true; + } + } else if (dtype == WITHDRAWAL) { + publ = tipc_nametbl_remove_publ(ntohl(i->type), ntohl(i->lower), + node, ntohl(i->ref), + ntohl(i->key)); + if (publ) { + tipc_nodesub_unsubscribe(&publ->subscr); + kfree(publ); + return true; + } + } else { + pr_warn("Unrecognized name table message received\n"); + } + return false; +} + +/** + * tipc_named_add_backlog - add a failed name table update to the backlog + * + */ +static void tipc_named_add_backlog(struct distr_item *i, u32 type, u32 node) +{ + struct distr_queue_item *e; + unsigned long now = get_jiffies_64(); + + e = kzalloc(sizeof(*e), GFP_ATOMIC); + if (!e) + return; + e->dtype = type; + e->node = node; + e->expires = now + msecs_to_jiffies(sysctl_tipc_named_timeout); + memcpy(e, i, sizeof(*i)); + list_add_tail(&e->next, &tipc_dist_queue); +} + +/** + * tipc_named_process_backlog - try to process any pending name table updates + * from the network. + */ +void tipc_named_process_backlog(void) +{ + struct distr_queue_item *e, *tmp; + char addr[16]; + unsigned long now = get_jiffies_64(); + + 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)) + continue; + } else { + tipc_addr_string_fill(addr, e->node); + pr_warn_ratelimited("Dropping name table update (%d) of {%u, %u, %u} from %s key=%u\n", + e->dtype, ntohl(e->i.type), + ntohl(e->i.lower), + ntohl(e->i.upper), + addr, ntohl(e->i.key)); + } + list_del(&e->next); + kfree(e); + } +} + +/** + * tipc_named_rcv - process name table update message sent by another node + */ +void tipc_named_rcv(struct sk_buff *buf) { - struct publication *publ; 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); write_lock_bh(&tipc_nametbl_lock); while (count--) { - if (msg_type(msg) == PUBLICATION) { - publ = tipc_nametbl_insert_publ(ntohl(item->type), - ntohl(item->lower), - ntohl(item->upper), - TIPC_CLUSTER_SCOPE, - msg_orignode(msg), - ntohl(item->ref), - ntohl(item->key)); - if (publ) { - tipc_nodesub_subscribe(&publ->subscr, - msg_orignode(msg), - publ, - (net_ev_handler) - named_purge_publ); - } - } else if (msg_type(msg) == WITHDRAWAL) { - publ = tipc_nametbl_remove_publ(ntohl(item->type), - ntohl(item->lower), - msg_orignode(msg), - ntohl(item->ref), - ntohl(item->key)); - - if (publ) { - tipc_nodesub_unsubscribe(&publ->subscr); - kfree(publ); - } else { - pr_err("Unable to remove publication by node 0x%x\n" - " (type=%u, lower=%u, ref=%u, key=%u)\n", - msg_orignode(msg), ntohl(item->type), - ntohl(item->lower), ntohl(item->ref), - ntohl(item->key)); - } - } else { - pr_warn("Unrecognized name table message received\n"); - } + if (!tipc_update_nametbl(item, node, msg_type(msg))) + tipc_named_add_backlog(item, msg_type(msg), node); item++; } + tipc_named_process_backlog(); write_unlock_bh(&tipc_nametbl_lock); kfree_skb(buf); } diff --git a/net/tipc/name_distr.h b/net/tipc/name_distr.h index 1e41bdd4f255..b9e75feb3434 100644 --- a/net/tipc/name_distr.h +++ b/net/tipc/name_distr.h @@ -39,10 +39,40 @@ #include "name_table.h" -void tipc_named_publish(struct publication *publ); -void tipc_named_withdraw(struct publication *publ); -void tipc_named_node_up(unsigned long node); -void tipc_named_recv(struct sk_buff *buf); +#define ITEM_SIZE sizeof(struct distr_item) + +/** + * struct distr_item - publication info distributed to other nodes + * @type: name sequence type + * @lower: name sequence lower bound + * @upper: name sequence upper bound + * @ref: publishing port reference + * @key: publication key + * + * ===> All fields are stored in network byte order. <=== + * + * First 3 fields identify (name or) name sequence being published. + * Reference field uniquely identifies port that published name sequence. + * Key field uniquely identifies publication, in the event a port has + * multiple publications of the same name sequence. + * + * Note: There is no field that identifies the publishing node because it is + * the same for all items contained within a publication message. + */ +struct distr_item { + __be32 type; + __be32 lower; + __be32 upper; + __be32 ref; + __be32 key; +}; + +struct sk_buff *tipc_named_publish(struct publication *publ); +struct sk_buff *tipc_named_withdraw(struct publication *publ); +void named_cluster_distribute(struct sk_buff *buf); +void tipc_named_node_up(u32 dnode); +void tipc_named_rcv(struct sk_buff *buf); void tipc_named_reinit(void); +void tipc_named_process_backlog(void); #endif diff --git a/net/tipc/name_table.c b/net/tipc/name_table.c index 09dcd54b04e1..3a6a0a7c0759 100644 --- a/net/tipc/name_table.c +++ b/net/tipc/name_table.c @@ -39,7 +39,6 @@ #include "name_table.h" #include "name_distr.h" #include "subscr.h" -#include "port.h" #define TIPC_NAMETBL_SIZE 1024 /* must be a power of 2 */ @@ -148,8 +147,7 @@ static struct publication *publ_create(u32 type, u32 lower, u32 upper, */ static struct sub_seq *tipc_subseq_alloc(u32 cnt) { - struct sub_seq *sseq = kcalloc(cnt, sizeof(struct sub_seq), GFP_ATOMIC); - return sseq; + return kcalloc(cnt, sizeof(struct sub_seq), GFP_ATOMIC); } /** @@ -263,8 +261,6 @@ static struct publication *tipc_nameseq_insert_publ(struct name_seq *nseq, /* Lower end overlaps existing entry => need an exact match */ if ((sseq->lower != lower) || (sseq->upper != upper)) { - pr_warn("Cannot publish {%u,%u,%u}, overlap error\n", - type, lower, upper); return NULL; } @@ -286,8 +282,6 @@ static struct publication *tipc_nameseq_insert_publ(struct name_seq *nseq, /* Fail if upper end overlaps into an existing entry */ if ((inspos < nseq->first_free) && (upper >= nseq->sseqs[inspos].lower)) { - pr_warn("Cannot publish {%u,%u,%u}, overlap error\n", - type, lower, upper); return NULL; } @@ -665,6 +659,7 @@ struct publication *tipc_nametbl_publish(u32 type, u32 lower, u32 upper, u32 scope, u32 port_ref, u32 key) { struct publication *publ; + struct sk_buff *buf = NULL; if (table.local_publ_count >= TIPC_MAX_PUBLICATIONS) { pr_warn("Publication failed, local publication limit reached (%u)\n", @@ -677,9 +672,14 @@ struct publication *tipc_nametbl_publish(u32 type, u32 lower, u32 upper, tipc_own_addr, port_ref, key); if (likely(publ)) { table.local_publ_count++; - tipc_named_publish(publ); + buf = tipc_named_publish(publ); + /* Any pending external events? */ + tipc_named_process_backlog(); } write_unlock_bh(&tipc_nametbl_lock); + + if (buf) + named_cluster_distribute(buf); return publ; } @@ -689,15 +689,21 @@ struct publication *tipc_nametbl_publish(u32 type, u32 lower, u32 upper, int tipc_nametbl_withdraw(u32 type, u32 lower, u32 ref, u32 key) { struct publication *publ; + struct sk_buff *buf; write_lock_bh(&tipc_nametbl_lock); publ = tipc_nametbl_remove_publ(type, lower, tipc_own_addr, ref, key); if (likely(publ)) { table.local_publ_count--; - tipc_named_withdraw(publ); + buf = tipc_named_withdraw(publ); + /* Any pending external events? */ + tipc_named_process_backlog(); write_unlock_bh(&tipc_nametbl_lock); list_del_init(&publ->pport_list); kfree(publ); + + if (buf) + named_cluster_distribute(buf); return 1; } write_unlock_bh(&tipc_nametbl_lock); @@ -942,20 +948,48 @@ int tipc_nametbl_init(void) return 0; } -void tipc_nametbl_stop(void) +/** + * tipc_purge_publications - remove all publications for a given type + * + * tipc_nametbl_lock must be held when calling this function + */ +static void tipc_purge_publications(struct name_seq *seq) { - u32 i; + struct publication *publ, *safe; + struct sub_seq *sseq; + struct name_info *info; - if (!table.types) + if (!seq->sseqs) { + nameseq_delete_empty(seq); return; + } + 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, + publ->ref, publ->key); + kfree(publ); + } +} - /* Verify name table is empty, then release it */ +void tipc_nametbl_stop(void) +{ + u32 i; + struct name_seq *seq; + struct hlist_head *seq_head; + struct hlist_node *safe; + + /* Verify name table is empty and purge any lingering + * publications, then release the name table + */ write_lock_bh(&tipc_nametbl_lock); for (i = 0; i < TIPC_NAMETBL_SIZE; i++) { if (hlist_empty(&table.types[i])) continue; - pr_err("nametbl_stop(): orphaned hash chain detected\n"); - break; + seq_head = &table.types[i]; + hlist_for_each_entry_safe(seq, safe, seq_head, ns_list) { + tipc_purge_publications(seq); + } } kfree(table.types); table.types = NULL; diff --git a/net/tipc/net.c b/net/tipc/net.c index 7d305ecc09c2..93b9944a6a8b 100644 --- a/net/tipc/net.c +++ b/net/tipc/net.c @@ -1,7 +1,7 @@ /* * net/tipc/net.c: TIPC network routing code * - * Copyright (c) 1995-2006, Ericsson AB + * Copyright (c) 1995-2006, 2014, Ericsson AB * Copyright (c) 2005, 2010-2011, Wind River Systems * All rights reserved. * @@ -38,46 +38,41 @@ #include "net.h" #include "name_distr.h" #include "subscr.h" -#include "port.h" +#include "socket.h" #include "node.h" #include "config.h" /* * The TIPC locking policy is designed to ensure a very fine locking * granularity, permitting complete parallel access to individual - * port and node/link instances. The code consists of three major + * port and node/link instances. The code consists of four major * locking domains, each protected with their own disjunct set of locks. * - * 1: The routing hierarchy. - * Comprises the structures 'zone', 'cluster', 'node', 'link' - * and 'bearer'. The whole hierarchy is protected by a big - * read/write lock, tipc_net_lock, to enssure that nothing is added - * or removed while code is accessing any of these structures. - * This layer must not be called from the two others while they - * hold any of their own locks. - * Neither must it itself do any upcalls to the other two before - * it has released tipc_net_lock and other protective locks. + * 1: The bearer level. + * RTNL lock is used to serialize the process of configuring bearer + * on update side, and RCU lock is applied on read side to make + * bearer instance valid on both paths of message transmission and + * reception. * - * Within the tipc_net_lock domain there are two sub-domains;'node' and - * 'bearer', where local write operations are permitted, - * provided that those are protected by individual spin_locks - * per instance. Code holding tipc_net_lock(read) and a node spin_lock - * is permitted to poke around in both the node itself and its - * subordinate links. I.e, it can update link counters and queues, - * change link state, send protocol messages, and alter the - * "active_links" array in the node; but it can _not_ remove a link - * or a node from the overall structure. - * Correspondingly, individual bearers may change status within a - * tipc_net_lock(read), protected by an individual spin_lock ber bearer - * instance, but it needs tipc_net_lock(write) to remove/add any bearers. + * 2: The node and link level. + * All node instances are saved into two tipc_node_list and node_htable + * lists. The two lists are protected by node_list_lock on write side, + * and they are guarded with RCU lock on read side. Especially node + * instance is destroyed only when TIPC module is removed, and we can + * confirm that there has no any user who is accessing the node at the + * moment. Therefore, Except for iterating the two lists within RCU + * protection, it's no needed to hold RCU that we access node instance + * in other places. * + * In addition, all members in node structure including link instances + * are protected by node spin lock. * - * 2: The transport level of the protocol. - * This consists of the structures port, (and its user level - * representations, such as user_port and tipc_sock), reference and - * tipc_user (port.c, reg.c, socket.c). + * 3: The transport level of the protocol. + * This consists of the structures port, (and its user level + * representations, such as user_port and tipc_sock), reference and + * tipc_user (port.c, reg.c, socket.c). * - * This layer has four different locks: + * This layer has four different locks: * - The tipc_port spin_lock. This is protecting each port instance * from parallel data access and removal. Since we can not place * this lock in the port itself, it has been placed in the @@ -96,7 +91,7 @@ * There are two such lists; 'port_list', which is used for management, * and 'wait_list', which is used to queue ports during congestion. * - * 3: The name table (name_table.c, name_distr.c, subscription.c) + * 4: The name table (name_table.c, name_distr.c, subscription.c) * - There is one big read/write-lock (tipc_nametbl_lock) protecting the * overall name table structure. Nothing must be added/removed to * this structure without holding write access to it. @@ -108,98 +103,38 @@ * - A local spin_lock protecting the queue of subscriber events. */ -DEFINE_RWLOCK(tipc_net_lock); - -static void net_route_named_msg(struct sk_buff *buf) -{ - struct tipc_msg *msg = buf_msg(buf); - u32 dnode; - u32 dport; - - if (!msg_named(msg)) { - kfree_skb(buf); - return; - } - - dnode = addr_domain(msg_lookup_scope(msg)); - dport = tipc_nametbl_translate(msg_nametype(msg), msg_nameinst(msg), &dnode); - if (dport) { - msg_set_destnode(msg, dnode); - msg_set_destport(msg, dport); - tipc_net_route_msg(buf); - return; - } - tipc_reject_msg(buf, TIPC_ERR_NO_NAME); -} - -void tipc_net_route_msg(struct sk_buff *buf) -{ - struct tipc_msg *msg; - u32 dnode; - - if (!buf) - return; - msg = buf_msg(buf); - - /* Handle message for this node */ - dnode = msg_short(msg) ? tipc_own_addr : msg_destnode(msg); - if (tipc_in_scope(dnode, tipc_own_addr)) { - if (msg_isdata(msg)) { - if (msg_mcast(msg)) - tipc_port_recv_mcast(buf, NULL); - else if (msg_destport(msg)) - tipc_port_recv_msg(buf); - else - net_route_named_msg(buf); - return; - } - switch (msg_user(msg)) { - case NAME_DISTRIBUTOR: - tipc_named_recv(buf); - break; - case CONN_MANAGER: - tipc_port_recv_proto_msg(buf); - break; - default: - kfree_skb(buf); - } - return; - } - - /* Handle message for another node */ - skb_trim(buf, msg_size(msg)); - tipc_link_send(buf, dnode, msg_link_selector(msg)); -} - -void tipc_net_start(u32 addr) +int tipc_net_start(u32 addr) { char addr_string[16]; + int res; - write_lock_bh(&tipc_net_lock); tipc_own_addr = addr; tipc_named_reinit(); - tipc_port_reinit(); - tipc_bclink_init(); - write_unlock_bh(&tipc_net_lock); + tipc_sk_reinit(); + res = tipc_bclink_init(); + if (res) + return res; - tipc_cfg_reinit(); + tipc_nametbl_publish(TIPC_CFG_SRV, tipc_own_addr, tipc_own_addr, + TIPC_ZONE_SCOPE, 0, tipc_own_addr); pr_info("Started in network mode\n"); pr_info("Own node address %s, network identity %u\n", tipc_addr_string_fill(addr_string, tipc_own_addr), tipc_net_id); + return 0; } void tipc_net_stop(void) { - struct tipc_node *node, *t_node; - if (!tipc_own_addr) return; - write_lock_bh(&tipc_net_lock); + + tipc_nametbl_withdraw(TIPC_CFG_SRV, tipc_own_addr, 0, tipc_own_addr); + rtnl_lock(); tipc_bearer_stop(); tipc_bclink_stop(); - list_for_each_entry_safe(node, t_node, &tipc_node_list, list) - tipc_node_delete(node); - write_unlock_bh(&tipc_net_lock); + tipc_node_stop(); + rtnl_unlock(); + pr_info("Left network mode\n"); } diff --git a/net/tipc/net.h b/net/tipc/net.h index 079daadb3f72..59ef3388be2c 100644 --- a/net/tipc/net.h +++ b/net/tipc/net.h @@ -37,11 +37,7 @@ #ifndef _TIPC_NET_H #define _TIPC_NET_H -extern rwlock_t tipc_net_lock; - -void tipc_net_route_msg(struct sk_buff *buf); - -void tipc_net_start(u32 addr); +int tipc_net_start(u32 addr); void tipc_net_stop(void); #endif diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c index 9f72a6376362..ad844d365340 100644 --- a/net/tipc/netlink.c +++ b/net/tipc/netlink.c @@ -47,7 +47,7 @@ static int handle_cmd(struct sk_buff *skb, struct genl_info *info) int hdr_space = nlmsg_total_size(GENL_HDRLEN + TIPC_GENL_HDRLEN); u16 cmd; - if ((req_userhdr->cmd & 0xC000) && (!capable(CAP_NET_ADMIN))) + if ((req_userhdr->cmd & 0xC000) && (!netlink_capable(skb, CAP_NET_ADMIN))) cmd = TIPC_CMD_NOT_NET_ADMIN; else cmd = req_userhdr->cmd; @@ -83,8 +83,6 @@ static struct genl_ops tipc_genl_ops[] = { }, }; -static int tipc_genl_family_registered; - int tipc_netlink_start(void) { int res; @@ -94,16 +92,10 @@ int tipc_netlink_start(void) pr_err("Failed to register netlink interface\n"); return res; } - - tipc_genl_family_registered = 1; return 0; } void tipc_netlink_stop(void) { - if (!tipc_genl_family_registered) - return; - genl_unregister_family(&tipc_genl_family); - tipc_genl_family_registered = 0; } diff --git a/net/tipc/node.c b/net/tipc/node.c index 25100c0a6fe8..90cee4a6fce4 100644 --- a/net/tipc/node.c +++ b/net/tipc/node.c @@ -1,8 +1,8 @@ /* * net/tipc/node.c: TIPC node management routines * - * Copyright (c) 2000-2006, 2012 Ericsson AB - * Copyright (c) 2005-2006, 2010-2011, Wind River Systems + * Copyright (c) 2000-2006, 2012-2014, Ericsson AB + * Copyright (c) 2005-2006, 2010-2014, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -38,19 +38,25 @@ #include "config.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 DEFINE_SPINLOCK(node_create_lock); - 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 atomic_t tipc_num_links = ATOMIC_INIT(0); +struct tipc_sock_conn { + u32 port; + u32 peer_port; + u32 peer_node; + struct list_head list; +}; /* * A trivial power-of-two bitmask technique is used for speed, since this @@ -73,37 +79,26 @@ struct tipc_node *tipc_node_find(u32 addr) if (unlikely(!in_own_cluster_exact(addr))) return NULL; - hlist_for_each_entry(node, &node_htable[tipc_hashfn(addr)], hash) { - if (node->addr == addr) + rcu_read_lock(); + hlist_for_each_entry_rcu(node, &node_htable[tipc_hashfn(addr)], hash) { + if (node->addr == addr) { + rcu_read_unlock(); return node; + } } + rcu_read_unlock(); return NULL; } -/** - * tipc_node_create - create neighboring node - * - * Currently, this routine is called by neighbor discovery code, which holds - * net_lock for reading only. We must take node_create_lock to ensure a node - * isn't created twice if two different bearers discover the node at the same - * time. (It would be preferable to switch to holding net_lock in write mode, - * but this is a non-trivial change.) - */ struct tipc_node *tipc_node_create(u32 addr) { struct tipc_node *n_ptr, *temp_node; - spin_lock_bh(&node_create_lock); - - n_ptr = tipc_node_find(addr); - if (n_ptr) { - spin_unlock_bh(&node_create_lock); - return n_ptr; - } + spin_lock_bh(&node_list_lock); n_ptr = kzalloc(sizeof(*n_ptr), GFP_ATOMIC); if (!n_ptr) { - spin_unlock_bh(&node_create_lock); + spin_unlock_bh(&node_list_lock); pr_warn("Node creation failed, no memory\n"); return NULL; } @@ -113,32 +108,109 @@ struct tipc_node *tipc_node_create(u32 addr) INIT_HLIST_NODE(&n_ptr->hash); INIT_LIST_HEAD(&n_ptr->list); INIT_LIST_HEAD(&n_ptr->nsub); + INIT_LIST_HEAD(&n_ptr->conn_sks); + __skb_queue_head_init(&n_ptr->waiting_sks); - hlist_add_head(&n_ptr->hash, &node_htable[tipc_hashfn(addr)]); + hlist_add_head_rcu(&n_ptr->hash, &node_htable[tipc_hashfn(addr)]); - list_for_each_entry(temp_node, &tipc_node_list, list) { + list_for_each_entry_rcu(temp_node, &tipc_node_list, list) { if (n_ptr->addr < temp_node->addr) break; } - list_add_tail(&n_ptr->list, &temp_node->list); - n_ptr->block_setup = WAIT_PEER_DOWN; + 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_create_lock); + spin_unlock_bh(&node_list_lock); return n_ptr; } -void tipc_node_delete(struct tipc_node *n_ptr) +static void tipc_node_delete(struct tipc_node *n_ptr) { - list_del(&n_ptr->list); - hlist_del(&n_ptr->hash); - kfree(n_ptr); + list_del_rcu(&n_ptr->list); + hlist_del_rcu(&n_ptr->hash); + kfree_rcu(n_ptr, rcu); tipc_num_nodes--; } +void tipc_node_stop(void) +{ + 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); +} + +int tipc_node_add_conn(u32 dnode, u32 port, u32 peer_port) +{ + struct tipc_node *node; + struct tipc_sock_conn *conn; + + if (in_own_node(dnode)) + return 0; + + node = tipc_node_find(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; + conn->peer_node = dnode; + conn->port = port; + conn->peer_port = peer_port; + + tipc_node_lock(node); + list_add_tail(&conn->list, &node->conn_sks); + tipc_node_unlock(node); + return 0; +} + +void tipc_node_remove_conn(u32 dnode, u32 port) +{ + struct tipc_node *node; + struct tipc_sock_conn *conn, *safe; + + if (in_own_node(dnode)) + return; + + node = tipc_node_find(dnode); + if (!node) + return; + + tipc_node_lock(node); + list_for_each_entry_safe(conn, safe, &node->conn_sks, list) { + if (port != conn->port) + continue; + list_del(&conn->list); + 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_link_up - handle addition of link * @@ -147,30 +219,36 @@ void tipc_node_delete(struct tipc_node *n_ptr) void tipc_node_link_up(struct tipc_node *n_ptr, struct tipc_link *l_ptr) { struct tipc_link **active = &n_ptr->active_links[0]; + u32 addr = n_ptr->addr; n_ptr->working_links++; - + tipc_nametbl_publish(TIPC_LINK_STATE, addr, addr, TIPC_NODE_SCOPE, + l_ptr->bearer_id, addr); pr_info("Established link <%s> on network plane %c\n", - l_ptr->name, l_ptr->b_ptr->net_plane); + l_ptr->name, l_ptr->net_plane); if (!active[0]) { active[0] = active[1] = l_ptr; node_established_contact(n_ptr); - return; + goto exit; } if (l_ptr->priority < active[0]->priority) { pr_info("New link <%s> becomes standby\n", l_ptr->name); - return; + goto exit; } - tipc_link_send_duplicate(active[0], l_ptr); + tipc_link_dup_queue_xmit(active[0], l_ptr); if (l_ptr->priority == active[0]->priority) { active[0] = l_ptr; - return; + goto exit; } pr_info("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); 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; } /** @@ -206,16 +284,18 @@ static void node_select_active_links(struct tipc_node *n_ptr) void tipc_node_link_down(struct tipc_node *n_ptr, struct tipc_link *l_ptr) { struct tipc_link **active; + u32 addr = n_ptr->addr; n_ptr->working_links--; + tipc_nametbl_withdraw(TIPC_LINK_STATE, addr, l_ptr->bearer_id, addr); if (!tipc_link_is_active(l_ptr)) { pr_info("Lost standby link <%s> on network plane %c\n", - l_ptr->name, l_ptr->b_ptr->net_plane); + l_ptr->name, l_ptr->net_plane); return; } pr_info("Lost link <%s> on network plane %c\n", - l_ptr->name, l_ptr->b_ptr->net_plane); + l_ptr->name, l_ptr->net_plane); active = &n_ptr->active_links[0]; if (active[0] == l_ptr) @@ -225,9 +305,22 @@ void tipc_node_link_down(struct tipc_node *n_ptr, struct tipc_link *l_ptr) if (active[0] == l_ptr) node_select_active_links(n_ptr); if (tipc_node_is_up(n_ptr)) - tipc_link_changeover(l_ptr); + tipc_link_failover_send_queue(l_ptr); else node_lost_contact(n_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; + return; + } + + /* Loopback link went down? No fragmentation needed from now on. */ + if (n_ptr->addr == tipc_own_addr) { + n_ptr->act_mtus[0] = MAX_MSG_SIZE; + n_ptr->act_mtus[1] = MAX_MSG_SIZE; + } } int tipc_node_active_links(struct tipc_node *n_ptr) @@ -235,11 +328,6 @@ int tipc_node_active_links(struct tipc_node *n_ptr) return n_ptr->active_links[0] != NULL; } -int tipc_node_redundant_links(struct tipc_node *n_ptr) -{ - return n_ptr->working_links > 1; -} - int tipc_node_is_up(struct tipc_node *n_ptr) { return tipc_node_active_links(n_ptr); @@ -247,40 +335,36 @@ int tipc_node_is_up(struct tipc_node *n_ptr) void tipc_node_attach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr) { - n_ptr->links[l_ptr->b_ptr->identity] = l_ptr; - atomic_inc(&tipc_num_links); + 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++; } void tipc_node_detach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr) { - n_ptr->links[l_ptr->b_ptr->identity] = NULL; - atomic_dec(&tipc_num_links); - n_ptr->link_cnt--; + int i; + + for (i = 0; i < MAX_BEARERS; i++) { + 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--; + } } static void node_established_contact(struct tipc_node *n_ptr) { - tipc_k_signal((Handler)tipc_named_node_up, n_ptr->addr); + n_ptr->action_flags |= TIPC_NOTIFY_NODE_UP; n_ptr->bclink.oos_state = 0; n_ptr->bclink.acked = tipc_bclink_get_last_sent(); tipc_bclink_add_node(n_ptr->addr); } -static void node_name_purge_complete(unsigned long node_addr) -{ - struct tipc_node *n_ptr; - - read_lock_bh(&tipc_net_lock); - n_ptr = tipc_node_find(node_addr); - if (n_ptr) { - tipc_node_lock(n_ptr); - n_ptr->block_setup &= ~WAIT_NAMES_GONE; - tipc_node_unlock(n_ptr); - } - read_unlock_bh(&tipc_net_lock); -} - static void node_lost_contact(struct tipc_node *n_ptr) { char addr_string[16]; @@ -291,17 +375,12 @@ static void node_lost_contact(struct tipc_node *n_ptr) /* Flush broadcast link info associated with lost node */ if (n_ptr->bclink.recv_permitted) { - while (n_ptr->bclink.deferred_head) { - struct sk_buff *buf = n_ptr->bclink.deferred_head; - n_ptr->bclink.deferred_head = buf->next; - kfree_skb(buf); - } + kfree_skb_list(n_ptr->bclink.deferred_head); n_ptr->bclink.deferred_size = 0; - if (n_ptr->bclink.reasm_head) { - kfree_skb(n_ptr->bclink.reasm_head); - n_ptr->bclink.reasm_head = NULL; - n_ptr->bclink.reasm_tail = NULL; + if (n_ptr->bclink.reasm_buf) { + kfree_skb(n_ptr->bclink.reasm_buf); + n_ptr->bclink.reasm_buf = NULL; } tipc_bclink_remove_node(n_ptr->addr); @@ -320,12 +399,13 @@ static void node_lost_contact(struct tipc_node *n_ptr) tipc_link_reset_fragments(l_ptr); } - /* Notify subscribers */ - tipc_nodesub_notify(n_ptr); + n_ptr->action_flags &= ~TIPC_WAIT_OWN_LINKS_DOWN; - /* Prevent re-contact with node until cleanup is done */ - n_ptr->block_setup = WAIT_PEER_DOWN | WAIT_NAMES_GONE; - tipc_k_signal((Handler)node_name_purge_complete, n_ptr->addr); + /* Notify subscribers and prevent re-contact with node until + * cleanup is done. + */ + n_ptr->action_flags |= TIPC_WAIT_PEER_LINKS_DOWN | + TIPC_NOTIFY_NODE_DOWN; } struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space) @@ -344,27 +424,28 @@ struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space) return tipc_cfg_reply_error_string(TIPC_CFG_INVALID_VALUE " (network address)"); - read_lock_bh(&tipc_net_lock); + spin_lock_bh(&node_list_lock); if (!tipc_num_nodes) { - read_unlock_bh(&tipc_net_lock); + spin_unlock_bh(&node_list_lock); return tipc_cfg_reply_none(); } /* For now, get space for all other nodes */ payload_size = TLV_SPACE(sizeof(node_info)) * tipc_num_nodes; if (payload_size > 32768u) { - read_unlock_bh(&tipc_net_lock); + 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) { - read_unlock_bh(&tipc_net_lock); + if (!buf) return NULL; - } /* Add TLVs for all nodes in scope */ - list_for_each_entry(n_ptr, &tipc_node_list, list) { + 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); @@ -372,8 +453,7 @@ struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space) tipc_cfg_append_tlv(buf, TIPC_TLV_NODE_INFO, &node_info, sizeof(node_info)); } - - read_unlock_bh(&tipc_net_lock); + rcu_read_unlock(); return buf; } @@ -396,21 +476,19 @@ struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) if (!tipc_own_addr) return tipc_cfg_reply_none(); - read_lock_bh(&tipc_net_lock); - + spin_lock_bh(&node_list_lock); /* Get space for all unicast links + broadcast link */ - payload_size = TLV_SPACE(sizeof(link_info)) * - (atomic_read(&tipc_num_links) + 1); + payload_size = TLV_SPACE((sizeof(link_info)) * (tipc_num_links + 1)); if (payload_size > 32768u) { - read_unlock_bh(&tipc_net_lock); + 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) { - read_unlock_bh(&tipc_net_lock); + if (!buf) return NULL; - } /* Add TLV for broadcast link */ link_info.dest = htonl(tipc_cluster_mask(tipc_own_addr)); @@ -419,7 +497,8 @@ struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) tipc_cfg_append_tlv(buf, TIPC_TLV_LINK_INFO, &link_info, sizeof(link_info)); /* Add TLVs for any other links in scope */ - list_for_each_entry(n_ptr, &tipc_node_list, list) { + rcu_read_lock(); + list_for_each_entry_rcu(n_ptr, &tipc_node_list, list) { u32 i; if (!tipc_in_scope(domain, n_ptr->addr)) @@ -436,7 +515,79 @@ struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space) } tipc_node_unlock(n_ptr); } - - read_unlock_bh(&tipc_net_lock); + rcu_read_unlock(); return buf; } + +/** + * tipc_node_get_linkname - get the name of a link + * + * @bearer_id: id of the bearer + * @node: peer node address + * @linkname: link name output buffer + * + * Returns 0 on success + */ +int tipc_node_get_linkname(u32 bearer_id, u32 addr, char *linkname, size_t len) +{ + struct tipc_link *link; + struct tipc_node *node = tipc_node_find(addr); + + if ((bearer_id >= MAX_BEARERS) || !node) + return -EINVAL; + tipc_node_lock(node); + link = node->links[bearer_id]; + if (link) { + strncpy(linkname, link->name, len); + tipc_node_unlock(node); + return 0; + } + tipc_node_unlock(node); + return -EINVAL; +} + +void tipc_node_unlock(struct tipc_node *node) +{ + LIST_HEAD(nsub_list); + LIST_HEAD(conn_sks); + struct sk_buff_head waiting_sks; + u32 addr = 0; + unsigned int flags = node->action_flags; + + if (likely(!node->action_flags)) { + spin_unlock_bh(&node->lock); + return; + } + + __skb_queue_head_init(&waiting_sks); + if (node->action_flags & TIPC_WAKEUP_USERS) { + skb_queue_splice_init(&node->waiting_sks, &waiting_sks); + node->action_flags &= ~TIPC_WAKEUP_USERS; + } + if (node->action_flags & TIPC_NOTIFY_NODE_DOWN) { + list_replace_init(&node->nsub, &nsub_list); + list_replace_init(&node->conn_sks, &conn_sks); + node->action_flags &= ~TIPC_NOTIFY_NODE_DOWN; + } + if (node->action_flags & TIPC_NOTIFY_NODE_UP) { + node->action_flags &= ~TIPC_NOTIFY_NODE_UP; + addr = node->addr; + } + node->action_flags &= ~TIPC_WAKEUP_BCAST_USERS; + 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_nodesub_notify(&nsub_list); + + if (flags & TIPC_WAKEUP_BCAST_USERS) + tipc_bclink_wakeup_users(); + + if (addr) + tipc_named_node_up(addr); +} diff --git a/net/tipc/node.h b/net/tipc/node.h index e5e96c04e167..67513c3c852c 100644 --- a/net/tipc/node.h +++ b/net/tipc/node.h @@ -2,7 +2,7 @@ * net/tipc/node.h: Include file for TIPC node management routines * * Copyright (c) 2000-2006, Ericsson AB - * Copyright (c) 2005, 2010-2011, Wind River Systems + * Copyright (c) 2005, 2010-2014, Wind River Systems * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -41,93 +41,129 @@ #include "addr.h" #include "net.h" #include "bearer.h" +#include "msg.h" /* * Out-of-range value for node signature */ #define INVALID_NODE_SIG 0x10000 -/* Flags used to block (re)establishment of contact with a neighboring node */ -#define WAIT_PEER_DOWN 0x0001 /* wait to see that peer's links are down */ -#define WAIT_NAMES_GONE 0x0002 /* wait for peer's publications to be purged */ -#define WAIT_NODE_DOWN 0x0004 /* wait until peer node is declared down */ +/* Flags used to take different actions according to flag type + * TIPC_WAIT_PEER_LINKS_DOWN: wait to see that peer's links are down + * TIPC_WAIT_OWN_LINKS_DOWN: wait until peer node is declared down + * TIPC_NOTIFY_NODE_DOWN: notify node is down + * TIPC_NOTIFY_NODE_UP: notify node is up + */ +enum { + TIPC_WAIT_PEER_LINKS_DOWN = (1 << 1), + TIPC_WAIT_OWN_LINKS_DOWN = (1 << 2), + TIPC_NOTIFY_NODE_DOWN = (1 << 3), + TIPC_NOTIFY_NODE_UP = (1 << 4), + TIPC_WAKEUP_USERS = (1 << 5), + TIPC_WAKEUP_BCAST_USERS = (1 << 6) +}; + +/** + * struct tipc_node_bclink - TIPC node bclink structure + * @acked: sequence # of last outbound b'cast message acknowledged by node + * @last_in: sequence # of last in-sequence b'cast message received from node + * @last_sent: sequence # of last b'cast message sent by node + * @oos_state: state tracker for handling OOS b'cast messages + * @deferred_size: number of OOS b'cast messages in deferred queue + * @deferred_head: oldest OOS b'cast message received from node + * @deferred_tail: newest OOS b'cast message received from node + * @reasm_buf: broadcast reassembly queue head from node + * @recv_permitted: true if node is allowed to receive b'cast messages + */ +struct tipc_node_bclink { + u32 acked; + u32 last_in; + u32 last_sent; + u32 oos_state; + u32 deferred_size; + struct sk_buff *deferred_head; + struct sk_buff *deferred_tail; + struct sk_buff *reasm_buf; + bool recv_permitted; +}; /** * struct tipc_node - TIPC node structure * @addr: network address of node * @lock: spinlock governing access to structure * @hash: links to adjacent nodes in unsorted hash chain - * @list: links to adjacent nodes in sorted list of cluster's nodes - * @nsub: list of "node down" subscriptions monitoring node * @active_links: pointers to active links to node * @links: pointers to all links to node + * @action_flags: bit mask of different types of node actions + * @bclink: broadcast-related info + * @list: links to adjacent nodes in sorted list of cluster's nodes * @working_links: number of working links to node (both active and standby) - * @block_setup: bit mask of conditions preventing link establishment to node * @link_cnt: number of links to node - * @permit_changeover: non-zero if node has redundant links to this system * @signature: node instance identifier - * @bclink: broadcast-related info - * @acked: sequence # of last outbound b'cast message acknowledged by node - * @last_in: sequence # of last in-sequence b'cast message received from node - * @last_sent: sequence # of last b'cast message sent by node - * @oos_state: state tracker for handling OOS b'cast messages - * @deferred_size: number of OOS b'cast messages in deferred queue - * @deferred_head: oldest OOS b'cast message received from node - * @deferred_tail: newest OOS b'cast message received from node - * @reasm_head: broadcast reassembly queue head from node - * @reasm_tail: last broadcast fragment received from node - * @recv_permitted: true if node is allowed to receive b'cast messages + * @nsub: list of "node down" subscriptions monitoring node + * @rcu: rcu struct for tipc_node */ struct tipc_node { u32 addr; spinlock_t lock; struct hlist_node hash; - struct list_head list; - struct list_head nsub; struct tipc_link *active_links[2]; + u32 act_mtus[2]; struct tipc_link *links[MAX_BEARERS]; + unsigned int action_flags; + struct tipc_node_bclink bclink; + struct list_head list; int link_cnt; int working_links; - int block_setup; - int permit_changeover; u32 signature; - struct { - u32 acked; - u32 last_in; - u32 last_sent; - u32 oos_state; - u32 deferred_size; - struct sk_buff *deferred_head; - struct sk_buff *deferred_tail; - struct sk_buff *reasm_head; - struct sk_buff *reasm_tail; - bool recv_permitted; - } bclink; + struct list_head nsub; + 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_delete(struct tipc_node *n_ptr); +void tipc_node_stop(void); 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_redundant_links(struct tipc_node *n_ptr); int tipc_node_is_up(struct tipc_node *n_ptr); struct sk_buff *tipc_node_get_links(const void *req_tlv_area, int req_tlv_space); struct sk_buff *tipc_node_get_nodes(const void *req_tlv_area, int req_tlv_space); +int tipc_node_get_linkname(u32 bearer_id, u32 node, char *linkname, size_t len); +void tipc_node_unlock(struct tipc_node *node); +int tipc_node_add_conn(u32 dnode, u32 port, u32 peer_port); +void tipc_node_remove_conn(u32 dnode, u32 port); -static inline void tipc_node_lock(struct tipc_node *n_ptr) +static inline void tipc_node_lock(struct tipc_node *node) { - spin_lock_bh(&n_ptr->lock); + spin_lock_bh(&node->lock); } -static inline void tipc_node_unlock(struct tipc_node *n_ptr) +static inline bool tipc_node_blocked(struct tipc_node *node) { - spin_unlock_bh(&n_ptr->lock); + return (node->action_flags & (TIPC_WAIT_PEER_LINKS_DOWN | + TIPC_NOTIFY_NODE_DOWN | TIPC_WAIT_OWN_LINKS_DOWN)); +} + +static inline uint tipc_node_get_mtu(u32 addr, u32 selector) +{ + struct tipc_node *node; + u32 mtu; + + node = tipc_node_find(addr); + + if (likely(node)) + mtu = node->act_mtus[selector & 1]; + else + mtu = MAX_MSG_SIZE; + + return mtu; } #endif diff --git a/net/tipc/node_subscr.c b/net/tipc/node_subscr.c index 8a7384c04add..2d13eea8574a 100644 --- a/net/tipc/node_subscr.c +++ b/net/tipc/node_subscr.c @@ -81,15 +81,16 @@ void tipc_nodesub_unsubscribe(struct tipc_node_subscr *node_sub) * * Note: node is locked by caller */ -void tipc_nodesub_notify(struct tipc_node *node) +void tipc_nodesub_notify(struct list_head *nsub_list) { - struct tipc_node_subscr *ns; + struct tipc_node_subscr *ns, *safe; + net_ev_handler handle_node_down; - list_for_each_entry(ns, &node->nsub, nodesub_list) { - if (ns->handle_node_down) { - tipc_k_signal((Handler)ns->handle_node_down, - (unsigned long)ns->usr_handle); + list_for_each_entry_safe(ns, safe, nsub_list, nodesub_list) { + handle_node_down = ns->handle_node_down; + if (handle_node_down) { ns->handle_node_down = NULL; + handle_node_down(ns->usr_handle); } } } diff --git a/net/tipc/node_subscr.h b/net/tipc/node_subscr.h index c95d20727ded..d91b8cc81e3d 100644 --- a/net/tipc/node_subscr.h +++ b/net/tipc/node_subscr.h @@ -58,6 +58,6 @@ struct tipc_node_subscr { void tipc_nodesub_subscribe(struct tipc_node_subscr *node_sub, u32 addr, void *usr_handle, net_ev_handler handle_down); void tipc_nodesub_unsubscribe(struct tipc_node_subscr *node_sub); -void tipc_nodesub_notify(struct tipc_node *node); +void tipc_nodesub_notify(struct list_head *nsub_list); #endif diff --git a/net/tipc/port.c b/net/tipc/port.c deleted file mode 100644 index c081a7632302..000000000000 --- a/net/tipc/port.c +++ /dev/null @@ -1,1054 +0,0 @@ -/* - * net/tipc/port.c: TIPC port code - * - * Copyright (c) 1992-2007, Ericsson AB - * Copyright (c) 2004-2008, 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 "config.h" -#include "port.h" -#include "name_table.h" - -/* Connection management: */ -#define PROBING_INTERVAL 3600000 /* [ms] => 1 h */ -#define CONFIRMED 0 -#define PROBING 1 - -#define MAX_REJECT_SIZE 1024 - -DEFINE_SPINLOCK(tipc_port_list_lock); - -static LIST_HEAD(ports); -static void port_handle_node_down(unsigned long ref); -static struct sk_buff *port_build_self_abort_msg(struct tipc_port *, u32 err); -static struct sk_buff *port_build_peer_abort_msg(struct tipc_port *, u32 err); -static void port_timeout(unsigned long ref); - - -static u32 port_peernode(struct tipc_port *p_ptr) -{ - return msg_destnode(&p_ptr->phdr); -} - -static u32 port_peerport(struct tipc_port *p_ptr) -{ - return msg_destport(&p_ptr->phdr); -} - -/** - * tipc_port_peer_msg - verify message was sent by connected port's peer - * - * Handles cases where the node's network address has changed from - * the default of <0.0.0> to its configured setting. - */ -int tipc_port_peer_msg(struct tipc_port *p_ptr, struct tipc_msg *msg) -{ - u32 peernode; - u32 orignode; - - if (msg_origport(msg) != port_peerport(p_ptr)) - return 0; - - orignode = msg_orignode(msg); - peernode = port_peernode(p_ptr); - return (orignode == peernode) || - (!orignode && (peernode == tipc_own_addr)) || - (!peernode && (orignode == tipc_own_addr)); -} - -/** - * tipc_multicast - send a multicast message to local and remote destinations - */ -int tipc_multicast(u32 ref, struct tipc_name_seq const *seq, - struct iovec const *msg_sect, unsigned int len) -{ - struct tipc_msg *hdr; - struct sk_buff *buf; - struct sk_buff *ibuf = NULL; - struct tipc_port_list dports = {0, NULL, }; - struct tipc_port *oport = tipc_port_deref(ref); - int ext_targets; - int res; - - if (unlikely(!oport)) - return -EINVAL; - - /* Create multicast message */ - hdr = &oport->phdr; - msg_set_type(hdr, TIPC_MCAST_MSG); - msg_set_lookup_scope(hdr, TIPC_CLUSTER_SCOPE); - msg_set_destport(hdr, 0); - msg_set_destnode(hdr, 0); - msg_set_nametype(hdr, seq->type); - msg_set_namelower(hdr, seq->lower); - msg_set_nameupper(hdr, seq->upper); - msg_set_hdr_sz(hdr, MCAST_H_SIZE); - res = tipc_msg_build(hdr, msg_sect, len, MAX_MSG_SIZE, &buf); - if (unlikely(!buf)) - return res; - - /* Figure out where to send multicast message */ - ext_targets = tipc_nametbl_mc_translate(seq->type, seq->lower, seq->upper, - TIPC_NODE_SCOPE, &dports); - - /* Send message to destinations (duplicate it only if necessary) */ - if (ext_targets) { - if (dports.count != 0) { - ibuf = skb_copy(buf, GFP_ATOMIC); - if (ibuf == NULL) { - tipc_port_list_free(&dports); - kfree_skb(buf); - return -ENOMEM; - } - } - res = tipc_bclink_send_msg(buf); - if ((res < 0) && (dports.count != 0)) - kfree_skb(ibuf); - } else { - ibuf = buf; - } - - if (res >= 0) { - if (ibuf) - tipc_port_recv_mcast(ibuf, &dports); - } else { - tipc_port_list_free(&dports); - } - return res; -} - -/** - * tipc_port_recv_mcast - deliver multicast message to all destination ports - * - * If there is no port list, perform a lookup to create one - */ -void tipc_port_recv_mcast(struct sk_buff *buf, struct tipc_port_list *dp) -{ - struct tipc_msg *msg; - struct tipc_port_list dports = {0, NULL, }; - struct tipc_port_list *item = dp; - int cnt = 0; - - msg = buf_msg(buf); - - /* Create destination port list, if one wasn't supplied */ - if (dp == NULL) { - tipc_nametbl_mc_translate(msg_nametype(msg), - msg_namelower(msg), - msg_nameupper(msg), - TIPC_CLUSTER_SCOPE, - &dports); - item = dp = &dports; - } - - /* Deliver a copy of message to each destination port */ - if (dp->count != 0) { - msg_set_destnode(msg, tipc_own_addr); - if (dp->count == 1) { - msg_set_destport(msg, dp->ports[0]); - tipc_port_recv_msg(buf); - tipc_port_list_free(dp); - return; - } - for (; cnt < dp->count; cnt++) { - int index = cnt % PLSIZE; - struct sk_buff *b = skb_clone(buf, GFP_ATOMIC); - - if (b == NULL) { - pr_warn("Unable to deliver multicast message(s)\n"); - goto exit; - } - if ((index == 0) && (cnt != 0)) - item = item->next; - msg_set_destport(buf_msg(b), item->ports[index]); - tipc_port_recv_msg(b); - } - } -exit: - kfree_skb(buf); - tipc_port_list_free(dp); -} - -/** - * tipc_createport - create a generic TIPC port - * - * Returns pointer to (locked) TIPC port, or NULL if unable to create it - */ -struct tipc_port *tipc_createport(struct sock *sk, - u32 (*dispatcher)(struct tipc_port *, - struct sk_buff *), - void (*wakeup)(struct tipc_port *), - const u32 importance) -{ - struct tipc_port *p_ptr; - struct tipc_msg *msg; - u32 ref; - - p_ptr = kzalloc(sizeof(*p_ptr), GFP_ATOMIC); - if (!p_ptr) { - pr_warn("Port creation failed, no memory\n"); - return NULL; - } - ref = tipc_ref_acquire(p_ptr, &p_ptr->lock); - if (!ref) { - pr_warn("Port creation failed, ref. table exhausted\n"); - kfree(p_ptr); - return NULL; - } - - p_ptr->sk = sk; - p_ptr->max_pkt = MAX_PKT_DEFAULT; - p_ptr->ref = ref; - INIT_LIST_HEAD(&p_ptr->wait_list); - INIT_LIST_HEAD(&p_ptr->subscription.nodesub_list); - p_ptr->dispatcher = dispatcher; - p_ptr->wakeup = wakeup; - k_init_timer(&p_ptr->timer, (Handler)port_timeout, ref); - INIT_LIST_HEAD(&p_ptr->publications); - INIT_LIST_HEAD(&p_ptr->port_list); - - /* - * Must hold port list lock while initializing message header template - * to ensure a change to node's own network address doesn't result - * in template containing out-dated network address information - */ - spin_lock_bh(&tipc_port_list_lock); - msg = &p_ptr->phdr; - tipc_msg_init(msg, importance, TIPC_NAMED_MSG, NAMED_H_SIZE, 0); - msg_set_origport(msg, ref); - list_add_tail(&p_ptr->port_list, &ports); - spin_unlock_bh(&tipc_port_list_lock); - return p_ptr; -} - -int tipc_deleteport(u32 ref) -{ - struct tipc_port *p_ptr; - struct sk_buff *buf = NULL; - - tipc_withdraw(ref, 0, NULL); - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - - tipc_ref_discard(ref); - tipc_port_unlock(p_ptr); - - k_cancel_timer(&p_ptr->timer); - if (p_ptr->connected) { - buf = port_build_peer_abort_msg(p_ptr, TIPC_ERR_NO_PORT); - tipc_nodesub_unsubscribe(&p_ptr->subscription); - } - - spin_lock_bh(&tipc_port_list_lock); - list_del(&p_ptr->port_list); - list_del(&p_ptr->wait_list); - spin_unlock_bh(&tipc_port_list_lock); - k_term_timer(&p_ptr->timer); - kfree(p_ptr); - tipc_net_route_msg(buf); - return 0; -} - -static int port_unreliable(struct tipc_port *p_ptr) -{ - return msg_src_droppable(&p_ptr->phdr); -} - -int tipc_portunreliable(u32 ref, unsigned int *isunreliable) -{ - struct tipc_port *p_ptr; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - *isunreliable = port_unreliable(p_ptr); - tipc_port_unlock(p_ptr); - return 0; -} - -int tipc_set_portunreliable(u32 ref, unsigned int isunreliable) -{ - struct tipc_port *p_ptr; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - msg_set_src_droppable(&p_ptr->phdr, (isunreliable != 0)); - tipc_port_unlock(p_ptr); - return 0; -} - -static int port_unreturnable(struct tipc_port *p_ptr) -{ - return msg_dest_droppable(&p_ptr->phdr); -} - -int tipc_portunreturnable(u32 ref, unsigned int *isunrejectable) -{ - struct tipc_port *p_ptr; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - *isunrejectable = port_unreturnable(p_ptr); - tipc_port_unlock(p_ptr); - return 0; -} - -int tipc_set_portunreturnable(u32 ref, unsigned int isunrejectable) -{ - struct tipc_port *p_ptr; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - msg_set_dest_droppable(&p_ptr->phdr, (isunrejectable != 0)); - tipc_port_unlock(p_ptr); - return 0; -} - -/* - * port_build_proto_msg(): create connection protocol message for port - * - * On entry the port must be locked and connected. - */ -static struct sk_buff *port_build_proto_msg(struct tipc_port *p_ptr, - u32 type, u32 ack) -{ - struct sk_buff *buf; - struct tipc_msg *msg; - - buf = tipc_buf_acquire(INT_H_SIZE); - if (buf) { - msg = buf_msg(buf); - tipc_msg_init(msg, CONN_MANAGER, type, INT_H_SIZE, - port_peernode(p_ptr)); - msg_set_destport(msg, port_peerport(p_ptr)); - msg_set_origport(msg, p_ptr->ref); - msg_set_msgcnt(msg, ack); - } - return buf; -} - -int tipc_reject_msg(struct sk_buff *buf, u32 err) -{ - struct tipc_msg *msg = buf_msg(buf); - struct sk_buff *rbuf; - struct tipc_msg *rmsg; - int hdr_sz; - u32 imp; - u32 data_sz = msg_data_sz(msg); - u32 src_node; - u32 rmsg_sz; - - /* discard rejected message if it shouldn't be returned to sender */ - if (WARN(!msg_isdata(msg), - "attempt to reject message with user=%u", msg_user(msg))) { - dump_stack(); - goto exit; - } - if (msg_errcode(msg) || msg_dest_droppable(msg)) - goto exit; - - /* - * construct returned message by copying rejected message header and - * data (or subset), then updating header fields that need adjusting - */ - hdr_sz = msg_hdr_sz(msg); - rmsg_sz = hdr_sz + min_t(u32, data_sz, MAX_REJECT_SIZE); - - rbuf = tipc_buf_acquire(rmsg_sz); - if (rbuf == NULL) - goto exit; - - rmsg = buf_msg(rbuf); - skb_copy_to_linear_data(rbuf, msg, rmsg_sz); - - if (msg_connected(rmsg)) { - imp = msg_importance(rmsg); - if (imp < TIPC_CRITICAL_IMPORTANCE) - msg_set_importance(rmsg, ++imp); - } - msg_set_non_seq(rmsg, 0); - msg_set_size(rmsg, rmsg_sz); - msg_set_errcode(rmsg, err); - msg_set_prevnode(rmsg, tipc_own_addr); - msg_swap_words(rmsg, 4, 5); - if (!msg_short(rmsg)) - msg_swap_words(rmsg, 6, 7); - - /* send self-abort message when rejecting on a connected port */ - if (msg_connected(msg)) { - struct tipc_port *p_ptr = tipc_port_lock(msg_destport(msg)); - - if (p_ptr) { - struct sk_buff *abuf = NULL; - - if (p_ptr->connected) - abuf = port_build_self_abort_msg(p_ptr, err); - tipc_port_unlock(p_ptr); - tipc_net_route_msg(abuf); - } - } - - /* send returned message & dispose of rejected message */ - src_node = msg_prevnode(msg); - if (in_own_node(src_node)) - tipc_port_recv_msg(rbuf); - else - tipc_link_send(rbuf, src_node, msg_link_selector(rmsg)); -exit: - kfree_skb(buf); - return data_sz; -} - -int tipc_port_reject_sections(struct tipc_port *p_ptr, struct tipc_msg *hdr, - struct iovec const *msg_sect, unsigned int len, - int err) -{ - struct sk_buff *buf; - int res; - - res = tipc_msg_build(hdr, msg_sect, len, MAX_MSG_SIZE, &buf); - if (!buf) - return res; - - return tipc_reject_msg(buf, err); -} - -static void port_timeout(unsigned long ref) -{ - struct tipc_port *p_ptr = tipc_port_lock(ref); - struct sk_buff *buf = NULL; - - if (!p_ptr) - return; - - if (!p_ptr->connected) { - tipc_port_unlock(p_ptr); - return; - } - - /* Last probe answered ? */ - if (p_ptr->probing_state == PROBING) { - buf = port_build_self_abort_msg(p_ptr, TIPC_ERR_NO_PORT); - } else { - buf = port_build_proto_msg(p_ptr, CONN_PROBE, 0); - p_ptr->probing_state = PROBING; - k_start_timer(&p_ptr->timer, p_ptr->probing_interval); - } - tipc_port_unlock(p_ptr); - tipc_net_route_msg(buf); -} - - -static void port_handle_node_down(unsigned long ref) -{ - struct tipc_port *p_ptr = tipc_port_lock(ref); - struct sk_buff *buf = NULL; - - if (!p_ptr) - return; - buf = port_build_self_abort_msg(p_ptr, TIPC_ERR_NO_NODE); - tipc_port_unlock(p_ptr); - tipc_net_route_msg(buf); -} - - -static struct sk_buff *port_build_self_abort_msg(struct tipc_port *p_ptr, u32 err) -{ - struct sk_buff *buf = port_build_peer_abort_msg(p_ptr, err); - - if (buf) { - struct tipc_msg *msg = buf_msg(buf); - msg_swap_words(msg, 4, 5); - msg_swap_words(msg, 6, 7); - } - return buf; -} - - -static struct sk_buff *port_build_peer_abort_msg(struct tipc_port *p_ptr, u32 err) -{ - struct sk_buff *buf; - struct tipc_msg *msg; - u32 imp; - - if (!p_ptr->connected) - return NULL; - - buf = tipc_buf_acquire(BASIC_H_SIZE); - if (buf) { - msg = buf_msg(buf); - memcpy(msg, &p_ptr->phdr, BASIC_H_SIZE); - msg_set_hdr_sz(msg, BASIC_H_SIZE); - msg_set_size(msg, BASIC_H_SIZE); - imp = msg_importance(msg); - if (imp < TIPC_CRITICAL_IMPORTANCE) - msg_set_importance(msg, ++imp); - msg_set_errcode(msg, err); - } - return buf; -} - -void tipc_port_recv_proto_msg(struct sk_buff *buf) -{ - struct tipc_msg *msg = buf_msg(buf); - struct tipc_port *p_ptr; - struct sk_buff *r_buf = NULL; - u32 destport = msg_destport(msg); - int wakeable; - - /* Validate connection */ - p_ptr = tipc_port_lock(destport); - if (!p_ptr || !p_ptr->connected || !tipc_port_peer_msg(p_ptr, msg)) { - r_buf = tipc_buf_acquire(BASIC_H_SIZE); - if (r_buf) { - msg = buf_msg(r_buf); - tipc_msg_init(msg, TIPC_HIGH_IMPORTANCE, TIPC_CONN_MSG, - BASIC_H_SIZE, msg_orignode(msg)); - msg_set_errcode(msg, TIPC_ERR_NO_PORT); - msg_set_origport(msg, destport); - msg_set_destport(msg, msg_origport(msg)); - } - if (p_ptr) - tipc_port_unlock(p_ptr); - goto exit; - } - - /* Process protocol message sent by peer */ - switch (msg_type(msg)) { - case CONN_ACK: - wakeable = tipc_port_congested(p_ptr) && p_ptr->congested && - p_ptr->wakeup; - p_ptr->acked += msg_msgcnt(msg); - if (!tipc_port_congested(p_ptr)) { - p_ptr->congested = 0; - if (wakeable) - p_ptr->wakeup(p_ptr); - } - break; - case CONN_PROBE: - r_buf = port_build_proto_msg(p_ptr, CONN_PROBE_REPLY, 0); - break; - default: - /* CONN_PROBE_REPLY or unrecognized - no action required */ - break; - } - p_ptr->probing_state = CONFIRMED; - tipc_port_unlock(p_ptr); -exit: - tipc_net_route_msg(r_buf); - kfree_skb(buf); -} - -static int port_print(struct tipc_port *p_ptr, 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), p_ptr->ref); - else - ret = tipc_snprintf(buf, len, "%-10u:", p_ptr->ref); - - if (p_ptr->connected) { - u32 dport = port_peerport(p_ptr); - u32 destnode = port_peernode(p_ptr); - - ret += tipc_snprintf(buf + ret, len - ret, - " connected to <%u.%u.%u:%u>", - tipc_zone(destnode), - tipc_cluster(destnode), - tipc_node(destnode), dport); - if (p_ptr->conn_type != 0) - ret += tipc_snprintf(buf + ret, len - ret, - " via {%u,%u}", p_ptr->conn_type, - p_ptr->conn_instance); - } else if (p_ptr->published) { - ret += tipc_snprintf(buf + ret, len - ret, " bound to"); - list_for_each_entry(publ, &p_ptr->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_port_get_ports(void) -{ - struct sk_buff *buf; - struct tlv_desc *rep_tlv; - char *pb; - int pb_len; - struct tipc_port *p_ptr; - int str_len = 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; - - spin_lock_bh(&tipc_port_list_lock); - list_for_each_entry(p_ptr, &ports, port_list) { - spin_lock_bh(p_ptr->lock); - str_len += port_print(p_ptr, pb, pb_len, 0); - spin_unlock_bh(p_ptr->lock); - } - spin_unlock_bh(&tipc_port_list_lock); - 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; -} - -void tipc_port_reinit(void) -{ - struct tipc_port *p_ptr; - struct tipc_msg *msg; - - spin_lock_bh(&tipc_port_list_lock); - list_for_each_entry(p_ptr, &ports, port_list) { - msg = &p_ptr->phdr; - msg_set_prevnode(msg, tipc_own_addr); - msg_set_orignode(msg, tipc_own_addr); - } - spin_unlock_bh(&tipc_port_list_lock); -} - -void tipc_acknowledge(u32 ref, u32 ack) -{ - struct tipc_port *p_ptr; - struct sk_buff *buf = NULL; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return; - if (p_ptr->connected) { - p_ptr->conn_unacked -= ack; - buf = port_build_proto_msg(p_ptr, CONN_ACK, ack); - } - tipc_port_unlock(p_ptr); - tipc_net_route_msg(buf); -} - -int tipc_portimportance(u32 ref, unsigned int *importance) -{ - struct tipc_port *p_ptr; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - *importance = (unsigned int)msg_importance(&p_ptr->phdr); - tipc_port_unlock(p_ptr); - return 0; -} - -int tipc_set_portimportance(u32 ref, unsigned int imp) -{ - struct tipc_port *p_ptr; - - if (imp > TIPC_CRITICAL_IMPORTANCE) - return -EINVAL; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - msg_set_importance(&p_ptr->phdr, (u32)imp); - tipc_port_unlock(p_ptr); - return 0; -} - - -int tipc_publish(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) -{ - struct tipc_port *p_ptr; - struct publication *publ; - u32 key; - int res = -EINVAL; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - - if (p_ptr->connected) - goto exit; - key = ref + p_ptr->pub_count + 1; - if (key == ref) { - res = -EADDRINUSE; - goto exit; - } - publ = tipc_nametbl_publish(seq->type, seq->lower, seq->upper, - scope, p_ptr->ref, key); - if (publ) { - list_add(&publ->pport_list, &p_ptr->publications); - p_ptr->pub_count++; - p_ptr->published = 1; - res = 0; - } -exit: - tipc_port_unlock(p_ptr); - return res; -} - -int tipc_withdraw(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) -{ - struct tipc_port *p_ptr; - struct publication *publ; - struct publication *tpubl; - int res = -EINVAL; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - if (!seq) { - list_for_each_entry_safe(publ, tpubl, - &p_ptr->publications, pport_list) { - tipc_nametbl_withdraw(publ->type, publ->lower, - publ->ref, publ->key); - } - res = 0; - } else { - list_for_each_entry_safe(publ, tpubl, - &p_ptr->publications, pport_list) { - if (publ->scope != scope) - continue; - if (publ->type != seq->type) - continue; - if (publ->lower != seq->lower) - continue; - if (publ->upper != seq->upper) - break; - tipc_nametbl_withdraw(publ->type, publ->lower, - publ->ref, publ->key); - res = 0; - break; - } - } - if (list_empty(&p_ptr->publications)) - p_ptr->published = 0; - tipc_port_unlock(p_ptr); - return res; -} - -int tipc_connect(u32 ref, struct tipc_portid const *peer) -{ - struct tipc_port *p_ptr; - int res; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - res = __tipc_connect(ref, p_ptr, peer); - tipc_port_unlock(p_ptr); - return res; -} - -/* - * __tipc_connect - connect to a remote peer - * - * Port must be locked. - */ -int __tipc_connect(u32 ref, struct tipc_port *p_ptr, - struct tipc_portid const *peer) -{ - struct tipc_msg *msg; - int res = -EINVAL; - - if (p_ptr->published || p_ptr->connected) - goto exit; - if (!peer->ref) - goto exit; - - msg = &p_ptr->phdr; - msg_set_destnode(msg, peer->node); - msg_set_destport(msg, peer->ref); - msg_set_type(msg, TIPC_CONN_MSG); - msg_set_lookup_scope(msg, 0); - msg_set_hdr_sz(msg, SHORT_H_SIZE); - - p_ptr->probing_interval = PROBING_INTERVAL; - p_ptr->probing_state = CONFIRMED; - p_ptr->connected = 1; - k_start_timer(&p_ptr->timer, p_ptr->probing_interval); - - tipc_nodesub_subscribe(&p_ptr->subscription, peer->node, - (void *)(unsigned long)ref, - (net_ev_handler)port_handle_node_down); - res = 0; -exit: - p_ptr->max_pkt = tipc_link_get_max_pkt(peer->node, ref); - return res; -} - -/* - * __tipc_disconnect - disconnect port from peer - * - * Port must be locked. - */ -int __tipc_disconnect(struct tipc_port *tp_ptr) -{ - int res; - - if (tp_ptr->connected) { - tp_ptr->connected = 0; - /* let timer expire on it's own to avoid deadlock! */ - tipc_nodesub_unsubscribe(&tp_ptr->subscription); - res = 0; - } else { - res = -ENOTCONN; - } - return res; -} - -/* - * tipc_disconnect(): Disconnect port form peer. - * This is a node local operation. - */ -int tipc_disconnect(u32 ref) -{ - struct tipc_port *p_ptr; - int res; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - res = __tipc_disconnect(p_ptr); - tipc_port_unlock(p_ptr); - return res; -} - -/* - * tipc_shutdown(): Send a SHUTDOWN msg to peer and disconnect - */ -int tipc_shutdown(u32 ref) -{ - struct tipc_port *p_ptr; - struct sk_buff *buf = NULL; - - p_ptr = tipc_port_lock(ref); - if (!p_ptr) - return -EINVAL; - - buf = port_build_peer_abort_msg(p_ptr, TIPC_CONN_SHUTDOWN); - tipc_port_unlock(p_ptr); - tipc_net_route_msg(buf); - return tipc_disconnect(ref); -} - -/** - * tipc_port_recv_msg - receive message from lower layer and deliver to port user - */ -int tipc_port_recv_msg(struct sk_buff *buf) -{ - struct tipc_port *p_ptr; - struct tipc_msg *msg = buf_msg(buf); - u32 destport = msg_destport(msg); - u32 dsz = msg_data_sz(msg); - u32 err; - - /* forward unresolved named message */ - if (unlikely(!destport)) { - tipc_net_route_msg(buf); - return dsz; - } - - /* validate destination & pass to port, otherwise reject message */ - p_ptr = tipc_port_lock(destport); - if (likely(p_ptr)) { - err = p_ptr->dispatcher(p_ptr, buf); - tipc_port_unlock(p_ptr); - if (likely(!err)) - return dsz; - } else { - err = TIPC_ERR_NO_PORT; - } - - return tipc_reject_msg(buf, err); -} - -/* - * tipc_port_recv_sections(): Concatenate and deliver sectioned - * message for this node. - */ -static int tipc_port_recv_sections(struct tipc_port *sender, - struct iovec const *msg_sect, - unsigned int len) -{ - struct sk_buff *buf; - int res; - - res = tipc_msg_build(&sender->phdr, msg_sect, len, MAX_MSG_SIZE, &buf); - if (likely(buf)) - tipc_port_recv_msg(buf); - return res; -} - -/** - * tipc_send - send message sections on connection - */ -int tipc_send(u32 ref, struct iovec const *msg_sect, unsigned int len) -{ - struct tipc_port *p_ptr; - u32 destnode; - int res; - - p_ptr = tipc_port_deref(ref); - if (!p_ptr || !p_ptr->connected) - return -EINVAL; - - p_ptr->congested = 1; - if (!tipc_port_congested(p_ptr)) { - destnode = port_peernode(p_ptr); - if (likely(!in_own_node(destnode))) - res = tipc_link_send_sections_fast(p_ptr, msg_sect, - len, destnode); - else - res = tipc_port_recv_sections(p_ptr, msg_sect, len); - - if (likely(res != -ELINKCONG)) { - p_ptr->congested = 0; - if (res > 0) - p_ptr->sent++; - return res; - } - } - if (port_unreliable(p_ptr)) { - p_ptr->congested = 0; - return len; - } - return -ELINKCONG; -} - -/** - * tipc_send2name - send message sections to port name - */ -int tipc_send2name(u32 ref, struct tipc_name const *name, unsigned int domain, - struct iovec const *msg_sect, unsigned int len) -{ - struct tipc_port *p_ptr; - struct tipc_msg *msg; - u32 destnode = domain; - u32 destport; - int res; - - p_ptr = tipc_port_deref(ref); - if (!p_ptr || p_ptr->connected) - return -EINVAL; - - msg = &p_ptr->phdr; - msg_set_type(msg, TIPC_NAMED_MSG); - msg_set_hdr_sz(msg, NAMED_H_SIZE); - msg_set_nametype(msg, name->type); - msg_set_nameinst(msg, name->instance); - msg_set_lookup_scope(msg, tipc_addr_scope(domain)); - destport = tipc_nametbl_translate(name->type, name->instance, &destnode); - msg_set_destnode(msg, destnode); - msg_set_destport(msg, destport); - - if (likely(destport || destnode)) { - if (likely(in_own_node(destnode))) - res = tipc_port_recv_sections(p_ptr, msg_sect, len); - else if (tipc_own_addr) - res = tipc_link_send_sections_fast(p_ptr, msg_sect, - len, destnode); - else - res = tipc_port_reject_sections(p_ptr, msg, msg_sect, - len, TIPC_ERR_NO_NODE); - if (likely(res != -ELINKCONG)) { - if (res > 0) - p_ptr->sent++; - return res; - } - if (port_unreliable(p_ptr)) { - return len; - } - return -ELINKCONG; - } - return tipc_port_reject_sections(p_ptr, msg, msg_sect, len, - TIPC_ERR_NO_NAME); -} - -/** - * tipc_send2port - send message sections to port identity - */ -int tipc_send2port(u32 ref, struct tipc_portid const *dest, - struct iovec const *msg_sect, unsigned int len) -{ - struct tipc_port *p_ptr; - struct tipc_msg *msg; - int res; - - p_ptr = tipc_port_deref(ref); - if (!p_ptr || p_ptr->connected) - return -EINVAL; - - msg = &p_ptr->phdr; - msg_set_type(msg, TIPC_DIRECT_MSG); - msg_set_lookup_scope(msg, 0); - msg_set_destnode(msg, dest->node); - msg_set_destport(msg, dest->ref); - msg_set_hdr_sz(msg, BASIC_H_SIZE); - - if (in_own_node(dest->node)) - res = tipc_port_recv_sections(p_ptr, msg_sect, len); - else if (tipc_own_addr) - res = tipc_link_send_sections_fast(p_ptr, msg_sect, len, - dest->node); - else - res = tipc_port_reject_sections(p_ptr, msg, msg_sect, len, - TIPC_ERR_NO_NODE); - if (likely(res != -ELINKCONG)) { - if (res > 0) - p_ptr->sent++; - return res; - } - if (port_unreliable(p_ptr)) { - return len; - } - return -ELINKCONG; -} diff --git a/net/tipc/port.h b/net/tipc/port.h deleted file mode 100644 index 912253597343..000000000000 --- a/net/tipc/port.h +++ /dev/null @@ -1,201 +0,0 @@ -/* - * net/tipc/port.h: Include file for TIPC port code - * - * Copyright (c) 1994-2007, 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. - */ - -#ifndef _TIPC_PORT_H -#define _TIPC_PORT_H - -#include "ref.h" -#include "net.h" -#include "msg.h" -#include "node_subscr.h" - -#define TIPC_FLOW_CONTROL_WIN 512 -#define CONN_OVERLOAD_LIMIT ((TIPC_FLOW_CONTROL_WIN * 2 + 1) * \ - SKB_TRUESIZE(TIPC_MAX_USER_MSG_SIZE)) - -/** - * struct tipc_port - TIPC port structure - * @sk: pointer to socket handle - * @lock: pointer to spinlock for controlling access to port - * @connected: non-zero if port is currently connected to a peer port - * @conn_type: TIPC type used when connection was established - * @conn_instance: TIPC instance used when connection was established - * @conn_unacked: number of unacknowledged messages received from peer port - * @published: non-zero if port has one or more associated names - * @congested: non-zero if cannot send because of link or port congestion - * @max_pkt: maximum packet size "hint" used when building messages sent by port - * @ref: unique reference to port in TIPC object registry - * @phdr: preformatted message header used when sending messages - * @port_list: adjacent ports in TIPC's global list of ports - * @dispatcher: ptr to routine which handles received messages - * @wakeup: ptr to routine to call when port is no longer congested - * @wait_list: adjacent ports in list of ports waiting on link congestion - * @waiting_pkts: - * @sent: # of non-empty messages sent by port - * @acked: # of non-empty message acknowledgements from connected port's peer - * @publications: list of publications for port - * @pub_count: total # of publications port has made during its lifetime - * @probing_state: - * @probing_interval: - * @timer_ref: - * @subscription: "node down" subscription used to terminate failed connections - */ -struct tipc_port { - struct sock *sk; - spinlock_t *lock; - int connected; - u32 conn_type; - u32 conn_instance; - u32 conn_unacked; - int published; - u32 congested; - u32 max_pkt; - u32 ref; - struct tipc_msg phdr; - struct list_head port_list; - u32 (*dispatcher)(struct tipc_port *, struct sk_buff *); - void (*wakeup)(struct tipc_port *); - struct list_head wait_list; - u32 waiting_pkts; - u32 sent; - u32 acked; - struct list_head publications; - u32 pub_count; - u32 probing_state; - u32 probing_interval; - struct timer_list timer; - struct tipc_node_subscr subscription; -}; - -extern spinlock_t tipc_port_list_lock; -struct tipc_port_list; - -/* - * TIPC port manipulation routines - */ -struct tipc_port *tipc_createport(struct sock *sk, - u32 (*dispatcher)(struct tipc_port *, - struct sk_buff *), - void (*wakeup)(struct tipc_port *), - const u32 importance); - -int tipc_reject_msg(struct sk_buff *buf, u32 err); - -void tipc_acknowledge(u32 port_ref, u32 ack); - -int tipc_deleteport(u32 portref); - -int tipc_portimportance(u32 portref, unsigned int *importance); -int tipc_set_portimportance(u32 portref, unsigned int importance); - -int tipc_portunreliable(u32 portref, unsigned int *isunreliable); -int tipc_set_portunreliable(u32 portref, unsigned int isunreliable); - -int tipc_portunreturnable(u32 portref, unsigned int *isunreturnable); -int tipc_set_portunreturnable(u32 portref, unsigned int isunreturnable); - -int tipc_publish(u32 portref, unsigned int scope, - struct tipc_name_seq const *name_seq); -int tipc_withdraw(u32 portref, unsigned int scope, - struct tipc_name_seq const *name_seq); - -int tipc_connect(u32 portref, struct tipc_portid const *port); - -int tipc_disconnect(u32 portref); - -int tipc_shutdown(u32 ref); - - -/* - * The following routines require that the port be locked on entry - */ -int __tipc_disconnect(struct tipc_port *tp_ptr); -int __tipc_connect(u32 ref, struct tipc_port *p_ptr, - struct tipc_portid const *peer); -int tipc_port_peer_msg(struct tipc_port *p_ptr, struct tipc_msg *msg); - -/* - * TIPC messaging routines - */ -int tipc_port_recv_msg(struct sk_buff *buf); -int tipc_send(u32 portref, struct iovec const *msg_sect, unsigned int len); - -int tipc_send2name(u32 portref, struct tipc_name const *name, u32 domain, - struct iovec const *msg_sect, unsigned int len); - -int tipc_send2port(u32 portref, struct tipc_portid const *dest, - struct iovec const *msg_sect, unsigned int len); - -int tipc_multicast(u32 portref, struct tipc_name_seq const *seq, - struct iovec const *msg, unsigned int len); - -int tipc_port_reject_sections(struct tipc_port *p_ptr, struct tipc_msg *hdr, - struct iovec const *msg_sect, unsigned int len, - int err); -struct sk_buff *tipc_port_get_ports(void); -void tipc_port_recv_proto_msg(struct sk_buff *buf); -void tipc_port_recv_mcast(struct sk_buff *buf, struct tipc_port_list *dp); -void tipc_port_reinit(void); - -/** - * tipc_port_lock - lock port instance referred to and return its pointer - */ -static inline struct tipc_port *tipc_port_lock(u32 ref) -{ - return (struct tipc_port *)tipc_ref_lock(ref); -} - -/** - * tipc_port_unlock - unlock a port instance - * - * Can use pointer instead of tipc_ref_unlock() since port is already locked. - */ -static inline void tipc_port_unlock(struct tipc_port *p_ptr) -{ - spin_unlock_bh(p_ptr->lock); -} - -static inline struct tipc_port *tipc_port_deref(u32 ref) -{ - return (struct tipc_port *)tipc_ref_deref(ref); -} - -static inline int tipc_port_congested(struct tipc_port *p_ptr) -{ - return (p_ptr->sent - p_ptr->acked) >= (TIPC_FLOW_CONTROL_WIN * 2); -} - -#endif diff --git a/net/tipc/ref.c b/net/tipc/ref.c deleted file mode 100644 index 2a2a938dc22c..000000000000 --- a/net/tipc/ref.c +++ /dev/null @@ -1,286 +0,0 @@ -/* - * net/tipc/ref.c: TIPC object registry code - * - * Copyright (c) 1991-2006, Ericsson AB - * Copyright (c) 2004-2007, 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 "ref.h" - -/** - * struct reference - TIPC object reference entry - * @object: pointer to object associated with reference entry - * @lock: spinlock controlling access to object - * @ref: reference value for object (combines instance & array index info) - */ -struct reference { - void *object; - spinlock_t lock; - u32 ref; -}; - -/** - * struct tipc_ref_table - table of TIPC object 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 object reference entry - * @last_free: array index of last unused object 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; -}; - -/* - * Object reference table consists of 2**N entries. - * - * State Object 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 objects - */ -int tipc_ref_table_init(u32 requested_size, u32 start) -{ - struct reference *table; - u32 actual_size; - - /* account for unused entry, then round up size to a power of 2 */ - - requested_size++; - for (actual_size = 16; actual_size < requested_size; actual_size <<= 1) - /* do nothing */ ; - - /* allocate table & mark all entries as uninitialized */ - table = vzalloc(actual_size * sizeof(struct reference)); - if (table == NULL) - return -ENOMEM; - - tipc_ref_table.entries = table; - tipc_ref_table.capacity = requested_size; - tipc_ref_table.init_point = 1; - tipc_ref_table.first_free = 0; - tipc_ref_table.last_free = 0; - tipc_ref_table.index_mask = actual_size - 1; - tipc_ref_table.start_mask = start & ~tipc_ref_table.index_mask; - - return 0; -} - -/** - * tipc_ref_table_stop - destroy reference table for objects - */ -void tipc_ref_table_stop(void) -{ - if (!tipc_ref_table.entries) - return; - - vfree(tipc_ref_table.entries); - tipc_ref_table.entries = NULL; -} - -/** - * tipc_ref_acquire - create reference to an object - * - * Register an object pointer in reference table and lock the object. - * Returns a unique reference value that is used from then on to retrieve the - * object pointer, or to determine that the object has been deregistered. - * - * Note: The object is returned in the locked state so that the caller can - * register a partially initialized object, without running the risk that - * the object will be accessed before initialization is complete. - */ -u32 tipc_ref_acquire(void *object, spinlock_t **lock) -{ - u32 index; - u32 index_mask; - u32 next_plus_upper; - u32 ref; - struct reference *entry = NULL; - - if (!object) { - pr_err("Attempt to acquire ref. to non-existent obj\n"); - return 0; - } - if (!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 entry */ - write_lock_bh(&ref_table_lock); - if (tipc_ref_table.first_free) { - 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; - } else if (tipc_ref_table.init_point < tipc_ref_table.capacity) { - index = tipc_ref_table.init_point++; - entry = &(tipc_ref_table.entries[index]); - spin_lock_init(&entry->lock); - ref = tipc_ref_table.start_mask + index; - } else { - ref = 0; - } - write_unlock_bh(&ref_table_lock); - - /* - * Grab the lock so no one else can modify this entry - * While we assign its ref value & object pointer - */ - if (entry) { - spin_lock_bh(&entry->lock); - entry->ref = ref; - entry->object = object; - *lock = &entry->lock; - /* - * keep it locked, the caller is responsible - * for unlocking this when they're done with it - */ - } - - return ref; -} - -/** - * tipc_ref_discard - invalidate references to an object - * - * Disallow future references to an object and free up the entry for re-use. - * Note: The entry's spin_lock may still be busy after discard - */ -void tipc_ref_discard(u32 ref) -{ - struct reference *entry; - u32 index; - u32 index_mask; - - if (!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); - - if (!entry->object) { - pr_err("Attempt to discard ref. to non-existent obj\n"); - goto exit; - } - if (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->object = NULL; - entry->ref = (ref & ~index_mask) + (index_mask + 1); - - /* append entry to free entry list */ - if (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_ref_lock - lock referenced object and return pointer to it - */ -void *tipc_ref_lock(u32 ref) -{ - if (likely(tipc_ref_table.entries)) { - struct reference *entry; - - entry = &tipc_ref_table.entries[ref & - tipc_ref_table.index_mask]; - if (likely(entry->ref != 0)) { - spin_lock_bh(&entry->lock); - if (likely((entry->ref == ref) && (entry->object))) - return entry->object; - spin_unlock_bh(&entry->lock); - } - } - return NULL; -} - - -/** - * tipc_ref_deref - return pointer referenced object (without locking it) - */ -void *tipc_ref_deref(u32 ref) -{ - if (likely(tipc_ref_table.entries)) { - struct reference *entry; - - entry = &tipc_ref_table.entries[ref & - tipc_ref_table.index_mask]; - if (likely(entry->ref == ref)) - return entry->object; - } - return NULL; -} diff --git a/net/tipc/server.c b/net/tipc/server.c index fd3fa57a410e..a538a02f869b 100644 --- a/net/tipc/server.c +++ b/net/tipc/server.c @@ -55,7 +55,7 @@ * @usr_data: user-specified field * @rx_action: what to do when connection socket is active * @outqueue: pointer to first outbound message in queue - * @outqueue_lock: controll access to the outqueue + * @outqueue_lock: control access to the outqueue * @outqueue: list of connection objects for its server * @swork: send work item */ @@ -87,7 +87,6 @@ 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 tipc_server *s = con->server; if (con->sock) { tipc_sock_release_local(con->sock); @@ -95,10 +94,6 @@ static void tipc_conn_kref_release(struct kref *kref) } tipc_clean_outqueues(con); - - if (con->conid) - s->tipc_conn_shutdown(con->conid, con->usr_data); - kfree(con); } @@ -124,7 +119,7 @@ static struct tipc_conn *tipc_conn_lookup(struct tipc_server *s, int conid) return con; } -static void sock_data_ready(struct sock *sk, int unused) +static void sock_data_ready(struct sock *sk) { struct tipc_conn *con; @@ -181,6 +176,9 @@ static void tipc_close_conn(struct tipc_conn *con) struct tipc_server *s = con->server; if (test_and_clear_bit(CF_CONNECTED, &con->flags)) { + if (con->conid) + s->tipc_conn_shutdown(con->conid, con->usr_data); + spin_lock_bh(&s->idr_lock); idr_remove(&s->conn_idr, con->conid); s->idr_in_use--; @@ -299,7 +297,7 @@ static int tipc_accept_from_sock(struct tipc_conn *con) newcon->usr_data = s->tipc_conn_new(newcon->conid); /* Wake up receive process in case of 'SYN+' message */ - newsock->sk->sk_data_ready(newsock->sk, 0); + newsock->sk->sk_data_ready(newsock->sk); return ret; } @@ -429,10 +427,12 @@ int tipc_conn_sendmsg(struct tipc_server *s, int conid, list_add_tail(&e->list, &con->outqueue); spin_unlock_bh(&con->outqueue_lock); - if (test_bit(CF_CONNECTED, &con->flags)) + if (test_bit(CF_CONNECTED, &con->flags)) { if (!queue_work(s->send_wq, &con->swork)) conn_put(con); - + } else { + conn_put(con); + } return 0; } @@ -573,7 +573,6 @@ int tipc_server_start(struct tipc_server *s) kmem_cache_destroy(s->rcvbuf_cache); return ret; } - s->enabled = 1; return ret; } @@ -583,10 +582,6 @@ void tipc_server_stop(struct tipc_server *s) int total = 0; int id; - if (!s->enabled) - return; - - s->enabled = 0; spin_lock_bh(&s->idr_lock); for (id = 0; total < s->idr_in_use; id++) { con = idr_find(&s->conn_idr, id); diff --git a/net/tipc/server.h b/net/tipc/server.h index 98b23f20bc0f..be817b0b547e 100644 --- a/net/tipc/server.h +++ b/net/tipc/server.h @@ -56,7 +56,6 @@ * @name: server name * @imp: message importance * @type: socket type - * @enabled: identify whether server is launched or not */ struct tipc_server { struct idr conn_idr; @@ -74,7 +73,6 @@ struct tipc_server { const char name[TIPC_SERVER_NAME_LEN]; int imp; int type; - int enabled; }; int tipc_conn_sendmsg(struct tipc_server *s, int conid, diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 3b61851bb927..75275c5cf929 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 Ericsson AB + * Copyright (c) 2001-2007, 2012-2014, Ericsson AB * Copyright (c) 2004-2008, 2010-2013, Wind River Systems * All rights reserved. * @@ -35,36 +35,84 @@ */ #include "core.h" -#include "port.h" - +#include "name_table.h" +#include "node.h" +#include "link.h" #include <linux/export.h> -#include <net/sock.h> +#include "config.h" +#include "socket.h" #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_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 +/** + * struct tipc_sock - TIPC socket structure + * @sk: socket - interacts with 'port' and with user via the socket API + * @connected: non-zero if port is currently connected to a peer port + * @conn_type: TIPC type used when connection was established + * @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 + * @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 + * @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 + */ struct tipc_sock { struct sock sk; - struct tipc_port *p; - struct tipc_portid peer_name; - unsigned int conn_timeout; + int connected; + u32 conn_type; + u32 conn_instance; + int published; + u32 max_pkt; + u32 ref; + 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; + uint conn_timeout; + atomic_t dupl_rcvcnt; + bool link_cong; + uint sent_unacked; + uint rcv_unacked; }; -#define tipc_sk(sk) ((struct tipc_sock *)(sk)) -#define tipc_sk_port(sk) (tipc_sk(sk)->p) - -#define tipc_rx_ready(sock) (!skb_queue_empty(&sock->sk->sk_receive_queue) || \ - (sock->state == SS_DISCONNECTING)) - -static int backlog_rcv(struct sock *sk, struct sk_buff *skb); -static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf); -static void wakeupdispatch(struct tipc_port *tport); -static void tipc_data_ready(struct sock *sk, int len); +static int tipc_backlog_rcv(struct sock *sk, struct sk_buff *skb); +static void tipc_data_ready(struct sock *sk); static void tipc_write_space(struct sock *sk); -static int release(struct socket *sock); -static int accept(struct socket *sock, struct socket *new_sock, int flags); +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 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 const struct proto_ops packet_ops; static const struct proto_ops stream_ops; @@ -73,8 +121,6 @@ static const struct proto_ops msg_ops; static struct proto tipc_proto; static struct proto tipc_proto_kern; -static int sockets_enabled; - /* * Revised TIPC socket locking policy: * @@ -120,27 +166,115 @@ static int sockets_enabled; * - port reference */ +static u32 tsk_peer_node(struct tipc_sock *tsk) +{ + return msg_destnode(&tsk->phdr); +} + +static u32 tsk_peer_port(struct tipc_sock *tsk) +{ + return msg_destport(&tsk->phdr); +} + +static bool tsk_unreliable(struct tipc_sock *tsk) +{ + return msg_src_droppable(&tsk->phdr) != 0; +} + +static void tsk_set_unreliable(struct tipc_sock *tsk, bool unreliable) +{ + msg_set_src_droppable(&tsk->phdr, unreliable ? 1 : 0); +} + +static bool tsk_unreturnable(struct tipc_sock *tsk) +{ + return msg_dest_droppable(&tsk->phdr) != 0; +} + +static void tsk_set_unreturnable(struct tipc_sock *tsk, bool unreturnable) +{ + msg_set_dest_droppable(&tsk->phdr, unreturnable ? 1 : 0); +} + +static int tsk_importance(struct tipc_sock *tsk) +{ + return msg_importance(&tsk->phdr); +} + +static int tsk_set_importance(struct tipc_sock *tsk, int imp) +{ + if (imp > TIPC_CRITICAL_IMPORTANCE) + return -EINVAL; + msg_set_importance(&tsk->phdr, (u32)imp); + return 0; +} + +static struct tipc_sock *tipc_sk(const struct sock *sk) +{ + return container_of(sk, struct tipc_sock, sk); +} + +static int tsk_conn_cong(struct tipc_sock *tsk) +{ + return tsk->sent_unacked >= TIPC_FLOWCTRL_WIN; +} + /** - * advance_rx_queue - discard first buffer in socket receive queue + * tsk_advance_rx_queue - discard first buffer in socket receive queue * * Caller must hold socket lock */ -static void advance_rx_queue(struct sock *sk) +static void tsk_advance_rx_queue(struct sock *sk) { kfree_skb(__skb_dequeue(&sk->sk_receive_queue)); } /** - * reject_rx_queue - reject all buffers in socket receive queue + * tsk_rej_rx_queue - reject all buffers in socket receive queue * * Caller must hold socket lock */ -static void reject_rx_queue(struct sock *sk) +static void tsk_rej_rx_queue(struct sock *sk) { struct sk_buff *buf; + u32 dnode; + + while ((buf = __skb_dequeue(&sk->sk_receive_queue))) { + if (tipc_msg_reverse(buf, &dnode, TIPC_ERR_NO_PORT)) + tipc_link_xmit(buf, dnode, 0); + } +} + +/* tsk_peer_msg - verify if message was sent by connected port's peer + * + * Handles cases where the node's network address has changed from + * the default of <0.0.0> to its configured setting. + */ +static bool tsk_peer_msg(struct tipc_sock *tsk, struct tipc_msg *msg) +{ + u32 peer_port = tsk_peer_port(tsk); + u32 orig_node; + u32 peer_node; + + if (unlikely(!tsk->connected)) + return false; + + if (unlikely(msg_origport(msg) != peer_port)) + return false; + + orig_node = msg_orignode(msg); + peer_node = tsk_peer_node(tsk); - while ((buf = __skb_dequeue(&sk->sk_receive_queue))) - tipc_reject_msg(buf, TIPC_ERR_NO_PORT); + if (likely(orig_node == peer_node)) + return true; + + if (!orig_node && (peer_node == tipc_own_addr)) + return true; + + if (!peer_node && (orig_node == tipc_own_addr)) + return true; + + return false; } /** @@ -155,13 +289,15 @@ static void reject_rx_queue(struct sock *sk) * * Returns 0 on success, errno otherwise */ -static int tipc_sk_create(struct net *net, struct socket *sock, int protocol, - int kern) +static int tipc_sk_create(struct net *net, struct socket *sock, + int protocol, int kern) { const struct proto_ops *ops; socket_state state; struct sock *sk; - struct tipc_port *tp_ptr; + struct tipc_sock *tsk; + struct tipc_msg *msg; + u32 ref; /* Validate arguments */ if (unlikely(protocol != 0)) @@ -194,34 +330,38 @@ static int tipc_sk_create(struct net *net, struct socket *sock, int protocol, if (sk == NULL) return -ENOMEM; - /* Allocate TIPC port for socket to use */ - tp_ptr = tipc_createport(sk, &dispatch, &wakeupdispatch, - TIPC_LOW_IMPORTANCE); - if (unlikely(!tp_ptr)) { - sk_free(sk); + 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, + 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); - sk->sk_backlog_rcv = backlog_rcv; + k_init_timer(&tsk->timer, (Handler)tipc_sk_timeout, ref); + sk->sk_backlog_rcv = tipc_backlog_rcv; sk->sk_rcvbuf = sysctl_tipc_rmem[1]; sk->sk_data_ready = tipc_data_ready; sk->sk_write_space = tipc_write_space; - tipc_sk(sk)->p = tp_ptr; - tipc_sk(sk)->conn_timeout = CONN_TIMEOUT_DEFAULT; - - spin_unlock_bh(tp_ptr->lock); + tsk->conn_timeout = CONN_TIMEOUT_DEFAULT; + tsk->sent_unacked = 0; + atomic_set(&tsk->dupl_rcvcnt, 0); if (sock->state == SS_READY) { - tipc_set_portunreturnable(tp_ptr->ref, 1); + tsk_set_unreturnable(tsk, true); if (sock->type == SOCK_DGRAM) - tipc_set_portunreliable(tp_ptr->ref, 1); + tsk_set_unreliable(tsk, true); } - return 0; } @@ -239,7 +379,6 @@ static int tipc_sk_create(struct net *net, struct socket *sock, int protocol, int tipc_sock_create_local(int type, struct socket **res) { int rc; - struct sock *sk; rc = sock_create_lite(AF_TIPC, type, 0, res); if (rc < 0) { @@ -248,8 +387,6 @@ int tipc_sock_create_local(int type, struct socket **res) } tipc_sk_create(&init_net, *res, 0, 1); - sk = (*res)->sk; - return 0; } @@ -262,7 +399,7 @@ int tipc_sock_create_local(int type, struct socket **res) */ void tipc_sock_release_local(struct socket *sock) { - release(sock); + tipc_release(sock); sock->ops = NULL; sock_release(sock); } @@ -288,7 +425,7 @@ int tipc_sock_accept_local(struct socket *sock, struct socket **newsock, if (ret < 0) return ret; - ret = accept(sock, *newsock, flags); + ret = tipc_accept(sock, *newsock, flags); if (ret < 0) { sock_release(*newsock); return ret; @@ -298,7 +435,7 @@ int tipc_sock_accept_local(struct socket *sock, struct socket **newsock, } /** - * release - destroy a TIPC socket + * tipc_release - destroy a TIPC socket * @sock: socket to destroy * * This routine cleans up any messages that are still queued on the socket. @@ -313,12 +450,12 @@ int tipc_sock_accept_local(struct socket *sock, struct socket **newsock, * * Returns 0 on success, errno otherwise */ -static int release(struct socket *sock) +static int tipc_release(struct socket *sock) { struct sock *sk = sock->sk; - struct tipc_port *tport; + struct tipc_sock *tsk; struct sk_buff *buf; - int res; + u32 dnode; /* * Exit if socket isn't fully initialized (occurs when a failed accept() @@ -327,13 +464,14 @@ static int release(struct socket *sock) if (sk == NULL) return 0; - tport = tipc_sk_port(sk); + tsk = tipc_sk(sk); lock_sock(sk); /* * Reject all unreceived messages, except on an active connection * (which disconnects locally & sends a 'FIN+' to peer) */ + dnode = tsk_peer_node(tsk); while (sock->state != SS_DISCONNECTING) { buf = __skb_dequeue(&sk->sk_receive_queue); if (buf == NULL) @@ -344,17 +482,27 @@ static int release(struct socket *sock) if ((sock->state == SS_CONNECTING) || (sock->state == SS_CONNECTED)) { sock->state = SS_DISCONNECTING; - tipc_disconnect(tport->ref); + tsk->connected = 0; + tipc_node_remove_conn(dnode, tsk->ref); } - tipc_reject_msg(buf, TIPC_ERR_NO_PORT); + if (tipc_msg_reverse(buf, &dnode, TIPC_ERR_NO_PORT)) + tipc_link_xmit(buf, dnode, 0); } } - /* - * Delete TIPC port; this ensures no more messages are queued - * (also disconnects an active connection & sends a 'FIN-' to peer) - */ - res = tipc_deleteport(tport->ref); + tipc_sk_withdraw(tsk, 0, NULL); + tipc_sk_ref_discard(tsk->ref); + k_cancel_timer(&tsk->timer); + if (tsk->connected) { + buf = 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); + if (buf) + tipc_link_xmit(buf, dnode, tsk->ref); + tipc_node_remove_conn(dnode, tsk->ref); + } + k_term_timer(&tsk->timer); /* Discard any remaining (connection-based) messages in receive queue */ __skb_queue_purge(&sk->sk_receive_queue); @@ -362,15 +510,14 @@ static int release(struct socket *sock) /* Reject any messages that accumulated in backlog queue */ sock->state = SS_DISCONNECTING; release_sock(sk); - sock_put(sk); sock->sk = NULL; - return res; + return 0; } /** - * bind - associate or disassocate TIPC name(s) with a socket + * tipc_bind - associate or disassocate TIPC name(s) with a socket * @sock: socket structure * @uaddr: socket address describing name(s) and desired operation * @uaddr_len: size of socket address data structure @@ -384,36 +531,53 @@ static int release(struct socket *sock) * NOTE: This routine doesn't need to take the socket lock since it doesn't * access any non-constant socket information. */ -static int bind(struct socket *sock, struct sockaddr *uaddr, int uaddr_len) +static int tipc_bind(struct socket *sock, struct sockaddr *uaddr, + int uaddr_len) { + struct sock *sk = sock->sk; struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; - u32 portref = tipc_sk_port(sock->sk)->ref; + struct tipc_sock *tsk = tipc_sk(sk); + int res = -EINVAL; - if (unlikely(!uaddr_len)) - return tipc_withdraw(portref, 0, NULL); + lock_sock(sk); + if (unlikely(!uaddr_len)) { + res = tipc_sk_withdraw(tsk, 0, NULL); + goto exit; + } - if (uaddr_len < sizeof(struct sockaddr_tipc)) - return -EINVAL; - if (addr->family != AF_TIPC) - return -EAFNOSUPPORT; + if (uaddr_len < sizeof(struct sockaddr_tipc)) { + res = -EINVAL; + goto exit; + } + if (addr->family != AF_TIPC) { + res = -EAFNOSUPPORT; + goto exit; + } if (addr->addrtype == TIPC_ADDR_NAME) addr->addr.nameseq.upper = addr->addr.nameseq.lower; - else if (addr->addrtype != TIPC_ADDR_NAMESEQ) - return -EAFNOSUPPORT; + else if (addr->addrtype != TIPC_ADDR_NAMESEQ) { + res = -EAFNOSUPPORT; + goto exit; + } if ((addr->addr.nameseq.type < TIPC_RESERVED_TYPES) && (addr->addr.nameseq.type != TIPC_TOP_SRV) && - (addr->addr.nameseq.type != TIPC_CFG_SRV)) - return -EACCES; + (addr->addr.nameseq.type != TIPC_CFG_SRV)) { + res = -EACCES; + goto exit; + } - return (addr->scope > 0) ? - tipc_publish(portref, addr->scope, &addr->addr.nameseq) : - tipc_withdraw(portref, -addr->scope, &addr->addr.nameseq); + res = (addr->scope > 0) ? + tipc_sk_publish(tsk, addr->scope, &addr->addr.nameseq) : + tipc_sk_withdraw(tsk, -addr->scope, &addr->addr.nameseq); +exit: + release_sock(sk); + return res; } /** - * get_name - get port ID of socket or peer socket + * tipc_getname - get port ID of socket or peer socket * @sock: socket structure * @uaddr: area for returned socket address * @uaddr_len: area for returned length of socket address @@ -425,21 +589,21 @@ static int bind(struct socket *sock, struct sockaddr *uaddr, int uaddr_len) * accesses socket information that is unchanging (or which changes in * a completely predictable manner). */ -static int get_name(struct socket *sock, struct sockaddr *uaddr, - int *uaddr_len, int peer) +static int tipc_getname(struct socket *sock, struct sockaddr *uaddr, + int *uaddr_len, int peer) { struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; - struct tipc_sock *tsock = tipc_sk(sock->sk); + struct tipc_sock *tsk = tipc_sk(sock->sk); memset(addr, 0, sizeof(*addr)); if (peer) { if ((sock->state != SS_CONNECTED) && ((peer != 2) || (sock->state != SS_DISCONNECTING))) return -ENOTCONN; - addr->addr.id.ref = tsock->peer_name.ref; - addr->addr.id.node = tsock->peer_name.node; + addr->addr.id.ref = tsk_peer_port(tsk); + addr->addr.id.node = tsk_peer_node(tsk); } else { - addr->addr.id.ref = tsock->p->ref; + addr->addr.id.ref = tsk->ref; addr->addr.id.node = tipc_own_addr; } @@ -453,7 +617,7 @@ static int get_name(struct socket *sock, struct sockaddr *uaddr, } /** - * poll - read and possibly block on pollmask + * tipc_poll - read and possibly block on pollmask * @file: file structure associated with the socket * @sock: socket for which to calculate the poll bits * @wait: ??? @@ -492,22 +656,23 @@ static int get_name(struct socket *sock, struct sockaddr *uaddr, * imply that the operation will succeed, merely that it should be performed * and will not block. */ -static unsigned int poll(struct file *file, struct socket *sock, - poll_table *wait) +static unsigned int tipc_poll(struct file *file, struct socket *sock, + poll_table *wait) { struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); u32 mask = 0; sock_poll_wait(file, sk_sleep(sk), wait); switch ((int)sock->state) { case SS_UNCONNECTED: - if (!tipc_sk_port(sk)->congested) + if (!tsk->link_cong) mask |= POLLOUT; break; case SS_READY: case SS_CONNECTED: - if (!tipc_sk_port(sk)->congested) + if (!tsk->link_cong && !tsk_conn_cong(tsk)) mask |= POLLOUT; /* fall thru' */ case SS_CONNECTING: @@ -524,6 +689,136 @@ static unsigned int poll(struct file *file, struct socket *sock, } /** + * tipc_sendmcast - send multicast message + * @sock: socket structure + * @seq: destination address + * @iov: message data to send + * @dsz: total length of message data + * @timeo: timeout to wait for wakeup + * + * Called from function tipc_sendmsg(), which has done all sanity checks + * Returns the number of bytes sent on success, or errno + */ +static int tipc_sendmcast(struct socket *sock, struct tipc_name_seq *seq, + struct iovec *iov, size_t dsz, long timeo) +{ + struct sock *sk = sock->sk; + struct tipc_msg *mhdr = &tipc_sk(sk)->phdr; + struct sk_buff *buf; + uint mtu; + int rc; + + msg_set_type(mhdr, TIPC_MCAST_MSG); + msg_set_lookup_scope(mhdr, TIPC_CLUSTER_SCOPE); + msg_set_destport(mhdr, 0); + msg_set_destnode(mhdr, 0); + msg_set_nametype(mhdr, seq->type); + msg_set_namelower(mhdr, seq->lower); + msg_set_nameupper(mhdr, seq->upper); + msg_set_hdr_sz(mhdr, MCAST_H_SIZE); + +new_mtu: + mtu = tipc_bclink_get_mtu(); + rc = tipc_msg_build(mhdr, iov, 0, dsz, mtu, &buf); + if (unlikely(rc < 0)) + return rc; + + do { + rc = tipc_bclink_xmit(buf); + if (likely(rc >= 0)) { + rc = dsz; + break; + } + if (rc == -EMSGSIZE) + goto new_mtu; + if (rc != -ELINKCONG) + break; + tipc_sk(sk)->link_cong = 1; + rc = tipc_wait_for_sndmsg(sock, &timeo); + if (rc) + kfree_skb_list(buf); + } while (!rc); + return rc; +} + +/* tipc_sk_mcast_rcv - Deliver multicast message to all destination sockets + */ +void tipc_sk_mcast_rcv(struct sk_buff *buf) +{ + 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; + 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"); + continue; + } + msg_set_destport(msg, item->ports[i]); + tipc_sk_rcv(b); + } + } + tipc_port_list_free(&dports); +} + +/** + * 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. + */ +static int tipc_sk_proto_rcv(struct tipc_sock *tsk, u32 *dnode, + struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + int conn_cong; + + /* Ignore if connection cannot be validated: */ + if (!tsk_peer_msg(tsk, msg)) + goto exit; + + tsk->probing_state = TIPC_CONN_OK; + + if (msg_type(msg) == CONN_ACK) { + conn_cong = tsk_conn_cong(tsk); + tsk->sent_unacked -= msg_msgcnt(msg); + 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; + } + /* Do nothing if msg_type() == CONN_PROBE_REPLY */ +exit: + kfree_skb(buf); + return TIPC_OK; +} + +/** * dest_name_check - verify user is permitted to send to specified port name * @dest: destination address * @m: descriptor for message to be sent @@ -537,6 +832,8 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m) { struct tipc_cfg_msg_hdr hdr; + if (unlikely(dest->addrtype == TIPC_ADDR_ID)) + return 0; if (likely(dest->addr.name.name.type >= TIPC_RESERVED_TYPES)) return 0; if (likely(dest->addr.name.name.type == TIPC_TOP_SRV)) @@ -554,12 +851,37 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m) return 0; } +static int tipc_wait_for_sndmsg(struct socket *sock, long *timeo_p) +{ + struct sock *sk = sock->sk; + struct tipc_sock *tsk = tipc_sk(sk); + DEFINE_WAIT(wait); + int done; + + do { + int err = sock_error(sk); + if (err) + return err; + if (sock->state == SS_DISCONNECTING) + return -EPIPE; + if (!*timeo_p) + return -EAGAIN; + if (signal_pending(current)) + return sock_intr_errno(*timeo_p); + + prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); + done = sk_wait_event(sk, timeo_p, !tsk->link_cong); + finish_wait(sk_sleep(sk), &wait); + } while (!done); + return 0; +} + /** - * send_msg - send message in connectionless manner + * tipc_sendmsg - send message in connectionless manner * @iocb: if NULL, indicates that socket lock is already held * @sock: socket structure * @m: message to send - * @total_len: length of message + * @dsz: amount of user data to be sent * * Message must have an destination specified explicitly. * Used for SOCK_RDM and SOCK_DGRAM messages, @@ -568,287 +890,272 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m) * * Returns the number of bytes sent on success, or errno otherwise */ -static int send_msg(struct kiocb *iocb, struct socket *sock, - struct msghdr *m, size_t total_len) +static int tipc_sendmsg(struct kiocb *iocb, 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_port *tport = tipc_sk_port(sk); - struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name; - int needs_conn; - long timeout_val; - int res = -EINVAL; + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_msg *mhdr = &tsk->phdr; + struct iovec *iov = m->msg_iov; + u32 dnode, dport; + struct sk_buff *buf; + struct tipc_name_seq *seq = &dest->addr.nameseq; + u32 mtu; + long timeo; + int rc = -EINVAL; if (unlikely(!dest)) return -EDESTADDRREQ; + if (unlikely((m->msg_namelen < sizeof(*dest)) || (dest->family != AF_TIPC))) return -EINVAL; - if (total_len > TIPC_MAX_USER_MSG_SIZE) + + if (dsz > TIPC_MAX_USER_MSG_SIZE) return -EMSGSIZE; if (iocb) lock_sock(sk); - needs_conn = (sock->state != SS_READY); - if (unlikely(needs_conn)) { + if (unlikely(sock->state != SS_READY)) { if (sock->state == SS_LISTENING) { - res = -EPIPE; + rc = -EPIPE; goto exit; } if (sock->state != SS_UNCONNECTED) { - res = -EISCONN; + rc = -EISCONN; goto exit; } - if (tport->published) { - res = -EOPNOTSUPP; + if (tsk->published) { + rc = -EOPNOTSUPP; goto exit; } if (dest->addrtype == TIPC_ADDR_NAME) { - tport->conn_type = dest->addr.name.name.type; - tport->conn_instance = dest->addr.name.name.instance; + tsk->conn_type = dest->addr.name.name.type; + tsk->conn_instance = dest->addr.name.name.instance; } + } + rc = dest_name_check(dest, m); + if (rc) + goto exit; + + timeo = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); - /* Abort any pending connection attempts (very unlikely) */ - reject_rx_queue(sk); + if (dest->addrtype == TIPC_ADDR_MCAST) { + rc = tipc_sendmcast(sock, seq, iov, dsz, timeo); + goto exit; + } else if (dest->addrtype == TIPC_ADDR_NAME) { + u32 type = dest->addr.name.name.type; + u32 inst = dest->addr.name.name.instance; + u32 domain = dest->addr.name.domain; + + dnode = domain; + msg_set_type(mhdr, TIPC_NAMED_MSG); + msg_set_hdr_sz(mhdr, NAMED_H_SIZE); + 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); + msg_set_destnode(mhdr, dnode); + msg_set_destport(mhdr, dport); + if (unlikely(!dport && !dnode)) { + rc = -EHOSTUNREACH; + goto exit; + } + } else if (dest->addrtype == TIPC_ADDR_ID) { + dnode = dest->addr.id.node; + msg_set_type(mhdr, TIPC_DIRECT_MSG); + msg_set_lookup_scope(mhdr, 0); + msg_set_destnode(mhdr, dnode); + msg_set_destport(mhdr, dest->addr.id.ref); + msg_set_hdr_sz(mhdr, BASIC_H_SIZE); } - timeout_val = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); +new_mtu: + mtu = tipc_node_get_mtu(dnode, tsk->ref); + rc = tipc_msg_build(mhdr, iov, 0, dsz, mtu, &buf); + if (rc < 0) + goto exit; do { - if (dest->addrtype == TIPC_ADDR_NAME) { - res = dest_name_check(dest, m); - if (res) - break; - res = tipc_send2name(tport->ref, - &dest->addr.name.name, - dest->addr.name.domain, - m->msg_iov, - total_len); - } else if (dest->addrtype == TIPC_ADDR_ID) { - res = tipc_send2port(tport->ref, - &dest->addr.id, - m->msg_iov, - total_len); - } else if (dest->addrtype == TIPC_ADDR_MCAST) { - if (needs_conn) { - res = -EOPNOTSUPP; - break; - } - res = dest_name_check(dest, m); - if (res) - break; - res = tipc_multicast(tport->ref, - &dest->addr.nameseq, - m->msg_iov, - total_len); - } - if (likely(res != -ELINKCONG)) { - if (needs_conn && (res >= 0)) + TIPC_SKB_CB(buf)->wakeup_pending = tsk->link_cong; + rc = tipc_link_xmit(buf, dnode, tsk->ref); + if (likely(rc >= 0)) { + if (sock->state != SS_READY) sock->state = SS_CONNECTING; + rc = dsz; break; } - if (timeout_val <= 0L) { - res = timeout_val ? timeout_val : -EWOULDBLOCK; + if (rc == -EMSGSIZE) + goto new_mtu; + if (rc != -ELINKCONG) break; - } - release_sock(sk); - timeout_val = wait_event_interruptible_timeout(*sk_sleep(sk), - !tport->congested, timeout_val); - lock_sock(sk); - } while (1); - + tsk->link_cong = 1; + rc = tipc_wait_for_sndmsg(sock, &timeo); + if (rc) + kfree_skb_list(buf); + } while (!rc); exit: if (iocb) release_sock(sk); - return res; + + return rc; } -/** - * send_packet - send a connection-oriented message - * @iocb: if NULL, indicates that socket lock is already held - * @sock: socket structure - * @m: message to send - * @total_len: length of message - * - * Used for SOCK_SEQPACKET messages and SOCK_STREAM data. - * - * Returns the number of bytes sent on success, or errno otherwise - */ -static int send_packet(struct kiocb *iocb, struct socket *sock, - struct msghdr *m, size_t total_len) +static int tipc_wait_for_sndpkt(struct socket *sock, long *timeo_p) { struct sock *sk = sock->sk; - struct tipc_port *tport = tipc_sk_port(sk); - struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name; - long timeout_val; - int res; - - /* Handle implied connection establishment */ - if (unlikely(dest)) - return send_msg(iocb, sock, m, total_len); - - if (total_len > TIPC_MAX_USER_MSG_SIZE) - return -EMSGSIZE; - - if (iocb) - lock_sock(sk); - - timeout_val = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); + struct tipc_sock *tsk = tipc_sk(sk); + DEFINE_WAIT(wait); + int done; do { - if (unlikely(sock->state != SS_CONNECTED)) { - if (sock->state == SS_DISCONNECTING) - res = -EPIPE; - else - res = -ENOTCONN; - break; - } - - res = tipc_send(tport->ref, m->msg_iov, total_len); - if (likely(res != -ELINKCONG)) - break; - if (timeout_val <= 0L) { - res = timeout_val ? timeout_val : -EWOULDBLOCK; - break; - } - release_sock(sk); - timeout_val = wait_event_interruptible_timeout(*sk_sleep(sk), - (!tport->congested || !tport->connected), timeout_val); - lock_sock(sk); - } while (1); - - if (iocb) - release_sock(sk); - return res; + int err = sock_error(sk); + if (err) + return err; + if (sock->state == SS_DISCONNECTING) + return -EPIPE; + else if (sock->state != SS_CONNECTED) + return -ENOTCONN; + if (!*timeo_p) + return -EAGAIN; + if (signal_pending(current)) + return sock_intr_errno(*timeo_p); + + prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); + done = sk_wait_event(sk, timeo_p, + (!tsk->link_cong && + !tsk_conn_cong(tsk)) || + !tsk->connected); + finish_wait(sk_sleep(sk), &wait); + } while (!done); + return 0; } /** - * send_stream - send stream-oriented data + * tipc_send_stream - send stream-oriented data * @iocb: (unused) * @sock: socket structure * @m: data to send - * @total_len: total length of data to be sent + * @dsz: total length of data to be transmitted * * Used for SOCK_STREAM data. * * Returns the number of bytes sent on success (or partial success), * or errno if no data sent */ -static int send_stream(struct kiocb *iocb, struct socket *sock, - struct msghdr *m, size_t total_len) +static int tipc_send_stream(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t dsz) { struct sock *sk = sock->sk; - struct tipc_port *tport = tipc_sk_port(sk); - struct msghdr my_msg; - struct iovec my_iov; - struct iovec *curr_iov; - int curr_iovlen; - char __user *curr_start; - u32 hdr_size; - int curr_left; - int bytes_to_send; - int bytes_sent; - int res; - - lock_sock(sk); + struct tipc_sock *tsk = tipc_sk(sk); + struct tipc_msg *mhdr = &tsk->phdr; + struct sk_buff *buf; + DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name); + u32 ref = tsk->ref; + int rc = -EINVAL; + long timeo; + u32 dnode; + uint mtu, send, sent = 0; - /* Handle special cases where there is no connection */ - if (unlikely(sock->state != SS_CONNECTED)) { - if (sock->state == SS_UNCONNECTED) { - res = send_packet(NULL, sock, m, total_len); - goto exit; - } else if (sock->state == SS_DISCONNECTING) { - res = -EPIPE; - goto exit; - } else { - res = -ENOTCONN; - goto exit; - } + /* Handle implied connection establishment */ + if (unlikely(dest)) { + rc = tipc_sendmsg(iocb, sock, m, dsz); + if (dsz && (dsz == rc)) + tsk->sent_unacked = 1; + return rc; } + if (dsz > (uint)INT_MAX) + return -EMSGSIZE; - if (unlikely(m->msg_name)) { - res = -EISCONN; - goto exit; - } + if (iocb) + lock_sock(sk); - if (total_len > (unsigned int)INT_MAX) { - res = -EMSGSIZE; + if (unlikely(sock->state != SS_CONNECTED)) { + if (sock->state == SS_DISCONNECTING) + rc = -EPIPE; + else + rc = -ENOTCONN; goto exit; } - /* - * Send each iovec entry using one or more messages - * - * Note: This algorithm is good for the most likely case - * (i.e. one large iovec entry), but could be improved to pass sets - * of small iovec entries into send_packet(). - */ - curr_iov = m->msg_iov; - curr_iovlen = m->msg_iovlen; - my_msg.msg_iov = &my_iov; - my_msg.msg_iovlen = 1; - my_msg.msg_flags = m->msg_flags; - my_msg.msg_name = NULL; - bytes_sent = 0; - - hdr_size = msg_hdr_sz(&tport->phdr); - - while (curr_iovlen--) { - curr_start = curr_iov->iov_base; - curr_left = curr_iov->iov_len; - - while (curr_left) { - bytes_to_send = tport->max_pkt - hdr_size; - if (bytes_to_send > TIPC_MAX_USER_MSG_SIZE) - bytes_to_send = TIPC_MAX_USER_MSG_SIZE; - if (curr_left < bytes_to_send) - bytes_to_send = curr_left; - my_iov.iov_base = curr_start; - my_iov.iov_len = bytes_to_send; - res = send_packet(NULL, sock, &my_msg, bytes_to_send); - if (res < 0) { - if (bytes_sent) - res = bytes_sent; - goto exit; + timeo = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT); + dnode = tsk_peer_node(tsk); + +next: + mtu = tsk->max_pkt; + send = min_t(uint, dsz - sent, TIPC_MAX_USER_MSG_SIZE); + rc = tipc_msg_build(mhdr, m->msg_iov, sent, send, mtu, &buf); + if (unlikely(rc < 0)) + goto exit; + do { + if (likely(!tsk_conn_cong(tsk))) { + rc = tipc_link_xmit(buf, dnode, ref); + if (likely(!rc)) { + tsk->sent_unacked++; + sent += send; + if (sent == dsz) + break; + goto next; } - curr_left -= bytes_to_send; - curr_start += bytes_to_send; - bytes_sent += bytes_to_send; + if (rc == -EMSGSIZE) { + tsk->max_pkt = tipc_node_get_mtu(dnode, ref); + goto next; + } + if (rc != -ELINKCONG) + break; + tsk->link_cong = 1; } - - curr_iov++; - } - res = bytes_sent; + rc = tipc_wait_for_sndpkt(sock, &timeo); + if (rc) + kfree_skb_list(buf); + } while (!rc); exit: - release_sock(sk); - return res; + if (iocb) + release_sock(sk); + return sent ? sent : rc; } /** - * auto_connect - complete connection setup to a remote port + * tipc_send_packet - send a connection-oriented message + * @iocb: if NULL, indicates that socket lock is already held * @sock: socket structure - * @msg: peer's response message + * @m: message to send + * @dsz: length of data to be transmitted * - * Returns 0 on success, errno otherwise + * Used for SOCK_SEQPACKET messages. + * + * Returns the number of bytes sent on success, or errno otherwise */ -static int auto_connect(struct socket *sock, struct tipc_msg *msg) +static int tipc_send_packet(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t dsz) { - struct tipc_sock *tsock = tipc_sk(sock->sk); - struct tipc_port *p_ptr; - - tsock->peer_name.ref = msg_origport(msg); - tsock->peer_name.node = msg_orignode(msg); - p_ptr = tipc_port_deref(tsock->p->ref); - if (!p_ptr) - return -EINVAL; + if (dsz > TIPC_MAX_USER_MSG_SIZE) + return -EMSGSIZE; - __tipc_connect(tsock->p->ref, p_ptr, &tsock->peer_name); + return tipc_send_stream(iocb, sock, m, dsz); +} - if (msg_importance(msg) > TIPC_CRITICAL_IMPORTANCE) - return -EINVAL; - msg_set_importance(&p_ptr->phdr, (u32)msg_importance(msg)); - sock->state = SS_CONNECTED; - return 0; +/* tipc_sk_finish_conn - complete the setup of a connection + */ +static void tipc_sk_finish_conn(struct tipc_sock *tsk, u32 peer_port, + u32 peer_node) +{ + struct tipc_msg *msg = &tsk->phdr; + + msg_set_destnode(msg, peer_node); + msg_set_destport(msg, peer_port); + msg_set_type(msg, TIPC_CONN_MSG); + msg_set_lookup_scope(msg, 0); + msg_set_hdr_sz(msg, SHORT_H_SIZE); + + tsk->probing_interval = 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); } /** @@ -860,7 +1167,7 @@ static int auto_connect(struct socket *sock, struct tipc_msg *msg) */ static void set_orig_addr(struct msghdr *m, struct tipc_msg *msg) { - struct sockaddr_tipc *addr = (struct sockaddr_tipc *)m->msg_name; + DECLARE_SOCKADDR(struct sockaddr_tipc *, addr, m->msg_name); if (addr) { addr->family = AF_TIPC; @@ -875,17 +1182,17 @@ static void set_orig_addr(struct msghdr *m, struct tipc_msg *msg) } /** - * anc_data_recv - optionally capture ancillary data for received message + * tipc_sk_anc_data_recv - optionally capture ancillary data for received message * @m: descriptor for message info * @msg: received message header - * @tport: TIPC port associated with message + * @tsk: TIPC port associated with message * * Note: Ancillary data is not captured if not requested by receiver. * * Returns 0 if successful, otherwise errno */ -static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, - struct tipc_port *tport) +static int tipc_sk_anc_data_recv(struct msghdr *m, struct tipc_msg *msg, + struct tipc_sock *tsk) { u32 anc_data[3]; u32 err; @@ -928,10 +1235,10 @@ static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, anc_data[2] = msg_nameupper(msg); break; case TIPC_CONN_MSG: - has_name = (tport->conn_type != 0); - anc_data[0] = tport->conn_type; - anc_data[1] = tport->conn_instance; - anc_data[2] = tport->conn_instance; + has_name = (tsk->conn_type != 0); + anc_data[0] = tsk->conn_type; + anc_data[1] = tsk->conn_instance; + anc_data[2] = tsk->conn_instance; break; default: has_name = 0; @@ -945,8 +1252,59 @@ static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, return 0; } +static void tipc_sk_send_ack(struct tipc_sock *tsk, uint ack) +{ + struct sk_buff *buf = NULL; + struct tipc_msg *msg; + u32 peer_port = tsk_peer_port(tsk); + u32 dnode = tsk_peer_node(tsk); + + if (!tsk->connected) + return; + buf = tipc_msg_create(CONN_MANAGER, CONN_ACK, INT_H_SIZE, 0, dnode, + tipc_own_addr, peer_port, tsk->ref, TIPC_OK); + if (!buf) + return; + msg = buf_msg(buf); + msg_set_msgcnt(msg, ack); + tipc_link_xmit(buf, dnode, msg_link_selector(msg)); +} + +static int tipc_wait_for_rcvmsg(struct socket *sock, long *timeop) +{ + struct sock *sk = sock->sk; + DEFINE_WAIT(wait); + long timeo = *timeop; + int err; + + for (;;) { + prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); + if (timeo && skb_queue_empty(&sk->sk_receive_queue)) { + if (sock->state == SS_DISCONNECTING) { + err = -ENOTCONN; + break; + } + release_sock(sk); + timeo = schedule_timeout(timeo); + lock_sock(sk); + } + 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; + } + finish_wait(sk_sleep(sk), &wait); + *timeop = timeo; + return err; +} + /** - * recv_msg - receive packet-oriented message + * tipc_recvmsg - receive packet-oriented message * @iocb: (unused) * @m: descriptor for message info * @buf_len: total size of user buffer area @@ -957,14 +1315,14 @@ static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, * * Returns size of returned message data, errno otherwise */ -static int recv_msg(struct kiocb *iocb, struct socket *sock, - struct msghdr *m, size_t buf_len, int flags) +static int tipc_recvmsg(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t buf_len, int flags) { struct sock *sk = sock->sk; - struct tipc_port *tport = tipc_sk_port(sk); + struct tipc_sock *tsk = tipc_sk(sk); struct sk_buff *buf; struct tipc_msg *msg; - long timeout; + long timeo; unsigned int sz; u32 err; int res; @@ -980,25 +1338,13 @@ static int recv_msg(struct kiocb *iocb, struct socket *sock, goto exit; } - timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); + timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); restart: /* Look for a message in receive queue; wait if necessary */ - while (skb_queue_empty(&sk->sk_receive_queue)) { - if (sock->state == SS_DISCONNECTING) { - res = -ENOTCONN; - goto exit; - } - if (timeout <= 0L) { - res = timeout ? timeout : -EWOULDBLOCK; - goto exit; - } - release_sock(sk); - timeout = wait_event_interruptible_timeout(*sk_sleep(sk), - tipc_rx_ready(sock), - timeout); - lock_sock(sk); - } + res = tipc_wait_for_rcvmsg(sock, &timeo); + if (res) + goto exit; /* Look at first message in receive queue */ buf = skb_peek(&sk->sk_receive_queue); @@ -1008,7 +1354,7 @@ restart: /* Discard an empty non-errored message & try again */ if ((!sz) && (!err)) { - advance_rx_queue(sk); + tsk_advance_rx_queue(sk); goto restart; } @@ -1016,7 +1362,7 @@ restart: set_orig_addr(m, msg); /* Capture ancillary data (optional) */ - res = anc_data_recv(m, msg, tport); + res = tipc_sk_anc_data_recv(m, msg, tsk); if (res) goto exit; @@ -1042,9 +1388,11 @@ restart: /* Consume received message (optional) */ if (likely(!(flags & MSG_PEEK))) { if ((sock->state != SS_READY) && - (++tport->conn_unacked >= TIPC_FLOW_CONTROL_WIN)) - tipc_acknowledge(tport->ref, tport->conn_unacked); - advance_rx_queue(sk); + (++tsk->rcv_unacked >= TIPC_CONNACK_INTV)) { + tipc_sk_send_ack(tsk, tsk->rcv_unacked); + tsk->rcv_unacked = 0; + } + tsk_advance_rx_queue(sk); } exit: release_sock(sk); @@ -1052,7 +1400,7 @@ exit: } /** - * recv_stream - receive stream-oriented data + * tipc_recv_stream - receive stream-oriented data * @iocb: (unused) * @m: descriptor for message info * @buf_len: total size of user buffer area @@ -1063,14 +1411,14 @@ exit: * * Returns size of returned message data, errno otherwise */ -static int recv_stream(struct kiocb *iocb, struct socket *sock, - struct msghdr *m, size_t buf_len, int flags) +static int tipc_recv_stream(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t buf_len, int flags) { struct sock *sk = sock->sk; - struct tipc_port *tport = tipc_sk_port(sk); + struct tipc_sock *tsk = tipc_sk(sk); struct sk_buff *buf; struct tipc_msg *msg; - long timeout; + long timeo; unsigned int sz; int sz_to_copy, target, needed; int sz_copied = 0; @@ -1083,31 +1431,19 @@ static int recv_stream(struct kiocb *iocb, struct socket *sock, lock_sock(sk); - if (unlikely((sock->state == SS_UNCONNECTED))) { + if (unlikely(sock->state == SS_UNCONNECTED)) { res = -ENOTCONN; goto exit; } target = sock_rcvlowat(sk, flags & MSG_WAITALL, buf_len); - timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); + timeo = sock_rcvtimeo(sk, flags & MSG_DONTWAIT); restart: /* Look for a message in receive queue; wait if necessary */ - while (skb_queue_empty(&sk->sk_receive_queue)) { - if (sock->state == SS_DISCONNECTING) { - res = -ENOTCONN; - goto exit; - } - if (timeout <= 0L) { - res = timeout ? timeout : -EWOULDBLOCK; - goto exit; - } - release_sock(sk); - timeout = wait_event_interruptible_timeout(*sk_sleep(sk), - tipc_rx_ready(sock), - timeout); - lock_sock(sk); - } + res = tipc_wait_for_rcvmsg(sock, &timeo); + if (res) + goto exit; /* Look at first message in receive queue */ buf = skb_peek(&sk->sk_receive_queue); @@ -1117,14 +1453,14 @@ restart: /* Discard an empty non-errored message & try again */ if ((!sz) && (!err)) { - advance_rx_queue(sk); + tsk_advance_rx_queue(sk); goto restart; } /* Optionally capture sender's address & ancillary data of first msg */ if (sz_copied == 0) { set_orig_addr(m, msg); - res = anc_data_recv(m, msg, tport); + res = tipc_sk_anc_data_recv(m, msg, tsk); if (res) goto exit; } @@ -1162,9 +1498,11 @@ restart: /* Consume received message (optional) */ if (likely(!(flags & MSG_PEEK))) { - if (unlikely(++tport->conn_unacked >= TIPC_FLOW_CONTROL_WIN)) - tipc_acknowledge(tport->ref, tport->conn_unacked); - advance_rx_queue(sk); + if (unlikely(++tsk->rcv_unacked >= TIPC_CONNACK_INTV)) { + tipc_sk_send_ack(tsk, tsk->rcv_unacked); + tsk->rcv_unacked = 0; + } + tsk_advance_rx_queue(sk); } /* Loop around if more data is required */ @@ -1201,7 +1539,7 @@ static void tipc_write_space(struct sock *sk) * @sk: socket * @len: the length of messages */ -static void tipc_data_ready(struct sock *sk, int len) +static void tipc_data_ready(struct sock *sk) { struct socket_wq *wq; @@ -1215,19 +1553,17 @@ static void tipc_data_ready(struct sock *sk, int len) /** * filter_connect - Handle all incoming messages for a connection-based socket - * @tsock: TIPC socket + * @tsk: TIPC socket * @msg: message * - * Returns TIPC error status code and socket error status code - * once it encounters some errors + * Returns 0 (TIPC_OK) if everyting ok, -TIPC_ERR_NO_PORT otherwise */ -static u32 filter_connect(struct tipc_sock *tsock, struct sk_buff **buf) +static int filter_connect(struct tipc_sock *tsk, struct sk_buff **buf) { - struct socket *sock = tsock->sk.sk_socket; + struct sock *sk = &tsk->sk; + struct socket *sock = sk->sk_socket; struct tipc_msg *msg = buf_msg(*buf); - struct sock *sk = &tsock->sk; - u32 retval = TIPC_ERR_NO_PORT; - int res; + int retval = -TIPC_ERR_NO_PORT; if (msg_mcast(msg)) return retval; @@ -1235,16 +1571,23 @@ static u32 filter_connect(struct tipc_sock *tsock, struct sk_buff **buf) switch ((int)sock->state) { case SS_CONNECTED: /* Accept only connection-based messages sent by peer */ - if (msg_connected(msg) && tipc_port_peer_msg(tsock->p, msg)) { + if (tsk_peer_msg(tsk, msg)) { if (unlikely(msg_errcode(msg))) { sock->state = SS_DISCONNECTING; - __tipc_disconnect(tsock->p); + tsk->connected = 0; + /* let timer expire on it's own */ + tipc_node_remove_conn(tsk_peer_node(tsk), + tsk->ref); } retval = TIPC_OK; } break; case SS_CONNECTING: /* Accept only ACK or NACK message */ + + if (unlikely(!msg_connected(msg))) + break; + if (unlikely(msg_errcode(msg))) { sock->state = SS_DISCONNECTING; sk->sk_err = ECONNREFUSED; @@ -1252,17 +1595,17 @@ static u32 filter_connect(struct tipc_sock *tsock, struct sk_buff **buf) break; } - if (unlikely(!msg_connected(msg))) - break; - - res = auto_connect(sock, msg); - if (res) { + if (unlikely(msg_importance(msg) > TIPC_CRITICAL_IMPORTANCE)) { sock->state = SS_DISCONNECTING; - sk->sk_err = -res; + sk->sk_err = EINVAL; retval = TIPC_OK; break; } + tipc_sk_finish_conn(tsk, msg_origport(msg), msg_orignode(msg)); + msg_set_importance(&tsk->phdr, msg_importance(msg)); + sock->state = SS_CONNECTED; + /* If an incoming message is an 'ACK-', it should be * discarded here because it doesn't contain useful * data. In addition, we should try to wake up @@ -1311,14 +1654,12 @@ static u32 filter_connect(struct tipc_sock *tsock, struct sk_buff **buf) static unsigned int rcvbuf_limit(struct sock *sk, struct sk_buff *buf) { struct tipc_msg *msg = buf_msg(buf); - unsigned int limit; if (msg_connected(msg)) - limit = sysctl_tipc_rmem[2]; - else - limit = sk->sk_rcvbuf >> TIPC_CRITICAL_IMPORTANCE << - msg_importance(msg); - return limit; + return sysctl_tipc_rmem[2]; + + return sk->sk_rcvbuf >> TIPC_CRITICAL_IMPORTANCE << + msg_importance(msg); } /** @@ -1331,43 +1672,56 @@ static unsigned int rcvbuf_limit(struct sock *sk, struct sk_buff *buf) * * Called with socket lock already taken; port lock may also be taken. * - * Returns TIPC error status code (TIPC_OK if message is not to be rejected) + * 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 */ -static u32 filter_rcv(struct sock *sk, struct sk_buff *buf) +static int filter_rcv(struct sock *sk, struct sk_buff *buf) { 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 res = TIPC_OK; + u32 onode; + int rc = TIPC_OK; + + if (unlikely(msg_user(msg) == CONN_MANAGER)) + return tipc_sk_proto_rcv(tsk, &onode, buf); + + if (unlikely(msg_user(msg) == SOCK_WAKEUP)) { + kfree_skb(buf); + tsk->link_cong = 0; + sk->sk_write_space(sk); + return TIPC_OK; + } /* Reject message if it is wrong sort of message for socket */ if (msg_type(msg) > TIPC_DIRECT_MSG) - return TIPC_ERR_NO_PORT; + return -TIPC_ERR_NO_PORT; if (sock->state == SS_READY) { if (msg_connected(msg)) - return TIPC_ERR_NO_PORT; + return -TIPC_ERR_NO_PORT; } else { - res = filter_connect(tipc_sk(sk), &buf); - if (res != TIPC_OK || buf == NULL) - return res; + rc = filter_connect(tsk, &buf); + if (rc != TIPC_OK || buf == NULL) + return rc; } /* Reject message if there isn't room to queue it */ if (sk_rmem_alloc_get(sk) + buf->truesize >= limit) - return TIPC_ERR_OVERLOAD; + 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); - sk->sk_data_ready(sk, 0); + sk->sk_data_ready(sk); return TIPC_OK; } /** - * backlog_rcv - handle incoming message from backlog queue + * tipc_backlog_rcv - handle incoming message from backlog queue * @sk: socket * @buf: message * @@ -1375,65 +1729,100 @@ static u32 filter_rcv(struct sock *sk, struct sk_buff *buf) * * Returns 0 */ -static int backlog_rcv(struct sock *sk, struct sk_buff *buf) +static int tipc_backlog_rcv(struct sock *sk, struct sk_buff *buf) { - u32 res; + int rc; + u32 onode; + struct tipc_sock *tsk = tipc_sk(sk); + uint truesize = buf->truesize; + + rc = filter_rcv(sk, buf); + + if (likely(!rc)) { + if (atomic_read(&tsk->dupl_rcvcnt) < TIPC_CONN_OVERLOAD_LIMIT) + atomic_add(truesize, &tsk->dupl_rcvcnt); + return 0; + } + + if ((rc < 0) && !tipc_msg_reverse(buf, &onode, -rc)) + return 0; + + tipc_link_xmit(buf, onode, 0); - res = filter_rcv(sk, buf); - if (res) - tipc_reject_msg(buf, res); return 0; } /** - * dispatch - handle incoming message - * @tport: TIPC port that received message - * @buf: message - * - * Called with port lock already taken. - * - * Returns TIPC error status code (TIPC_OK if message is not to be rejected) + * tipc_sk_rcv - handle incoming message + * @buf: buffer containing arriving message + * Consumes buffer + * Returns 0 if success, or errno: -EHOSTUNREACH */ -static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf) +int tipc_sk_rcv(struct sk_buff *buf) { - struct sock *sk = tport->sk; - u32 res; + struct tipc_sock *tsk; + struct sock *sk; + u32 dport = msg_destport(buf_msg(buf)); + int rc = TIPC_OK; + uint limit; + u32 dnode; + + /* Validate destination and message */ + tsk = tipc_sk_get(dport); + if (unlikely(!tsk)) { + rc = tipc_msg_eval(buf, &dnode); + goto exit; + } + sk = &tsk->sk; - /* - * Process message if socket is unlocked; otherwise add to backlog queue - * - * This code is based on sk_receive_skb(), but must be distinct from it - * since a TIPC-specific filter/reject mechanism is utilized - */ + /* Queue message */ bh_lock_sock(sk); + if (!sock_owned_by_user(sk)) { - res = filter_rcv(sk, buf); + rc = filter_rcv(sk, buf); } else { - if (sk_add_backlog(sk, buf, rcvbuf_limit(sk, buf))) - res = TIPC_ERR_OVERLOAD; - else - res = TIPC_OK; + if (sk->sk_backlog.len == 0) + atomic_set(&tsk->dupl_rcvcnt, 0); + limit = rcvbuf_limit(sk, buf) + atomic_read(&tsk->dupl_rcvcnt); + if (sk_add_backlog(sk, buf, limit)) + rc = -TIPC_ERR_OVERLOAD; } bh_unlock_sock(sk); + tipc_sk_put(tsk); + if (likely(!rc)) + return 0; +exit: + if ((rc < 0) && !tipc_msg_reverse(buf, &dnode, -rc)) + return -EHOSTUNREACH; - return res; + tipc_link_xmit(buf, dnode, 0); + return (rc < 0) ? -EHOSTUNREACH : 0; } -/** - * wakeupdispatch - wake up port after congestion - * @tport: port to wakeup - * - * Called with port lock already taken. - */ -static void wakeupdispatch(struct tipc_port *tport) +static int tipc_wait_for_connect(struct socket *sock, long *timeo_p) { - struct sock *sk = tport->sk; + struct sock *sk = sock->sk; + DEFINE_WAIT(wait); + int done; - sk->sk_write_space(sk); + do { + int err = sock_error(sk); + if (err) + return err; + if (!*timeo_p) + return -ETIMEDOUT; + if (signal_pending(current)) + return sock_intr_errno(*timeo_p); + + prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); + done = sk_wait_event(sk, timeo_p, sock->state != SS_CONNECTING); + finish_wait(sk_sleep(sk), &wait); + } while (!done); + return 0; } /** - * connect - establish a connection to another TIPC port + * tipc_connect - establish a connection to another TIPC port * @sock: socket structure * @dest: socket address for destination port * @destlen: size of socket address data structure @@ -1441,13 +1830,14 @@ static void wakeupdispatch(struct tipc_port *tport) * * Returns 0 on success, errno otherwise */ -static int connect(struct socket *sock, struct sockaddr *dest, int destlen, - int flags) +static int tipc_connect(struct socket *sock, struct sockaddr *dest, + int destlen, int flags) { struct sock *sk = sock->sk; struct sockaddr_tipc *dst = (struct sockaddr_tipc *)dest; struct msghdr m = {NULL,}; - unsigned int timeout; + long timeout = (flags & O_NONBLOCK) ? 0 : tipc_sk(sk)->conn_timeout; + socket_state previous; int res; lock_sock(sk); @@ -1469,8 +1859,7 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen, goto exit; } - timeout = (flags & O_NONBLOCK) ? 0 : tipc_sk(sk)->conn_timeout; - + previous = sock->state; switch (sock->state) { case SS_UNCONNECTED: /* Send a 'SYN-' to destination */ @@ -1483,7 +1872,7 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen, if (!timeout) m.msg_flags = MSG_DONTWAIT; - res = send_msg(NULL, sock, &m, 0); + res = tipc_sendmsg(NULL, sock, &m, 0); if ((res < 0) && (res != -EWOULDBLOCK)) goto exit; @@ -1492,56 +1881,35 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen, * case is EINPROGRESS, rather than EALREADY. */ res = -EINPROGRESS; - break; case SS_CONNECTING: - res = -EALREADY; + if (previous == SS_CONNECTING) + res = -EALREADY; + if (!timeout) + goto exit; + timeout = msecs_to_jiffies(timeout); + /* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */ + res = tipc_wait_for_connect(sock, &timeout); break; case SS_CONNECTED: res = -EISCONN; break; default: res = -EINVAL; - goto exit; - } - - if (sock->state == SS_CONNECTING) { - if (!timeout) - goto exit; - - /* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */ - release_sock(sk); - res = wait_event_interruptible_timeout(*sk_sleep(sk), - sock->state != SS_CONNECTING, - timeout ? (long)msecs_to_jiffies(timeout) - : MAX_SCHEDULE_TIMEOUT); - lock_sock(sk); - if (res <= 0) { - if (res == 0) - res = -ETIMEDOUT; - else - ; /* leave "res" unchanged */ - goto exit; - } + break; } - - if (unlikely(sock->state == SS_DISCONNECTING)) - res = sock_error(sk); - else - res = 0; - exit: release_sock(sk); return res; } /** - * listen - allow socket to listen for incoming connections + * tipc_listen - allow socket to listen for incoming connections * @sock: socket structure * @len: (unused) * * Returns 0 on success, errno otherwise */ -static int listen(struct socket *sock, int len) +static int tipc_listen(struct socket *sock, int len) { struct sock *sk = sock->sk; int res; @@ -1559,23 +1927,57 @@ static int listen(struct socket *sock, int len) return res; } +static int tipc_wait_for_accept(struct socket *sock, long timeo) +{ + struct sock *sk = sock->sk; + DEFINE_WAIT(wait); + int err; + + /* True wake-one mechanism for incoming connections: only + * one process gets woken up, not the 'whole herd'. + * Since we do not 'race & poll' for established sockets + * anymore, the common case will execute the loop only once. + */ + for (;;) { + prepare_to_wait_exclusive(sk_sleep(sk), &wait, + TASK_INTERRUPTIBLE); + if (timeo && skb_queue_empty(&sk->sk_receive_queue)) { + release_sock(sk); + timeo = schedule_timeout(timeo); + lock_sock(sk); + } + err = 0; + if (!skb_queue_empty(&sk->sk_receive_queue)) + break; + err = -EINVAL; + if (sock->state != SS_LISTENING) + break; + err = sock_intr_errno(timeo); + if (signal_pending(current)) + break; + err = -EAGAIN; + if (!timeo) + break; + } + finish_wait(sk_sleep(sk), &wait); + return err; +} + /** - * accept - wait for connection request + * tipc_accept - wait for connection request * @sock: listening socket * @newsock: new socket that is to be connected * @flags: file-related flags associated with socket * * Returns 0 on success, errno otherwise */ -static int accept(struct socket *sock, struct socket *new_sock, int flags) +static int tipc_accept(struct socket *sock, struct socket *new_sock, int flags) { struct sock *new_sk, *sk = sock->sk; struct sk_buff *buf; struct tipc_sock *new_tsock; - struct tipc_port *new_tport; struct tipc_msg *msg; - u32 new_ref; - + long timeo; int res; lock_sock(sk); @@ -1584,19 +1986,10 @@ static int accept(struct socket *sock, struct socket *new_sock, int flags) res = -EINVAL; goto exit; } - - while (skb_queue_empty(&sk->sk_receive_queue)) { - if (flags & O_NONBLOCK) { - res = -EWOULDBLOCK; - goto exit; - } - release_sock(sk); - res = wait_event_interruptible(*sk_sleep(sk), - (!skb_queue_empty(&sk->sk_receive_queue))); - lock_sock(sk); - if (res) - goto exit; - } + timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); + res = tipc_wait_for_accept(sock, timeo); + if (res) + goto exit; buf = skb_peek(&sk->sk_receive_queue); @@ -1606,8 +1999,6 @@ static int accept(struct socket *sock, struct socket *new_sock, int flags) new_sk = new_sock->sk; new_tsock = tipc_sk(new_sk); - new_tport = new_tsock->p; - new_ref = new_tport->ref; msg = buf_msg(buf); /* we lock on new_sk; but lockdep sees the lock on sk */ @@ -1617,18 +2008,16 @@ static int accept(struct socket *sock, struct socket *new_sock, int flags) * Reject any stray messages received by new socket * before the socket lock was taken (very, very unlikely) */ - reject_rx_queue(new_sk); + tsk_rej_rx_queue(new_sk); /* Connect new socket to it's peer */ - new_tsock->peer_name.ref = msg_origport(msg); - new_tsock->peer_name.node = msg_orignode(msg); - tipc_connect(new_ref, &new_tsock->peer_name); + tipc_sk_finish_conn(new_tsock, msg_origport(msg), msg_orignode(msg)); new_sock->state = SS_CONNECTED; - tipc_set_portimportance(new_ref, msg_importance(msg)); + tsk_set_importance(new_tsock, msg_importance(msg)); if (msg_named(msg)) { - new_tport->conn_type = msg_nametype(msg); - new_tport->conn_instance = msg_nameinst(msg); + new_tsock->conn_type = msg_nametype(msg); + new_tsock->conn_instance = msg_nameinst(msg); } /* @@ -1638,22 +2027,21 @@ static int accept(struct socket *sock, struct socket *new_sock, int flags) if (!msg_data_sz(msg)) { struct msghdr m = {NULL,}; - advance_rx_queue(sk); - send_packet(NULL, new_sock, &m, 0); + tsk_advance_rx_queue(sk); + tipc_send_packet(NULL, new_sock, &m, 0); } else { __skb_dequeue(&sk->sk_receive_queue); __skb_queue_head(&new_sk->sk_receive_queue, buf); skb_set_owner_r(buf, new_sk); } release_sock(new_sk); - exit: release_sock(sk); return res; } /** - * shutdown - shutdown socket connection + * tipc_shutdown - shutdown socket connection * @sock: socket structure * @how: direction to close (must be SHUT_RDWR) * @@ -1661,11 +2049,12 @@ exit: * * Returns 0 on success, errno otherwise */ -static int shutdown(struct socket *sock, int how) +static int tipc_shutdown(struct socket *sock, int how) { struct sock *sk = sock->sk; - struct tipc_port *tport = tipc_sk_port(sk); + struct tipc_sock *tsk = tipc_sk(sk); struct sk_buff *buf; + u32 dnode; int res; if (how != SHUT_RDWR) @@ -1685,14 +2074,21 @@ restart: kfree_skb(buf); goto restart; } - tipc_disconnect(tport->ref); - tipc_reject_msg(buf, TIPC_CONN_SHUTDOWN); + if (tipc_msg_reverse(buf, &dnode, TIPC_CONN_SHUTDOWN)) + tipc_link_xmit(buf, dnode, tsk->ref); + tipc_node_remove_conn(dnode, tsk->ref); } else { - tipc_shutdown(tport->ref); + dnode = tsk_peer_node(tsk); + buf = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, + TIPC_CONN_MSG, SHORT_H_SIZE, + 0, dnode, tipc_own_addr, + tsk_peer_port(tsk), + tsk->ref, TIPC_CONN_SHUTDOWN); + tipc_link_xmit(buf, dnode, tsk->ref); } - + tsk->connected = 0; sock->state = SS_DISCONNECTING; - + tipc_node_remove_conn(dnode, tsk->ref); /* fall through */ case SS_DISCONNECTING: @@ -1713,8 +2109,434 @@ restart: return res; } +static void tipc_sk_timeout(unsigned long ref) +{ + struct tipc_sock *tsk; + struct sock *sk; + struct sk_buff *buf = NULL; + u32 peer_port, peer_node; + + tsk = tipc_sk_get(ref); + if (!tsk) + return; + + sk = &tsk->sk; + bh_lock_sock(sk); + if (!tsk->connected) { + bh_unlock_sock(sk); + goto exit; + } + peer_port = tsk_peer_port(tsk); + peer_node = tsk_peer_node(tsk); + + if (tsk->probing_state == TIPC_CONN_PROBING) { + /* Previous probe not answered -> self abort */ + buf = 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); + } else { + buf = tipc_msg_create(CONN_MANAGER, CONN_PROBE, INT_H_SIZE, + 0, peer_node, tipc_own_addr, + peer_port, ref, TIPC_OK); + tsk->probing_state = TIPC_CONN_PROBING; + k_start_timer(&tsk->timer, tsk->probing_interval); + } + bh_unlock_sock(sk); + if (buf) + tipc_link_xmit(buf, peer_node, ref); +exit: + tipc_sk_put(tsk); +} + +static int tipc_sk_publish(struct tipc_sock *tsk, uint scope, + struct tipc_name_seq const *seq) +{ + struct publication *publ; + u32 key; + + if (tsk->connected) + return -EINVAL; + key = tsk->ref + tsk->pub_count + 1; + if (key == tsk->ref) + return -EADDRINUSE; + + publ = tipc_nametbl_publish(seq->type, seq->lower, seq->upper, + scope, tsk->ref, key); + if (unlikely(!publ)) + return -EINVAL; + + list_add(&publ->pport_list, &tsk->publications); + tsk->pub_count++; + tsk->published = 1; + return 0; +} + +static int tipc_sk_withdraw(struct tipc_sock *tsk, uint scope, + struct tipc_name_seq const *seq) +{ + struct publication *publ; + struct publication *safe; + int rc = -EINVAL; + + list_for_each_entry_safe(publ, safe, &tsk->publications, pport_list) { + if (seq) { + if (publ->scope != scope) + continue; + if (publ->type != seq->type) + continue; + if (publ->lower != seq->lower) + continue; + if (publ->upper != seq->upper) + break; + tipc_nametbl_withdraw(publ->type, publ->lower, + publ->ref, publ->key); + rc = 0; + break; + } + tipc_nametbl_withdraw(publ->type, publ->lower, + publ->ref, publ->key); + rc = 0; + } + if (list_empty(&tsk->publications)) + tsk->published = 0; + 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) +{ + struct tipc_msg *msg; + u32 ref = 0; + struct tipc_sock *tsk = tipc_sk_get_next(&ref); + + 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); + } +} + +/** + * 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); + /** - * setsockopt - set socket option + * tipc_ref_table_init - create reference table for sockets + */ +int tipc_sk_ref_table_init(u32 req_sz, u32 start) +{ + 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; + + return 0; +} + +/** + * 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; +} + +/* 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) +{ + 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; + } + + if (ref) { + entry->ref = ref; + entry->tsk = tsk; + } + write_unlock_bh(&ref_table_lock); + return ref; +} + +/* 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) +{ + 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); + + if (unlikely(!entry->tsk)) { + pr_err("Attempt to discard ref. to non-existent socket\n"); + goto exit; + } + 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) +{ + struct reference *entry; + struct tipc_sock *tsk; + + 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; +} + +/* tipc_sk_get_next - lock & return next socket after referenced one +*/ +struct tipc_sock *tipc_sk_get_next(u32 *ref) +{ + struct reference *entry; + struct tipc_sock *tsk = NULL; + uint index = *ref & tipc_ref_table.index_mask; + + 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; +} + +static void tipc_sk_put(struct tipc_sock *tsk) +{ + sock_put(&tsk->sk); +} + +/** + * tipc_setsockopt - set socket option * @sock: socket structure * @lvl: option level * @opt: option identifier @@ -1726,11 +2548,11 @@ restart: * * Returns 0 on success, errno otherwise */ -static int setsockopt(struct socket *sock, int lvl, int opt, char __user *ov, - unsigned int ol) +static int tipc_setsockopt(struct socket *sock, int lvl, int opt, + char __user *ov, unsigned int ol) { struct sock *sk = sock->sk; - struct tipc_port *tport = tipc_sk_port(sk); + struct tipc_sock *tsk = tipc_sk(sk); u32 value; int res; @@ -1748,16 +2570,16 @@ static int setsockopt(struct socket *sock, int lvl, int opt, char __user *ov, switch (opt) { case TIPC_IMPORTANCE: - res = tipc_set_portimportance(tport->ref, value); + res = tsk_set_importance(tsk, value); break; case TIPC_SRC_DROPPABLE: if (sock->type != SOCK_STREAM) - res = tipc_set_portunreliable(tport->ref, value); + tsk_set_unreliable(tsk, value); else res = -ENOPROTOOPT; break; case TIPC_DEST_DROPPABLE: - res = tipc_set_portunreturnable(tport->ref, value); + tsk_set_unreturnable(tsk, value); break; case TIPC_CONN_TIMEOUT: tipc_sk(sk)->conn_timeout = value; @@ -1773,7 +2595,7 @@ static int setsockopt(struct socket *sock, int lvl, int opt, char __user *ov, } /** - * getsockopt - get socket option + * tipc_getsockopt - get socket option * @sock: socket structure * @lvl: option level * @opt: option identifier @@ -1785,11 +2607,11 @@ static int setsockopt(struct socket *sock, int lvl, int opt, char __user *ov, * * Returns 0 on success, errno otherwise */ -static int getsockopt(struct socket *sock, int lvl, int opt, char __user *ov, - int __user *ol) +static int tipc_getsockopt(struct socket *sock, int lvl, int opt, + char __user *ov, int __user *ol) { struct sock *sk = sock->sk; - struct tipc_port *tport = tipc_sk_port(sk); + struct tipc_sock *tsk = tipc_sk(sk); int len; u32 value; int res; @@ -1806,16 +2628,16 @@ static int getsockopt(struct socket *sock, int lvl, int opt, char __user *ov, switch (opt) { case TIPC_IMPORTANCE: - res = tipc_portimportance(tport->ref, &value); + value = tsk_importance(tsk); break; case TIPC_SRC_DROPPABLE: - res = tipc_portunreliable(tport->ref, &value); + value = tsk_unreliable(tsk); break; case TIPC_DEST_DROPPABLE: - res = tipc_portunreturnable(tport->ref, &value); + value = tsk_unreturnable(tsk); break; case TIPC_CONN_TIMEOUT: - value = tipc_sk(sk)->conn_timeout; + value = tsk->conn_timeout; /* no need to set "res", since already 0 at this point */ break; case TIPC_NODE_RECVQ_DEPTH: @@ -1842,25 +2664,46 @@ static int getsockopt(struct socket *sock, int lvl, int opt, char __user *ov, return put_user(sizeof(value), ol); } +static int tipc_ioctl(struct socket *sk, unsigned int cmd, unsigned long arg) +{ + struct tipc_sioc_ln_req lnr; + void __user *argp = (void __user *)arg; + + switch (cmd) { + case SIOCGETLINKNAME: + if (copy_from_user(&lnr, argp, sizeof(lnr))) + return -EFAULT; + if (!tipc_node_get_linkname(lnr.bearer_id, lnr.peer, + lnr.linkname, TIPC_MAX_LINK_NAME)) { + if (copy_to_user(argp, &lnr, sizeof(lnr))) + return -EFAULT; + return 0; + } + return -EADDRNOTAVAIL; + default: + return -ENOIOCTLCMD; + } +} + /* Protocol switches for the various types of TIPC sockets */ static const struct proto_ops msg_ops = { .owner = THIS_MODULE, .family = AF_TIPC, - .release = release, - .bind = bind, - .connect = connect, + .release = tipc_release, + .bind = tipc_bind, + .connect = tipc_connect, .socketpair = sock_no_socketpair, .accept = sock_no_accept, - .getname = get_name, - .poll = poll, - .ioctl = sock_no_ioctl, + .getname = tipc_getname, + .poll = tipc_poll, + .ioctl = tipc_ioctl, .listen = sock_no_listen, - .shutdown = shutdown, - .setsockopt = setsockopt, - .getsockopt = getsockopt, - .sendmsg = send_msg, - .recvmsg = recv_msg, + .shutdown = tipc_shutdown, + .setsockopt = tipc_setsockopt, + .getsockopt = tipc_getsockopt, + .sendmsg = tipc_sendmsg, + .recvmsg = tipc_recvmsg, .mmap = sock_no_mmap, .sendpage = sock_no_sendpage }; @@ -1868,20 +2711,20 @@ static const struct proto_ops msg_ops = { static const struct proto_ops packet_ops = { .owner = THIS_MODULE, .family = AF_TIPC, - .release = release, - .bind = bind, - .connect = connect, + .release = tipc_release, + .bind = tipc_bind, + .connect = tipc_connect, .socketpair = sock_no_socketpair, - .accept = accept, - .getname = get_name, - .poll = poll, - .ioctl = sock_no_ioctl, - .listen = listen, - .shutdown = shutdown, - .setsockopt = setsockopt, - .getsockopt = getsockopt, - .sendmsg = send_packet, - .recvmsg = recv_msg, + .accept = tipc_accept, + .getname = tipc_getname, + .poll = tipc_poll, + .ioctl = tipc_ioctl, + .listen = tipc_listen, + .shutdown = tipc_shutdown, + .setsockopt = tipc_setsockopt, + .getsockopt = tipc_getsockopt, + .sendmsg = tipc_send_packet, + .recvmsg = tipc_recvmsg, .mmap = sock_no_mmap, .sendpage = sock_no_sendpage }; @@ -1889,20 +2732,20 @@ static const struct proto_ops packet_ops = { static const struct proto_ops stream_ops = { .owner = THIS_MODULE, .family = AF_TIPC, - .release = release, - .bind = bind, - .connect = connect, + .release = tipc_release, + .bind = tipc_bind, + .connect = tipc_connect, .socketpair = sock_no_socketpair, - .accept = accept, - .getname = get_name, - .poll = poll, - .ioctl = sock_no_ioctl, - .listen = listen, - .shutdown = shutdown, - .setsockopt = setsockopt, - .getsockopt = getsockopt, - .sendmsg = send_stream, - .recvmsg = recv_stream, + .accept = tipc_accept, + .getname = tipc_getname, + .poll = tipc_poll, + .ioctl = tipc_ioctl, + .listen = tipc_listen, + .shutdown = tipc_shutdown, + .setsockopt = tipc_setsockopt, + .getsockopt = tipc_getsockopt, + .sendmsg = tipc_send_stream, + .recvmsg = tipc_recv_stream, .mmap = sock_no_mmap, .sendpage = sock_no_sendpage }; @@ -1947,8 +2790,6 @@ int tipc_socket_init(void) proto_unregister(&tipc_proto); goto out; } - - sockets_enabled = 1; out: return res; } @@ -1958,10 +2799,6 @@ int tipc_socket_init(void) */ void tipc_socket_stop(void) { - if (!sockets_enabled) - return; - - sockets_enabled = 0; sock_unregister(tipc_family_ops.family); proto_unregister(&tipc_proto); } diff --git a/net/tipc/ref.h b/net/tipc/socket.h index 5bc8e7ab84de..baa43d03901e 100644 --- a/net/tipc/ref.h +++ b/net/tipc/socket.h @@ -1,8 +1,6 @@ -/* - * net/tipc/ref.h: Include file for TIPC object registry code +/* net/tipc/socket.h: Include file for TIPC socket code * - * Copyright (c) 1991-2006, Ericsson AB - * Copyright (c) 2005-2006, Wind River Systems + * Copyright (c) 2014, Ericsson AB * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -34,16 +32,20 @@ * POSSIBILITY OF SUCH DAMAGE. */ -#ifndef _TIPC_REF_H -#define _TIPC_REF_H +#ifndef _TIPC_SOCK_H +#define _TIPC_SOCK_H -int tipc_ref_table_init(u32 requested_size, u32 start); -void tipc_ref_table_stop(void); +#include <net/sock.h> -u32 tipc_ref_acquire(void *object, spinlock_t **lock); -void tipc_ref_discard(u32 ref); - -void *tipc_ref_lock(u32 ref); -void *tipc_ref_deref(u32 ref); +#define TIPC_CONNACK_INTV 256 +#define TIPC_FLOWCTRL_WIN (TIPC_CONNACK_INTV * 2) +#define TIPC_CONN_OVERLOAD_LIMIT ((TIPC_FLOWCTRL_WIN * 2 + 1) * \ + SKB_TRUESIZE(TIPC_MAX_USER_MSG_SIZE)) +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); #endif diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c index d38bb45d82e9..31b5cb232a43 100644 --- a/net/tipc/subscr.c +++ b/net/tipc/subscr.c @@ -36,13 +36,12 @@ #include "core.h" #include "name_table.h" -#include "port.h" #include "subscr.h" /** * struct tipc_subscriber - TIPC network topology subscriber * @conid: connection identifier to server connecting to subscriber - * @lock: controll access to subscriber + * @lock: control access to subscriber * @subscription_list: list of subscription objects for this subscriber */ struct tipc_subscriber { @@ -96,20 +95,16 @@ static void subscr_send_event(struct tipc_subscription *sub, u32 found_lower, { struct tipc_subscriber *subscriber = sub->subscriber; struct kvec msg_sect; - int ret; msg_sect.iov_base = (void *)&sub->evt; msg_sect.iov_len = sizeof(struct tipc_event); - sub->evt.event = htohl(event, sub->swap); sub->evt.found_lower = htohl(found_lower, sub->swap); 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); - ret = tipc_conn_sendmsg(&topsrv, subscriber->conid, NULL, - msg_sect.iov_base, msg_sect.iov_len); - if (ret < 0) - pr_err("Sending subscription event failed, no memory\n"); + tipc_conn_sendmsg(&topsrv, subscriber->conid, NULL, msg_sect.iov_base, + msg_sect.iov_len); } /** @@ -153,14 +148,6 @@ static void subscr_timeout(struct tipc_subscription *sub) /* The spin lock per subscriber is used to protect its members */ spin_lock_bh(&subscriber->lock); - /* Validate if the connection related to the subscriber is - * closed (in case subscriber is terminating) - */ - if (subscriber->conid == 0) { - spin_unlock_bh(&subscriber->lock); - return; - } - /* Validate timeout (in case subscription is being cancelled) */ if (sub->timeout == TIPC_WAIT_FOREVER) { spin_unlock_bh(&subscriber->lock); @@ -215,9 +202,6 @@ static void subscr_release(struct tipc_subscriber *subscriber) spin_lock_bh(&subscriber->lock); - /* Invalidate subscriber reference */ - subscriber->conid = 0; - /* Destroy any existing subscriptions for subscriber */ list_for_each_entry_safe(sub, sub_temp, &subscriber->subscription_list, subscription_list) { @@ -278,9 +262,9 @@ static void subscr_cancel(struct tipc_subscr *s, * * Called with subscriber lock held. */ -static struct tipc_subscription *subscr_subscribe(struct tipc_subscr *s, - struct tipc_subscriber *subscriber) -{ +static int subscr_subscribe(struct tipc_subscr *s, + struct tipc_subscriber *subscriber, + struct tipc_subscription **sub_p) { struct tipc_subscription *sub; int swap; @@ -291,23 +275,21 @@ static struct tipc_subscription *subscr_subscribe(struct tipc_subscr *s, if (s->filter & htohl(TIPC_SUB_CANCEL, swap)) { s->filter &= ~htohl(TIPC_SUB_CANCEL, swap); subscr_cancel(s, subscriber); - return NULL; + return 0; } /* Refuse subscription if global limit exceeded */ if (atomic_read(&subscription_count) >= TIPC_MAX_SUBSCRIPTIONS) { pr_warn("Subscription rejected, limit reached (%u)\n", TIPC_MAX_SUBSCRIPTIONS); - subscr_terminate(subscriber); - return NULL; + return -EINVAL; } /* Allocate subscription object */ sub = kmalloc(sizeof(*sub), GFP_ATOMIC); if (!sub) { pr_warn("Subscription rejected, no memory\n"); - subscr_terminate(subscriber); - return NULL; + return -ENOMEM; } /* Initialize subscription object */ @@ -321,8 +303,7 @@ static struct tipc_subscription *subscr_subscribe(struct tipc_subscr *s, (sub->seq.lower > sub->seq.upper)) { pr_warn("Subscription rejected, illegal request\n"); kfree(sub); - subscr_terminate(subscriber); - return NULL; + return -EINVAL; } INIT_LIST_HEAD(&sub->nameseq_list); list_add(&sub->subscription_list, &subscriber->subscription_list); @@ -335,8 +316,8 @@ static struct tipc_subscription *subscr_subscribe(struct tipc_subscr *s, (Handler)subscr_timeout, (unsigned long)sub); k_start_timer(&sub->timer, sub->timeout); } - - return sub; + *sub_p = sub; + return 0; } /* Handle one termination request for the subscriber */ @@ -350,10 +331,14 @@ static void subscr_conn_msg_event(int conid, struct sockaddr_tipc *addr, void *usr_data, void *buf, size_t len) { struct tipc_subscriber *subscriber = usr_data; - struct tipc_subscription *sub; + struct tipc_subscription *sub = NULL; spin_lock_bh(&subscriber->lock); - sub = subscr_subscribe((struct tipc_subscr *)buf, subscriber); + if (subscr_subscribe((struct tipc_subscr *)buf, subscriber, &sub) < 0) { + spin_unlock_bh(&subscriber->lock); + subscr_terminate(subscriber); + return; + } if (sub) tipc_nametbl_subscribe(sub); spin_unlock_bh(&subscriber->lock); diff --git a/net/tipc/sysctl.c b/net/tipc/sysctl.c index f3fef93325a8..1a779b1e8510 100644 --- a/net/tipc/sysctl.c +++ b/net/tipc/sysctl.c @@ -47,6 +47,13 @@ static struct ctl_table tipc_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .procname = "named_timeout", + .data = &sysctl_tipc_named_timeout, + .maxlen = sizeof(sysctl_tipc_named_timeout), + .mode = 0644, + .proc_handler = proc_dointvec, + }, {} }; |