@@ -6,7 +6,7 @@ obj-$(CONFIG_TIPC) := tipc.o
tipc-y += addr.o bcast.o bearer.o \
core.o link.o discover.o msg.o \
- name_distr.o subscr.o name_table.o net.o \
+ name_distr.o subscr.o monitor.o name_table.o net.o \
netlink.o netlink_compat.o node.o socket.o eth_media.o \
server.o socket.o
@@ -73,4 +73,5 @@ int tipc_addr_node_valid(u32 addr);
int tipc_in_scope(u32 domain, u32 addr);
int tipc_addr_scope(u32 domain);
char *tipc_addr_string_fill(char *string, u32 addr);
+
#endif
@@ -1,7 +1,7 @@
/*
* net/tipc/bearer.c: TIPC bearer code
*
- * Copyright (c) 1996-2006, 2013-2014, Ericsson AB
+ * Copyright (c) 1996-2006, 2013-2016, Ericsson AB
* Copyright (c) 2004-2006, 2010-2013, Wind River Systems
* All rights reserved.
*
@@ -39,6 +39,7 @@
#include "bearer.h"
#include "link.h"
#include "discover.h"
+#include "monitor.h"
#include "bcast.h"
#include "netlink.h"
@@ -313,6 +314,10 @@ restart:
rcu_assign_pointer(tn->bearer_list[bearer_id], b);
if (skb)
tipc_bearer_xmit_skb(net, bearer_id, skb, &b->bcast_addr);
+
+ if (tipc_mon_create(net, bearer_id))
+ return -ENOMEM;
+
pr_info("Enabled bearer <%s>, discovery domain %s, priority %u\n",
name,
tipc_addr_string_fill(addr_string, disc_domain), priority);
@@ -348,6 +353,7 @@ static void bearer_disable(struct net *net, struct tipc_bearer *b)
tipc_disc_delete(b->link_req);
RCU_INIT_POINTER(tn->bearer_list[bearer_id], NULL);
kfree_rcu(b, rcu);
+ tipc_mon_delete(net, bearer_id);
}
int tipc_enable_l2_media(struct net *net, struct tipc_bearer *b,
@@ -1,7 +1,7 @@
/*
* net/tipc/bearer.h: Include file for TIPC bearer code
*
- * Copyright (c) 1996-2006, 2013-2014, Ericsson AB
+ * Copyright (c) 1996-2006, 2013-2016, Ericsson AB
* Copyright (c) 2005, 2010-2011, Wind River Systems
* All rights reserved.
*
@@ -57,6 +57,7 @@ static int __net_init tipc_init_net(struct net *net)
tn->net_id = 4711;
tn->own_addr = 0;
+ tn->mon_threshold = TIPC_DEF_MON_THRESHOLD;
get_random_bytes(&tn->random, sizeof(int));
INIT_LIST_HEAD(&tn->node_list);
spin_lock_init(&tn->node_list_lock);
@@ -66,11 +66,13 @@ struct tipc_bc_base;
struct tipc_link;
struct tipc_name_table;
struct tipc_server;
+struct tipc_monitor;
#define TIPC_MOD_VER "2.0.0"
-#define NODE_HTABLE_SIZE 512
-#define MAX_BEARERS 3
+#define NODE_HTABLE_SIZE 512
+#define MAX_BEARERS 3
+#define TIPC_DEF_MON_THRESHOLD 32
extern int tipc_net_id __read_mostly;
extern int sysctl_tipc_rmem[3] __read_mostly;
@@ -88,6 +90,10 @@ struct tipc_net {
u32 num_nodes;
u32 num_links;
+ /* Neighbor monitoring list */
+ struct tipc_monitor *monitors[MAX_BEARERS];
+ int mon_threshold;
+
/* Bearer list */
struct tipc_bearer __rcu *bearer_list[MAX_BEARERS + 1];
@@ -126,6 +132,11 @@ static inline struct list_head *tipc_nodes(struct net *net)
return &tipc_net(net)->node_list;
}
+static inline unsigned int tipc_hashfn(u32 addr)
+{
+ return addr & (NODE_HTABLE_SIZE - 1);
+}
+
static inline u16 mod(u16 x)
{
return x & 0xffffu;
@@ -42,6 +42,7 @@
#include "name_distr.h"
#include "discover.h"
#include "netlink.h"
+#include "monitor.h"
#include <linux/pkt_sched.h>
@@ -95,6 +96,7 @@ struct tipc_stats {
* @pmsg: convenience pointer to "proto_msg" field
* @priority: current link priority
* @net_plane: current link network plane ('A' through 'H')
+ * @mon_state: cookie with information needed by link monitor
* @backlog_limit: backlog queue congestion thresholds (indexed by importance)
* @exp_msg_count: # of tunnelled messages expected during link changeover
* @reset_rcv_checkpt: seq # of last acknowledged message at time of link reset
@@ -138,6 +140,7 @@ struct tipc_link {
char if_name[TIPC_MAX_IF_NAME];
u32 priority;
char net_plane;
+ struct tipc_mon_state mon_state;
u16 rst_cnt;
/* Failover/synch */
@@ -708,18 +711,25 @@ int tipc_link_timeout(struct tipc_link *l, struct sk_buff_head *xmitq)
bool setup = false;
u16 bc_snt = l->bc_sndlink->snd_nxt - 1;
u16 bc_acked = l->bc_rcvlink->acked;
-
- link_profile_stats(l);
+ struct tipc_mon_state *mstate = &l->mon_state;
switch (l->state) {
case LINK_ESTABLISHED:
case LINK_SYNCHING:
- if (l->silent_intv_cnt > l->abort_limit)
- return tipc_link_fsm_evt(l, LINK_FAILURE_EVT);
mtyp = STATE_MSG;
+ link_profile_stats(l);
+ tipc_mon_get_state(l->net, l->addr, mstate, l->bearer_id);
+ if (mstate->reset || (l->silent_intv_cnt > l->abort_limit))
+ return tipc_link_fsm_evt(l, LINK_FAILURE_EVT);
state = bc_acked != bc_snt;
- probe = l->silent_intv_cnt;
- l->silent_intv_cnt++;
+ state |= l->bc_rcvlink->rcv_unacked;
+ state |= l->rcv_unacked;
+ state |= !skb_queue_empty(&l->transmq);
+ state |= !skb_queue_empty(&l->deferdq);
+ probe = mstate->probing;
+ probe |= l->silent_intv_cnt;
+ if (probe || mstate->monitoring)
+ l->silent_intv_cnt++;
break;
case LINK_RESET:
setup = l->rst_cnt++ <= 4;
@@ -830,6 +840,7 @@ void tipc_link_reset(struct tipc_link *l)
l->stats.recv_info = 0;
l->stale_count = 0;
l->bc_peer_is_up = false;
+ memset(&l->mon_state, 0, sizeof(l->mon_state));
tipc_link_reset_stats(l);
}
@@ -1238,6 +1249,9 @@ static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe,
struct tipc_msg *hdr;
struct sk_buff_head *dfq = &l->deferdq;
bool node_up = link_is_up(l->bc_rcvlink);
+ struct tipc_mon_state *mstate = &l->mon_state;
+ int dlen = 0;
+ void *data;
/* Don't send protocol message during reset or link failover */
if (tipc_link_is_blocked(l))
@@ -1250,12 +1264,13 @@ static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe,
rcvgap = buf_seqno(skb_peek(dfq)) - l->rcv_nxt;
skb = tipc_msg_create(LINK_PROTOCOL, mtyp, INT_H_SIZE,
- TIPC_MAX_IF_NAME, l->addr,
+ tipc_max_domain_size, l->addr,
tipc_own_addr(l->net), 0, 0, 0);
if (!skb)
return;
hdr = buf_msg(skb);
+ data = msg_data(hdr);
msg_set_session(hdr, l->session);
msg_set_bearer_id(hdr, l->bearer_id);
msg_set_net_plane(hdr, l->net_plane);
@@ -1271,14 +1286,18 @@ static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe,
if (mtyp == STATE_MSG) {
msg_set_seq_gap(hdr, rcvgap);
- msg_set_size(hdr, INT_H_SIZE);
msg_set_probe(hdr, probe);
+ tipc_mon_prep(l->net, data, &dlen, mstate, l->bearer_id);
+ msg_set_size(hdr, INT_H_SIZE + dlen);
+ skb_trim(skb, INT_H_SIZE + dlen);
l->stats.sent_states++;
l->rcv_unacked = 0;
} else {
/* RESET_MSG or ACTIVATE_MSG */
msg_set_max_pkt(hdr, l->advertised_mtu);
- strcpy(msg_data(hdr), l->if_name);
+ strcpy(data, l->if_name);
+ msg_set_size(hdr, INT_H_SIZE + TIPC_MAX_IF_NAME);
+ skb_trim(skb, INT_H_SIZE + TIPC_MAX_IF_NAME);
}
if (probe)
l->stats.sent_probes++;
@@ -1371,7 +1390,9 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
u16 peers_tol = msg_link_tolerance(hdr);
u16 peers_prio = msg_linkprio(hdr);
u16 rcv_nxt = l->rcv_nxt;
+ u16 dlen = msg_data_sz(hdr);
int mtyp = msg_type(hdr);
+ void *data;
char *if_name;
int rc = 0;
@@ -1381,6 +1402,10 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
if (tipc_own_addr(l->net) > msg_prevnode(hdr))
l->net_plane = msg_net_plane(hdr);
+ skb_linearize(skb);
+ hdr = buf_msg(skb);
+ data = msg_data(hdr);
+
switch (mtyp) {
case RESET_MSG:
@@ -1391,8 +1416,6 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
/* fall thru' */
case ACTIVATE_MSG:
- skb_linearize(skb);
- hdr = buf_msg(skb);
/* Complete own link name with peer's interface name */
if_name = strrchr(l->name, ':') + 1;
@@ -1400,7 +1423,7 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
break;
if (msg_data_sz(hdr) < TIPC_MAX_IF_NAME)
break;
- strncpy(if_name, msg_data(hdr), TIPC_MAX_IF_NAME);
+ strncpy(if_name, data, TIPC_MAX_IF_NAME);
/* Update own tolerance if peer indicates a non-zero value */
if (in_range(peers_tol, TIPC_MIN_LINK_TOL, TIPC_MAX_LINK_TOL))
@@ -1448,6 +1471,8 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
rc = TIPC_LINK_UP_EVT;
break;
}
+ tipc_mon_rcv(l->net, data, dlen, l->addr,
+ &l->mon_state, l->bearer_id);
/* Send NACK if peer has sent pkts we haven't received yet */
if (more(peers_snd_nxt, rcv_nxt) && !tipc_link_is_synching(l))
new file mode 100644
@@ -0,0 +1,651 @@
+/*
+ * net/tipc/monitor.c
+ *
+ * Copyright (c) 2016, Ericsson AB
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * 3. Neither the names of the copyright holders nor the names of its
+ * contributors may be used to endorse or promote products derived from
+ * this software without specific prior written permission.
+ *
+ * Alternatively, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") version 2 as published by the Free
+ * Software Foundation.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include "core.h"
+#include "addr.h"
+#include "monitor.h"
+
+#define MAX_MON_DOMAIN 64
+#define MON_TIMEOUT 120000
+#define MAX_PEER_DOWN_EVENTS 4
+
+/* struct tipc_mon_domain: domain record to be transferred between peers
+ * @len: actual size of domain record
+ * @gen: current generation of sender's domain
+ * @ack_gen: most recent generation of self's domain acked by peer
+ * @member_cnt: number of domain member nodes described in this record
+ * @up_map: bit map indicating which of the members the sender considers up
+ * @members: identity of the domain members
+ */
+struct tipc_mon_domain {
+ u16 len;
+ u16 gen;
+ u16 ack_gen;
+ u16 member_cnt;
+ u64 up_map;
+ u32 members[MAX_MON_DOMAIN];
+};
+
+/* struct tipc_peer: state of a peer node and its domain
+ * @addr: tipc node identity of peer
+ * @head_map: shows which other nodes currently consider peer 'up'
+ * @domain: most recent domain record from peer
+ * @hash: position in hashed lookup list
+ * @list: position in linked list, in circular ascending order by 'addr'
+ * @applied: number of reported domain members applied on this monitor list
+ * @is_up: peer is up as seen from this node
+ * @is_head: peer is assigned domain head as seen from this node
+ * @is_local: peer is in local domain and should be continuously monitored
+ * @down_cnt: - numbers of other peers which have reported this on lost
+ */
+struct tipc_peer {
+ u32 addr;
+ struct tipc_mon_domain *domain;
+ struct hlist_node hash;
+ struct list_head list;
+ u8 applied;
+ u8 down_cnt;
+ bool is_up;
+ bool is_head;
+ bool is_local;
+};
+
+struct tipc_monitor {
+ struct hlist_head peers[NODE_HTABLE_SIZE];
+ int peer_cnt;
+ struct tipc_peer *self;
+ rwlock_t lock;
+ struct tipc_mon_domain cache;
+ u16 list_gen;
+ u16 dom_gen;
+ struct net *net;
+ struct timer_list timer;
+ unsigned long timer_intv;
+};
+
+static struct tipc_monitor *tipc_monitor(struct net *net, int bearer_id)
+{
+ return tipc_net(net)->monitors[bearer_id];
+}
+
+const int tipc_max_domain_size = sizeof(struct tipc_mon_domain);
+
+/* dom_rec_len(): actual length of domain record for transport
+ */
+static int dom_rec_len(struct tipc_mon_domain *dom, u16 mcnt)
+{
+ return ((void *)&dom->members - (void *)dom) + (mcnt * sizeof(u32));
+}
+
+/* dom_size() : calculate size of own domain based on number of peers
+ */
+static int dom_size(int peers)
+{
+ int i = 0;
+
+ while ((i * i) < peers)
+ i++;
+ return i < MAX_MON_DOMAIN ? i : MAX_MON_DOMAIN;
+}
+
+static void map_set(u64 *up_map, int i, unsigned int v)
+{
+ *up_map &= ~(1 << i);
+ *up_map |= (v << i);
+}
+
+static int map_get(u64 up_map, int i)
+{
+ return (up_map & (1 << i)) >> i;
+}
+
+static struct tipc_peer *peer_prev(struct tipc_peer *peer)
+{
+ return list_last_entry(&peer->list, struct tipc_peer, list);
+}
+
+static struct tipc_peer *peer_nxt(struct tipc_peer *peer)
+{
+ return list_first_entry(&peer->list, struct tipc_peer, list);
+}
+
+static struct tipc_peer *peer_head(struct tipc_peer *peer)
+{
+ while (!peer->is_head)
+ peer = peer_prev(peer);
+ return peer;
+}
+
+static struct tipc_peer *get_peer(struct tipc_monitor *mon, u32 addr)
+{
+ struct tipc_peer *peer;
+ unsigned int thash = tipc_hashfn(addr);
+
+ hlist_for_each_entry(peer, &mon->peers[thash], hash) {
+ if (peer->addr == addr)
+ return peer;
+ }
+ return NULL;
+}
+
+static struct tipc_peer *get_self(struct net *net, int bearer_id)
+{
+ struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
+
+ return mon->self;
+}
+
+static inline bool tipc_mon_is_active(struct net *net, struct tipc_monitor *mon)
+{
+ struct tipc_net *tn = tipc_net(net);
+
+ return mon->peer_cnt > tn->mon_threshold;
+}
+
+/* mon_identify_lost_members() : - identify amd mark potentially lost members
+ */
+static void mon_identify_lost_members(struct tipc_peer *peer,
+ struct tipc_mon_domain *dom_bef,
+ int applied_bef)
+{
+ struct tipc_peer *member = peer;
+ struct tipc_mon_domain *dom_aft = peer->domain;
+ int applied_aft = peer->applied;
+ int i;
+
+ for (i = 0; i < applied_bef; i++) {
+ member = peer_nxt(member);
+
+ /* Do nothing if self or peer already see member as down */
+ if (!member->is_up || !map_get(dom_bef->up_map, i))
+ continue;
+
+ /* Loss of local node must be detected by active probing */
+ if (member->is_local)
+ continue;
+
+ /* Start probing if member was removed from applied domain */
+ if (!applied_aft || (applied_aft < i)) {
+ member->down_cnt = 1;
+ continue;
+ }
+
+ /* Member loss is confirmed if it is still in applied domain */
+ if (!map_get(dom_aft->up_map, i))
+ member->down_cnt++;
+ }
+}
+
+/* mon_apply_domain() : match a peer's domain record against monitor list
+ */
+static void mon_apply_domain(struct tipc_monitor *mon,
+ struct tipc_peer *peer)
+{
+ struct tipc_mon_domain *dom = peer->domain;
+ struct tipc_peer *member;
+ u32 addr;
+ int i;
+
+ if (!dom || !peer->is_up)
+ return;
+
+ /* Scan across domain members and match against monitor list */
+ peer->applied = 0;
+ member = peer_nxt(peer);
+ for (i = 0; i < dom->member_cnt; i++) {
+ addr = dom->members[i];
+ if (addr != member->addr)
+ return;
+ peer->applied++;
+ member = peer_nxt(member);
+ }
+}
+
+/* mon_update_local_domain() : update after peer addition/removal/up/down
+ */
+static void mon_update_local_domain(struct tipc_monitor *mon)
+{
+ struct tipc_peer *self = mon->self;
+ struct tipc_mon_domain *cache = &mon->cache;
+ struct tipc_mon_domain *dom = self->domain;
+ struct tipc_peer *peer = self;
+ u64 prev_up_map = dom->up_map;
+ u16 member_cnt, i;
+ bool diff;
+
+ /* Update local domain size based on current size of cluster */
+ member_cnt = dom_size(mon->peer_cnt) - 1;
+ self->applied = member_cnt;
+
+ /* Update native and cached outgoing local domain records */
+ dom->len = dom_rec_len(dom, member_cnt);
+ diff = dom->member_cnt != member_cnt;
+ dom->member_cnt = member_cnt;
+ for (i = 0; i < member_cnt; i++) {
+ peer = peer_nxt(peer);
+ diff |= dom->members[i] != peer->addr;
+ dom->members[i] = peer->addr;
+ map_set(&dom->up_map, i, peer->is_up);
+ cache->members[i] = htonl(peer->addr);
+ }
+ diff |= dom->up_map != prev_up_map;
+ if (!diff)
+ return;
+ dom->gen = ++mon->dom_gen;
+ cache->len = htons(dom->len);
+ cache->gen = htons(dom->gen);
+ cache->member_cnt = htons(member_cnt);
+ cache->up_map = cpu_to_be64(dom->up_map);
+ mon_apply_domain(mon, self);
+}
+
+/* mon_update_neighbors() : update preceding neighbors of added/removed peer
+ */
+static void mon_update_neighbors(struct tipc_monitor *mon,
+ struct tipc_peer *peer)
+{
+ int dz, i;
+
+ dz = dom_size(mon->peer_cnt);
+ for (i = 0; i < dz; i++) {
+ mon_apply_domain(mon, peer);
+ peer = peer_prev(peer);
+ }
+}
+
+/* mon_assign_roles() : reassign peer roles after a network change
+ * The monitor list is consistent at this stage; i.e., each peer is monitoring
+ * a set of domain members as matched between domain record and the monitor list
+ */
+static void mon_assign_roles(struct tipc_monitor *mon, struct tipc_peer *head)
+{
+ struct tipc_peer *peer = peer_nxt(head);
+ struct tipc_peer *self = mon->self;
+ int i = 0;
+
+ for (; peer != self; peer = peer_nxt(peer)) {
+ peer->is_local = false;
+
+ /* Update domain member */
+ if (i++ < head->applied) {
+ peer->is_head = false;
+ if (head == self)
+ peer->is_local = true;
+ continue;
+ }
+ /* Assign next domain head */
+ if (!peer->is_up)
+ continue;
+ if (peer->is_head)
+ break;
+ head = peer;
+ head->is_head = true;
+ i = 0;
+ }
+ mon->list_gen++;
+}
+
+void tipc_mon_remove_peer(struct net *net, u32 addr, int bearer_id)
+{
+ struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
+ struct tipc_peer *self = get_self(net, bearer_id);
+ struct tipc_peer *peer, *prev, *head;
+
+ write_lock_bh(&mon->lock);
+ peer = get_peer(mon, addr);
+ if (!peer)
+ goto exit;
+ prev = peer_prev(peer);
+ list_del(&peer->list);
+ hlist_del(&peer->hash);
+ kfree(peer->domain);
+ kfree(peer);
+ mon->peer_cnt--;
+ head = peer_head(prev);
+ if (head == self)
+ mon_update_local_domain(mon);
+ mon_update_neighbors(mon, prev);
+
+ /* Revert to full-mesh monitoring if we reach threshold */
+ if (!tipc_mon_is_active(net, mon)) {
+ list_for_each_entry(peer, &self->list, list) {
+ kfree(peer->domain);
+ peer->domain = NULL;
+ peer->applied = 0;
+ }
+ }
+ mon_assign_roles(mon, head);
+exit:
+ write_unlock_bh(&mon->lock);
+}
+
+static bool tipc_mon_add_peer(struct tipc_monitor *mon, u32 addr,
+ struct tipc_peer **peer)
+{
+ struct tipc_peer *self = mon->self;
+ struct tipc_peer *cur, *prev, *p;
+
+ p = kzalloc(sizeof(*p), GFP_ATOMIC);
+ *peer = p;
+ if (!p)
+ return false;
+ p->addr = addr;
+
+ /* Add new peer to lookup list */
+ INIT_LIST_HEAD(&p->list);
+ hlist_add_head(&p->hash, &mon->peers[tipc_hashfn(addr)]);
+
+ /* Sort new peer into iterator list, in ascending circular order */
+ prev = self;
+ list_for_each_entry(cur, &self->list, list) {
+ if ((addr > prev->addr) && (addr < cur->addr))
+ break;
+ if (((addr < cur->addr) || (addr > prev->addr)) &&
+ (prev->addr > cur->addr))
+ break;
+ prev = cur;
+ }
+ list_add_tail(&p->list, &cur->list);
+ mon->peer_cnt++;
+ mon_update_neighbors(mon, p);
+ return true;
+}
+
+void tipc_mon_peer_up(struct net *net, u32 addr, int bearer_id)
+{
+ struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
+ struct tipc_peer *self = get_self(net, bearer_id);
+ struct tipc_peer *peer, *head;
+
+ write_lock_bh(&mon->lock);
+ peer = get_peer(mon, addr);
+ if (!peer && !tipc_mon_add_peer(mon, addr, &peer))
+ goto exit;
+ peer->is_up = true;
+ head = peer_head(peer);
+ if (head == self)
+ mon_update_local_domain(mon);
+ mon_assign_roles(mon, head);
+exit:
+ write_unlock_bh(&mon->lock);
+}
+
+void tipc_mon_peer_down(struct net *net, u32 addr, int bearer_id)
+{
+ struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
+ struct tipc_peer *self = get_self(net, bearer_id);
+ struct tipc_peer *peer, *head;
+ struct tipc_mon_domain *dom;
+ int applied;
+
+ write_lock_bh(&mon->lock);
+ peer = get_peer(mon, addr);
+ if (!peer) {
+ pr_warn("Mon: unknown link %x/%u DOWN\n", addr, bearer_id);
+ goto exit;
+ }
+ applied = peer->applied;
+ peer->applied = 0;
+ dom = peer->domain;
+ peer->domain = NULL;
+ if (peer->is_head)
+ mon_identify_lost_members(peer, dom, applied);
+ kfree(dom);
+ peer->is_up = false;
+ peer->is_head = false;
+ peer->is_local = false;
+ peer->down_cnt = 0;
+ head = peer_head(peer);
+ if (head == self)
+ mon_update_local_domain(mon);
+ mon_assign_roles(mon, head);
+exit:
+ write_unlock_bh(&mon->lock);
+}
+
+/* tipc_mon_rcv - process monitor domain event message
+ */
+void tipc_mon_rcv(struct net *net, void *data, u16 dlen, u32 addr,
+ struct tipc_mon_state *state, int bearer_id)
+{
+ struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
+ struct tipc_mon_domain *arrv_dom = data;
+ struct tipc_mon_domain dom_bef;
+ struct tipc_mon_domain *dom;
+ struct tipc_peer *peer;
+ u16 new_member_cnt = ntohs(arrv_dom->member_cnt);
+ int new_dlen = dom_rec_len(arrv_dom, new_member_cnt);
+ u16 new_gen = ntohs(arrv_dom->gen);
+ u16 acked_gen = ntohs(arrv_dom->ack_gen);
+ bool probing = state->probing;
+ int i, applied_bef;
+
+ state->probing = false;
+ if (!dlen)
+ return;
+
+ /* Sanity check received domain record */
+ if ((dlen != ntohs(arrv_dom->len)) || (dlen != new_dlen)) {
+ pr_warn_ratelimited("Received illegal domain record\n");
+ return;
+ }
+
+ /* Synch generation numbers with peer if link just came up */
+ if (!state->synched) {
+ state->peer_gen = new_gen - 1;
+ state->acked_gen = acked_gen;
+ state->synched = true;
+ }
+
+ if (more(acked_gen, state->acked_gen))
+ state->acked_gen = acked_gen;
+
+ /* Drop duplicate unless we are waiting for a probe response */
+ if (!more(new_gen, state->peer_gen) && !probing)
+ return;
+
+ write_lock_bh(&mon->lock);
+ peer = get_peer(mon, addr);
+ if (!peer || !peer->is_up)
+ goto exit;
+
+ /* Peer is confirmed, stop any ongoing probing */
+ peer->down_cnt = 0;
+
+ /* Task is done for duplicate record */
+ if (!more(new_gen, state->peer_gen))
+ goto exit;
+
+ state->peer_gen = new_gen;
+
+ /* Cache current domain record for later use */
+ dom_bef.member_cnt = 0;
+ dom = peer->domain;
+ if (dom)
+ memcpy(&dom_bef, dom, dom->len);
+
+ /* Transform and store received domain record */
+ if (!dom || (dom->len < new_dlen)) {
+ kfree(dom);
+ dom = kmalloc(new_dlen, GFP_ATOMIC);
+ peer->domain = dom;
+ if (!dom)
+ goto exit;
+ }
+ dom->len = new_dlen;
+ dom->gen = new_gen;
+ dom->member_cnt = new_member_cnt;
+ dom->up_map = be64_to_cpu(arrv_dom->up_map);
+ for (i = 0; i < new_member_cnt; i++)
+ dom->members[i] = ntohl(arrv_dom->members[i]);
+
+ /* Update peers affected by this domain record */
+ applied_bef = peer->applied;
+ mon_apply_domain(mon, peer);
+ mon_identify_lost_members(peer, &dom_bef, applied_bef);
+ mon_assign_roles(mon, peer_head(peer));
+exit:
+ write_unlock_bh(&mon->lock);
+}
+
+void tipc_mon_prep(struct net *net, void *data, int *dlen,
+ struct tipc_mon_state *state, int bearer_id)
+{
+ struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
+ struct tipc_mon_domain *dom = data;
+ u16 gen = mon->dom_gen;
+ u16 len;
+
+ if (!tipc_mon_is_active(net, mon))
+ return;
+
+ /* Send only a dummy record with ack if peer has acked our last sent */
+ if (likely(state->acked_gen == gen)) {
+ len = dom_rec_len(dom, 0);
+ *dlen = len;
+ dom->len = htons(len);
+ dom->gen = htons(gen);
+ dom->ack_gen = htons(state->peer_gen);
+ dom->member_cnt = 0;
+ return;
+ }
+ /* Send the full record */
+ read_lock_bh(&mon->lock);
+ len = ntohs(mon->cache.len);
+ *dlen = len;
+ memcpy(data, &mon->cache, len);
+ read_unlock_bh(&mon->lock);
+ dom->ack_gen = htons(state->peer_gen);
+}
+
+void tipc_mon_get_state(struct net *net, u32 addr,
+ struct tipc_mon_state *state,
+ int bearer_id)
+{
+ struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
+ struct tipc_peer *peer;
+
+ /* Used cached state if table has not changed */
+ if (!state->probing &&
+ (state->list_gen == mon->list_gen) &&
+ (state->acked_gen == mon->dom_gen))
+ return;
+
+ read_lock_bh(&mon->lock);
+ peer = get_peer(mon, addr);
+ if (peer) {
+ state->probing = state->acked_gen != mon->dom_gen;
+ state->probing |= peer->down_cnt;
+ state->reset |= peer->down_cnt >= MAX_PEER_DOWN_EVENTS;
+ state->monitoring = peer->is_local;
+ state->monitoring |= peer->is_head;
+ state->list_gen = mon->list_gen;
+ }
+ read_unlock_bh(&mon->lock);
+}
+
+static void mon_timeout(unsigned long m)
+{
+ struct tipc_monitor *mon = (void *)m;
+ struct tipc_peer *self;
+ int best_member_cnt = dom_size(mon->peer_cnt) - 1;
+
+ write_lock_bh(&mon->lock);
+ self = mon->self;
+ if (self && (best_member_cnt != self->applied)) {
+ mon_update_local_domain(mon);
+ mon_assign_roles(mon, self);
+ }
+ write_unlock_bh(&mon->lock);
+ mod_timer(&mon->timer, jiffies + mon->timer_intv);
+}
+
+int tipc_mon_create(struct net *net, int bearer_id)
+{
+ struct tipc_net *tn = tipc_net(net);
+ struct tipc_monitor *mon;
+ struct tipc_peer *self;
+ struct tipc_mon_domain *dom;
+
+ if (tn->monitors[bearer_id])
+ return 0;
+
+ mon = kzalloc(sizeof(*mon), GFP_ATOMIC);
+ self = kzalloc(sizeof(*self), GFP_ATOMIC);
+ dom = kzalloc(sizeof(*dom), GFP_ATOMIC);
+ if (!mon || !self || !dom) {
+ kfree(mon);
+ kfree(self);
+ kfree(dom);
+ return -ENOMEM;
+ }
+ tn->monitors[bearer_id] = mon;
+ rwlock_init(&mon->lock);
+ mon->net = net;
+ mon->peer_cnt = 1;
+ mon->self = self;
+ self->domain = dom;
+ self->addr = tipc_own_addr(net);
+ self->is_up = true;
+ self->is_head = true;
+ INIT_LIST_HEAD(&self->list);
+ setup_timer(&mon->timer, mon_timeout, (unsigned long)mon);
+ mon->timer_intv = msecs_to_jiffies(MON_TIMEOUT + (tn->random & 0xffff));
+ mod_timer(&mon->timer, jiffies + mon->timer_intv);
+ return 0;
+}
+
+void tipc_mon_delete(struct net *net, int bearer_id)
+{
+ struct tipc_net *tn = tipc_net(net);
+ struct tipc_monitor *mon = tipc_monitor(net, bearer_id);
+ struct tipc_peer *self = get_self(net, bearer_id);
+ struct tipc_peer *peer, *tmp;
+
+ write_lock_bh(&mon->lock);
+ tn->monitors[bearer_id] = NULL;
+ list_for_each_entry_safe(peer, tmp, &self->list, list) {
+ list_del(&peer->list);
+ hlist_del(&peer->hash);
+ kfree(peer->domain);
+ kfree(peer);
+ }
+ mon->self = NULL;
+ write_unlock_bh(&mon->lock);
+ del_timer_sync(&mon->timer);
+ kfree(self->domain);
+ kfree(self);
+ kfree(mon);
+}
new file mode 100644
@@ -0,0 +1,73 @@
+/*
+ * net/tipc/monitor.h
+ *
+ * Copyright (c) 2015, Ericsson AB
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * 3. Neither the names of the copyright holders nor the names of its
+ * contributors may be used to endorse or promote products derived from
+ * this software without specific prior written permission.
+ *
+ * Alternatively, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") version 2 as published by the Free
+ * Software Foundation.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef _TIPC_MONITOR_H
+#define _TIPC_MONITOR_H
+
+/* struct tipc_mon_state: link instance's cache of monitor list and domain state
+ * @list_gen: current generation of this node's monitor list
+ * @gen: current generation of this node's local domain
+ * @peer_gen: most recent domain generation received from peer
+ * @acked_gen: most recent generation of self's domain acked by peer
+ * @monitoring: this peer endpoint should continuously monitored
+ * @probing: peer endpoint should be temporarily probed for potential loss
+ * @synched: domain record's generation has been synched with peer after reset
+ */
+struct tipc_mon_state {
+ u16 list_gen;
+ u16 peer_gen;
+ u16 acked_gen;
+ bool monitoring :1;
+ bool probing :1;
+ bool reset :1;
+ bool synched :1;
+};
+
+int tipc_mon_create(struct net *net, int bearer_id);
+void tipc_mon_delete(struct net *net, int bearer_id);
+
+void tipc_mon_peer_up(struct net *net, u32 addr, int bearer_id);
+void tipc_mon_peer_down(struct net *net, u32 addr, int bearer_id);
+void tipc_mon_prep(struct net *net, void *data, int *dlen,
+ struct tipc_mon_state *state, int bearer_id);
+void tipc_mon_rcv(struct net *net, void *data, u16 dlen, u32 addr,
+ struct tipc_mon_state *state, int bearer_id);
+void tipc_mon_get_state(struct net *net, u32 addr,
+ struct tipc_mon_state *state,
+ int bearer_id);
+void tipc_mon_remove_peer(struct net *net, u32 addr, int bearer_id);
+
+extern const int tipc_max_domain_size;
+#endif
@@ -40,6 +40,7 @@
#include "name_distr.h"
#include "socket.h"
#include "bcast.h"
+#include "monitor.h"
#include "discover.h"
#include "netlink.h"
@@ -205,17 +206,6 @@ u16 tipc_node_get_capabilities(struct net *net, u32 addr)
return caps;
}
-/*
- * A trivial power-of-two bitmask technique is used for speed, since this
- * operation is done for every incoming TIPC packet. The number of hash table
- * entries has been chosen so that no hash chain exceeds 8 nodes and will
- * usually be much smaller (typically only a single node).
- */
-static unsigned int tipc_hashfn(u32 addr)
-{
- return addr & (NODE_HTABLE_SIZE - 1);
-}
-
static void tipc_node_kref_release(struct kref *kref)
{
struct tipc_node *n = container_of(kref, struct tipc_node, kref);
@@ -279,6 +269,7 @@ static void tipc_node_write_unlock(struct tipc_node *n)
u32 addr = 0;
u32 flags = n->action_flags;
u32 link_id = 0;
+ u32 bearer_id;
struct list_head *publ_list;
if (likely(!flags)) {
@@ -288,6 +279,7 @@ static void tipc_node_write_unlock(struct tipc_node *n)
addr = n->addr;
link_id = n->link_id;
+ bearer_id = link_id & 0xffff;
publ_list = &n->publ_list;
n->action_flags &= ~(TIPC_NOTIFY_NODE_DOWN | TIPC_NOTIFY_NODE_UP |
@@ -301,13 +293,16 @@ static void tipc_node_write_unlock(struct tipc_node *n)
if (flags & TIPC_NOTIFY_NODE_UP)
tipc_named_node_up(net, addr);
- if (flags & TIPC_NOTIFY_LINK_UP)
+ if (flags & TIPC_NOTIFY_LINK_UP) {
+ tipc_mon_peer_up(net, addr, bearer_id);
tipc_nametbl_publish(net, TIPC_LINK_STATE, addr, addr,
TIPC_NODE_SCOPE, link_id, addr);
-
- if (flags & TIPC_NOTIFY_LINK_DOWN)
+ }
+ if (flags & TIPC_NOTIFY_LINK_DOWN) {
+ tipc_mon_peer_down(net, addr, bearer_id);
tipc_nametbl_withdraw(net, TIPC_LINK_STATE, addr,
link_id, addr);
+ }
}
struct tipc_node *tipc_node_create(struct net *net, u32 addr, u16 capabilities)
@@ -691,6 +686,7 @@ static void tipc_node_link_down(struct tipc_node *n, int bearer_id, bool delete)
struct tipc_link *l = le->link;
struct tipc_media_addr *maddr;
struct sk_buff_head xmitq;
+ int old_bearer_id = bearer_id;
if (!l)
return;
@@ -710,6 +706,8 @@ static void tipc_node_link_down(struct tipc_node *n, int bearer_id, bool delete)
tipc_link_fsm_evt(l, LINK_RESET_EVT);
}
tipc_node_write_unlock(n);
+ if (delete)
+ tipc_mon_remove_peer(n->net, n->addr, old_bearer_id);
tipc_bearer_xmit(n->net, bearer_id, &xmitq, maddr);
tipc_sk_rcv(n->net, &le->inputq);
}