4 files changed, 394 insertions, 680 deletions

diff --git a/kernel/fs/dlm/config.c b/kernel/fs/dlm/config.c
index d521bddf8..8e294fbba 100644
--- a/kernel/fs/dlm/config.c
+++ b/kernel/fs/dlm/config.c
@@ -61,35 +61,8 @@ static struct config_item *make_node(struct config_group *, const char *);
 static void drop_node(struct config_group *, struct config_item *);
 static void release_node(struct config_item *);
 
-static ssize_t show_cluster(struct config_item *i, struct configfs_attribute *a,
-			    char *buf);
-static ssize_t store_cluster(struct config_item *i,
-			     struct configfs_attribute *a,
-			     const char *buf, size_t len);
-static ssize_t show_comm(struct config_item *i, struct configfs_attribute *a,
-			 char *buf);
-static ssize_t store_comm(struct config_item *i, struct configfs_attribute *a,
-			  const char *buf, size_t len);
-static ssize_t show_node(struct config_item *i, struct configfs_attribute *a,
-			 char *buf);
-static ssize_t store_node(struct config_item *i, struct configfs_attribute *a,
-			  const char *buf, size_t len);
-
-static ssize_t comm_nodeid_read(struct dlm_comm *cm, char *buf);
-static ssize_t comm_nodeid_write(struct dlm_comm *cm, const char *buf,
-				size_t len);
-static ssize_t comm_local_read(struct dlm_comm *cm, char *buf);
-static ssize_t comm_local_write(struct dlm_comm *cm, const char *buf,
-				size_t len);
-static ssize_t comm_addr_write(struct dlm_comm *cm, const char *buf,
-				size_t len);
-static ssize_t comm_addr_list_read(struct dlm_comm *cm, char *buf);
-static ssize_t node_nodeid_read(struct dlm_node *nd, char *buf);
-static ssize_t node_nodeid_write(struct dlm_node *nd, const char *buf,
-				size_t len);
-static ssize_t node_weight_read(struct dlm_node *nd, char *buf);
-static ssize_t node_weight_write(struct dlm_node *nd, const char *buf,
-				size_t len);
+static struct configfs_attribute *comm_attrs[];
+static struct configfs_attribute *node_attrs[];
 
 struct dlm_cluster {
 	struct config_group group;
@@ -108,6 +81,12 @@ struct dlm_cluster {
 	char cl_cluster_name[DLM_LOCKSPACE_LEN];
 };
 
+static struct dlm_cluster *config_item_to_cluster(struct config_item *i)
+{
+	return i ? container_of(to_config_group(i), struct dlm_cluster, group) :
+		   NULL;
+}
+
 enum {
 	CLUSTER_ATTR_TCP_PORT = 0,
 	CLUSTER_ATTR_BUFFER_SIZE,
@@ -124,33 +103,24 @@ enum {
 	CLUSTER_ATTR_CLUSTER_NAME,
 };
 
-struct cluster_attribute {
-	struct configfs_attribute attr;
-	ssize_t (*show)(struct dlm_cluster *, char *);
-	ssize_t (*store)(struct dlm_cluster *, const char *, size_t);
-};
-
-static ssize_t cluster_cluster_name_read(struct dlm_cluster *cl, char *buf)
+static ssize_t cluster_cluster_name_show(struct config_item *item, char *buf)
 {
+	struct dlm_cluster *cl = config_item_to_cluster(item);
 	return sprintf(buf, "%s\n", cl->cl_cluster_name);
 }
 
-static ssize_t cluster_cluster_name_write(struct dlm_cluster *cl,
+static ssize_t cluster_cluster_name_store(struct config_item *item,
 					  const char *buf, size_t len)
 {
+	struct dlm_cluster *cl = config_item_to_cluster(item);
+
 	strlcpy(dlm_config.ci_cluster_name, buf,
 				sizeof(dlm_config.ci_cluster_name));
 	strlcpy(cl->cl_cluster_name, buf, sizeof(cl->cl_cluster_name));
 	return len;
 }
 
-static struct cluster_attribute cluster_attr_cluster_name = {
-	.attr   = { .ca_owner = THIS_MODULE,
-                    .ca_name = "cluster_name",
-                    .ca_mode = S_IRUGO | S_IWUSR },
-	.show   = cluster_cluster_name_read,
-	.store  = cluster_cluster_name_write,
-};
+CONFIGFS_ATTR(cluster_, cluster_name);
 
 static ssize_t cluster_set(struct dlm_cluster *cl, unsigned int *cl_field,
 			   int *info_field, int check_zero,
@@ -175,17 +145,19 @@ static ssize_t cluster_set(struct dlm_cluster *cl, unsigned int *cl_field,
 }
 
 #define CLUSTER_ATTR(name, check_zero)                                        \
-static ssize_t name##_write(struct dlm_cluster *cl, const char *buf, size_t len) \
+static ssize_t cluster_##name##_store(struct config_item *item, \
+		const char *buf, size_t len) \
 {                                                                             \
+	struct dlm_cluster *cl = config_item_to_cluster(item);		      \
 	return cluster_set(cl, &cl->cl_##name, &dlm_config.ci_##name,         \
 			   check_zero, buf, len);                             \
 }                                                                             \
-static ssize_t name##_read(struct dlm_cluster *cl, char *buf)                 \
+static ssize_t cluster_##name##_show(struct config_item *item, char *buf)     \
 {                                                                             \
+	struct dlm_cluster *cl = config_item_to_cluster(item);		      \
 	return snprintf(buf, PAGE_SIZE, "%u\n", cl->cl_##name);               \
 }                                                                             \
-static struct cluster_attribute cluster_attr_##name =                         \
-__CONFIGFS_ATTR(name, 0644, name##_read, name##_write)
+CONFIGFS_ATTR(cluster_, name);
 
 CLUSTER_ATTR(tcp_port, 1);
 CLUSTER_ATTR(buffer_size, 1);
@@ -201,19 +173,19 @@ CLUSTER_ATTR(new_rsb_count, 0);
 CLUSTER_ATTR(recover_callbacks, 0);
 
 static struct configfs_attribute *cluster_attrs[] = {
-	[CLUSTER_ATTR_TCP_PORT] = &cluster_attr_tcp_port.attr,
-	[CLUSTER_ATTR_BUFFER_SIZE] = &cluster_attr_buffer_size.attr,
-	[CLUSTER_ATTR_RSBTBL_SIZE] = &cluster_attr_rsbtbl_size.attr,
-	[CLUSTER_ATTR_RECOVER_TIMER] = &cluster_attr_recover_timer.attr,
-	[CLUSTER_ATTR_TOSS_SECS] = &cluster_attr_toss_secs.attr,
-	[CLUSTER_ATTR_SCAN_SECS] = &cluster_attr_scan_secs.attr,
-	[CLUSTER_ATTR_LOG_DEBUG] = &cluster_attr_log_debug.attr,
-	[CLUSTER_ATTR_PROTOCOL] = &cluster_attr_protocol.attr,
-	[CLUSTER_ATTR_TIMEWARN_CS] = &cluster_attr_timewarn_cs.attr,
-	[CLUSTER_ATTR_WAITWARN_US] = &cluster_attr_waitwarn_us.attr,
-	[CLUSTER_ATTR_NEW_RSB_COUNT] = &cluster_attr_new_rsb_count.attr,
-	[CLUSTER_ATTR_RECOVER_CALLBACKS] = &cluster_attr_recover_callbacks.attr,
-	[CLUSTER_ATTR_CLUSTER_NAME] = &cluster_attr_cluster_name.attr,
+	[CLUSTER_ATTR_TCP_PORT] = &cluster_attr_tcp_port,
+	[CLUSTER_ATTR_BUFFER_SIZE] = &cluster_attr_buffer_size,
+	[CLUSTER_ATTR_RSBTBL_SIZE] = &cluster_attr_rsbtbl_size,
+	[CLUSTER_ATTR_RECOVER_TIMER] = &cluster_attr_recover_timer,
+	[CLUSTER_ATTR_TOSS_SECS] = &cluster_attr_toss_secs,
+	[CLUSTER_ATTR_SCAN_SECS] = &cluster_attr_scan_secs,
+	[CLUSTER_ATTR_LOG_DEBUG] = &cluster_attr_log_debug,
+	[CLUSTER_ATTR_PROTOCOL] = &cluster_attr_protocol,
+	[CLUSTER_ATTR_TIMEWARN_CS] = &cluster_attr_timewarn_cs,
+	[CLUSTER_ATTR_WAITWARN_US] = &cluster_attr_waitwarn_us,
+	[CLUSTER_ATTR_NEW_RSB_COUNT] = &cluster_attr_new_rsb_count,
+	[CLUSTER_ATTR_RECOVER_CALLBACKS] = &cluster_attr_recover_callbacks,
+	[CLUSTER_ATTR_CLUSTER_NAME] = &cluster_attr_cluster_name,
 	NULL,
 };
 
@@ -224,83 +196,11 @@ enum {
 	COMM_ATTR_ADDR_LIST,
 };
 
-struct comm_attribute {
-	struct configfs_attribute attr;
-	ssize_t (*show)(struct dlm_comm *, char *);
-	ssize_t (*store)(struct dlm_comm *, const char *, size_t);
-};
-
-static struct comm_attribute comm_attr_nodeid = {
-	.attr   = { .ca_owner = THIS_MODULE,
-                    .ca_name = "nodeid",
-                    .ca_mode = S_IRUGO | S_IWUSR },
-	.show   = comm_nodeid_read,
-	.store  = comm_nodeid_write,
-};
-
-static struct comm_attribute comm_attr_local = {
-	.attr   = { .ca_owner = THIS_MODULE,
-                    .ca_name = "local",
-                    .ca_mode = S_IRUGO | S_IWUSR },
-	.show   = comm_local_read,
-	.store  = comm_local_write,
-};
-
-static struct comm_attribute comm_attr_addr = {
-	.attr   = { .ca_owner = THIS_MODULE,
-                    .ca_name = "addr",
-                    .ca_mode = S_IWUSR },
-	.store  = comm_addr_write,
-};
-
-static struct comm_attribute comm_attr_addr_list = {
-	.attr   = { .ca_owner = THIS_MODULE,
-                    .ca_name = "addr_list",
-                    .ca_mode = S_IRUGO },
-	.show   = comm_addr_list_read,
-};
-
-static struct configfs_attribute *comm_attrs[] = {
-	[COMM_ATTR_NODEID] = &comm_attr_nodeid.attr,
-	[COMM_ATTR_LOCAL] = &comm_attr_local.attr,
-	[COMM_ATTR_ADDR] = &comm_attr_addr.attr,
-	[COMM_ATTR_ADDR_LIST] = &comm_attr_addr_list.attr,
-	NULL,
-};
-
 enum {
 	NODE_ATTR_NODEID = 0,
 	NODE_ATTR_WEIGHT,
 };
 
-struct node_attribute {
-	struct configfs_attribute attr;
-	ssize_t (*show)(struct dlm_node *, char *);
-	ssize_t (*store)(struct dlm_node *, const char *, size_t);
-};
-
-static struct node_attribute node_attr_nodeid = {
-	.attr   = { .ca_owner = THIS_MODULE,
-                    .ca_name = "nodeid",
-                    .ca_mode = S_IRUGO | S_IWUSR },
-	.show   = node_nodeid_read,
-	.store  = node_nodeid_write,
-};
-
-static struct node_attribute node_attr_weight = {
-	.attr   = { .ca_owner = THIS_MODULE,
-                    .ca_name = "weight",
-                    .ca_mode = S_IRUGO | S_IWUSR },
-	.show   = node_weight_read,
-	.store  = node_weight_write,
-};
-
-static struct configfs_attribute *node_attrs[] = {
-	[NODE_ATTR_NODEID] = &node_attr_nodeid.attr,
-	[NODE_ATTR_WEIGHT] = &node_attr_weight.attr,
-	NULL,
-};
-
 struct dlm_clusters {
 	struct configfs_subsystem subsys;
 };
@@ -349,8 +249,6 @@ static struct configfs_group_operations clusters_ops = {
 
 static struct configfs_item_operations cluster_ops = {
 	.release = release_cluster,
-	.show_attribute = show_cluster,
-	.store_attribute = store_cluster,
 };
 
 static struct configfs_group_operations spaces_ops = {
@@ -369,8 +267,6 @@ static struct configfs_group_operations comms_ops = {
 
 static struct configfs_item_operations comm_ops = {
 	.release = release_comm,
-	.show_attribute = show_comm,
-	.store_attribute = store_comm,
 };
 
 static struct configfs_group_operations nodes_ops = {
@@ -380,8 +276,6 @@ static struct configfs_group_operations nodes_ops = {
 
 static struct configfs_item_operations node_ops = {
 	.release = release_node,
-	.show_attribute = show_node,
-	.store_attribute = store_node,
 };
 
 static struct config_item_type clusters_type = {
@@ -427,12 +321,6 @@ static struct config_item_type node_type = {
 	.ct_owner = THIS_MODULE,
 };
 
-static struct dlm_cluster *config_item_to_cluster(struct config_item *i)
-{
-	return i ? container_of(to_config_group(i), struct dlm_cluster, group) :
-		   NULL;
-}
-
 static struct dlm_space *config_item_to_space(struct config_item *i)
 {
 	return i ? container_of(to_config_group(i), struct dlm_space, group) :
@@ -687,66 +575,30 @@ void dlm_config_exit(void)
  * Functions for user space to read/write attributes
  */
 
-static ssize_t show_cluster(struct config_item *i, struct configfs_attribute *a,
-			    char *buf)
-{
-	struct dlm_cluster *cl = config_item_to_cluster(i);
-	struct cluster_attribute *cla =
-			container_of(a, struct cluster_attribute, attr);
-	return cla->show ? cla->show(cl, buf) : 0;
-}
-
-static ssize_t store_cluster(struct config_item *i,
-			     struct configfs_attribute *a,
-			     const char *buf, size_t len)
+static ssize_t comm_nodeid_show(struct config_item *item, char *buf)
 {
-	struct dlm_cluster *cl = config_item_to_cluster(i);
-	struct cluster_attribute *cla =
-		container_of(a, struct cluster_attribute, attr);
-	return cla->store ? cla->store(cl, buf, len) : -EINVAL;
-}
-
-static ssize_t show_comm(struct config_item *i, struct configfs_attribute *a,
-			 char *buf)
-{
-	struct dlm_comm *cm = config_item_to_comm(i);
-	struct comm_attribute *cma =
-			container_of(a, struct comm_attribute, attr);
-	return cma->show ? cma->show(cm, buf) : 0;
-}
-
-static ssize_t store_comm(struct config_item *i, struct configfs_attribute *a,
-			  const char *buf, size_t len)
-{
-	struct dlm_comm *cm = config_item_to_comm(i);
-	struct comm_attribute *cma =
-		container_of(a, struct comm_attribute, attr);
-	return cma->store ? cma->store(cm, buf, len) : -EINVAL;
+	return sprintf(buf, "%d\n", config_item_to_comm(item)->nodeid);
 }
 
-static ssize_t comm_nodeid_read(struct dlm_comm *cm, char *buf)
-{
-	return sprintf(buf, "%d\n", cm->nodeid);
-}
-
-static ssize_t comm_nodeid_write(struct dlm_comm *cm, const char *buf,
+static ssize_t comm_nodeid_store(struct config_item *item, const char *buf,
 				 size_t len)
 {
-	int rc = kstrtoint(buf, 0, &cm->nodeid);
+	int rc = kstrtoint(buf, 0, &config_item_to_comm(item)->nodeid);
 
 	if (rc)
 		return rc;
 	return len;
 }
 
-static ssize_t comm_local_read(struct dlm_comm *cm, char *buf)
+static ssize_t comm_local_show(struct config_item *item, char *buf)
 {
-	return sprintf(buf, "%d\n", cm->local);
+	return sprintf(buf, "%d\n", config_item_to_comm(item)->local);
 }
 
-static ssize_t comm_local_write(struct dlm_comm *cm, const char *buf,
+static ssize_t comm_local_store(struct config_item *item, const char *buf,
 				size_t len)
 {
+	struct dlm_comm *cm = config_item_to_comm(item);
 	int rc = kstrtoint(buf, 0, &cm->local);
 
 	if (rc)
@@ -756,8 +608,10 @@ static ssize_t comm_local_write(struct dlm_comm *cm, const char *buf,
 	return len;
 }
 
-static ssize_t comm_addr_write(struct dlm_comm *cm, const char *buf, size_t len)
+static ssize_t comm_addr_store(struct config_item *item, const char *buf,
+		size_t len)
 {
+	struct dlm_comm *cm = config_item_to_comm(item);
 	struct sockaddr_storage *addr;
 	int rv;
 
@@ -783,8 +637,9 @@ static ssize_t comm_addr_write(struct dlm_comm *cm, const char *buf, size_t len)
 	return len;
 }
 
-static ssize_t comm_addr_list_read(struct dlm_comm *cm, char *buf)
+static ssize_t comm_addr_list_show(struct config_item *item, char *buf)
 {
+	struct dlm_comm *cm = config_item_to_comm(item);
 	ssize_t s;
 	ssize_t allowance;
 	int i;
@@ -827,32 +682,28 @@ static ssize_t comm_addr_list_read(struct dlm_comm *cm, char *buf)
 	return 4096 - allowance;
 }
 
-static ssize_t show_node(struct config_item *i, struct configfs_attribute *a,
-			 char *buf)
-{
-	struct dlm_node *nd = config_item_to_node(i);
-	struct node_attribute *nda =
-			container_of(a, struct node_attribute, attr);
-	return nda->show ? nda->show(nd, buf) : 0;
-}
+CONFIGFS_ATTR(comm_, nodeid);
+CONFIGFS_ATTR(comm_, local);
+CONFIGFS_ATTR_WO(comm_, addr);
+CONFIGFS_ATTR_RO(comm_, addr_list);
 
-static ssize_t store_node(struct config_item *i, struct configfs_attribute *a,
-			  const char *buf, size_t len)
-{
-	struct dlm_node *nd = config_item_to_node(i);
-	struct node_attribute *nda =
-		container_of(a, struct node_attribute, attr);
-	return nda->store ? nda->store(nd, buf, len) : -EINVAL;
-}
+static struct configfs_attribute *comm_attrs[] = {
+	[COMM_ATTR_NODEID] = &comm_attr_nodeid,
+	[COMM_ATTR_LOCAL] = &comm_attr_local,
+	[COMM_ATTR_ADDR] = &comm_attr_addr,
+	[COMM_ATTR_ADDR_LIST] = &comm_attr_addr_list,
+	NULL,
+};
 
-static ssize_t node_nodeid_read(struct dlm_node *nd, char *buf)
+static ssize_t node_nodeid_show(struct config_item *item, char *buf)
 {
-	return sprintf(buf, "%d\n", nd->nodeid);
+	return sprintf(buf, "%d\n", config_item_to_node(item)->nodeid);
 }
 
-static ssize_t node_nodeid_write(struct dlm_node *nd, const char *buf,
+static ssize_t node_nodeid_store(struct config_item *item, const char *buf,
 				 size_t len)
 {
+	struct dlm_node *nd = config_item_to_node(item);
 	uint32_t seq = 0;
 	int rc = kstrtoint(buf, 0, &nd->nodeid);
 
@@ -863,21 +714,30 @@ static ssize_t node_nodeid_write(struct dlm_node *nd, const char *buf,
 	return len;
 }
 
-static ssize_t node_weight_read(struct dlm_node *nd, char *buf)
+static ssize_t node_weight_show(struct config_item *item, char *buf)
 {
-	return sprintf(buf, "%d\n", nd->weight);
+	return sprintf(buf, "%d\n", config_item_to_node(item)->weight);
 }
 
-static ssize_t node_weight_write(struct dlm_node *nd, const char *buf,
+static ssize_t node_weight_store(struct config_item *item, const char *buf,
 				 size_t len)
 {
-	int rc = kstrtoint(buf, 0, &nd->weight);
+	int rc = kstrtoint(buf, 0, &config_item_to_node(item)->weight);
 
 	if (rc)
 		return rc;
 	return len;
 }
 
+CONFIGFS_ATTR(node_, nodeid);
+CONFIGFS_ATTR(node_, weight);
+
+static struct configfs_attribute *node_attrs[] = {
+	[NODE_ATTR_NODEID] = &node_attr_nodeid,
+	[NODE_ATTR_WEIGHT] = &node_attr_weight,
+	NULL,
+};
+
 /*
  * Functions for the dlm to get the info that's been configured
  */
diff --git a/kernel/fs/dlm/lowcomms.c b/kernel/fs/dlm/lowcomms.c
index d08e079ea..3a37bd3f9 100644
--- a/kernel/fs/dlm/lowcomms.c
+++ b/kernel/fs/dlm/lowcomms.c
@@ -120,12 +120,11 @@ struct connection {
 	struct cbuf cb;
 	int retries;
 #define MAX_CONNECT_RETRIES 3
-	int sctp_assoc;
 	struct hlist_node list;
 	struct connection *othercon;
 	struct work_struct rwork; /* Receive workqueue */
 	struct work_struct swork; /* Send workqueue */
-	bool try_new_addr;
+	void (*orig_error_report)(struct sock *sk);
 };
 #define sock2con(x) ((struct connection *)(x)->sk_user_data)
 
@@ -252,26 +251,6 @@ static struct connection *nodeid2con(int nodeid, gfp_t allocation)
 	return con;
 }
 
-/* This is a bit drastic, but only called when things go wrong */
-static struct connection *assoc2con(int assoc_id)
-{
-	int i;
-	struct connection *con;
-
-	mutex_lock(&connections_lock);
-
-	for (i = 0 ; i < CONN_HASH_SIZE; i++) {
-		hlist_for_each_entry(con, &connection_hash[i], list) {
-			if (con->sctp_assoc == assoc_id) {
-				mutex_unlock(&connections_lock);
-				return con;
-			}
-		}
-	}
-	mutex_unlock(&connections_lock);
-	return NULL;
-}
-
 static struct dlm_node_addr *find_node_addr(int nodeid)
 {
 	struct dlm_node_addr *na;
@@ -322,14 +301,14 @@ static int nodeid_to_addr(int nodeid, struct sockaddr_storage *sas_out,
 	spin_lock(&dlm_node_addrs_spin);
 	na = find_node_addr(nodeid);
 	if (na && na->addr_count) {
+		memcpy(&sas, na->addr[na->curr_addr_index],
+		       sizeof(struct sockaddr_storage));
+
 		if (try_new_addr) {
 			na->curr_addr_index++;
 			if (na->curr_addr_index == na->addr_count)
 				na->curr_addr_index = 0;
 		}
-
-		memcpy(&sas, na->addr[na->curr_addr_index ],
-			sizeof(struct sockaddr_storage));
 	}
 	spin_unlock(&dlm_node_addrs_spin);
 
@@ -442,7 +421,7 @@ static void lowcomms_write_space(struct sock *sk)
 
 	if (test_and_clear_bit(CF_APP_LIMITED, &con->flags)) {
 		con->sock->sk->sk_write_pending--;
-		clear_bit(SOCK_ASYNC_NOSPACE, &con->sock->flags);
+		clear_bit(SOCKWQ_ASYNC_NOSPACE, &con->sock->flags);
 	}
 
 	if (!test_and_set_bit(CF_WRITE_PENDING, &con->flags))
@@ -459,18 +438,23 @@ static inline void lowcomms_connect_sock(struct connection *con)
 
 static void lowcomms_state_change(struct sock *sk)
 {
-	if (sk->sk_state == TCP_ESTABLISHED)
+	/* SCTP layer is not calling sk_data_ready when the connection
+	 * is done, so we catch the signal through here. Also, it
+	 * doesn't switch socket state when entering shutdown, so we
+	 * skip the write in that case.
+	 */
+	if (sk->sk_shutdown) {
+		if (sk->sk_shutdown == RCV_SHUTDOWN)
+			lowcomms_data_ready(sk);
+	} else if (sk->sk_state == TCP_ESTABLISHED) {
 		lowcomms_write_space(sk);
+	}
 }
 
 int dlm_lowcomms_connect_node(int nodeid)
 {
 	struct connection *con;
 
-	/* with sctp there's no connecting without sending */
-	if (dlm_config.ci_protocol != 0)
-		return 0;
-
 	if (nodeid == dlm_our_nodeid())
 		return 0;
 
@@ -481,6 +465,43 @@ int dlm_lowcomms_connect_node(int nodeid)
 	return 0;
 }
 
+static void lowcomms_error_report(struct sock *sk)
+{
+	struct connection *con = sock2con(sk);
+	struct sockaddr_storage saddr;
+
+	if (nodeid_to_addr(con->nodeid, &saddr, NULL, false)) {
+		printk_ratelimited(KERN_ERR "dlm: node %d: socket error "
+				   "sending to node %d, port %d, "
+				   "sk_err=%d/%d\n", dlm_our_nodeid(),
+				   con->nodeid, dlm_config.ci_tcp_port,
+				   sk->sk_err, sk->sk_err_soft);
+		return;
+	} else if (saddr.ss_family == AF_INET) {
+		struct sockaddr_in *sin4 = (struct sockaddr_in *)&saddr;
+
+		printk_ratelimited(KERN_ERR "dlm: node %d: socket error "
+				   "sending to node %d at %pI4, port %d, "
+				   "sk_err=%d/%d\n", dlm_our_nodeid(),
+				   con->nodeid, &sin4->sin_addr.s_addr,
+				   dlm_config.ci_tcp_port, sk->sk_err,
+				   sk->sk_err_soft);
+	} else {
+		struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)&saddr;
+
+		printk_ratelimited(KERN_ERR "dlm: node %d: socket error "
+				   "sending to node %d at %u.%u.%u.%u, "
+				   "port %d, sk_err=%d/%d\n", dlm_our_nodeid(),
+				   con->nodeid, sin6->sin6_addr.s6_addr32[0],
+				   sin6->sin6_addr.s6_addr32[1],
+				   sin6->sin6_addr.s6_addr32[2],
+				   sin6->sin6_addr.s6_addr32[3],
+				   dlm_config.ci_tcp_port, sk->sk_err,
+				   sk->sk_err_soft);
+	}
+	con->orig_error_report(sk);
+}
+
 /* Make a socket active */
 static void add_sock(struct socket *sock, struct connection *con)
 {
@@ -492,6 +513,8 @@ static void add_sock(struct socket *sock, struct connection *con)
 	con->sock->sk->sk_state_change = lowcomms_state_change;
 	con->sock->sk->sk_user_data = con;
 	con->sock->sk->sk_allocation = GFP_NOFS;
+	con->orig_error_report = con->sock->sk->sk_error_report;
+	con->sock->sk->sk_error_report = lowcomms_error_report;
 }
 
 /* Add the port number to an IPv6 or 4 sockaddr and return the address
@@ -514,17 +537,24 @@ static void make_sockaddr(struct sockaddr_storage *saddr, uint16_t port,
 }
 
 /* Close a remote connection and tidy up */
-static void close_connection(struct connection *con, bool and_other)
+static void close_connection(struct connection *con, bool and_other,
+			     bool tx, bool rx)
 {
-	mutex_lock(&con->sock_mutex);
+	clear_bit(CF_CONNECT_PENDING, &con->flags);
+	clear_bit(CF_WRITE_PENDING, &con->flags);
+	if (tx && cancel_work_sync(&con->swork))
+		log_print("canceled swork for node %d", con->nodeid);
+	if (rx && cancel_work_sync(&con->rwork))
+		log_print("canceled rwork for node %d", con->nodeid);
 
+	mutex_lock(&con->sock_mutex);
 	if (con->sock) {
 		sock_release(con->sock);
 		con->sock = NULL;
 	}
 	if (con->othercon && and_other) {
 		/* Will only re-enter once. */
-		close_connection(con->othercon, false);
+		close_connection(con->othercon, false, true, true);
 	}
 	if (con->rx_page) {
 		__free_page(con->rx_page);
@@ -535,254 +565,6 @@ static void close_connection(struct connection *con, bool and_other)
 	mutex_unlock(&con->sock_mutex);
 }
 
-/* We only send shutdown messages to nodes that are not part of the cluster */
-static void sctp_send_shutdown(sctp_assoc_t associd)
-{
-	static char outcmsg[CMSG_SPACE(sizeof(struct sctp_sndrcvinfo))];
-	struct msghdr outmessage;
-	struct cmsghdr *cmsg;
-	struct sctp_sndrcvinfo *sinfo;
-	int ret;
-	struct connection *con;
-
-	con = nodeid2con(0,0);
-	BUG_ON(con == NULL);
-
-	outmessage.msg_name = NULL;
-	outmessage.msg_namelen = 0;
-	outmessage.msg_control = outcmsg;
-	outmessage.msg_controllen = sizeof(outcmsg);
-	outmessage.msg_flags = MSG_EOR;
-
-	cmsg = CMSG_FIRSTHDR(&outmessage);
-	cmsg->cmsg_level = IPPROTO_SCTP;
-	cmsg->cmsg_type = SCTP_SNDRCV;
-	cmsg->cmsg_len = CMSG_LEN(sizeof(struct sctp_sndrcvinfo));
-	outmessage.msg_controllen = cmsg->cmsg_len;
-	sinfo = CMSG_DATA(cmsg);
-	memset(sinfo, 0x00, sizeof(struct sctp_sndrcvinfo));
-
-	sinfo->sinfo_flags |= MSG_EOF;
-	sinfo->sinfo_assoc_id = associd;
-
-	ret = kernel_sendmsg(con->sock, &outmessage, NULL, 0, 0);
-
-	if (ret != 0)
-		log_print("send EOF to node failed: %d", ret);
-}
-
-static void sctp_init_failed_foreach(struct connection *con)
-{
-
-	/*
-	 * Don't try to recover base con and handle race where the
-	 * other node's assoc init creates a assoc and we get that
-	 * notification, then we get a notification that our attempt
-	 * failed due. This happens when we are still trying the primary
-	 * address, but the other node has already tried secondary addrs
-	 * and found one that worked.
-	 */
-	if (!con->nodeid || con->sctp_assoc)
-		return;
-
-	log_print("Retrying SCTP association init for node %d\n", con->nodeid);
-
-	con->try_new_addr = true;
-	con->sctp_assoc = 0;
-	if (test_and_clear_bit(CF_INIT_PENDING, &con->flags)) {
-		if (!test_and_set_bit(CF_WRITE_PENDING, &con->flags))
-			queue_work(send_workqueue, &con->swork);
-	}
-}
-
-/* INIT failed but we don't know which node...
-   restart INIT on all pending nodes */
-static void sctp_init_failed(void)
-{
-	mutex_lock(&connections_lock);
-
-	foreach_conn(sctp_init_failed_foreach);
-
-	mutex_unlock(&connections_lock);
-}
-
-static void retry_failed_sctp_send(struct connection *recv_con,
-				   struct sctp_send_failed *sn_send_failed,
-				   char *buf)
-{
-	int len = sn_send_failed->ssf_length - sizeof(struct sctp_send_failed);
-	struct dlm_mhandle *mh;
-	struct connection *con;
-	char *retry_buf;
-	int nodeid = sn_send_failed->ssf_info.sinfo_ppid;
-
-	log_print("Retry sending %d bytes to node id %d", len, nodeid);
-	
-	if (!nodeid) {
-		log_print("Shouldn't resend data via listening connection.");
-		return;
-	}
-
-	con = nodeid2con(nodeid, 0);
-	if (!con) {
-		log_print("Could not look up con for nodeid %d\n",
-			  nodeid);
-		return;
-	}
-
-	mh = dlm_lowcomms_get_buffer(nodeid, len, GFP_NOFS, &retry_buf);
-	if (!mh) {
-		log_print("Could not allocate buf for retry.");
-		return;
-	}
-	memcpy(retry_buf, buf + sizeof(struct sctp_send_failed), len);
-	dlm_lowcomms_commit_buffer(mh);
-
-	/*
-	 * If we got a assoc changed event before the send failed event then
-	 * we only need to retry the send.
-	 */
-	if (con->sctp_assoc) {
-		if (!test_and_set_bit(CF_WRITE_PENDING, &con->flags))
-			queue_work(send_workqueue, &con->swork);
-	} else
-		sctp_init_failed_foreach(con);
-}
-
-/* Something happened to an association */
-static void process_sctp_notification(struct connection *con,
-				      struct msghdr *msg, char *buf)
-{
-	union sctp_notification *sn = (union sctp_notification *)buf;
-	struct linger linger;
-
-	switch (sn->sn_header.sn_type) {
-	case SCTP_SEND_FAILED:
-		retry_failed_sctp_send(con, &sn->sn_send_failed, buf);
-		break;
-	case SCTP_ASSOC_CHANGE:
-		switch (sn->sn_assoc_change.sac_state) {
-		case SCTP_COMM_UP:
-		case SCTP_RESTART:
-		{
-			/* Check that the new node is in the lockspace */
-			struct sctp_prim prim;
-			int nodeid;
-			int prim_len, ret;
-			int addr_len;
-			struct connection *new_con;
-
-			/*
-			 * We get this before any data for an association.
-			 * We verify that the node is in the cluster and
-			 * then peel off a socket for it.
-			 */
-			if ((int)sn->sn_assoc_change.sac_assoc_id <= 0) {
-				log_print("COMM_UP for invalid assoc ID %d",
-					 (int)sn->sn_assoc_change.sac_assoc_id);
-				sctp_init_failed();
-				return;
-			}
-			memset(&prim, 0, sizeof(struct sctp_prim));
-			prim_len = sizeof(struct sctp_prim);
-			prim.ssp_assoc_id = sn->sn_assoc_change.sac_assoc_id;
-
-			ret = kernel_getsockopt(con->sock,
-						IPPROTO_SCTP,
-						SCTP_PRIMARY_ADDR,
-						(char*)&prim,
-						&prim_len);
-			if (ret < 0) {
-				log_print("getsockopt/sctp_primary_addr on "
-					  "new assoc %d failed : %d",
-					  (int)sn->sn_assoc_change.sac_assoc_id,
-					  ret);
-
-				/* Retry INIT later */
-				new_con = assoc2con(sn->sn_assoc_change.sac_assoc_id);
-				if (new_con)
-					clear_bit(CF_CONNECT_PENDING, &con->flags);
-				return;
-			}
-			make_sockaddr(&prim.ssp_addr, 0, &addr_len);
-			if (addr_to_nodeid(&prim.ssp_addr, &nodeid)) {
-				unsigned char *b=(unsigned char *)&prim.ssp_addr;
-				log_print("reject connect from unknown addr");
-				print_hex_dump_bytes("ss: ", DUMP_PREFIX_NONE, 
-						     b, sizeof(struct sockaddr_storage));
-				sctp_send_shutdown(prim.ssp_assoc_id);
-				return;
-			}
-
-			new_con = nodeid2con(nodeid, GFP_NOFS);
-			if (!new_con)
-				return;
-
-			/* Peel off a new sock */
-			lock_sock(con->sock->sk);
-			ret = sctp_do_peeloff(con->sock->sk,
-				sn->sn_assoc_change.sac_assoc_id,
-				&new_con->sock);
-			release_sock(con->sock->sk);
-			if (ret < 0) {
-				log_print("Can't peel off a socket for "
-					  "connection %d to node %d: err=%d",
-					  (int)sn->sn_assoc_change.sac_assoc_id,
-					  nodeid, ret);
-				return;
-			}
-			add_sock(new_con->sock, new_con);
-
-			linger.l_onoff = 1;
-			linger.l_linger = 0;
-			ret = kernel_setsockopt(new_con->sock, SOL_SOCKET, SO_LINGER,
-						(char *)&linger, sizeof(linger));
-			if (ret < 0)
-				log_print("set socket option SO_LINGER failed");
-
-			log_print("connecting to %d sctp association %d",
-				 nodeid, (int)sn->sn_assoc_change.sac_assoc_id);
-
-			new_con->sctp_assoc = sn->sn_assoc_change.sac_assoc_id;
-			new_con->try_new_addr = false;
-			/* Send any pending writes */
-			clear_bit(CF_CONNECT_PENDING, &new_con->flags);
-			clear_bit(CF_INIT_PENDING, &new_con->flags);
-			if (!test_and_set_bit(CF_WRITE_PENDING, &new_con->flags)) {
-				queue_work(send_workqueue, &new_con->swork);
-			}
-			if (!test_and_set_bit(CF_READ_PENDING, &new_con->flags))
-				queue_work(recv_workqueue, &new_con->rwork);
-		}
-		break;
-
-		case SCTP_COMM_LOST:
-		case SCTP_SHUTDOWN_COMP:
-		{
-			con = assoc2con(sn->sn_assoc_change.sac_assoc_id);
-			if (con) {
-				con->sctp_assoc = 0;
-			}
-		}
-		break;
-
-		case SCTP_CANT_STR_ASSOC:
-		{
-			/* Will retry init when we get the send failed notification */
-			log_print("Can't start SCTP association - retrying");
-		}
-		break;
-
-		default:
-			log_print("unexpected SCTP assoc change id=%d state=%d",
-				  (int)sn->sn_assoc_change.sac_assoc_id,
-				  sn->sn_assoc_change.sac_state);
-		}
-	default:
-		; /* fall through */
-	}
-}
-
 /* Data received from remote end */
 static int receive_from_sock(struct connection *con)
 {
@@ -793,7 +575,6 @@ static int receive_from_sock(struct connection *con)
 	int r;
 	int call_again_soon = 0;
 	int nvec;
-	char incmsg[CMSG_SPACE(sizeof(struct sctp_sndrcvinfo))];
 
 	mutex_lock(&con->sock_mutex);
 
@@ -801,6 +582,10 @@ static int receive_from_sock(struct connection *con)
 		ret = -EAGAIN;
 		goto out_close;
 	}
+	if (con->nodeid == 0) {
+		ret = -EINVAL;
+		goto out_close;
+	}
 
 	if (con->rx_page == NULL) {
 		/*
@@ -813,11 +598,6 @@ static int receive_from_sock(struct connection *con)
 		cbuf_init(&con->cb, PAGE_CACHE_SIZE);
 	}
 
-	/* Only SCTP needs these really */
-	memset(&incmsg, 0, sizeof(incmsg));
-	msg.msg_control = incmsg;
-	msg.msg_controllen = sizeof(incmsg);
-
 	/*
 	 * iov[0] is the bit of the circular buffer between the current end
 	 * point (cb.base + cb.len) and the end of the buffer.
@@ -843,31 +623,18 @@ static int receive_from_sock(struct connection *con)
 			       MSG_DONTWAIT | MSG_NOSIGNAL);
 	if (ret <= 0)
 		goto out_close;
-
-	/* Process SCTP notifications */
-	if (msg.msg_flags & MSG_NOTIFICATION) {
-		msg.msg_control = incmsg;
-		msg.msg_controllen = sizeof(incmsg);
-
-		process_sctp_notification(con, &msg,
-				page_address(con->rx_page) + con->cb.base);
-		mutex_unlock(&con->sock_mutex);
-		return 0;
-	}
-	BUG_ON(con->nodeid == 0);
-
-	if (ret == len)
+	else if (ret == len)
 		call_again_soon = 1;
+
 	cbuf_add(&con->cb, ret);
 	ret = dlm_process_incoming_buffer(con->nodeid,
 					  page_address(con->rx_page),
 					  con->cb.base, con->cb.len,
 					  PAGE_CACHE_SIZE);
 	if (ret == -EBADMSG) {
-		log_print("lowcomms: addr=%p, base=%u, len=%u, "
-			  "iov_len=%u, iov_base[0]=%p, read=%d",
-			  page_address(con->rx_page), con->cb.base, con->cb.len,
-			  len, iov[0].iov_base, r);
+		log_print("lowcomms: addr=%p, base=%u, len=%u, read=%d",
+			  page_address(con->rx_page), con->cb.base,
+			  con->cb.len, r);
 	}
 	if (ret < 0)
 		goto out_close;
@@ -892,7 +659,7 @@ out_resched:
 out_close:
 	mutex_unlock(&con->sock_mutex);
 	if (ret != -EAGAIN) {
-		close_connection(con, false);
+		close_connection(con, false, true, false);
 		/* Reconnect when there is something to send */
 	}
 	/* Don't return success if we really got EOF */
@@ -921,8 +688,8 @@ static int tcp_accept_from_sock(struct connection *con)
 	mutex_unlock(&connections_lock);
 
 	memset(&peeraddr, 0, sizeof(peeraddr));
-	result = sock_create_kern(dlm_local_addr[0]->ss_family, SOCK_STREAM,
-				  IPPROTO_TCP, &newsock);
+	result = sock_create_kern(&init_net, dlm_local_addr[0]->ss_family,
+				  SOCK_STREAM, IPPROTO_TCP, &newsock);
 	if (result < 0)
 		return -ENOMEM;
 
@@ -1033,6 +800,120 @@ accept_err:
 	return result;
 }
 
+static int sctp_accept_from_sock(struct connection *con)
+{
+	/* Check that the new node is in the lockspace */
+	struct sctp_prim prim;
+	int nodeid;
+	int prim_len, ret;
+	int addr_len;
+	struct connection *newcon;
+	struct connection *addcon;
+	struct socket *newsock;
+
+	mutex_lock(&connections_lock);
+	if (!dlm_allow_conn) {
+		mutex_unlock(&connections_lock);
+		return -1;
+	}
+	mutex_unlock(&connections_lock);
+
+	mutex_lock_nested(&con->sock_mutex, 0);
+
+	ret = kernel_accept(con->sock, &newsock, O_NONBLOCK);
+	if (ret < 0)
+		goto accept_err;
+
+	memset(&prim, 0, sizeof(struct sctp_prim));
+	prim_len = sizeof(struct sctp_prim);
+
+	ret = kernel_getsockopt(newsock, IPPROTO_SCTP, SCTP_PRIMARY_ADDR,
+				(char *)&prim, &prim_len);
+	if (ret < 0) {
+		log_print("getsockopt/sctp_primary_addr failed: %d", ret);
+		goto accept_err;
+	}
+
+	make_sockaddr(&prim.ssp_addr, 0, &addr_len);
+	if (addr_to_nodeid(&prim.ssp_addr, &nodeid)) {
+		unsigned char *b = (unsigned char *)&prim.ssp_addr;
+
+		log_print("reject connect from unknown addr");
+		print_hex_dump_bytes("ss: ", DUMP_PREFIX_NONE,
+				     b, sizeof(struct sockaddr_storage));
+		goto accept_err;
+	}
+
+	newcon = nodeid2con(nodeid, GFP_NOFS);
+	if (!newcon) {
+		ret = -ENOMEM;
+		goto accept_err;
+	}
+
+	mutex_lock_nested(&newcon->sock_mutex, 1);
+
+	if (newcon->sock) {
+		struct connection *othercon = newcon->othercon;
+
+		if (!othercon) {
+			othercon = kmem_cache_zalloc(con_cache, GFP_NOFS);
+			if (!othercon) {
+				log_print("failed to allocate incoming socket");
+				mutex_unlock(&newcon->sock_mutex);
+				ret = -ENOMEM;
+				goto accept_err;
+			}
+			othercon->nodeid = nodeid;
+			othercon->rx_action = receive_from_sock;
+			mutex_init(&othercon->sock_mutex);
+			INIT_WORK(&othercon->swork, process_send_sockets);
+			INIT_WORK(&othercon->rwork, process_recv_sockets);
+			set_bit(CF_IS_OTHERCON, &othercon->flags);
+		}
+		if (!othercon->sock) {
+			newcon->othercon = othercon;
+			othercon->sock = newsock;
+			newsock->sk->sk_user_data = othercon;
+			add_sock(newsock, othercon);
+			addcon = othercon;
+		} else {
+			printk("Extra connection from node %d attempted\n", nodeid);
+			ret = -EAGAIN;
+			mutex_unlock(&newcon->sock_mutex);
+			goto accept_err;
+		}
+	} else {
+		newsock->sk->sk_user_data = newcon;
+		newcon->rx_action = receive_from_sock;
+		add_sock(newsock, newcon);
+		addcon = newcon;
+	}
+
+	log_print("connected to %d", nodeid);
+
+	mutex_unlock(&newcon->sock_mutex);
+
+	/*
+	 * Add it to the active queue in case we got data
+	 * between processing the accept adding the socket
+	 * to the read_sockets list
+	 */
+	if (!test_and_set_bit(CF_READ_PENDING, &addcon->flags))
+		queue_work(recv_workqueue, &addcon->rwork);
+	mutex_unlock(&con->sock_mutex);
+
+	return 0;
+
+accept_err:
+	mutex_unlock(&con->sock_mutex);
+	if (newsock)
+		sock_release(newsock);
+	if (ret != -EAGAIN)
+		log_print("error accepting connection from node: %d", ret);
+
+	return ret;
+}
+
 static void free_entry(struct writequeue_entry *e)
 {
 	__free_page(e->page);
@@ -1057,97 +938,129 @@ static void writequeue_entry_complete(struct writequeue_entry *e, int completed)
 	}
 }
 
+/*
+ * sctp_bind_addrs - bind a SCTP socket to all our addresses
+ */
+static int sctp_bind_addrs(struct connection *con, uint16_t port)
+{
+	struct sockaddr_storage localaddr;
+	int i, addr_len, result = 0;
+
+	for (i = 0; i < dlm_local_count; i++) {
+		memcpy(&localaddr, dlm_local_addr[i], sizeof(localaddr));
+		make_sockaddr(&localaddr, port, &addr_len);
+
+		if (!i)
+			result = kernel_bind(con->sock,
+					     (struct sockaddr *)&localaddr,
+					     addr_len);
+		else
+			result = kernel_setsockopt(con->sock, SOL_SCTP,
+						   SCTP_SOCKOPT_BINDX_ADD,
+						   (char *)&localaddr, addr_len);
+
+		if (result < 0) {
+			log_print("Can't bind to %d addr number %d, %d.\n",
+				  port, i + 1, result);
+			break;
+		}
+	}
+	return result;
+}
+
 /* Initiate an SCTP association.
    This is a special case of send_to_sock() in that we don't yet have a
    peeled-off socket for this association, so we use the listening socket
    and add the primary IP address of the remote node.
  */
-static void sctp_init_assoc(struct connection *con)
+static void sctp_connect_to_sock(struct connection *con)
 {
-	struct sockaddr_storage rem_addr;
-	char outcmsg[CMSG_SPACE(sizeof(struct sctp_sndrcvinfo))];
-	struct msghdr outmessage;
-	struct cmsghdr *cmsg;
-	struct sctp_sndrcvinfo *sinfo;
-	struct connection *base_con;
-	struct writequeue_entry *e;
-	int len, offset;
-	int ret;
-	int addrlen;
-	struct kvec iov[1];
+	struct sockaddr_storage daddr;
+	int one = 1;
+	int result;
+	int addr_len;
+	struct socket *sock;
+
+	if (con->nodeid == 0) {
+		log_print("attempt to connect sock 0 foiled");
+		return;
+	}
 
 	mutex_lock(&con->sock_mutex);
-	if (test_and_set_bit(CF_INIT_PENDING, &con->flags))
-		goto unlock;
 
-	if (nodeid_to_addr(con->nodeid, NULL, (struct sockaddr *)&rem_addr,
-			   con->try_new_addr)) {
+	/* Some odd races can cause double-connects, ignore them */
+	if (con->retries++ > MAX_CONNECT_RETRIES)
+		goto out;
+
+	if (con->sock) {
+		log_print("node %d already connected.", con->nodeid);
+		goto out;
+	}
+
+	memset(&daddr, 0, sizeof(daddr));
+	result = nodeid_to_addr(con->nodeid, &daddr, NULL, true);
+	if (result < 0) {
 		log_print("no address for nodeid %d", con->nodeid);
-		goto unlock;
+		goto out;
 	}
-	base_con = nodeid2con(0, 0);
-	BUG_ON(base_con == NULL);
 
-	make_sockaddr(&rem_addr, dlm_config.ci_tcp_port, &addrlen);
+	/* Create a socket to communicate with */
+	result = sock_create_kern(&init_net, dlm_local_addr[0]->ss_family,
+				  SOCK_STREAM, IPPROTO_SCTP, &sock);
+	if (result < 0)
+		goto socket_err;
 
-	outmessage.msg_name = &rem_addr;
-	outmessage.msg_namelen = addrlen;
-	outmessage.msg_control = outcmsg;
-	outmessage.msg_controllen = sizeof(outcmsg);
-	outmessage.msg_flags = MSG_EOR;
+	sock->sk->sk_user_data = con;
+	con->rx_action = receive_from_sock;
+	con->connect_action = sctp_connect_to_sock;
+	add_sock(sock, con);
 
-	spin_lock(&con->writequeue_lock);
+	/* Bind to all addresses. */
+	if (sctp_bind_addrs(con, 0))
+		goto bind_err;
 
-	if (list_empty(&con->writequeue)) {
-		spin_unlock(&con->writequeue_lock);
-		log_print("writequeue empty for nodeid %d", con->nodeid);
-		goto unlock;
-	}
+	make_sockaddr(&daddr, dlm_config.ci_tcp_port, &addr_len);
 
-	e = list_first_entry(&con->writequeue, struct writequeue_entry, list);
-	len = e->len;
-	offset = e->offset;
+	log_print("connecting to %d", con->nodeid);
 
-	/* Send the first block off the write queue */
-	iov[0].iov_base = page_address(e->page)+offset;
-	iov[0].iov_len = len;
-	spin_unlock(&con->writequeue_lock);
+	/* Turn off Nagle's algorithm */
+	kernel_setsockopt(sock, SOL_TCP, TCP_NODELAY, (char *)&one,
+			  sizeof(one));
 
-	if (rem_addr.ss_family == AF_INET) {
-		struct sockaddr_in *sin = (struct sockaddr_in *)&rem_addr;
-		log_print("Trying to connect to %pI4", &sin->sin_addr.s_addr);
-	} else {
-		struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)&rem_addr;
-		log_print("Trying to connect to %pI6", &sin6->sin6_addr);
-	}
+	result = sock->ops->connect(sock, (struct sockaddr *)&daddr, addr_len,
+				   O_NONBLOCK);
+	if (result == -EINPROGRESS)
+		result = 0;
+	if (result == 0)
+		goto out;
 
-	cmsg = CMSG_FIRSTHDR(&outmessage);
-	cmsg->cmsg_level = IPPROTO_SCTP;
-	cmsg->cmsg_type = SCTP_SNDRCV;
-	cmsg->cmsg_len = CMSG_LEN(sizeof(struct sctp_sndrcvinfo));
-	sinfo = CMSG_DATA(cmsg);
-	memset(sinfo, 0x00, sizeof(struct sctp_sndrcvinfo));
-	sinfo->sinfo_ppid = cpu_to_le32(con->nodeid);
-	outmessage.msg_controllen = cmsg->cmsg_len;
-	sinfo->sinfo_flags |= SCTP_ADDR_OVER;
 
-	ret = kernel_sendmsg(base_con->sock, &outmessage, iov, 1, len);
-	if (ret < 0) {
-		log_print("Send first packet to node %d failed: %d",
-			  con->nodeid, ret);
+bind_err:
+	con->sock = NULL;
+	sock_release(sock);
 
-		/* Try again later */
+socket_err:
+	/*
+	 * Some errors are fatal and this list might need adjusting. For other
+	 * errors we try again until the max number of retries is reached.
+	 */
+	if (result != -EHOSTUNREACH &&
+	    result != -ENETUNREACH &&
+	    result != -ENETDOWN &&
+	    result != -EINVAL &&
+	    result != -EPROTONOSUPPORT) {
+		log_print("connect %d try %d error %d", con->nodeid,
+			  con->retries, result);
+		mutex_unlock(&con->sock_mutex);
+		msleep(1000);
 		clear_bit(CF_CONNECT_PENDING, &con->flags);
-		clear_bit(CF_INIT_PENDING, &con->flags);
-	}
-	else {
-		spin_lock(&con->writequeue_lock);
-		writequeue_entry_complete(e, ret);
-		spin_unlock(&con->writequeue_lock);
+		lowcomms_connect_sock(con);
+		return;
 	}
 
-unlock:
+out:
 	mutex_unlock(&con->sock_mutex);
+	set_bit(CF_WRITE_PENDING, &con->flags);
 }
 
 /* Connect a new socket to its peer */
@@ -1173,8 +1086,8 @@ static void tcp_connect_to_sock(struct connection *con)
 		goto out;
 
 	/* Create a socket to communicate with */
-	result = sock_create_kern(dlm_local_addr[0]->ss_family, SOCK_STREAM,
-				  IPPROTO_TCP, &sock);
+	result = sock_create_kern(&init_net, dlm_local_addr[0]->ss_family,
+				  SOCK_STREAM, IPPROTO_TCP, &sock);
 	if (result < 0)
 		goto out_err;
 
@@ -1236,11 +1149,13 @@ out_err:
 			  con->retries, result);
 		mutex_unlock(&con->sock_mutex);
 		msleep(1000);
+		clear_bit(CF_CONNECT_PENDING, &con->flags);
 		lowcomms_connect_sock(con);
 		return;
 	}
 out:
 	mutex_unlock(&con->sock_mutex);
+	set_bit(CF_WRITE_PENDING, &con->flags);
 	return;
 }
 
@@ -1258,8 +1173,8 @@ static struct socket *tcp_create_listen_sock(struct connection *con,
 		addr_len = sizeof(struct sockaddr_in6);
 
 	/* Create a socket to communicate with */
-	result = sock_create_kern(dlm_local_addr[0]->ss_family, SOCK_STREAM,
-				  IPPROTO_TCP, &sock);
+	result = sock_create_kern(&init_net, dlm_local_addr[0]->ss_family,
+				  SOCK_STREAM, IPPROTO_TCP, &sock);
 	if (result < 0) {
 		log_print("Can't create listening comms socket");
 		goto create_out;
@@ -1325,37 +1240,11 @@ static void init_local(void)
 	}
 }
 
-/* Bind to an IP address. SCTP allows multiple address so it can do
-   multi-homing */
-static int add_sctp_bind_addr(struct connection *sctp_con,
-			      struct sockaddr_storage *addr,
-			      int addr_len, int num)
-{
-	int result = 0;
-
-	if (num == 1)
-		result = kernel_bind(sctp_con->sock,
-				     (struct sockaddr *) addr,
-				     addr_len);
-	else
-		result = kernel_setsockopt(sctp_con->sock, SOL_SCTP,
-					   SCTP_SOCKOPT_BINDX_ADD,
-					   (char *)addr, addr_len);
-
-	if (result < 0)
-		log_print("Can't bind to port %d addr number %d",
-			  dlm_config.ci_tcp_port, num);
-
-	return result;
-}
-
 /* Initialise SCTP socket and bind to all interfaces */
 static int sctp_listen_for_all(void)
 {
 	struct socket *sock = NULL;
-	struct sockaddr_storage localaddr;
-	struct sctp_event_subscribe subscribe;
-	int result = -EINVAL, num = 1, i, addr_len;
+	int result = -EINVAL;
 	struct connection *con = nodeid2con(0, GFP_NOFS);
 	int bufsize = NEEDED_RMEM;
 	int one = 1;
@@ -1365,34 +1254,18 @@ static int sctp_listen_for_all(void)
 
 	log_print("Using SCTP for communications");
 
-	result = sock_create_kern(dlm_local_addr[0]->ss_family, SOCK_SEQPACKET,
-				  IPPROTO_SCTP, &sock);
+	result = sock_create_kern(&init_net, dlm_local_addr[0]->ss_family,
+				  SOCK_STREAM, IPPROTO_SCTP, &sock);
 	if (result < 0) {
 		log_print("Can't create comms socket, check SCTP is loaded");
 		goto out;
 	}
 
-	/* Listen for events */
-	memset(&subscribe, 0, sizeof(subscribe));
-	subscribe.sctp_data_io_event = 1;
-	subscribe.sctp_association_event = 1;
-	subscribe.sctp_send_failure_event = 1;
-	subscribe.sctp_shutdown_event = 1;
-	subscribe.sctp_partial_delivery_event = 1;
-
 	result = kernel_setsockopt(sock, SOL_SOCKET, SO_RCVBUFFORCE,
 				 (char *)&bufsize, sizeof(bufsize));
 	if (result)
 		log_print("Error increasing buffer space on socket %d", result);
 
-	result = kernel_setsockopt(sock, SOL_SCTP, SCTP_EVENTS,
-				   (char *)&subscribe, sizeof(subscribe));
-	if (result < 0) {
-		log_print("Failed to set SCTP_EVENTS on socket: result=%d",
-			  result);
-		goto create_delsock;
-	}
-
 	result = kernel_setsockopt(sock, SOL_SCTP, SCTP_NODELAY, (char *)&one,
 				   sizeof(one));
 	if (result < 0)
@@ -1402,19 +1275,12 @@ static int sctp_listen_for_all(void)
 	sock->sk->sk_user_data = con;
 	con->sock = sock;
 	con->sock->sk->sk_data_ready = lowcomms_data_ready;
-	con->rx_action = receive_from_sock;
-	con->connect_action = sctp_init_assoc;
+	con->rx_action = sctp_accept_from_sock;
+	con->connect_action = sctp_connect_to_sock;
 
-	/* Bind to all interfaces. */
-	for (i = 0; i < dlm_local_count; i++) {
-		memcpy(&localaddr, dlm_local_addr[i], sizeof(localaddr));
-		make_sockaddr(&localaddr, dlm_config.ci_tcp_port, &addr_len);
-
-		result = add_sctp_bind_addr(con, &localaddr, addr_len, num);
-		if (result)
-			goto create_delsock;
-		++num;
-	}
+	/* Bind to all addresses. */
+	if (sctp_bind_addrs(con, dlm_config.ci_tcp_port))
+		goto create_delsock;
 
 	result = sock->ops->listen(sock, 5);
 	if (result < 0) {
@@ -1582,7 +1448,7 @@ static void send_to_sock(struct connection *con)
 					      msg_flags);
 			if (ret == -EAGAIN || ret == 0) {
 				if (ret == -EAGAIN &&
-				    test_bit(SOCK_ASYNC_NOSPACE, &con->sock->flags) &&
+				    test_bit(SOCKWQ_ASYNC_NOSPACE, &con->sock->flags) &&
 				    !test_and_set_bit(CF_APP_LIMITED, &con->flags)) {
 					/* Notify TCP that we're limited by the
 					 * application window size.
@@ -1612,14 +1478,13 @@ out:
 
 send_error:
 	mutex_unlock(&con->sock_mutex);
-	close_connection(con, false);
+	close_connection(con, false, false, true);
 	lowcomms_connect_sock(con);
 	return;
 
 out_connect:
 	mutex_unlock(&con->sock_mutex);
-	if (!test_bit(CF_INIT_PENDING, &con->flags))
-		lowcomms_connect_sock(con);
+	lowcomms_connect_sock(con);
 }
 
 static void clean_one_writequeue(struct connection *con)
@@ -1644,15 +1509,9 @@ int dlm_lowcomms_close(int nodeid)
 	log_print("closing connection to node %d", nodeid);
 	con = nodeid2con(nodeid, 0);
 	if (con) {
-		clear_bit(CF_CONNECT_PENDING, &con->flags);
-		clear_bit(CF_WRITE_PENDING, &con->flags);
 		set_bit(CF_CLOSE, &con->flags);
-		if (cancel_work_sync(&con->swork))
-			log_print("canceled swork for node %d", nodeid);
-		if (cancel_work_sync(&con->rwork))
-			log_print("canceled rwork for node %d", nodeid);
+		close_connection(con, true, true, true);
 		clean_one_writequeue(con);
-		close_connection(con, true);
 	}
 
 	spin_lock(&dlm_node_addrs_spin);
@@ -1685,10 +1544,8 @@ static void process_send_sockets(struct work_struct *work)
 {
 	struct connection *con = container_of(work, struct connection, swork);
 
-	if (test_and_clear_bit(CF_CONNECT_PENDING, &con->flags)) {
+	if (test_and_clear_bit(CF_CONNECT_PENDING, &con->flags))
 		con->connect_action(con);
-		set_bit(CF_WRITE_PENDING, &con->flags);
-	}
 	if (test_and_clear_bit(CF_WRITE_PENDING, &con->flags))
 		send_to_sock(con);
 }
@@ -1735,7 +1592,7 @@ static void stop_conn(struct connection *con)
 
 static void free_conn(struct connection *con)
 {
-	close_connection(con, true);
+	close_connection(con, true, true, true);
 	if (con->othercon)
 		kmem_cache_free(con_cache, con->othercon);
 	hlist_del(&con->list);
@@ -1806,7 +1663,7 @@ fail_unlisten:
 	dlm_allow_conn = 0;
 	con = nodeid2con(0,0);
 	if (con) {
-		close_connection(con, false);
+		close_connection(con, false, true, true);
 		kmem_cache_free(con_cache, con);
 	}
 fail_destroy:
diff --git a/kernel/fs/dlm/plock.c b/kernel/fs/dlm/plock.c
index e0ab3a93e..d401425f6 100644
--- a/kernel/fs/dlm/plock.c
+++ b/kernel/fs/dlm/plock.c
@@ -145,7 +145,7 @@ int dlm_posix_lock(dlm_lockspace_t *lockspace, u64 number, struct file *file,
 	send_op(op);
 
 	if (xop->callback == NULL) {
-		rv = wait_event_killable(recv_wq, (op->done != 0));
+		rv = wait_event_interruptible(recv_wq, (op->done != 0));
 		if (rv == -ERESTARTSYS) {
 			log_debug(ls, "dlm_posix_lock: wait killed %llx",
 				  (unsigned long long)number);
@@ -172,7 +172,7 @@ int dlm_posix_lock(dlm_lockspace_t *lockspace, u64 number, struct file *file,
 	rv = op->info.rv;
 
 	if (!rv) {
-		if (posix_lock_file_wait(file, fl) < 0)
+		if (locks_lock_file_wait(file, fl) < 0)
 			log_error(ls, "dlm_posix_lock: vfs lock error %llx",
 				  (unsigned long long)number);
 	}
@@ -262,7 +262,7 @@ int dlm_posix_unlock(dlm_lockspace_t *lockspace, u64 number, struct file *file,
 	/* cause the vfs unlock to return ENOENT if lock is not found */
 	fl->fl_flags |= FL_EXISTS;
 
-	rv = posix_lock_file_wait(file, fl);
+	rv = locks_lock_file_wait(file, fl);
 	if (rv == -ENOENT) {
 		rv = 0;
 		goto out_free;
@@ -509,7 +509,6 @@ int dlm_plock_init(void)
 
 void dlm_plock_exit(void)
 {
-	if (misc_deregister(&plock_dev_misc) < 0)
-		log_print("dlm_plock_exit: misc_deregister failed");
+	misc_deregister(&plock_dev_misc);
 }
 
diff --git a/kernel/fs/dlm/user.c b/kernel/fs/dlm/user.c
index fb85f32e9..173b3873a 100644
--- a/kernel/fs/dlm/user.c
+++ b/kernel/fs/dlm/user.c
@@ -362,18 +362,15 @@ fail:
 
 int dlm_device_deregister(struct dlm_ls *ls)
 {
-	int error;
-
 	/* The device is not registered.  This happens when the lockspace
 	   was never used from userspace, or when device_create_lockspace()
 	   calls dlm_release_lockspace() after the register fails. */
 	if (!ls->ls_device.name)
 		return 0;
 
-	error = misc_deregister(&ls->ls_device);
-	if (!error)
-		kfree(ls->ls_device.name);
-	return error;
+	misc_deregister(&ls->ls_device);
+	kfree(ls->ls_device.name);
+	return 0;
 }
 
 static int device_user_purge(struct dlm_user_proc *proc,
@@ -785,6 +782,7 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count,
 	DECLARE_WAITQUEUE(wait, current);
 	struct dlm_callback cb;
 	int rv, resid, copy_lvb = 0;
+	int old_mode, new_mode;
 
 	if (count == sizeof(struct dlm_device_version)) {
 		rv = copy_version_to_user(buf, count);
@@ -841,6 +839,9 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count,
 
 	lkb = list_entry(proc->asts.next, struct dlm_lkb, lkb_cb_list);
 
+	/* rem_lkb_callback sets a new lkb_last_cast */
+	old_mode = lkb->lkb_last_cast.mode;
+
 	rv = dlm_rem_lkb_callback(lkb->lkb_resource->res_ls, lkb, &cb, &resid);
 	if (rv < 0) {
 		/* this shouldn't happen; lkb should have been removed from
@@ -864,9 +865,6 @@ static ssize_t device_read(struct file *file, char __user *buf, size_t count,
 	}
 
 	if (cb.flags & DLM_CB_CAST) {
-		int old_mode, new_mode;
-
-		old_mode = lkb->lkb_last_cast.mode;
 		new_mode = cb.mode;
 
 		if (!cb.sb_status && lkb->lkb_lksb->sb_lvbptr &&