tipc: update a binding service via broadcast

Currently, updating binding table (add service binding to
name table/withdraw a service binding) is being sent over replicast.
However, if we are scaling up clusters to > 100 nodes/containers this
method is less affection because of looping through nodes in a cluster one
by one.

It is worth to use broadcast to update a binding service. This way, the
binding table can be updated on all peer nodes in one shot.

Broadcast is used when all peer nodes, as indicated by a new capability
flag TIPC_NAMED_BCAST, support reception of this message type.

Four problems need to be considered when introducing this feature.
1) When establishing a link to a new peer node we still update this by a
unicast 'bulk' update. This may lead to race conditions, where a later
broadcast publication/withdrawal bypass the 'bulk', resulting in
disordered publications, or even that a withdrawal may arrive before the
corresponding publication. We solve this by adding an 'is_last_bulk' bit
in the last bulk messages so that it can be distinguished from all other
messages. Only when this message has arrived do we open up for reception
of broadcast publications/withdrawals.

2) When a first legacy node is added to the cluster all distribution
will switch over to use the legacy 'replicast' method, while the
opposite happens when the last legacy node leaves the cluster. This
entails another risk of message disordering that has to be handled. We
solve this by adding a sequence number to the broadcast/replicast
messages, so that disordering can be discovered and corrected. Note
however that we don't need to consider potential message loss or
duplication at this protocol level.

3) Bulk messages don't contain any sequence numbers, and will always
arrive in order. Hence we must exempt those from the sequence number
control and deliver them unconditionally. We solve this by adding a new
'is_bulk' bit in those messages so that they can be recognized.

4) Legacy messages, which don't contain any new bits or sequence
numbers, but neither can arrive out of order, also need to be exempt
from the initial synchronization and sequence number check, and
delivered unconditionally. Therefore, we add another 'is_not_legacy' bit
to all new messages so that those can be distinguished from legacy
messages and the latter delivered directly.

v1->v2:
 - fix warning issue reported by kbuild test robot <lkp@intel.com>
 - add santiy check to drop the publication message with a sequence
number that is lower than the agreed synch point

Signed-off-by: kernel test robot <lkp@intel.com>
Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au>
Acked-by: Jon Maloy <jmaloy@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 87 insertions, 29 deletions

diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c
index 5feaf3b67380..2f9c148f17e2 100644
--- a/net/tipc/name_distr.c
+++ b/net/tipc/name_distr.c
@@ -102,7 +102,8 @@ struct sk_buff *tipc_named_publish(struct net *net, struct publication *publ)
 		pr_warn("Publication distribution failure\n");
 		return NULL;
 	}
-
+	msg_set_named_seqno(buf_msg(skb), nt->snd_nxt++);
+	msg_set_non_legacy(buf_msg(skb));
 	item = (struct distr_item *)msg_data(buf_msg(skb));
 	publ_to_item(item, publ);
 	return skb;
@@ -114,8 +115,8 @@ struct sk_buff *tipc_named_publish(struct net *net, struct publication *publ)
 struct sk_buff *tipc_named_withdraw(struct net *net, struct publication *publ)
 {
 	struct name_table *nt = tipc_name_table(net);
-	struct sk_buff *buf;
 	struct distr_item *item;
+	struct sk_buff *skb;
 
 	write_lock_bh(&nt->cluster_scope_lock);
 	list_del(&publ->binding_node);
@@ -123,15 +124,16 @@ struct sk_buff *tipc_named_withdraw(struct net *net, struct publication *publ)
 	if (publ->scope == TIPC_NODE_SCOPE)
 		return NULL;
 
-	buf = named_prepare_buf(net, WITHDRAWAL, ITEM_SIZE, 0);
-	if (!buf) {
+	skb = named_prepare_buf(net, WITHDRAWAL, ITEM_SIZE, 0);
+	if (!skb) {
 		pr_warn("Withdrawal distribution failure\n");
 		return NULL;
 	}
-
-	item = (struct distr_item *)msg_data(buf_msg(buf));
+	msg_set_named_seqno(buf_msg(skb), nt->snd_nxt++);
+	msg_set_non_legacy(buf_msg(skb));
+	item = (struct distr_item *)msg_data(buf_msg(skb));
 	publ_to_item(item, publ);
-	return buf;
+	return skb;
 }
 
 /**
@@ -141,7 +143,7 @@ struct sk_buff *tipc_named_withdraw(struct net *net, struct publication *publ)
  * @pls: linked list of publication items to be packed into buffer chain
  */
 static void named_distribute(struct net *net, struct sk_buff_head *list,
-			     u32 dnode, struct list_head *pls)
+			     u32 dnode, struct list_head *pls, u16 seqno)
 {
 	struct publication *publ;
 	struct sk_buff *skb = NULL;
@@ -149,6 +151,7 @@ static void named_distribute(struct net *net, struct sk_buff_head *list,
 	u32 msg_dsz = ((tipc_node_get_mtu(net, dnode, 0, false) - INT_H_SIZE) /
 			ITEM_SIZE) * ITEM_SIZE;
 	u32 msg_rem = msg_dsz;
+	struct tipc_msg *hdr;
 
 	list_for_each_entry(publ, pls, binding_node) {
 		/* Prepare next buffer: */
@@ -159,8 +162,11 @@ static void named_distribute(struct net *net, struct sk_buff_head *list,
 				pr_warn("Bulk publication failure\n");
 				return;
 			}
-			msg_set_bc_ack_invalid(buf_msg(skb), true);
-			item = (struct distr_item *)msg_data(buf_msg(skb));
+			hdr = buf_msg(skb);
+			msg_set_bc_ack_invalid(hdr, true);
+			msg_set_bulk(hdr);
+			msg_set_non_legacy(hdr);
+			item = (struct distr_item *)msg_data(hdr);
 		}
 
 		/* Pack publication into message: */
@@ -176,24 +182,35 @@ static void named_distribute(struct net *net, struct sk_buff_head *list,
 		}
 	}
 	if (skb) {
-		msg_set_size(buf_msg(skb), INT_H_SIZE + (msg_dsz - msg_rem));
+		hdr = buf_msg(skb);
+		msg_set_size(hdr, INT_H_SIZE + (msg_dsz - msg_rem));
 		skb_trim(skb, INT_H_SIZE + (msg_dsz - msg_rem));
 		__skb_queue_tail(list, skb);
 	}
+	hdr = buf_msg(skb_peek_tail(list));
+	msg_set_last_bulk(hdr);
+	msg_set_named_seqno(hdr, seqno);
 }
 
 /**
  * tipc_named_node_up - tell specified node about all publications by this node
  */
-void tipc_named_node_up(struct net *net, u32 dnode)
+void tipc_named_node_up(struct net *net, u32 dnode, u16 capabilities)
 {
 	struct name_table *nt = tipc_name_table(net);
+	struct tipc_net *tn = tipc_net(net);
 	struct sk_buff_head head;
+	u16 seqno;
 
 	__skb_queue_head_init(&head);
+	spin_lock_bh(&tn->nametbl_lock);
+	if (!(capabilities & TIPC_NAMED_BCAST))
+		nt->rc_dests++;
+	seqno = nt->snd_nxt;
+	spin_unlock_bh(&tn->nametbl_lock);
 
 	read_lock_bh(&nt->cluster_scope_lock);
-	named_distribute(net, &head, dnode, &nt->cluster_scope);
+	named_distribute(net, &head, dnode, &nt->cluster_scope, seqno);
 	tipc_node_xmit(net, &head, dnode, 0);
 	read_unlock_bh(&nt->cluster_scope_lock);
 }
@@ -245,13 +262,21 @@ static void tipc_dist_queue_purge(struct net *net, u32 addr)
 	spin_unlock_bh(&tn->nametbl_lock);
 }
 
-void tipc_publ_notify(struct net *net, struct list_head *nsub_list, u32 addr)
+void tipc_publ_notify(struct net *net, struct list_head *nsub_list,
+		      u32 addr, u16 capabilities)
 {
+	struct name_table *nt = tipc_name_table(net);
+	struct tipc_net *tn = tipc_net(net);
+
 	struct publication *publ, *tmp;
 
 	list_for_each_entry_safe(publ, tmp, nsub_list, binding_node)
 		tipc_publ_purge(net, publ, addr);
 	tipc_dist_queue_purge(net, addr);
+	spin_lock_bh(&tn->nametbl_lock);
+	if (!(capabilities & TIPC_NAMED_BCAST))
+		nt->rc_dests--;
+	spin_unlock_bh(&tn->nametbl_lock);
 }
 
 /**
@@ -295,29 +320,62 @@ static bool tipc_update_nametbl(struct net *net, struct distr_item *i,
 	return false;
 }
 
+static struct sk_buff *tipc_named_dequeue(struct sk_buff_head *namedq,
+					  u16 *rcv_nxt, bool *open)
+{
+	struct sk_buff *skb, *tmp;
+	struct tipc_msg *hdr;
+	u16 seqno;
+
+	skb_queue_walk_safe(namedq, skb, tmp) {
+		skb_linearize(skb);
+		hdr = buf_msg(skb);
+		seqno = msg_named_seqno(hdr);
+		if (msg_is_last_bulk(hdr)) {
+			*rcv_nxt = seqno;
+			*open = true;
+		}
+
+		if (msg_is_bulk(hdr) || msg_is_legacy(hdr)) {
+			__skb_unlink(skb, namedq);
+			return skb;
+		}
+
+		if (*open && (*rcv_nxt == seqno)) {
+			(*rcv_nxt)++;
+			__skb_unlink(skb, namedq);
+			return skb;
+		}
+
+		if (less(seqno, *rcv_nxt)) {
+			__skb_unlink(skb, namedq);
+			kfree_skb(skb);
+			continue;
+		}
+	}
+	return NULL;
+}
+
 /**
  * tipc_named_rcv - process name table update messages sent by another node
  */
-void tipc_named_rcv(struct net *net, struct sk_buff_head *inputq)
+void tipc_named_rcv(struct net *net, struct sk_buff_head *namedq,
+		    u16 *rcv_nxt, bool *open)
 {
-	struct tipc_net *tn = net_generic(net, tipc_net_id);
-	struct tipc_msg *msg;
+	struct tipc_net *tn = tipc_net(net);
 	struct distr_item *item;
-	uint count;
-	u32 node;
+	struct tipc_msg *hdr;
 	struct sk_buff *skb;
-	int mtype;
+	u32 count, node;
 
 	spin_lock_bh(&tn->nametbl_lock);
-	for (skb = skb_dequeue(inputq); skb; skb = skb_dequeue(inputq)) {
-		skb_linearize(skb);
-		msg = buf_msg(skb);
-		mtype = msg_type(msg);
-		item = (struct distr_item *)msg_data(msg);
-		count = msg_data_sz(msg) / ITEM_SIZE;
-		node = msg_orignode(msg);
+	while ((skb = tipc_named_dequeue(namedq, rcv_nxt, open))) {
+		hdr = buf_msg(skb);
+		node = msg_orignode(hdr);
+		item = (struct distr_item *)msg_data(hdr);
+		count = msg_data_sz(hdr) / ITEM_SIZE;
 		while (count--) {
-			tipc_update_nametbl(net, item, node, mtype);
+			tipc_update_nametbl(net, item, node, msg_type(hdr));
 			item++;
 		}
 		kfree_skb(skb);
@@ -345,6 +403,6 @@ void tipc_named_reinit(struct net *net)
 		publ->node = self;
 	list_for_each_entry_rcu(publ, &nt->cluster_scope, binding_node)
 		publ->node = self;
-
+	nt->rc_dests = 0;
 	spin_unlock_bh(&tn->nametbl_lock);
 }