1 files changed, 143 insertions, 56 deletions

diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c
index ea9e5817b9e9..3e4ad801786f 100644
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -143,11 +143,13 @@ static bool lookup_subflow_by_daddr(const struct list_head *list,
 	return false;
 }
 
-static struct mptcp_pm_addr_entry *
+static bool
 select_local_address(const struct pm_nl_pernet *pernet,
-		     const struct mptcp_sock *msk)
+		     const struct mptcp_sock *msk,
+		     struct mptcp_pm_addr_entry *new_entry)
 {
-	struct mptcp_pm_addr_entry *entry, *ret = NULL;
+	struct mptcp_pm_addr_entry *entry;
+	bool found = false;
 
 	msk_owned_by_me(msk);
 
@@ -159,17 +161,21 @@ select_local_address(const struct pm_nl_pernet *pernet,
 		if (!test_bit(entry->addr.id, msk->pm.id_avail_bitmap))
 			continue;
 
-		ret = entry;
+		*new_entry = *entry;
+		found = true;
 		break;
 	}
 	rcu_read_unlock();
-	return ret;
+
+	return found;
 }
 
-static struct mptcp_pm_addr_entry *
-select_signal_address(struct pm_nl_pernet *pernet, const struct mptcp_sock *msk)
+static bool
+select_signal_address(struct pm_nl_pernet *pernet, const struct mptcp_sock *msk,
+		      struct mptcp_pm_addr_entry *new_entry)
 {
-	struct mptcp_pm_addr_entry *entry, *ret = NULL;
+	struct mptcp_pm_addr_entry *entry;
+	bool found = false;
 
 	rcu_read_lock();
 	/* do not keep any additional per socket state, just signal
@@ -184,11 +190,13 @@ select_signal_address(struct pm_nl_pernet *pernet, const struct mptcp_sock *msk)
 		if (!(entry->flags & MPTCP_PM_ADDR_FLAG_SIGNAL))
 			continue;
 
-		ret = entry;
+		*new_entry = *entry;
+		found = true;
 		break;
 	}
 	rcu_read_unlock();
-	return ret;
+
+	return found;
 }
 
 unsigned int mptcp_pm_get_add_addr_signal_max(const struct mptcp_sock *msk)
@@ -348,7 +356,7 @@ bool mptcp_pm_alloc_anno_list(struct mptcp_sock *msk,
 	add_entry = mptcp_lookup_anno_list_by_saddr(msk, addr);
 
 	if (add_entry) {
-		if (mptcp_pm_is_kernel(msk))
+		if (WARN_ON_ONCE(mptcp_pm_is_kernel(msk)))
 			return false;
 
 		sk_reset_timer(sk, &add_entry->add_timer,
@@ -471,7 +479,6 @@ static void __mptcp_pm_send_ack(struct mptcp_sock *msk, struct mptcp_subflow_con
 	slow = lock_sock_fast(ssk);
 	if (prio) {
 		subflow->send_mp_prio = 1;
-		subflow->backup = backup;
 		subflow->request_bkup = backup;
 	}
 
@@ -514,8 +521,9 @@ __lookup_addr(struct pm_nl_pernet *pernet, const struct mptcp_addr_info *info)
 static void mptcp_pm_create_subflow_or_signal_addr(struct mptcp_sock *msk)
 {
 	struct sock *sk = (struct sock *)msk;
-	struct mptcp_pm_addr_entry *local;
+	struct mptcp_pm_addr_entry local;
 	unsigned int add_addr_signal_max;
+	bool signal_and_subflow = false;
 	unsigned int local_addr_max;
 	struct pm_nl_pernet *pernet;
 	unsigned int subflows_max;
@@ -556,8 +564,6 @@ static void mptcp_pm_create_subflow_or_signal_addr(struct mptcp_sock *msk)
 
 	/* check first for announce */
 	if (msk->pm.add_addr_signaled < add_addr_signal_max) {
-		local = select_signal_address(pernet, msk);
-
 		/* due to racing events on both ends we can reach here while
 		 * previous add address is still running: if we invoke now
 		 * mptcp_pm_announce_addr(), that will fail and the
@@ -568,16 +574,25 @@ static void mptcp_pm_create_subflow_or_signal_addr(struct mptcp_sock *msk)
 		if (msk->pm.addr_signal & BIT(MPTCP_ADD_ADDR_SIGNAL))
 			return;
 
-		if (local) {
-			if (mptcp_pm_alloc_anno_list(msk, &local->addr)) {
-				__clear_bit(local->addr.id, msk->pm.id_avail_bitmap);
-				msk->pm.add_addr_signaled++;
-				mptcp_pm_announce_addr(msk, &local->addr, false);
-				mptcp_pm_nl_addr_send_ack(msk);
-			}
-		}
+		if (!select_signal_address(pernet, msk, &local))
+			goto subflow;
+
+		/* If the alloc fails, we are on memory pressure, not worth
+		 * continuing, and trying to create subflows.
+		 */
+		if (!mptcp_pm_alloc_anno_list(msk, &local.addr))
+			return;
+
+		__clear_bit(local.addr.id, msk->pm.id_avail_bitmap);
+		msk->pm.add_addr_signaled++;
+		mptcp_pm_announce_addr(msk, &local.addr, false);
+		mptcp_pm_nl_addr_send_ack(msk);
+
+		if (local.flags & MPTCP_PM_ADDR_FLAG_SUBFLOW)
+			signal_and_subflow = true;
 	}
 
+subflow:
 	/* check if should create a new subflow */
 	while (msk->pm.local_addr_used < local_addr_max &&
 	       msk->pm.subflows < subflows_max) {
@@ -585,21 +600,22 @@ static void mptcp_pm_create_subflow_or_signal_addr(struct mptcp_sock *msk)
 		bool fullmesh;
 		int i, nr;
 
-		local = select_local_address(pernet, msk);
-		if (!local)
+		if (signal_and_subflow)
+			signal_and_subflow = false;
+		else if (!select_local_address(pernet, msk, &local))
 			break;
 
-		fullmesh = !!(local->flags & MPTCP_PM_ADDR_FLAG_FULLMESH);
+		fullmesh = !!(local.flags & MPTCP_PM_ADDR_FLAG_FULLMESH);
 
 		msk->pm.local_addr_used++;
-		__clear_bit(local->addr.id, msk->pm.id_avail_bitmap);
-		nr = fill_remote_addresses_vec(msk, &local->addr, fullmesh, addrs);
+		__clear_bit(local.addr.id, msk->pm.id_avail_bitmap);
+		nr = fill_remote_addresses_vec(msk, &local.addr, fullmesh, addrs);
 		if (nr == 0)
 			continue;
 
 		spin_unlock_bh(&msk->pm.lock);
 		for (i = 0; i < nr; i++)
-			__mptcp_subflow_connect(sk, &local->addr, &addrs[i]);
+			__mptcp_subflow_connect(sk, &local.addr, &addrs[i]);
 		spin_lock_bh(&msk->pm.lock);
 	}
 	mptcp_pm_nl_check_work_pending(msk);
@@ -624,6 +640,7 @@ static unsigned int fill_local_addresses_vec(struct mptcp_sock *msk,
 {
 	struct sock *sk = (struct sock *)msk;
 	struct mptcp_pm_addr_entry *entry;
+	struct mptcp_addr_info mpc_addr;
 	struct pm_nl_pernet *pernet;
 	unsigned int subflows_max;
 	int i = 0;
@@ -631,6 +648,8 @@ static unsigned int fill_local_addresses_vec(struct mptcp_sock *msk,
 	pernet = pm_nl_get_pernet_from_msk(msk);
 	subflows_max = mptcp_pm_get_subflows_max(msk);
 
+	mptcp_local_address((struct sock_common *)msk, &mpc_addr);
+
 	rcu_read_lock();
 	list_for_each_entry_rcu(entry, &pernet->local_addr_list, list) {
 		if (!(entry->flags & MPTCP_PM_ADDR_FLAG_FULLMESH))
@@ -641,7 +660,13 @@ static unsigned int fill_local_addresses_vec(struct mptcp_sock *msk,
 
 		if (msk->pm.subflows < subflows_max) {
 			msk->pm.subflows++;
-			addrs[i++] = entry->addr;
+			addrs[i] = entry->addr;
+
+			/* Special case for ID0: set the correct ID */
+			if (mptcp_addresses_equal(&entry->addr, &mpc_addr, entry->addr.port))
+				addrs[i].id = 0;
+
+			i++;
 		}
 	}
 	rcu_read_unlock();
@@ -817,25 +842,27 @@ static void mptcp_pm_nl_rm_addr_or_subflow(struct mptcp_sock *msk,
 			mptcp_close_ssk(sk, ssk, subflow);
 			spin_lock_bh(&msk->pm.lock);
 
-			removed = true;
+			removed |= subflow->request_join;
 			if (rm_type == MPTCP_MIB_RMSUBFLOW)
 				__MPTCP_INC_STATS(sock_net(sk), rm_type);
 		}
-		if (rm_type == MPTCP_MIB_RMSUBFLOW)
-			__set_bit(rm_id ? rm_id : msk->mpc_endpoint_id, msk->pm.id_avail_bitmap);
-		else if (rm_type == MPTCP_MIB_RMADDR)
+
+		if (rm_type == MPTCP_MIB_RMADDR)
 			__MPTCP_INC_STATS(sock_net(sk), rm_type);
+
 		if (!removed)
 			continue;
 
 		if (!mptcp_pm_is_kernel(msk))
 			continue;
 
-		if (rm_type == MPTCP_MIB_RMADDR) {
-			msk->pm.add_addr_accepted--;
-			WRITE_ONCE(msk->pm.accept_addr, true);
-		} else if (rm_type == MPTCP_MIB_RMSUBFLOW) {
-			msk->pm.local_addr_used--;
+		if (rm_type == MPTCP_MIB_RMADDR && rm_id &&
+		    !WARN_ON_ONCE(msk->pm.add_addr_accepted == 0)) {
+			/* Note: if the subflow has been closed before, this
+			 * add_addr_accepted counter will not be decremented.
+			 */
+			if (--msk->pm.add_addr_accepted < mptcp_pm_get_add_addr_accept_max(msk))
+				WRITE_ONCE(msk->pm.accept_addr, true);
 		}
 	}
 }
@@ -845,8 +872,8 @@ static void mptcp_pm_nl_rm_addr_received(struct mptcp_sock *msk)
 	mptcp_pm_nl_rm_addr_or_subflow(msk, &msk->pm.rm_list_rx, MPTCP_MIB_RMADDR);
 }
 
-void mptcp_pm_nl_rm_subflow_received(struct mptcp_sock *msk,
-				     const struct mptcp_rm_list *rm_list)
+static void mptcp_pm_nl_rm_subflow_received(struct mptcp_sock *msk,
+					    const struct mptcp_rm_list *rm_list)
 {
 	mptcp_pm_nl_rm_addr_or_subflow(msk, rm_list, MPTCP_MIB_RMSUBFLOW);
 }
@@ -1102,6 +1129,24 @@ int mptcp_pm_nl_get_local_id(struct mptcp_sock *msk, struct mptcp_addr_info *skc
 	return ret;
 }
 
+bool mptcp_pm_nl_is_backup(struct mptcp_sock *msk, struct mptcp_addr_info *skc)
+{
+	struct pm_nl_pernet *pernet = pm_nl_get_pernet_from_msk(msk);
+	struct mptcp_pm_addr_entry *entry;
+	bool backup = false;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(entry, &pernet->local_addr_list, list) {
+		if (mptcp_addresses_equal(&entry->addr, skc, entry->addr.port)) {
+			backup = !!(entry->flags & MPTCP_PM_ADDR_FLAG_BACKUP);
+			break;
+		}
+	}
+	rcu_read_unlock();
+
+	return backup;
+}
+
 #define MPTCP_PM_CMD_GRP_OFFSET       0
 #define MPTCP_PM_EV_GRP_OFFSET        1
 
@@ -1311,8 +1356,8 @@ int mptcp_pm_nl_add_addr_doit(struct sk_buff *skb, struct genl_info *info)
 	if (ret < 0)
 		return ret;
 
-	if (addr.addr.port && !(addr.flags & MPTCP_PM_ADDR_FLAG_SIGNAL)) {
-		GENL_SET_ERR_MSG(info, "flags must have signal when using port");
+	if (addr.addr.port && !address_use_port(&addr)) {
+		GENL_SET_ERR_MSG(info, "flags must have signal and not subflow when using port");
 		return -EINVAL;
 	}
 
@@ -1363,6 +1408,10 @@ int mptcp_pm_nl_get_flags_and_ifindex_by_id(struct mptcp_sock *msk, unsigned int
 	struct sock *sk = (struct sock *)msk;
 	struct net *net = sock_net(sk);
 
+	/* No entries with ID 0 */
+	if (id == 0)
+		return 0;
+
 	rcu_read_lock();
 	entry = __lookup_addr_by_id(pm_nl_get_pernet(net), id);
 	if (entry) {
@@ -1401,12 +1450,24 @@ static bool mptcp_pm_remove_anno_addr(struct mptcp_sock *msk,
 	ret = remove_anno_list_by_saddr(msk, addr);
 	if (ret || force) {
 		spin_lock_bh(&msk->pm.lock);
+		if (ret) {
+			__set_bit(addr->id, msk->pm.id_avail_bitmap);
+			msk->pm.add_addr_signaled--;
+		}
 		mptcp_pm_remove_addr(msk, &list);
 		spin_unlock_bh(&msk->pm.lock);
 	}
 	return ret;
 }
 
+static void __mark_subflow_endp_available(struct mptcp_sock *msk, u8 id)
+{
+	/* If it was marked as used, and not ID 0, decrement local_addr_used */
+	if (!__test_and_set_bit(id ? : msk->mpc_endpoint_id, msk->pm.id_avail_bitmap) &&
+	    id && !WARN_ON_ONCE(msk->pm.local_addr_used == 0))
+		msk->pm.local_addr_used--;
+}
+
 static int mptcp_nl_remove_subflow_and_signal_addr(struct net *net,
 						   const struct mptcp_pm_addr_entry *entry)
 {
@@ -1435,8 +1496,19 @@ static int mptcp_nl_remove_subflow_and_signal_addr(struct net *net,
 		remove_subflow = lookup_subflow_by_saddr(&msk->conn_list, addr);
 		mptcp_pm_remove_anno_addr(msk, addr, remove_subflow &&
 					  !(entry->flags & MPTCP_PM_ADDR_FLAG_IMPLICIT));
-		if (remove_subflow)
-			mptcp_pm_remove_subflow(msk, &list);
+
+		if (remove_subflow) {
+			spin_lock_bh(&msk->pm.lock);
+			mptcp_pm_nl_rm_subflow_received(msk, &list);
+			spin_unlock_bh(&msk->pm.lock);
+		}
+
+		if (entry->flags & MPTCP_PM_ADDR_FLAG_SUBFLOW) {
+			spin_lock_bh(&msk->pm.lock);
+			__mark_subflow_endp_available(msk, list.ids[0]);
+			spin_unlock_bh(&msk->pm.lock);
+		}
+
 		release_sock(sk);
 
 next:
@@ -1471,6 +1543,7 @@ static int mptcp_nl_remove_id_zero_address(struct net *net,
 		spin_lock_bh(&msk->pm.lock);
 		mptcp_pm_remove_addr(msk, &list);
 		mptcp_pm_nl_rm_subflow_received(msk, &list);
+		__mark_subflow_endp_available(msk, 0);
 		spin_unlock_bh(&msk->pm.lock);
 		release_sock(sk);
 
@@ -1534,16 +1607,25 @@ void mptcp_pm_remove_addrs(struct mptcp_sock *msk, struct list_head *rm_list)
 {
 	struct mptcp_rm_list alist = { .nr = 0 };
 	struct mptcp_pm_addr_entry *entry;
+	int anno_nr = 0;
 
 	list_for_each_entry(entry, rm_list, list) {
-		if ((remove_anno_list_by_saddr(msk, &entry->addr) ||
-		     lookup_subflow_by_saddr(&msk->conn_list, &entry->addr)) &&
-		    alist.nr < MPTCP_RM_IDS_MAX)
-			alist.ids[alist.nr++] = entry->addr.id;
+		if (alist.nr >= MPTCP_RM_IDS_MAX)
+			break;
+
+		/* only delete if either announced or matching a subflow */
+		if (remove_anno_list_by_saddr(msk, &entry->addr))
+			anno_nr++;
+		else if (!lookup_subflow_by_saddr(&msk->conn_list,
+						  &entry->addr))
+			continue;
+
+		alist.ids[alist.nr++] = entry->addr.id;
 	}
 
 	if (alist.nr) {
 		spin_lock_bh(&msk->pm.lock);
+		msk->pm.add_addr_signaled -= anno_nr;
 		mptcp_pm_remove_addr(msk, &alist);
 		spin_unlock_bh(&msk->pm.lock);
 	}
@@ -1556,22 +1638,26 @@ static void mptcp_pm_remove_addrs_and_subflows(struct mptcp_sock *msk,
 	struct mptcp_pm_addr_entry *entry;
 
 	list_for_each_entry(entry, rm_list, list) {
-		if (lookup_subflow_by_saddr(&msk->conn_list, &entry->addr) &&
-		    slist.nr < MPTCP_RM_IDS_MAX)
+		if (slist.nr < MPTCP_RM_IDS_MAX &&
+		    lookup_subflow_by_saddr(&msk->conn_list, &entry->addr))
 			slist.ids[slist.nr++] = entry->addr.id;
 
-		if (remove_anno_list_by_saddr(msk, &entry->addr) &&
-		    alist.nr < MPTCP_RM_IDS_MAX)
+		if (alist.nr < MPTCP_RM_IDS_MAX &&
+		    remove_anno_list_by_saddr(msk, &entry->addr))
 			alist.ids[alist.nr++] = entry->addr.id;
 	}
 
+	spin_lock_bh(&msk->pm.lock);
 	if (alist.nr) {
-		spin_lock_bh(&msk->pm.lock);
+		msk->pm.add_addr_signaled -= alist.nr;
 		mptcp_pm_remove_addr(msk, &alist);
-		spin_unlock_bh(&msk->pm.lock);
 	}
 	if (slist.nr)
-		mptcp_pm_remove_subflow(msk, &slist);
+		mptcp_pm_nl_rm_subflow_received(msk, &slist);
+	/* Reset counters: maybe some subflows have been removed before */
+	bitmap_fill(msk->pm.id_avail_bitmap, MPTCP_PM_MAX_ADDR_ID + 1);
+	msk->pm.local_addr_used = 0;
+	spin_unlock_bh(&msk->pm.lock);
 }
 
 static void mptcp_nl_remove_addrs_list(struct net *net,
@@ -1859,6 +1945,7 @@ static void mptcp_pm_nl_fullmesh(struct mptcp_sock *msk,
 
 	spin_lock_bh(&msk->pm.lock);
 	mptcp_pm_nl_rm_subflow_received(msk, &list);
+	__mark_subflow_endp_available(msk, list.ids[0]);
 	mptcp_pm_create_subflow_or_signal_addr(msk);
 	spin_unlock_bh(&msk->pm.lock);
 }