Merge branch 'mptcp-Include-multiple-address-ids-in-RM_ADDR'

Mat Martineau says:

====================
mptcp: Include multiple address ids in RM_ADDR

Here's a patch series from the MPTCP tree that extends the capabilities
of the MPTCP RM_ADDR header.

MPTCP peers can exchange information about their IP addresses that are
available for additional MPTCP subflows. IP addresses are advertised
with an ADD_ADDR header type, and those advertisements are revoked with
the RM_ADDR header type. RFC 8684 allows the RM_ADDR header to include
more than one address ID, so multiple advertisements can be revoked in a
single header. Previous kernel versions have only used RM_ADDR with a
single address ID, so multiple removals required multiple packets.

Patches 1-4 plumb address id list structures around the MPTCP code,
where before only a single address ID was passed.

Patches 5-8 make use of the address lists at the path manager layer that
tracks available addresses for both peers.

Patches 9-11 update the selftests to cover the new use of RM_ADDR with
multiple address IDs.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

include/net/mptcp.h

@@ -34,6 +34,13 @@ struct mptcp_ext {
 	/* one byte hole */
 };
 
+#define MPTCP_RM_IDS_MAX	8
+
+struct mptcp_rm_list {
+	u8 ids[MPTCP_RM_IDS_MAX];
+	u8 nr;
+};
+
 struct mptcp_out_options {
 #if IS_ENABLED(CONFIG_MPTCP)
 	u16 suboptions;
@@ -48,7 +55,7 @@ struct mptcp_out_options {
 	u8 addr_id;
 	u16 port;
 	u64 ahmac;
-	u8 rm_id;
+	struct mptcp_rm_list rm_list;
 	u8 join_id;
 	u8 backup;
 	u32 nonce;

net/mptcp/options.c

@@ -26,6 +26,7 @@ static void mptcp_parse_option(const struct sk_buff *skb,
 	int expected_opsize;
 	u8 version;
 	u8 flags;
+	u8 i;
 
 	switch (subtype) {
 	case MPTCPOPT_MP_CAPABLE:
@@ -272,14 +273,17 @@ static void mptcp_parse_option(const struct sk_buff *skb,
 		break;
 
 	case MPTCPOPT_RM_ADDR:
-		if (opsize != TCPOLEN_MPTCP_RM_ADDR_BASE)
+		if (opsize < TCPOLEN_MPTCP_RM_ADDR_BASE + 1 ||
+		    opsize > TCPOLEN_MPTCP_RM_ADDR_BASE + MPTCP_RM_IDS_MAX)
 			break;
 
 		ptr++;
 
 		mp_opt->rm_addr = 1;
-		mp_opt->rm_id = *ptr++;
-		pr_debug("RM_ADDR: id=%d", mp_opt->rm_id);
+		mp_opt->rm_list.nr = opsize - TCPOLEN_MPTCP_RM_ADDR_BASE;
+		for (i = 0; i < mp_opt->rm_list.nr; i++)
+			mp_opt->rm_list.ids[i] = *ptr++;
+		pr_debug("RM_ADDR: rm_list_nr=%d", mp_opt->rm_list.nr);
 		break;
 
 	case MPTCPOPT_MP_PRIO:
@@ -674,20 +678,25 @@ static bool mptcp_established_options_rm_addr(struct sock *sk,
 {
 	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
 	struct mptcp_sock *msk = mptcp_sk(subflow->conn);
-	u8 rm_id;
+	struct mptcp_rm_list rm_list;
+	int i, len;
 
 	if (!mptcp_pm_should_rm_signal(msk) ||
-	    !(mptcp_pm_rm_addr_signal(msk, remaining, &rm_id)))
+	    !(mptcp_pm_rm_addr_signal(msk, remaining, &rm_list)))
 		return false;
 
-	if (remaining < TCPOLEN_MPTCP_RM_ADDR_BASE)
+	len = mptcp_rm_addr_len(&rm_list);
+	if (len < 0)
+		return false;
+	if (remaining < len)
 		return false;
 
-	*size = TCPOLEN_MPTCP_RM_ADDR_BASE;
+	*size = len;
 	opts->suboptions |= OPTION_MPTCP_RM_ADDR;
-	opts->rm_id = rm_id;
+	opts->rm_list = rm_list;
 
-	pr_debug("rm_id=%d", opts->rm_id);
+	for (i = 0; i < opts->rm_list.nr; i++)
+		pr_debug("rm_list_ids[%d]=%d", i, opts->rm_list.ids[i]);
 
 	return true;
 }
@@ -1038,7 +1047,7 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb)
 	}
 
 	if (mp_opt.rm_addr) {
-		mptcp_pm_rm_addr_received(msk, mp_opt.rm_id);
+		mptcp_pm_rm_addr_received(msk, &mp_opt.rm_list);
 		mp_opt.rm_addr = 0;
 	}
 
@@ -1217,9 +1226,23 @@ void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp,
 	}
 
 	if (OPTION_MPTCP_RM_ADDR & opts->suboptions) {
+		u8 i = 1;
+
 		*ptr++ = mptcp_option(MPTCPOPT_RM_ADDR,
-				      TCPOLEN_MPTCP_RM_ADDR_BASE,
-				      0, opts->rm_id);
+				      TCPOLEN_MPTCP_RM_ADDR_BASE + opts->rm_list.nr,
+				      0, opts->rm_list.ids[0]);
+
+		while (i < opts->rm_list.nr) {
+			u8 id1, id2, id3, id4;
+
+			id1 = opts->rm_list.ids[i];
+			id2 = i + 1 < opts->rm_list.nr ? opts->rm_list.ids[i + 1] : TCPOPT_NOP;
+			id3 = i + 2 < opts->rm_list.nr ? opts->rm_list.ids[i + 2] : TCPOPT_NOP;
+			id4 = i + 3 < opts->rm_list.nr ? opts->rm_list.ids[i + 3] : TCPOPT_NOP;
+			put_unaligned_be32(id1 << 24 | id2 << 16 | id3 << 8 | id4, ptr);
+			ptr += 1;
+			i += 4;
+		}
 	}
 
 	if (OPTION_MPTCP_PRIO & opts->suboptions) {

net/mptcp/pm.c

@@ -39,29 +39,29 @@ int mptcp_pm_announce_addr(struct mptcp_sock *msk,
 	return 0;
 }
 
-int mptcp_pm_remove_addr(struct mptcp_sock *msk, u8 local_id)
+int mptcp_pm_remove_addr(struct mptcp_sock *msk, const struct mptcp_rm_list *rm_list)
 {
 	u8 rm_addr = READ_ONCE(msk->pm.addr_signal);
 
-	pr_debug("msk=%p, local_id=%d", msk, local_id);
+	pr_debug("msk=%p, rm_list_nr=%d", msk, rm_list->nr);
 
 	if (rm_addr) {
 		pr_warn("addr_signal error, rm_addr=%d", rm_addr);
 		return -EINVAL;
 	}
 
-	msk->pm.rm_id = local_id;
+	msk->pm.rm_list_tx = *rm_list;
 	rm_addr |= BIT(MPTCP_RM_ADDR_SIGNAL);
 	WRITE_ONCE(msk->pm.addr_signal, rm_addr);
 	return 0;
 }
 
-int mptcp_pm_remove_subflow(struct mptcp_sock *msk, u8 local_id)
+int mptcp_pm_remove_subflow(struct mptcp_sock *msk, const struct mptcp_rm_list *rm_list)
 {
-	pr_debug("msk=%p, local_id=%d", msk, local_id);
+	pr_debug("msk=%p, rm_list_nr=%d", msk, rm_list->nr);
 
 	spin_lock_bh(&msk->pm.lock);
-	mptcp_pm_nl_rm_subflow_received(msk, local_id);
+	mptcp_pm_nl_rm_subflow_received(msk, rm_list);
 	spin_unlock_bh(&msk->pm.lock);
 	return 0;
 }
@@ -205,17 +205,20 @@ void mptcp_pm_add_addr_send_ack(struct mptcp_sock *msk)
 	mptcp_pm_schedule_work(msk, MPTCP_PM_ADD_ADDR_SEND_ACK);
 }
 
-void mptcp_pm_rm_addr_received(struct mptcp_sock *msk, u8 rm_id)
+void mptcp_pm_rm_addr_received(struct mptcp_sock *msk,
+			       const struct mptcp_rm_list *rm_list)
 {
 	struct mptcp_pm_data *pm = &msk->pm;
+	u8 i;
 
-	pr_debug("msk=%p remote_id=%d", msk, rm_id);
+	pr_debug("msk=%p remote_ids_nr=%d", msk, rm_list->nr);
 
-	mptcp_event_addr_removed(msk, rm_id);
+	for (i = 0; i < rm_list->nr; i++)
+		mptcp_event_addr_removed(msk, rm_list->ids[i]);
 
 	spin_lock_bh(&pm->lock);
 	mptcp_pm_schedule_work(msk, MPTCP_PM_RM_ADDR_RECEIVED);
-	pm->rm_id = rm_id;
+	pm->rm_list_rx = *rm_list;
 	spin_unlock_bh(&pm->lock);
 }
 
@@ -258,20 +261,25 @@ bool mptcp_pm_add_addr_signal(struct mptcp_sock *msk, unsigned int remaining,
 }
 
 bool mptcp_pm_rm_addr_signal(struct mptcp_sock *msk, unsigned int remaining,
-			     u8 *rm_id)
+			     struct mptcp_rm_list *rm_list)
 {
-	int ret = false;
+	int ret = false, len;
 
 	spin_lock_bh(&msk->pm.lock);
 
 	/* double check after the lock is acquired */
 	if (!mptcp_pm_should_rm_signal(msk))
 		goto out_unlock;
 
-	if (remaining < TCPOLEN_MPTCP_RM_ADDR_BASE)
+	len = mptcp_rm_addr_len(&msk->pm.rm_list_tx);
+	if (len < 0) {
+		WRITE_ONCE(msk->pm.addr_signal, 0);
+		goto out_unlock;
+	}
+	if (remaining < len)
 		goto out_unlock;
 
-	*rm_id = msk->pm.rm_id;
+	*rm_list = msk->pm.rm_list_tx;
 	WRITE_ONCE(msk->pm.addr_signal, 0);
 	ret = true;
 
@@ -291,7 +299,8 @@ void mptcp_pm_data_init(struct mptcp_sock *msk)
 	msk->pm.add_addr_accepted = 0;
 	msk->pm.local_addr_used = 0;
 	msk->pm.subflows = 0;
-	msk->pm.rm_id = 0;
+	msk->pm.rm_list_tx.nr = 0;
+	msk->pm.rm_list_rx.nr = 0;
 	WRITE_ONCE(msk->pm.work_pending, false);
 	WRITE_ONCE(msk->pm.addr_signal, 0);
 	WRITE_ONCE(msk->pm.accept_addr, false);