Merge branch 'mptcp-RM_ADDR-ADD_ADDR-enhancements'

Geliang Tang says:

====================
mptcp: RM_ADDR/ADD_ADDR enhancements

This series include two enhancements for the MPTCP path management,
namely RM_ADDR support and ADD_ADDR echo support, as specified by RFC
sections 3.4.1 and 3.4.2.

1 RM_ADDR support include 9 patches (1-3 and 8-13):

Patch 1 is the helper for patch 2, these two patches add the RM_ADDR
outgoing functions, which are derived from ADD_ADDR's corresponding
functions.

Patch 3 adds the RM_ADDR incoming logic, when RM_ADDR suboption is
received, close the subflow matching the rm_id, and update PM counter.

Patch 8 is the main remove routine. When the PM netlink removes an address,
we traverse all the existing msk sockets to find the relevant sockets. Then
trigger the RM_ADDR signal and remove the subflow which using this local
address, this subflow removing functions has been implemented in patch 9.

Finally, patches 10-13 are the self-tests for RM_ADDR.

2 ADD_ADDR echo support include 7 patches (4-7 and 14-16).

Patch 4 adds the ADD_ADDR echo logic, when the ADD_ADDR suboption has been
received, send out the same ADD_ADDR suboption with echo-flag, and no HMAC
included.

Patches 5 and 6 are the self-tests for ADD_ADDR echo. Patch 7 is a little
cleaning up.

Patch 14 and 15 are the helpers for patch 16. These three patches add
the ADD_ADDR retransmition when no ADD_ADDR echo is received.
====================

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

include/net/sock.h

@@ -2195,6 +2195,8 @@ void sk_reset_timer(struct sock *sk, struct timer_list *timer,
 
 void sk_stop_timer(struct sock *sk, struct timer_list *timer);
 
+void sk_stop_timer_sync(struct sock *sk, struct timer_list *timer);
+
 int __sk_queue_drop_skb(struct sock *sk, struct sk_buff_head *sk_queue,
 			struct sk_buff *skb, unsigned int flags,
 			void (*destructor)(struct sock *sk,

net/core/sock.c

@@ -2947,6 +2947,13 @@ void sk_stop_timer(struct sock *sk, struct timer_list* timer)
 }
 EXPORT_SYMBOL(sk_stop_timer);
 
+void sk_stop_timer_sync(struct sock *sk, struct timer_list *timer)
+{
+	if (del_timer_sync(timer))
+		__sock_put(sk);
+}
+EXPORT_SYMBOL(sk_stop_timer_sync);
+
 void sock_init_data(struct socket *sock, struct sock *sk)
 {
 	sk_init_common(sk);

net/mptcp/mib.c

@@ -27,6 +27,10 @@ static const struct snmp_mib mptcp_snmp_list[] = {
 	SNMP_MIB_ITEM("OFOMerge", MPTCP_MIB_OFOMERGE),
 	SNMP_MIB_ITEM("NoDSSInWindow", MPTCP_MIB_NODSSWINDOW),
 	SNMP_MIB_ITEM("DuplicateData", MPTCP_MIB_DUPDATA),
+	SNMP_MIB_ITEM("AddAddr", MPTCP_MIB_ADDADDR),
+	SNMP_MIB_ITEM("EchoAdd", MPTCP_MIB_ECHOADD),
+	SNMP_MIB_ITEM("RmAddr", MPTCP_MIB_RMADDR),
+	SNMP_MIB_ITEM("RmSubflow", MPTCP_MIB_RMSUBFLOW),
 	SNMP_MIB_SENTINEL
 };
 

net/mptcp/mib.h

@@ -20,6 +20,10 @@ enum linux_mptcp_mib_field {
 	MPTCP_MIB_OFOMERGE,		/* Segments merged in OoO queue */
 	MPTCP_MIB_NODSSWINDOW,		/* Segments not in MPTCP windows */
 	MPTCP_MIB_DUPDATA,		/* Segments discarded due to duplicate DSS */
+	MPTCP_MIB_ADDADDR,		/* Received ADD_ADDR with echo-flag=0 */
+	MPTCP_MIB_ECHOADD,		/* Received ADD_ADDR with echo-flag=1 */
+	MPTCP_MIB_RMADDR,		/* Received RM_ADDR */
+	MPTCP_MIB_RMSUBFLOW,		/* Remove a subflow */
 	__MPTCP_MIB_MAX
 };
 

net/mptcp/options.c

@@ -11,6 +11,7 @@
 #include <net/tcp.h>
 #include <net/mptcp.h>
 #include "protocol.h"
+#include "mib.h"
 
 static bool mptcp_cap_flag_sha256(u8 flags)
 {
@@ -242,7 +243,7 @@ static void mptcp_parse_option(const struct sk_buff *skb,
 		mp_opt->add_addr = 1;
 		mp_opt->port = 0;
 		mp_opt->addr_id = *ptr++;
-		pr_debug("ADD_ADDR: id=%d", mp_opt->addr_id);
+		pr_debug("ADD_ADDR: id=%d, echo=%d", mp_opt->addr_id, mp_opt->echo);
 		if (mp_opt->family == MPTCP_ADDR_IPVERSION_4) {
 			memcpy((u8 *)&mp_opt->addr.s_addr, (u8 *)ptr, 4);
 			ptr += 4;
@@ -571,18 +572,19 @@ static u64 add_addr6_generate_hmac(u64 key1, u64 key2, u8 addr_id,
 }
 #endif
 
-static bool mptcp_established_options_addr(struct sock *sk,
-					   unsigned int *size,
-					   unsigned int remaining,
-					   struct mptcp_out_options *opts)
+static bool mptcp_established_options_add_addr(struct sock *sk,
+					       unsigned int *size,
+					       unsigned int remaining,
+					       struct mptcp_out_options *opts)
 {
 	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
 	struct mptcp_sock *msk = mptcp_sk(subflow->conn);
 	struct mptcp_addr_info saddr;
+	bool echo;
 	int len;
 
-	if (!mptcp_pm_should_signal(msk) ||
-	    !(mptcp_pm_addr_signal(msk, remaining, &saddr)))
+	if (!mptcp_pm_should_add_signal(msk) ||
+	    !(mptcp_pm_add_addr_signal(msk, remaining, &saddr, &echo)))
 		return false;
 
 	len = mptcp_add_addr_len(saddr.family);
@@ -594,22 +596,51 @@ static bool mptcp_established_options_addr(struct sock *sk,
 	if (saddr.family == AF_INET) {
 		opts->suboptions |= OPTION_MPTCP_ADD_ADDR;
 		opts->addr = saddr.addr;
-		opts->ahmac = add_addr_generate_hmac(msk->local_key,
-						     msk->remote_key,
-						     opts->addr_id,
-						     &opts->addr);
+		if (!echo) {
+			opts->ahmac = add_addr_generate_hmac(msk->local_key,
+							     msk->remote_key,
+							     opts->addr_id,
+							     &opts->addr);
+		}
 	}
 #if IS_ENABLED(CONFIG_MPTCP_IPV6)
 	else if (saddr.family == AF_INET6) {
 		opts->suboptions |= OPTION_MPTCP_ADD_ADDR6;
 		opts->addr6 = saddr.addr6;
-		opts->ahmac = add_addr6_generate_hmac(msk->local_key,
-						      msk->remote_key,
-						      opts->addr_id,
-						      &opts->addr6);
+		if (!echo) {
+			opts->ahmac = add_addr6_generate_hmac(msk->local_key,
+							      msk->remote_key,
+							      opts->addr_id,
+							      &opts->addr6);
+		}
 	}
 #endif
-	pr_debug("addr_id=%d, ahmac=%llu", opts->addr_id, opts->ahmac);
+	pr_debug("addr_id=%d, ahmac=%llu, echo=%d", opts->addr_id, opts->ahmac, echo);
+
+	return true;
+}
+
+static bool mptcp_established_options_rm_addr(struct sock *sk,
+					      unsigned int *size,
+					      unsigned int remaining,
+					      struct mptcp_out_options *opts)
+{
+	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
+	struct mptcp_sock *msk = mptcp_sk(subflow->conn);
+	u8 rm_id;
+
+	if (!mptcp_pm_should_rm_signal(msk) ||
+	    !(mptcp_pm_rm_addr_signal(msk, remaining, &rm_id)))
+		return false;
+
+	if (remaining < TCPOLEN_MPTCP_RM_ADDR_BASE)
+		return false;
+
+	*size = TCPOLEN_MPTCP_RM_ADDR_BASE;
+	opts->suboptions |= OPTION_MPTCP_RM_ADDR;
+	opts->rm_id = rm_id;
+
+	pr_debug("rm_id=%d", opts->rm_id);
 
 	return true;
 }
@@ -640,7 +671,11 @@ bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
 
 	*size += opt_size;
 	remaining -= opt_size;
-	if (mptcp_established_options_addr(sk, &opt_size, remaining, opts)) {
+	if (mptcp_established_options_add_addr(sk, &opt_size, remaining, opts)) {
+		*size += opt_size;
+		remaining -= opt_size;
+		ret = true;
+	} else if (mptcp_established_options_rm_addr(sk, &opt_size, remaining, opts)) {
 		*size += opt_size;
 		remaining -= opt_size;
 		ret = true;
@@ -854,11 +889,21 @@ void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb,
 			addr.addr6 = mp_opt.addr6;
 		}
 #endif
-		if (!mp_opt.echo)
+		if (!mp_opt.echo) {
 			mptcp_pm_add_addr_received(msk, &addr);
+			MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_ADDADDR);
+		} else {
+			mptcp_pm_del_add_timer(msk, &addr);
+			MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_ECHOADD);
+		}
 		mp_opt.add_addr = 0;
 	}
 
+	if (mp_opt.rm_addr) {
+		mptcp_pm_rm_addr_received(msk, mp_opt.rm_id);
+		mp_opt.rm_addr = 0;
+	}
+
 	if (!mp_opt.dss)
 		return;
 

net/mptcp/pm.c

@@ -13,23 +13,34 @@
 /* path manager command handlers */
 
 int mptcp_pm_announce_addr(struct mptcp_sock *msk,
-			   const struct mptcp_addr_info *addr)
+			   const struct mptcp_addr_info *addr,
+			   bool echo)
 {
 	pr_debug("msk=%p, local_id=%d", msk, addr->id);
 
 	msk->pm.local = *addr;
-	WRITE_ONCE(msk->pm.addr_signal, true);
+	WRITE_ONCE(msk->pm.add_addr_echo, echo);
+	WRITE_ONCE(msk->pm.add_addr_signal, true);
 	return 0;
 }
 
 int mptcp_pm_remove_addr(struct mptcp_sock *msk, u8 local_id)
 {
-	return -ENOTSUPP;
+	pr_debug("msk=%p, local_id=%d", msk, local_id);
+
+	msk->pm.rm_id = local_id;
+	WRITE_ONCE(msk->pm.rm_addr_signal, true);
+	return 0;
 }
 
-int mptcp_pm_remove_subflow(struct mptcp_sock *msk, u8 remote_id)
+int mptcp_pm_remove_subflow(struct mptcp_sock *msk, u8 local_id)
 {
-	return -ENOTSUPP;
+	pr_debug("msk=%p, local_id=%d", msk, local_id);
+
+	spin_lock_bh(&msk->pm.lock);
+	mptcp_pm_nl_rm_subflow_received(msk, local_id);
+	spin_unlock_bh(&msk->pm.lock);
+	return 0;
 }
 
 /* path manager event handlers */
@@ -46,7 +57,7 @@ void mptcp_pm_new_connection(struct mptcp_sock *msk, int server_side)
 bool mptcp_pm_allow_new_subflow(struct mptcp_sock *msk)
 {
 	struct mptcp_pm_data *pm = &msk->pm;
-	int ret;
+	int ret = 0;
 
 	pr_debug("msk=%p subflows=%d max=%d allow=%d", msk, pm->subflows,
 		 pm->subflows_max, READ_ONCE(pm->accept_subflow));
@@ -56,9 +67,11 @@ bool mptcp_pm_allow_new_subflow(struct mptcp_sock *msk)
 		return false;
 
 	spin_lock_bh(&pm->lock);
-	ret = pm->subflows < pm->subflows_max;
-	if (ret && ++pm->subflows == pm->subflows_max)
-		WRITE_ONCE(pm->accept_subflow, false);
+	if (READ_ONCE(pm->accept_subflow)) {
+		ret = pm->subflows < pm->subflows_max;
+		if (ret && ++pm->subflows == pm->subflows_max)
+			WRITE_ONCE(pm->accept_subflow, false);
+	}
 	spin_unlock_bh(&pm->lock);
 
 	return ret;
@@ -135,38 +148,70 @@ void mptcp_pm_add_addr_received(struct mptcp_sock *msk,
 	pr_debug("msk=%p remote_id=%d accept=%d", msk, addr->id,
 		 READ_ONCE(pm->accept_addr));
 
-	/* avoid acquiring the lock if there is no room for fouther addresses */
-	if (!READ_ONCE(pm->accept_addr))
-		return;
-
 	spin_lock_bh(&pm->lock);
 
-	/* be sure there is something to signal re-checking under PM lock */
-	if (READ_ONCE(pm->accept_addr) &&
-	    mptcp_pm_schedule_work(msk, MPTCP_PM_ADD_ADDR_RECEIVED))
+	if (!READ_ONCE(pm->accept_addr))
+		mptcp_pm_announce_addr(msk, addr, true);
+	else if (mptcp_pm_schedule_work(msk, MPTCP_PM_ADD_ADDR_RECEIVED))
 		pm->remote = *addr;
 
 	spin_unlock_bh(&pm->lock);
 }
 
+void mptcp_pm_rm_addr_received(struct mptcp_sock *msk, u8 rm_id)
+{
+	struct mptcp_pm_data *pm = &msk->pm;
+
+	pr_debug("msk=%p remote_id=%d", msk, rm_id);
+
+	spin_lock_bh(&pm->lock);
+	mptcp_pm_schedule_work(msk, MPTCP_PM_RM_ADDR_RECEIVED);
+	pm->rm_id = rm_id;
+	spin_unlock_bh(&pm->lock);
+}
+
 /* path manager helpers */
 
-bool mptcp_pm_addr_signal(struct mptcp_sock *msk, unsigned int remaining,
-			  struct mptcp_addr_info *saddr)
+bool mptcp_pm_add_addr_signal(struct mptcp_sock *msk, unsigned int remaining,
+			      struct mptcp_addr_info *saddr, bool *echo)
 {
 	int ret = false;
 
 	spin_lock_bh(&msk->pm.lock);
 
 	/* double check after the lock is acquired */
-	if (!mptcp_pm_should_signal(msk))
+	if (!mptcp_pm_should_add_signal(msk))
 		goto out_unlock;
 
 	if (remaining < mptcp_add_addr_len(msk->pm.local.family))
 		goto out_unlock;
 
 	*saddr = msk->pm.local;
-	WRITE_ONCE(msk->pm.addr_signal, false);
+	*echo = READ_ONCE(msk->pm.add_addr_echo);
+	WRITE_ONCE(msk->pm.add_addr_signal, false);
+	ret = true;
+
+out_unlock:
+	spin_unlock_bh(&msk->pm.lock);
+	return ret;
+}
+
+bool mptcp_pm_rm_addr_signal(struct mptcp_sock *msk, unsigned int remaining,
+			     u8 *rm_id)
+{
+	int ret = false;
+
+	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)
+		goto out_unlock;
+
+	*rm_id = msk->pm.rm_id;
+	WRITE_ONCE(msk->pm.rm_addr_signal, false);
 	ret = true;
 
 out_unlock:
@@ -185,13 +230,17 @@ 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;
 	WRITE_ONCE(msk->pm.work_pending, false);
-	WRITE_ONCE(msk->pm.addr_signal, false);
+	WRITE_ONCE(msk->pm.add_addr_signal, false);
+	WRITE_ONCE(msk->pm.rm_addr_signal, false);
 	WRITE_ONCE(msk->pm.accept_addr, false);
 	WRITE_ONCE(msk->pm.accept_subflow, false);
+	WRITE_ONCE(msk->pm.add_addr_echo, false);
 	msk->pm.status = 0;
 
 	spin_lock_init(&msk->pm.lock);
+	INIT_LIST_HEAD(&msk->pm.anno_list);
 
 	mptcp_pm_nl_data_init(msk);
 }