Merging upstream version 6.9.7.

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

4 files changed, 237 insertions, 41 deletions

diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index c99ffe6c68..c6807e473a 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -82,10 +82,6 @@ enum ad_link_speed_type {
 #define MAC_ADDRESS_EQUAL(A, B)	\
 	ether_addr_equal_64bits((const u8 *)A, (const u8 *)B)
 
-static const u8 null_mac_addr[ETH_ALEN + 2] __long_aligned = {
-	0, 0, 0, 0, 0, 0
-};
-
 static const u16 ad_ticks_per_sec = 1000 / AD_TIMER_INTERVAL;
 static const int ad_delta_in_ticks = (AD_TIMER_INTERVAL * HZ) / 1000;
 
@@ -106,6 +102,9 @@ static void ad_agg_selection_logic(struct aggregator *aggregator,
 static void ad_clear_agg(struct aggregator *aggregator);
 static void ad_initialize_agg(struct aggregator *aggregator);
 static void ad_initialize_port(struct port *port, int lacp_fast);
+static void ad_enable_collecting(struct port *port);
+static void ad_disable_distributing(struct port *port,
+				    bool *update_slave_arr);
 static void ad_enable_collecting_distributing(struct port *port,
 					      bool *update_slave_arr);
 static void ad_disable_collecting_distributing(struct port *port,
@@ -172,8 +171,37 @@ static inline int __agg_has_partner(struct aggregator *agg)
 }
 
 /**
+ * __disable_distributing_port - disable the port's slave for distributing.
+ * Port will still be able to collect.
+ * @port: the port we're looking at
+ *
+ * This will disable only distributing on the port's slave.
+ */
+static void __disable_distributing_port(struct port *port)
+{
+	bond_set_slave_tx_disabled_flags(port->slave, BOND_SLAVE_NOTIFY_LATER);
+}
+
+/**
+ * __enable_collecting_port - enable the port's slave for collecting,
+ * if it's up
+ * @port: the port we're looking at
+ *
+ * This will enable only collecting on the port's slave.
+ */
+static void __enable_collecting_port(struct port *port)
+{
+	struct slave *slave = port->slave;
+
+	if (slave->link == BOND_LINK_UP && bond_slave_is_up(slave))
+		bond_set_slave_rx_enabled_flags(slave, BOND_SLAVE_NOTIFY_LATER);
+}
+
+/**
  * __disable_port - disable the port's slave
  * @port: the port we're looking at
+ *
+ * This will disable both collecting and distributing on the port's slave.
  */
 static inline void __disable_port(struct port *port)
 {
@@ -183,6 +211,8 @@ static inline void __disable_port(struct port *port)
 /**
  * __enable_port - enable the port's slave, if it's up
  * @port: the port we're looking at
+ *
+ * This will enable both collecting and distributing on the port's slave.
  */
 static inline void __enable_port(struct port *port)
 {
@@ -193,10 +223,27 @@ static inline void __enable_port(struct port *port)
 }
 
 /**
- * __port_is_enabled - check if the port's slave is in active state
+ * __port_move_to_attached_state - check if port should transition back to attached
+ * state.
+ * @port: the port we're looking at
+ */
+static bool __port_move_to_attached_state(struct port *port)
+{
+	if (!(port->sm_vars & AD_PORT_SELECTED) ||
+	    (port->sm_vars & AD_PORT_STANDBY) ||
+	    !(port->partner_oper.port_state & LACP_STATE_SYNCHRONIZATION) ||
+	    !(port->actor_oper_port_state & LACP_STATE_SYNCHRONIZATION))
+		port->sm_mux_state = AD_MUX_ATTACHED;
+
+	return port->sm_mux_state == AD_MUX_ATTACHED;
+}
+
+/**
+ * __port_is_collecting_distributing - check if the port's slave is in the
+ * combined collecting/distributing state
  * @port: the port we're looking at
  */
-static inline int __port_is_enabled(struct port *port)
+static int __port_is_collecting_distributing(struct port *port)
 {
 	return bond_is_active_slave(port->slave);
 }
@@ -942,6 +989,7 @@ static int ad_marker_send(struct port *port, struct bond_marker *marker)
  */
 static void ad_mux_machine(struct port *port, bool *update_slave_arr)
 {
+	struct bonding *bond = __get_bond_by_port(port);
 	mux_states_t last_state;
 
 	/* keep current State Machine state to compare later if it was
@@ -999,9 +1047,13 @@ static void ad_mux_machine(struct port *port, bool *update_slave_arr)
 			if ((port->sm_vars & AD_PORT_SELECTED) &&
 			    (port->partner_oper.port_state & LACP_STATE_SYNCHRONIZATION) &&
 			    !__check_agg_selection_timer(port)) {
-				if (port->aggregator->is_active)
-					port->sm_mux_state =
-					    AD_MUX_COLLECTING_DISTRIBUTING;
+				if (port->aggregator->is_active) {
+					int state = AD_MUX_COLLECTING_DISTRIBUTING;
+
+					if (!bond->params.coupled_control)
+						state = AD_MUX_COLLECTING;
+					port->sm_mux_state = state;
+				}
 			} else if (!(port->sm_vars & AD_PORT_SELECTED) ||
 				   (port->sm_vars & AD_PORT_STANDBY)) {
 				/* if UNSELECTED or STANDBY */
@@ -1019,11 +1071,45 @@ static void ad_mux_machine(struct port *port, bool *update_slave_arr)
 			}
 			break;
 		case AD_MUX_COLLECTING_DISTRIBUTING:
+			if (!__port_move_to_attached_state(port)) {
+				/* if port state hasn't changed make
+				 * sure that a collecting distributing
+				 * port in an active aggregator is enabled
+				 */
+				if (port->aggregator->is_active &&
+				    !__port_is_collecting_distributing(port)) {
+					__enable_port(port);
+					*update_slave_arr = true;
+				}
+			}
+			break;
+		case AD_MUX_COLLECTING:
+			if (!__port_move_to_attached_state(port)) {
+				if ((port->sm_vars & AD_PORT_SELECTED) &&
+				    (port->partner_oper.port_state & LACP_STATE_SYNCHRONIZATION) &&
+				    (port->partner_oper.port_state & LACP_STATE_COLLECTING)) {
+					port->sm_mux_state = AD_MUX_DISTRIBUTING;
+				} else {
+					/* If port state hasn't changed, make sure that a collecting
+					 * port is enabled for an active aggregator.
+					 */
+					struct slave *slave = port->slave;
+
+					if (port->aggregator->is_active &&
+					    bond_is_slave_rx_disabled(slave)) {
+						ad_enable_collecting(port);
+						*update_slave_arr = true;
+					}
+				}
+			}
+			break;
+		case AD_MUX_DISTRIBUTING:
 			if (!(port->sm_vars & AD_PORT_SELECTED) ||
 			    (port->sm_vars & AD_PORT_STANDBY) ||
+			    !(port->partner_oper.port_state & LACP_STATE_COLLECTING) ||
 			    !(port->partner_oper.port_state & LACP_STATE_SYNCHRONIZATION) ||
 			    !(port->actor_oper_port_state & LACP_STATE_SYNCHRONIZATION)) {
-				port->sm_mux_state = AD_MUX_ATTACHED;
+				port->sm_mux_state = AD_MUX_COLLECTING;
 			} else {
 				/* if port state hasn't changed make
 				 * sure that a collecting distributing
@@ -1031,7 +1117,7 @@ static void ad_mux_machine(struct port *port, bool *update_slave_arr)
 				 */
 				if (port->aggregator &&
 				    port->aggregator->is_active &&
-				    !__port_is_enabled(port)) {
+				    !__port_is_collecting_distributing(port)) {
 					__enable_port(port);
 					*update_slave_arr = true;
 				}
@@ -1082,6 +1168,20 @@ static void ad_mux_machine(struct port *port, bool *update_slave_arr)
 							  update_slave_arr);
 			port->ntt = true;
 			break;
+		case AD_MUX_COLLECTING:
+			port->actor_oper_port_state |= LACP_STATE_COLLECTING;
+			port->actor_oper_port_state &= ~LACP_STATE_DISTRIBUTING;
+			port->actor_oper_port_state |= LACP_STATE_SYNCHRONIZATION;
+			ad_enable_collecting(port);
+			ad_disable_distributing(port, update_slave_arr);
+			port->ntt = true;
+			break;
+		case AD_MUX_DISTRIBUTING:
+			port->actor_oper_port_state |= LACP_STATE_DISTRIBUTING;
+			port->actor_oper_port_state |= LACP_STATE_SYNCHRONIZATION;
+			ad_enable_collecting_distributing(port,
+							  update_slave_arr);
+			break;
 		default:
 			break;
 		}
@@ -1484,7 +1584,7 @@ static void ad_port_selection_logic(struct port *port, bool *update_slave_arr)
 		     (aggregator->partner_system_priority == port->partner_oper.system_priority) &&
 		     (aggregator->partner_oper_aggregator_key == port->partner_oper.key)
 		    ) &&
-		    ((!MAC_ADDRESS_EQUAL(&(port->partner_oper.system), &(null_mac_addr)) && /* partner answers */
+		    ((__agg_has_partner(aggregator) && /* partner answers */
 		      !aggregator->is_individual)  /* but is not individual OR */
 		    )
 		   ) {
@@ -1907,6 +2007,43 @@ static void ad_initialize_port(struct port *port, int lacp_fast)
 }
 
 /**
+ * ad_enable_collecting - enable a port's receive
+ * @port: the port we're looking at
+ *
+ * Enable @port if it's in an active aggregator
+ */
+static void ad_enable_collecting(struct port *port)
+{
+	if (port->aggregator->is_active) {
+		struct slave *slave = port->slave;
+
+		slave_dbg(slave->bond->dev, slave->dev,
+			  "Enabling collecting on port %d (LAG %d)\n",
+			  port->actor_port_number,
+			  port->aggregator->aggregator_identifier);
+		__enable_collecting_port(port);
+	}
+}
+
+/**
+ * ad_disable_distributing - disable a port's transmit
+ * @port: the port we're looking at
+ * @update_slave_arr: Does slave array need update?
+ */
+static void ad_disable_distributing(struct port *port, bool *update_slave_arr)
+{
+	if (port->aggregator && __agg_has_partner(port->aggregator)) {
+		slave_dbg(port->slave->bond->dev, port->slave->dev,
+			  "Disabling distributing on port %d (LAG %d)\n",
+			  port->actor_port_number,
+			  port->aggregator->aggregator_identifier);
+		__disable_distributing_port(port);
+		/* Slave array needs an update */
+		*update_slave_arr = true;
+	}
+}
+
+/**
  * ad_enable_collecting_distributing - enable a port's transmit/receive
  * @port: the port we're looking at
  * @update_slave_arr: Does slave array need update?
@@ -1935,9 +2072,7 @@ static void ad_enable_collecting_distributing(struct port *port,
 static void ad_disable_collecting_distributing(struct port *port,
 					       bool *update_slave_arr)
 {
-	if (port->aggregator &&
-	    !MAC_ADDRESS_EQUAL(&(port->aggregator->partner_system),
-			       &(null_mac_addr))) {
+	if (port->aggregator && __agg_has_partner(port->aggregator)) {
 		slave_dbg(port->slave->bond->dev, port->slave->dev,
 			  "Disabling port %d (LAG %d)\n",
 			  port->actor_port_number,
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index cd0683bcca..bceda85f0d 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -2611,7 +2611,7 @@ static int bond_miimon_inspect(struct bonding *bond)
 			bond_propose_link_state(slave, BOND_LINK_FAIL);
 			commit++;
 			slave->delay = bond->params.downdelay;
-			if (slave->delay) {
+			if (slave->delay && net_ratelimit()) {
 				slave_info(bond->dev, slave->dev, "link status down for %sinterface, disabling it in %d ms\n",
 					   (BOND_MODE(bond) ==
 					    BOND_MODE_ACTIVEBACKUP) ?
@@ -2625,9 +2625,10 @@ static int bond_miimon_inspect(struct bonding *bond)
 				/* recovered before downdelay expired */
 				bond_propose_link_state(slave, BOND_LINK_UP);
 				slave->last_link_up = jiffies;
-				slave_info(bond->dev, slave->dev, "link status up again after %d ms\n",
-					   (bond->params.downdelay - slave->delay) *
-					   bond->params.miimon);
+				if (net_ratelimit())
+					slave_info(bond->dev, slave->dev, "link status up again after %d ms\n",
+						   (bond->params.downdelay - slave->delay) *
+						   bond->params.miimon);
 				commit++;
 				continue;
 			}
@@ -2649,7 +2650,7 @@ static int bond_miimon_inspect(struct bonding *bond)
 			commit++;
 			slave->delay = bond->params.updelay;
 
-			if (slave->delay) {
+			if (slave->delay && net_ratelimit()) {
 				slave_info(bond->dev, slave->dev, "link status up, enabling it in %d ms\n",
 					   ignore_updelay ? 0 :
 					   bond->params.updelay *
@@ -2659,9 +2660,10 @@ static int bond_miimon_inspect(struct bonding *bond)
 		case BOND_LINK_BACK:
 			if (!link_state) {
 				bond_propose_link_state(slave, BOND_LINK_DOWN);
-				slave_info(bond->dev, slave->dev, "link status down again after %d ms\n",
-					   (bond->params.updelay - slave->delay) *
-					   bond->params.miimon);
+				if (net_ratelimit())
+					slave_info(bond->dev, slave->dev, "link status down again after %d ms\n",
+						   (bond->params.updelay - slave->delay) *
+						   bond->params.miimon);
 				commit++;
 				continue;
 			}
@@ -6305,6 +6307,7 @@ static int __init bond_check_params(struct bond_params *params)
 	params->ad_actor_sys_prio = ad_actor_sys_prio;
 	eth_zero_addr(params->ad_actor_system);
 	params->ad_user_port_key = ad_user_port_key;
+	params->coupled_control = 1;
 	if (packets_per_slave > 0) {
 		params->reciprocal_packets_per_slave =
 			reciprocal_value(packets_per_slave);
@@ -6414,28 +6417,41 @@ static int __net_init bond_net_init(struct net *net)
 	return 0;
 }
 
-static void __net_exit bond_net_exit_batch(struct list_head *net_list)
+/* According to commit 69b0216ac255 ("bonding: fix bonding_masters
+ * race condition in bond unloading") we need to remove sysfs files
+ * before we remove our devices (done later in bond_net_exit_batch_rtnl())
+ */
+static void __net_exit bond_net_pre_exit(struct net *net)
+{
+	struct bond_net *bn = net_generic(net, bond_net_id);
+
+	bond_destroy_sysfs(bn);
+}
+
+static void __net_exit bond_net_exit_batch_rtnl(struct list_head *net_list,
+						struct list_head *dev_kill_list)
 {
 	struct bond_net *bn;
 	struct net *net;
-	LIST_HEAD(list);
-
-	list_for_each_entry(net, net_list, exit_list) {
-		bn = net_generic(net, bond_net_id);
-		bond_destroy_sysfs(bn);
-	}
 
 	/* Kill off any bonds created after unregistering bond rtnl ops */
-	rtnl_lock();
 	list_for_each_entry(net, net_list, exit_list) {
 		struct bonding *bond, *tmp_bond;
 
 		bn = net_generic(net, bond_net_id);
 		list_for_each_entry_safe(bond, tmp_bond, &bn->dev_list, bond_list)
-			unregister_netdevice_queue(bond->dev, &list);
+			unregister_netdevice_queue(bond->dev, dev_kill_list);
 	}
-	unregister_netdevice_many(&list);
-	rtnl_unlock();
+}
+
+/* According to commit 23fa5c2caae0 ("bonding: destroy proc directory
+ * only after all bonds are gone") bond_destroy_proc_dir() is called
+ * after bond_net_exit_batch_rtnl() has completed.
+ */
+static void __net_exit bond_net_exit_batch(struct list_head *net_list)
+{
+	struct bond_net *bn;
+	struct net *net;
 
 	list_for_each_entry(net, net_list, exit_list) {
 		bn = net_generic(net, bond_net_id);
@@ -6445,6 +6461,8 @@ static void __net_exit bond_net_exit_batch(struct list_head *net_list)
 
 static struct pernet_operations bond_net_ops = {
 	.init = bond_net_init,
+	.pre_exit = bond_net_pre_exit,
+	.exit_batch_rtnl = bond_net_exit_batch_rtnl,
 	.exit_batch = bond_net_exit_batch,
 	.id   = &bond_net_id,
 	.size = sizeof(struct bond_net),
@@ -6459,16 +6477,16 @@ static int __init bonding_init(void)
 	if (res)
 		goto out;
 
+	bond_create_debugfs();
+
 	res = register_pernet_subsys(&bond_net_ops);
 	if (res)
-		goto out;
+		goto err_net_ops;
 
 	res = bond_netlink_init();
 	if (res)
 		goto err_link;
 
-	bond_create_debugfs();
-
 	for (i = 0; i < max_bonds; i++) {
 		res = bond_create(&init_net, NULL);
 		if (res)
@@ -6483,10 +6501,11 @@ static int __init bonding_init(void)
 out:
 	return res;
 err:
-	bond_destroy_debugfs();
 	bond_netlink_fini();
 err_link:
 	unregister_pernet_subsys(&bond_net_ops);
+err_net_ops:
+	bond_destroy_debugfs();
 	goto out;
 
 }
@@ -6495,11 +6514,11 @@ static void __exit bonding_exit(void)
 {
 	unregister_netdevice_notifier(&bond_netdev_notifier);
 
-	bond_destroy_debugfs();
-
 	bond_netlink_fini();
 	unregister_pernet_subsys(&bond_net_ops);
 
+	bond_destroy_debugfs();
+
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	/* Make sure we don't have an imbalance on our netpoll blocking */
 	WARN_ON(atomic_read(&netpoll_block_tx));
diff --git a/drivers/net/bonding/bond_netlink.c b/drivers/net/bonding/bond_netlink.c
index cfa74cf8bb..29b4c3d1b9 100644
--- a/drivers/net/bonding/bond_netlink.c
+++ b/drivers/net/bonding/bond_netlink.c
@@ -122,6 +122,7 @@ static const struct nla_policy bond_policy[IFLA_BOND_MAX + 1] = {
 	[IFLA_BOND_PEER_NOTIF_DELAY]    = NLA_POLICY_FULL_RANGE(NLA_U32, &delay_range),
 	[IFLA_BOND_MISSED_MAX]		= { .type = NLA_U8 },
 	[IFLA_BOND_NS_IP6_TARGET]	= { .type = NLA_NESTED },
+	[IFLA_BOND_COUPLED_CONTROL]	= { .type = NLA_U8 },
 };
 
 static const struct nla_policy bond_slave_policy[IFLA_BOND_SLAVE_MAX + 1] = {
@@ -549,6 +550,16 @@ static int bond_changelink(struct net_device *bond_dev, struct nlattr *tb[],
 			return err;
 	}
 
+	if (data[IFLA_BOND_COUPLED_CONTROL]) {
+		int coupled_control = nla_get_u8(data[IFLA_BOND_COUPLED_CONTROL]);
+
+		bond_opt_initval(&newval, coupled_control);
+		err = __bond_opt_set(bond, BOND_OPT_COUPLED_CONTROL, &newval,
+				     data[IFLA_BOND_COUPLED_CONTROL], extack);
+		if (err)
+			return err;
+	}
+
 	return 0;
 }
 
@@ -615,6 +626,7 @@ static size_t bond_get_size(const struct net_device *bond_dev)
 						/* IFLA_BOND_NS_IP6_TARGET */
 		nla_total_size(sizeof(struct nlattr)) +
 		nla_total_size(sizeof(struct in6_addr)) * BOND_MAX_NS_TARGETS +
+		nla_total_size(sizeof(u8)) +	/* IFLA_BOND_COUPLED_CONTROL */
 		0;
 }
 
@@ -774,6 +786,10 @@ static int bond_fill_info(struct sk_buff *skb,
 		       bond->params.missed_max))
 		goto nla_put_failure;
 
+	if (nla_put_u8(skb, IFLA_BOND_COUPLED_CONTROL,
+		       bond->params.coupled_control))
+		goto nla_put_failure;
+
 	if (BOND_MODE(bond) == BOND_MODE_8023AD) {
 		struct ad_info info;
 
diff --git a/drivers/net/bonding/bond_options.c b/drivers/net/bonding/bond_options.c
index f3f27f0bd2..4cdbc7e084 100644
--- a/drivers/net/bonding/bond_options.c
+++ b/drivers/net/bonding/bond_options.c
@@ -84,7 +84,8 @@ static int bond_option_ad_user_port_key_set(struct bonding *bond,
 					    const struct bond_opt_value *newval);
 static int bond_option_missed_max_set(struct bonding *bond,
 				      const struct bond_opt_value *newval);
-
+static int bond_option_coupled_control_set(struct bonding *bond,
+					   const struct bond_opt_value *newval);
 
 static const struct bond_opt_value bond_mode_tbl[] = {
 	{ "balance-rr",    BOND_MODE_ROUNDROBIN,   BOND_VALFLAG_DEFAULT},
@@ -232,6 +233,12 @@ static const struct bond_opt_value bond_missed_max_tbl[] = {
 	{ NULL,		-1,	0},
 };
 
+static const struct bond_opt_value bond_coupled_control_tbl[] = {
+	{ "on",  1,  BOND_VALFLAG_DEFAULT},
+	{ "off", 0,  0},
+	{ NULL,  -1, 0},
+};
+
 static const struct bond_option bond_opts[BOND_OPT_LAST] = {
 	[BOND_OPT_MODE] = {
 		.id = BOND_OPT_MODE,
@@ -496,6 +503,15 @@ static const struct bond_option bond_opts[BOND_OPT_LAST] = {
 		.desc = "Delay between each peer notification on failover event, in milliseconds",
 		.values = bond_peer_notif_delay_tbl,
 		.set = bond_option_peer_notif_delay_set
+	},
+	[BOND_OPT_COUPLED_CONTROL] = {
+		.id = BOND_OPT_COUPLED_CONTROL,
+		.name = "coupled_control",
+		.desc = "Opt into using coupled control MUX for LACP states",
+		.unsuppmodes = BOND_MODE_ALL_EX(BIT(BOND_MODE_8023AD)),
+		.flags = BOND_OPTFLAG_IFDOWN,
+		.values = bond_coupled_control_tbl,
+		.set = bond_option_coupled_control_set,
 	}
 };
 
@@ -1692,3 +1708,13 @@ static int bond_option_ad_user_port_key_set(struct bonding *bond,
 	bond->params.ad_user_port_key = newval->value;
 	return 0;
 }
+
+static int bond_option_coupled_control_set(struct bonding *bond,
+					   const struct bond_opt_value *newval)
+{
+	netdev_info(bond->dev, "Setting coupled_control to %s (%llu)\n",
+		    newval->string, newval->value);
+
+	bond->params.coupled_control = newval->value;
+	return 0;
+}