Merge branch 'bond_rcu'

Nikolay Aleksandrov says:

====================
 This patchset aims to lay the groundwork, and do the initial conversion to
RCUism. I decided that it'll be much better to make the bonding RCU
conversion gradual, so patches can be reviewed and tested better rather
than having one huge patch (which I did in the beginning, before this).
The first patch is straightforward and it converts the bonding to the
standard list API, simplifying a lot of code, removing unnecessary local
variables and allowing to use the nice rculist API later. It also takes
care of some minor styling issues (re-arranging local variables longest ->
shortest, removing brackets for single statement if/else, leaving new line
before return statement etc.).
 The second patch simplifies the conversion by removing unnecessary
read_lock(&bond->curr_slave_lock) in xmit paths that are to be converted
later, because we only care if the pointer is NULL or a slave there, since
we already have bond->lock the slave can't go away.
 The third patch simplifies the broadcast xmit function by removing
the use of curr_active_slave and converting to standard list API. Also this
design of the broadcast xmit function avoids a subtle double packet tx race
when converted to RCU.
 The fourth patch factors out the code that transmits skb through a slave
with given id (i.e. rr_tx_counter in rr mode, hashed value in xor mode) and
simplifies the active-backup xmit path because bond_dev_queue_xmit always
consumes the skb. The new bond_xmit_slave_id function is used in rr and xor
modes currently, but the plans are to use it in 3ad mode as well thus it's
made global. I've left the function prototype to be 81 chars so I wouldn't
break it, if this is an issue I can always break it in more lines.
 The fifth patch introduces RCU by converting attach/detach and release to
RCU. It also converts dereferencing of curr_active_slave to rcu_dereference
although it's not fully converted to RCU, that is needed for the converted
xmit paths. And it converts roundrobin, broadcast, xor and active-backup
xmit paths to RCU. The 3ad and ALB/TLB modes acquire read_lock(&bond->lock)
to make sure that no slave will be removed and to sync properly with
enslave and release as before.
 This way for the price of a little complexity, we'll be able to convert
individual parts of the bonding to RCU, and test them easier in the
process. If this patchset is accepted in some form, I'll post followups
in the next weeks that gradually convert the bonding to RCU and remove the
need for the rwlocks.
 For performance notes please refer to patch 5 (RCU conversion one).
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/bonding/bond_3ad.c

@@ -143,10 +143,9 @@ static inline struct bonding *__get_bond_by_port(struct port *port)
  */
 static inline struct port *__get_first_port(struct bonding *bond)
 {
-	if (bond->slave_cnt == 0)
-		return NULL;
+	struct slave *first_slave = bond_first_slave(bond);
 
-	return &(SLAVE_AD_INFO(bond->first_slave).port);
+	return first_slave ? &(SLAVE_AD_INFO(first_slave).port) : NULL;
 }
 
 /**
@@ -159,13 +158,16 @@ static inline struct port *__get_first_port(struct bonding *bond)
 static inline struct port *__get_next_port(struct port *port)
 {
 	struct bonding *bond = __get_bond_by_port(port);
-	struct slave *slave = port->slave;
+	struct slave *slave = port->slave, *slave_next;
 
 	// If there's no bond for this port, or this is the last slave
-	if ((bond == NULL) || (slave->next == bond->first_slave))
+	if (bond == NULL)
+		return NULL;
+	slave_next = bond_next_slave(bond, slave);
+	if (!slave_next || bond_is_first_slave(bond, slave_next))
 		return NULL;
 
-	return &(SLAVE_AD_INFO(slave->next).port);
+	return &(SLAVE_AD_INFO(slave_next).port);
 }
 
 /**
@@ -178,12 +180,14 @@ static inline struct port *__get_next_port(struct port *port)
 static inline struct aggregator *__get_first_agg(struct port *port)
 {
 	struct bonding *bond = __get_bond_by_port(port);
+	struct slave *first_slave;
 
 	// If there's no bond for this port, or bond has no slaves
-	if ((bond == NULL) || (bond->slave_cnt == 0))
+	if (bond == NULL)
 		return NULL;
+	first_slave = bond_first_slave(bond);
 
-	return &(SLAVE_AD_INFO(bond->first_slave).aggregator);
+	return first_slave ? &(SLAVE_AD_INFO(first_slave).aggregator) : NULL;
 }
 
 /**
@@ -195,14 +199,17 @@ static inline struct aggregator *__get_first_agg(struct port *port)
  */
 static inline struct aggregator *__get_next_agg(struct aggregator *aggregator)
 {
-	struct slave *slave = aggregator->slave;
+	struct slave *slave = aggregator->slave, *slave_next;
 	struct bonding *bond = bond_get_bond_by_slave(slave);
 
 	// If there's no bond for this aggregator, or this is the last slave
-	if ((bond == NULL) || (slave->next == bond->first_slave))
+	if (bond == NULL)
+		return NULL;
+	slave_next = bond_next_slave(bond, slave);
+	if (!slave_next || bond_is_first_slave(bond, slave_next))
 		return NULL;
 
-	return &(SLAVE_AD_INFO(slave->next).aggregator);
+	return &(SLAVE_AD_INFO(slave_next).aggregator);
 }
 
 /*
@@ -2110,7 +2117,7 @@ void bond_3ad_state_machine_handler(struct work_struct *work)
 	read_lock(&bond->lock);
 
 	//check if there are any slaves
-	if (bond->slave_cnt == 0)
+	if (list_empty(&bond->slave_list))
 		goto re_arm;
 
 	// check if agg_select_timer timer after initialize is timed out
@@ -2336,8 +2343,12 @@ void bond_3ad_handle_link_change(struct slave *slave, char link)
 int bond_3ad_set_carrier(struct bonding *bond)
 {
 	struct aggregator *active;
+	struct slave *first_slave;
 
-	active = __get_active_agg(&(SLAVE_AD_INFO(bond->first_slave).aggregator));
+	first_slave = bond_first_slave(bond);
+	if (!first_slave)
+		return 0;
+	active = __get_active_agg(&(SLAVE_AD_INFO(first_slave).aggregator));
 	if (active) {
 		/* are enough slaves available to consider link up? */
 		if (active->num_of_ports < bond->params.min_links) {
@@ -2415,6 +2426,7 @@ int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev)
 	struct ad_info ad_info;
 	int res = 1;
 
+	read_lock(&bond->lock);
 	if (__bond_3ad_get_active_agg_info(bond, &ad_info)) {
 		pr_debug("%s: Error: __bond_3ad_get_active_agg_info failed\n",
 			 dev->name);
@@ -2432,7 +2444,7 @@ int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev)
 
 	slave_agg_no = bond->xmit_hash_policy(skb, slaves_in_agg);
 
-	bond_for_each_slave(bond, slave, i) {
+	bond_for_each_slave(bond, slave) {
 		struct aggregator *agg = SLAVE_AD_INFO(slave).port.aggregator;
 
 		if (agg && (agg->aggregator_identifier == agg_id)) {
@@ -2464,6 +2476,7 @@ int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev)
 	}
 
 out:
+	read_unlock(&bond->lock);
 	if (res) {
 		/* no suitable interface, frame not sent */
 		kfree_skb(skb);
@@ -2501,7 +2514,6 @@ int bond_3ad_lacpdu_recv(const struct sk_buff *skb, struct bonding *bond,
  */
 void bond_3ad_update_lacp_rate(struct bonding *bond)
 {
-	int i;
 	struct slave *slave;
 	struct port *port = NULL;
 	int lacp_fast;
@@ -2509,7 +2521,7 @@ void bond_3ad_update_lacp_rate(struct bonding *bond)
 	write_lock_bh(&bond->lock);
 	lacp_fast = bond->params.lacp_fast;
 
-	bond_for_each_slave(bond, slave, i) {
+	bond_for_each_slave(bond, slave) {
 		port = &(SLAVE_AD_INFO(slave).port);
 		if (port->slave == NULL)
 			continue;

drivers/net/bonding/bond_alb.c

@@ -224,13 +224,12 @@ static struct slave *tlb_get_least_loaded_slave(struct bonding *bond)
 {
 	struct slave *slave, *least_loaded;
 	long long max_gap;
-	int i;
 
 	least_loaded = NULL;
 	max_gap = LLONG_MIN;
 
 	/* Find the slave with the largest gap */
-	bond_for_each_slave(bond, slave, i) {
+	bond_for_each_slave(bond, slave) {
 		if (SLAVE_IS_OK(slave)) {
 			long long gap = compute_gap(slave);
 
@@ -386,11 +385,10 @@ static struct slave *rlb_next_rx_slave(struct bonding *bond)
 	struct slave *rx_slave, *slave, *start_at;
 	int i = 0;
 
-	if (bond_info->next_rx_slave) {
+	if (bond_info->next_rx_slave)
 		start_at = bond_info->next_rx_slave;
-	} else {
-		start_at = bond->first_slave;
-	}
+	else
+		start_at = bond_first_slave(bond);
 
 	rx_slave = NULL;
 
@@ -405,7 +403,8 @@ static struct slave *rlb_next_rx_slave(struct bonding *bond)
 	}
 
 	if (rx_slave) {
-		bond_info->next_rx_slave = rx_slave->next;
+		slave = bond_next_slave(bond, rx_slave);
+		bond_info->next_rx_slave = slave;
 	}
 
 	return rx_slave;
@@ -1173,9 +1172,8 @@ static int alb_handle_addr_collision_on_attach(struct bonding *bond, struct slav
 {
 	struct slave *tmp_slave1, *free_mac_slave = NULL;
 	struct slave *has_bond_addr = bond->curr_active_slave;
-	int i;
 
-	if (bond->slave_cnt == 0) {
+	if (list_empty(&bond->slave_list)) {
 		/* this is the first slave */
 		return 0;
 	}
@@ -1196,7 +1194,7 @@ static int alb_handle_addr_collision_on_attach(struct bonding *bond, struct slav
 	/* The slave's address is equal to the address of the bond.
 	 * Search for a spare address in the bond for this slave.
 	 */
-	bond_for_each_slave(bond, tmp_slave1, i) {
+	bond_for_each_slave(bond, tmp_slave1) {
 		if (!bond_slave_has_mac(bond, tmp_slave1->perm_hwaddr)) {
 			/* no slave has tmp_slave1's perm addr
 			 * as its curr addr
@@ -1246,17 +1244,15 @@ static int alb_handle_addr_collision_on_attach(struct bonding *bond, struct slav
  */
 static int alb_set_mac_address(struct bonding *bond, void *addr)
 {
-	struct sockaddr sa;
-	struct slave *slave, *stop_at;
 	char tmp_addr[ETH_ALEN];
+	struct slave *slave;
+	struct sockaddr sa;
 	int res;
-	int i;
 
-	if (bond->alb_info.rlb_enabled) {
+	if (bond->alb_info.rlb_enabled)
 		return 0;
-	}
 
-	bond_for_each_slave(bond, slave, i) {
+	bond_for_each_slave(bond, slave) {
 		/* save net_device's current hw address */
 		memcpy(tmp_addr, slave->dev->dev_addr, ETH_ALEN);
 
@@ -1276,8 +1272,7 @@ static int alb_set_mac_address(struct bonding *bond, void *addr)
 	sa.sa_family = bond->dev->type;
 
 	/* unwind from head to the slave that failed */
-	stop_at = slave;
-	bond_for_each_slave_from_to(bond, slave, i, bond->first_slave, stop_at) {
+	bond_for_each_slave_continue_reverse(bond, slave) {
 		memcpy(tmp_addr, slave->dev->dev_addr, ETH_ALEN);
 		dev_set_mac_address(slave->dev, &sa);
 		memcpy(slave->dev->dev_addr, tmp_addr, ETH_ALEN);
@@ -1342,6 +1337,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
 
 	/* make sure that the curr_active_slave do not change during tx
 	 */
+	read_lock(&bond->lock);
 	read_lock(&bond->curr_slave_lock);
 
 	switch (ntohs(skb->protocol)) {
@@ -1446,11 +1442,12 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
 	}
 
 	read_unlock(&bond->curr_slave_lock);
-
+	read_unlock(&bond->lock);
 	if (res) {
 		/* no suitable interface, frame not sent */
 		kfree_skb(skb);
 	}
+
 	return NETDEV_TX_OK;
 }
 
@@ -1460,11 +1457,10 @@ void bond_alb_monitor(struct work_struct *work)
 					    alb_work.work);
 	struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
 	struct slave *slave;
-	int i;
 
 	read_lock(&bond->lock);
 
-	if (bond->slave_cnt == 0) {
+	if (list_empty(&bond->slave_list)) {
 		bond_info->tx_rebalance_counter = 0;
 		bond_info->lp_counter = 0;
 		goto re_arm;
@@ -1482,9 +1478,8 @@ void bond_alb_monitor(struct work_struct *work)
 		 */
 		read_lock(&bond->curr_slave_lock);
 
-		bond_for_each_slave(bond, slave, i) {
+		bond_for_each_slave(bond, slave)
 			alb_send_learning_packets(slave, slave->dev->dev_addr);
-		}
 
 		read_unlock(&bond->curr_slave_lock);
 
@@ -1496,7 +1491,7 @@ void bond_alb_monitor(struct work_struct *work)
 
 		read_lock(&bond->curr_slave_lock);
 
-		bond_for_each_slave(bond, slave, i) {
+		bond_for_each_slave(bond, slave) {
 			tlb_clear_slave(bond, slave, 1);
 			if (slave == bond->curr_active_slave) {
 				SLAVE_TLB_INFO(slave).load =
@@ -1602,9 +1597,8 @@ int bond_alb_init_slave(struct bonding *bond, struct slave *slave)
  */
 void bond_alb_deinit_slave(struct bonding *bond, struct slave *slave)
 {
-	if (bond->slave_cnt > 1) {
+	if (!list_empty(&bond->slave_list))
 		alb_change_hw_addr_on_detach(bond, slave);
-	}
 
 	tlb_clear_slave(bond, slave, 0);
 
@@ -1661,9 +1655,8 @@ void bond_alb_handle_active_change(struct bonding *bond, struct slave *new_slave
 {
 	struct slave *swap_slave;
 
-	if (bond->curr_active_slave == new_slave) {
+	if (bond->curr_active_slave == new_slave)
 		return;
-	}
 
 	if (bond->curr_active_slave && bond->alb_info.primary_is_promisc) {
 		dev_set_promiscuity(bond->curr_active_slave->dev, -1);
@@ -1672,11 +1665,10 @@ void bond_alb_handle_active_change(struct bonding *bond, struct slave *new_slave
 	}
 
 	swap_slave = bond->curr_active_slave;
-	bond->curr_active_slave = new_slave;
+	rcu_assign_pointer(bond->curr_active_slave, new_slave);
 
-	if (!new_slave || (bond->slave_cnt == 0)) {
+	if (!new_slave || list_empty(&bond->slave_list))
 		return;
-	}
 
 	/* set the new curr_active_slave to the bonds mac address
 	 * i.e. swap mac addresses of old curr_active_slave and new curr_active_slave
@@ -1689,9 +1681,8 @@ void bond_alb_handle_active_change(struct bonding *bond, struct slave *new_slave
 	 * ignored so we can mess with their MAC addresses without
 	 * fear of interference from transmit activity.
 	 */
-	if (swap_slave) {
+	if (swap_slave)
 		tlb_clear_slave(bond, swap_slave, 1);
-	}
 	tlb_clear_slave(bond, new_slave, 1);
 
 	write_unlock_bh(&bond->curr_slave_lock);

drivers/net/bonding/bond_procfs.c

@@ -12,7 +12,6 @@ static void *bond_info_seq_start(struct seq_file *seq, loff_t *pos)
 	struct bonding *bond = seq->private;
 	loff_t off = 0;
 	struct slave *slave;
-	int i;
 
 	/* make sure the bond won't be taken away */
 	rcu_read_lock();
@@ -21,10 +20,9 @@ static void *bond_info_seq_start(struct seq_file *seq, loff_t *pos)
 	if (*pos == 0)
 		return SEQ_START_TOKEN;
 
-	bond_for_each_slave(bond, slave, i) {
+	bond_for_each_slave(bond, slave)
 		if (++off == *pos)
 			return slave;
-	}
 
 	return NULL;
 }
@@ -36,11 +34,13 @@ static void *bond_info_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 
 	++*pos;
 	if (v == SEQ_START_TOKEN)
-		return bond->first_slave;
+		return bond_first_slave(bond);
 
-	slave = slave->next;
+	if (bond_is_last_slave(bond, slave))
+		return NULL;
+	slave = bond_next_slave(bond, slave);
 
-	return (slave == bond->first_slave) ? NULL : slave;
+	return slave;
 }
 
 static void bond_info_seq_stop(struct seq_file *seq, void *v)

drivers/net/bonding/bond_sysfs.c

@@ -209,12 +209,12 @@ void bond_destroy_slave_symlinks(struct net_device *master,
 static ssize_t bonding_show_slaves(struct device *d,
 				   struct device_attribute *attr, char *buf)
 {
-	struct slave *slave;
-	int i, res = 0;
 	struct bonding *bond = to_bond(d);
+	struct slave *slave;
+	int res = 0;
 
 	read_lock(&bond->lock);
-	bond_for_each_slave(bond, slave, i) {
+	bond_for_each_slave(bond, slave) {
 		if (res > (PAGE_SIZE - IFNAMSIZ)) {
 			/* not enough space for another interface name */
 			if ((PAGE_SIZE - res) > 10)
@@ -227,6 +227,7 @@ static ssize_t bonding_show_slaves(struct device *d,
 	read_unlock(&bond->lock);
 	if (res)
 		buf[res-1] = '\n'; /* eat the leftover space */
+
 	return res;
 }
 
@@ -325,7 +326,7 @@ static ssize_t bonding_store_mode(struct device *d,
 		goto out;
 	}
 
-	if (bond->slave_cnt > 0) {
+	if (!list_empty(&bond->slave_list)) {
 		pr_err("unable to update mode of %s because it has slaves.\n",
 			bond->dev->name);
 		ret = -EPERM;
@@ -507,7 +508,7 @@ static ssize_t bonding_store_fail_over_mac(struct device *d,
 	if (!rtnl_trylock())
 		return restart_syscall();
 
-	if (bond->slave_cnt != 0) {
+	if (!list_empty(&bond->slave_list)) {
 		pr_err("%s: Can't alter fail_over_mac with slaves in bond.\n",
 		       bond->dev->name);
 		ret = -EPERM;
@@ -668,7 +669,7 @@ static ssize_t bonding_store_arp_targets(struct device *d,
 			 &newtarget);
 		/* not to race with bond_arp_rcv */
 		write_lock_bh(&bond->lock);
-		bond_for_each_slave(bond, slave, i)
+		bond_for_each_slave(bond, slave)
 			slave->target_last_arp_rx[ind] = jiffies;
 		targets[ind] = newtarget;
 		write_unlock_bh(&bond->lock);
@@ -694,7 +695,7 @@ static ssize_t bonding_store_arp_targets(struct device *d,
 			&newtarget);
 
 		write_lock_bh(&bond->lock);
-		bond_for_each_slave(bond, slave, i) {
+		bond_for_each_slave(bond, slave) {
 			targets_rx = slave->target_last_arp_rx;
 			j = ind;
 			for (; (j < BOND_MAX_ARP_TARGETS-1) && targets[j+1]; j++)
@@ -1085,10 +1086,9 @@ static ssize_t bonding_store_primary(struct device *d,
 				     struct device_attribute *attr,
 				     const char *buf, size_t count)
 {
-	int i;
-	struct slave *slave;
 	struct bonding *bond = to_bond(d);
 	char ifname[IFNAMSIZ];
+	struct slave *slave;
 
 	if (!rtnl_trylock())
 		return restart_syscall();
@@ -1114,7 +1114,7 @@ static ssize_t bonding_store_primary(struct device *d,
 		goto out;
 	}
 
-	bond_for_each_slave(bond, slave, i) {
+	bond_for_each_slave(bond, slave) {
 		if (strncmp(slave->dev->name, ifname, IFNAMSIZ) == 0) {
 			pr_info("%s: Setting %s as primary slave.\n",
 				bond->dev->name, slave->dev->name);
@@ -1243,16 +1243,16 @@ static ssize_t bonding_show_active_slave(struct device *d,
 					 struct device_attribute *attr,
 					 char *buf)
 {
-	struct slave *curr;
 	struct bonding *bond = to_bond(d);
+	struct slave *curr;
 	int count = 0;
 
-	read_lock(&bond->curr_slave_lock);
-	curr = bond->curr_active_slave;
-	read_unlock(&bond->curr_slave_lock);
-
+	rcu_read_lock();
+	curr = rcu_dereference(bond->curr_active_slave);
 	if (USES_PRIMARY(bond->params.mode) && curr)
 		count = sprintf(buf, "%s\n", curr->dev->name);
+	rcu_read_unlock();
+
 	return count;
 }
 
@@ -1260,16 +1260,14 @@ static ssize_t bonding_store_active_slave(struct device *d,
 					  struct device_attribute *attr,
 					  const char *buf, size_t count)
 {
-	int i;
-	struct slave *slave;
-	struct slave *old_active = NULL;
-	struct slave *new_active = NULL;
+	struct slave *slave, *old_active, *new_active;
 	struct bonding *bond = to_bond(d);
 	char ifname[IFNAMSIZ];
 
 	if (!rtnl_trylock())
 		return restart_syscall();
 
+	old_active = new_active = NULL;
 	block_netpoll_tx();
 	read_lock(&bond->lock);
 	write_lock_bh(&bond->curr_slave_lock);
@@ -1286,12 +1284,12 @@ static ssize_t bonding_store_active_slave(struct device *d,
 	if (!strlen(ifname) || buf[0] == '\n') {
 		pr_info("%s: Clearing current active slave.\n",
 			bond->dev->name);
-		bond->curr_active_slave = NULL;
+		rcu_assign_pointer(bond->curr_active_slave, NULL);
 		bond_select_active_slave(bond);
 		goto out;
 	}
 
-	bond_for_each_slave(bond, slave, i) {
+	bond_for_each_slave(bond, slave) {
 		if (strncmp(slave->dev->name, ifname, IFNAMSIZ) == 0) {
 			old_active = bond->curr_active_slave;
 			new_active = slave;
@@ -1349,14 +1347,9 @@ static ssize_t bonding_show_mii_status(struct device *d,
 				       struct device_attribute *attr,
 				       char *buf)
 {
-	struct slave *curr;
 	struct bonding *bond = to_bond(d);
 
-	read_lock(&bond->curr_slave_lock);
-	curr = bond->curr_active_slave;
-	read_unlock(&bond->curr_slave_lock);
-
-	return sprintf(buf, "%s\n", curr ? "up" : "down");
+	return sprintf(buf, "%s\n", bond->curr_active_slave ? "up" : "down");
 }
 static DEVICE_ATTR(mii_status, S_IRUGO, bonding_show_mii_status, NULL);
 
@@ -1475,15 +1468,15 @@ static ssize_t bonding_show_queue_id(struct device *d,
 				     struct device_attribute *attr,
 				     char *buf)
 {
-	struct slave *slave;
-	int i, res = 0;
 	struct bonding *bond = to_bond(d);
+	struct slave *slave;
+	int res = 0;
 
 	if (!rtnl_trylock())
 		return restart_syscall();
 
 	read_lock(&bond->lock);
-	bond_for_each_slave(bond, slave, i) {
+	bond_for_each_slave(bond, slave) {
 		if (res > (PAGE_SIZE - IFNAMSIZ - 6)) {
 			/* not enough space for another interface_name:queue_id pair */
 			if ((PAGE_SIZE - res) > 10)
@@ -1498,6 +1491,7 @@ static ssize_t bonding_show_queue_id(struct device *d,
 	if (res)
 		buf[res-1] = '\n'; /* eat the leftover space */
 	rtnl_unlock();
+
 	return res;
 }
 
@@ -1512,7 +1506,7 @@ static ssize_t bonding_store_queue_id(struct device *d,
 	struct slave *slave, *update_slave;
 	struct bonding *bond = to_bond(d);
 	u16 qid;
-	int i, ret = count;
+	int ret = count;
 	char *delim;
 	struct net_device *sdev = NULL;
 
@@ -1547,7 +1541,7 @@ static ssize_t bonding_store_queue_id(struct device *d,
 
 	/* Search for thes slave and check for duplicate qids */
 	update_slave = NULL;
-	bond_for_each_slave(bond, slave, i) {
+	bond_for_each_slave(bond, slave) {
 		if (sdev == slave->dev)
 			/*
 			 * We don't need to check the matching
@@ -1599,8 +1593,8 @@ static ssize_t bonding_store_slaves_active(struct device *d,
 					   struct device_attribute *attr,
 					   const char *buf, size_t count)
 {
-	int i, new_value, ret = count;
 	struct bonding *bond = to_bond(d);
+	int new_value, ret = count;
 	struct slave *slave;
 
 	if (sscanf(buf, "%d", &new_value) != 1) {
@@ -1623,7 +1617,7 @@ static ssize_t bonding_store_slaves_active(struct device *d,
 	}
 
 	read_lock(&bond->lock);
-	bond_for_each_slave(bond, slave, i) {
+	bond_for_each_slave(bond, slave) {
 		if (!bond_is_active_slave(slave)) {
 			if (new_value)
 				slave->inactive = 0;

drivers/net/bonding/bonding.h

@@ -71,6 +71,28 @@
 	set_fs(fs);			\
 	res; })
 
+/* slave list primitives */
+#define bond_to_slave(ptr) list_entry(ptr, struct slave, list)
+
+/* IMPORTANT: bond_first/last_slave can return NULL in case of an empty list */
+#define bond_first_slave(bond) \
+	list_first_entry_or_null(&(bond)->slave_list, struct slave, list)
+#define bond_last_slave(bond) \
+	(list_empty(&(bond)->slave_list) ? NULL : \
+					   bond_to_slave((bond)->slave_list.prev))
+
+#define bond_is_first_slave(bond, pos) ((pos)->list.prev == &(bond)->slave_list)
+#define bond_is_last_slave(bond, pos) ((pos)->list.next == &(bond)->slave_list)
+
+/* Since bond_first/last_slave can return NULL, these can return NULL too */
+#define bond_next_slave(bond, pos) \
+	(bond_is_last_slave(bond, pos) ? bond_first_slave(bond) : \
+					 bond_to_slave((pos)->list.next))
+
+#define bond_prev_slave(bond, pos) \
+	(bond_is_first_slave(bond, pos) ? bond_last_slave(bond) : \
+					  bond_to_slave((pos)->list.prev))
+
 /**
  * bond_for_each_slave_from - iterate the slaves list from a starting point
  * @bond:	the bond holding this list.
@@ -80,37 +102,33 @@
  *
  * Caller must hold bond->lock
  */
-#define bond_for_each_slave_from(bond, pos, cnt, start)	\
-	for (cnt = 0, pos = start;				\
-	     cnt < (bond)->slave_cnt;				\
-             cnt++, pos = (pos)->next)
+#define bond_for_each_slave_from(bond, pos, cnt, start) \
+	for (cnt = 0, pos = start; pos && cnt < (bond)->slave_cnt; \
+	     cnt++, pos = bond_next_slave(bond, pos))
 
 /**
- * bond_for_each_slave_from_to - iterate the slaves list from start point to stop point
- * @bond:	the bond holding this list.
- * @pos:	current slave.
- * @cnt:	counter for number max of moves
- * @start:	start point.
- * @stop:	stop point.
+ * bond_for_each_slave - iterate over all slaves
+ * @bond:	the bond holding this list
+ * @pos:	current slave
  *
  * Caller must hold bond->lock
  */
-#define bond_for_each_slave_from_to(bond, pos, cnt, start, stop)	\
-	for (cnt = 0, pos = start;					\
-	     ((cnt < (bond)->slave_cnt) && (pos != (stop)->next));	\
-             cnt++, pos = (pos)->next)
+#define bond_for_each_slave(bond, pos) \
+	list_for_each_entry(pos, &(bond)->slave_list, list)
+
+/* Caller must have rcu_read_lock */
+#define bond_for_each_slave_rcu(bond, pos) \
+	list_for_each_entry_rcu(pos, &(bond)->slave_list, list)
 
 /**
- * bond_for_each_slave - iterate the slaves list from head
- * @bond:	the bond holding this list.
- * @pos:	current slave.
- * @cnt:	counter for max number of moves
+ * bond_for_each_slave_reverse - iterate in reverse from a given position
+ * @bond:	the bond holding this list
+ * @pos:	slave to continue from
  *
  * Caller must hold bond->lock
  */
-#define bond_for_each_slave(bond, pos, cnt)	\
-		bond_for_each_slave_from(bond, pos, cnt, (bond)->first_slave)
-
+#define bond_for_each_slave_continue_reverse(bond, pos) \
+	list_for_each_entry_continue_reverse(pos, &(bond)->slave_list, list)
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
 extern atomic_t netpoll_block_tx;
@@ -174,8 +192,7 @@ struct vlan_entry {
 
 struct slave {
 	struct net_device *dev; /* first - useful for panic debug */
-	struct slave *next;
-	struct slave *prev;
+	struct list_head list;
 	struct bonding *bond; /* our master */
 	int    delay;
 	unsigned long jiffies;
@@ -215,7 +232,7 @@ struct slave {
  */
 struct bonding {
 	struct   net_device *dev; /* first - useful for panic debug */
-	struct   slave *first_slave;
+	struct   list_head slave_list;
 	struct   slave *curr_active_slave;
 	struct   slave *current_arp_slave;
 	struct   slave *primary_slave;
@@ -270,13 +287,10 @@ static inline struct slave *bond_get_slave_by_dev(struct bonding *bond,
 						  struct net_device *slave_dev)
 {
 	struct slave *slave = NULL;
-	int i;
 
-	bond_for_each_slave(bond, slave, i) {
-		if (slave->dev == slave_dev) {
+	bond_for_each_slave(bond, slave)
+		if (slave->dev == slave_dev)
 			return slave;
-		}
-	}
 
 	return NULL;
 }
@@ -416,10 +430,20 @@ static inline __be32 bond_confirm_addr(struct net_device *dev, __be32 dst, __be3
 	return addr;
 }
 
+static inline bool slave_can_tx(struct slave *slave)
+{
+	if (IS_UP(slave->dev) && slave->link == BOND_LINK_UP &&
+	    bond_is_active_slave(slave))
+		return true;
+	else
+		return false;
+}
+
 struct bond_net;
 
 struct vlan_entry *bond_next_vlan(struct bonding *bond, struct vlan_entry *curr);
 int bond_dev_queue_xmit(struct bonding *bond, struct sk_buff *skb, struct net_device *slave_dev);
+void bond_xmit_slave_id(struct bonding *bond, struct sk_buff *skb, int slave_id);
 int bond_create(struct net *net, const char *name);
 int bond_create_sysfs(struct bond_net *net);
 void bond_destroy_sysfs(struct bond_net *net);
@@ -477,10 +501,9 @@ static inline void bond_destroy_proc_dir(struct bond_net *bn)
 static inline struct slave *bond_slave_has_mac(struct bonding *bond,
 					       const u8 *mac)
 {
-	int i = 0;
 	struct slave *tmp;
 
-	bond_for_each_slave(bond, tmp, i)
+	bond_for_each_slave(bond, tmp)
 		if (ether_addr_equal_64bits(mac, tmp->dev->dev_addr))
 			return tmp;