diff options
Diffstat (limited to 'drivers/net/bonding')
| -rw-r--r-- | drivers/net/bonding/bond_3ad.c | 158 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_3ad.h | 176 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_alb.c | 325 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_alb.h | 1 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_debugfs.c | 12 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_main.c | 1084 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_netlink.c | 136 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_options.c | 1284 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_options.h | 130 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_procfs.c | 53 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_sysfs.c | 979 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_sysfs_slave.c | 8 | ||||
| -rw-r--r-- | drivers/net/bonding/bonding.h | 272 |
13 files changed, 2428 insertions, 2190 deletions
diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c index cce1f1bf90b..0dfeaf5da3f 100644 --- a/drivers/net/bonding/bond_3ad.c +++ b/drivers/net/bonding/bond_3ad.c @@ -157,7 +157,7 @@ static inline struct aggregator *__get_first_agg(struct port *port) rcu_read_lock(); first_slave = bond_first_slave_rcu(bond); - agg = first_slave ? &(SLAVE_AD_INFO(first_slave).aggregator) : NULL; + agg = first_slave ? &(SLAVE_AD_INFO(first_slave)->aggregator) : NULL; rcu_read_unlock(); return agg; @@ -181,7 +181,7 @@ static inline int __agg_has_partner(struct aggregator *agg) */ static inline void __disable_port(struct port *port) { - bond_set_slave_inactive_flags(port->slave); + bond_set_slave_inactive_flags(port->slave, BOND_SLAVE_NOTIFY_LATER); } /** @@ -192,8 +192,8 @@ static inline void __enable_port(struct port *port) { struct slave *slave = port->slave; - if ((slave->link == BOND_LINK_UP) && IS_UP(slave->dev)) - bond_set_slave_active_flags(slave); + if ((slave->link == BOND_LINK_UP) && bond_slave_is_up(slave)) + bond_set_slave_active_flags(slave, BOND_SLAVE_NOTIFY_LATER); } /** @@ -241,7 +241,7 @@ static inline int __check_agg_selection_timer(struct port *port) */ static inline void __get_state_machine_lock(struct port *port) { - spin_lock_bh(&(SLAVE_AD_INFO(port->slave).state_machine_lock)); + spin_lock_bh(&(SLAVE_AD_INFO(port->slave)->state_machine_lock)); } /** @@ -250,7 +250,7 @@ static inline void __get_state_machine_lock(struct port *port) */ static inline void __release_state_machine_lock(struct port *port) { - spin_unlock_bh(&(SLAVE_AD_INFO(port->slave).state_machine_lock)); + spin_unlock_bh(&(SLAVE_AD_INFO(port->slave)->state_machine_lock)); } /** @@ -350,7 +350,7 @@ static u8 __get_duplex(struct port *port) static inline void __initialize_port_locks(struct slave *slave) { /* make sure it isn't called twice */ - spin_lock_init(&(SLAVE_AD_INFO(slave).state_machine_lock)); + spin_lock_init(&(SLAVE_AD_INFO(slave)->state_machine_lock)); } /* Conversions */ @@ -688,8 +688,8 @@ static struct aggregator *__get_active_agg(struct aggregator *aggregator) struct slave *slave; bond_for_each_slave_rcu(bond, slave, iter) - if (SLAVE_AD_INFO(slave).aggregator.is_active) - return &(SLAVE_AD_INFO(slave).aggregator); + if (SLAVE_AD_INFO(slave)->aggregator.is_active) + return &(SLAVE_AD_INFO(slave)->aggregator); return NULL; } @@ -768,11 +768,11 @@ static int ad_lacpdu_send(struct port *port) lacpdu_header = (struct lacpdu_header *)skb_put(skb, length); - memcpy(lacpdu_header->hdr.h_dest, lacpdu_mcast_addr, ETH_ALEN); + ether_addr_copy(lacpdu_header->hdr.h_dest, lacpdu_mcast_addr); /* Note: source address is set to be the member's PERMANENT address, * because we use it to identify loopback lacpdus in receive. */ - memcpy(lacpdu_header->hdr.h_source, slave->perm_hwaddr, ETH_ALEN); + ether_addr_copy(lacpdu_header->hdr.h_source, slave->perm_hwaddr); lacpdu_header->hdr.h_proto = PKT_TYPE_LACPDU; lacpdu_header->lacpdu = port->lacpdu; @@ -810,11 +810,11 @@ static int ad_marker_send(struct port *port, struct bond_marker *marker) marker_header = (struct bond_marker_header *)skb_put(skb, length); - memcpy(marker_header->hdr.h_dest, lacpdu_mcast_addr, ETH_ALEN); + ether_addr_copy(marker_header->hdr.h_dest, lacpdu_mcast_addr); /* Note: source address is set to be the member's PERMANENT address, * because we use it to identify loopback MARKERs in receive. */ - memcpy(marker_header->hdr.h_source, slave->perm_hwaddr, ETH_ALEN); + ether_addr_copy(marker_header->hdr.h_source, slave->perm_hwaddr); marker_header->hdr.h_proto = PKT_TYPE_LACPDU; marker_header->marker = *marker; @@ -1079,7 +1079,8 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port) /* detect loopback situation */ if (MAC_ADDRESS_EQUAL(&(lacpdu->actor_system), &(port->actor_system))) { - pr_err("%s: An illegal loopback occurred on adapter (%s).\nCheck the configuration to verify that all adapters are connected to 802.3ad compliant switch ports\n", + pr_err("%s: An illegal loopback occurred on adapter (%s)\n" + "Check the configuration to verify that all adapters are connected to 802.3ad compliant switch ports\n", port->slave->bond->dev->name, port->slave->dev->name); return; @@ -1283,16 +1284,16 @@ static void ad_port_selection_logic(struct port *port) /* meaning: the port was related to an aggregator * but was not on the aggregator port list */ - pr_warn("%s: Warning: Port %d (on %s) was related to aggregator %d but was not on its port list\n", - port->slave->bond->dev->name, - port->actor_port_number, - port->slave->dev->name, - port->aggregator->aggregator_identifier); + pr_warn_ratelimited("%s: Warning: Port %d (on %s) was related to aggregator %d but was not on its port list\n", + port->slave->bond->dev->name, + port->actor_port_number, + port->slave->dev->name, + port->aggregator->aggregator_identifier); } } /* search on all aggregators for a suitable aggregator for this port */ bond_for_each_slave(bond, slave, iter) { - aggregator = &(SLAVE_AD_INFO(slave).aggregator); + aggregator = &(SLAVE_AD_INFO(slave)->aggregator); /* keep a free aggregator for later use(if needed) */ if (!aggregator->lag_ports) { @@ -1444,9 +1445,9 @@ static struct aggregator *ad_agg_selection_test(struct aggregator *best, break; default: - pr_warn("%s: Impossible agg select mode %d\n", - curr->slave->bond->dev->name, - __get_agg_selection_mode(curr->lag_ports)); + pr_warn_ratelimited("%s: Impossible agg select mode %d\n", + curr->slave->bond->dev->name, + __get_agg_selection_mode(curr->lag_ports)); break; } @@ -1503,7 +1504,7 @@ static void ad_agg_selection_logic(struct aggregator *agg) best = (active && agg_device_up(active)) ? active : NULL; bond_for_each_slave_rcu(bond, slave, iter) { - agg = &(SLAVE_AD_INFO(slave).aggregator); + agg = &(SLAVE_AD_INFO(slave)->aggregator); agg->is_active = 0; @@ -1548,7 +1549,7 @@ static void ad_agg_selection_logic(struct aggregator *agg) best->slave ? best->slave->dev->name : "NULL"); bond_for_each_slave_rcu(bond, slave, iter) { - agg = &(SLAVE_AD_INFO(slave).aggregator); + agg = &(SLAVE_AD_INFO(slave)->aggregator); pr_debug("Agg=%d; P=%d; a k=%d; p k=%d; Ind=%d; Act=%d\n", agg->aggregator_identifier, agg->num_of_ports, @@ -1559,9 +1560,9 @@ static void ad_agg_selection_logic(struct aggregator *agg) /* check if any partner replys */ if (best->is_individual) { - pr_warn("%s: Warning: No 802.3ad response from the link partner for any adapters in the bond\n", - best->slave ? - best->slave->bond->dev->name : "NULL"); + pr_warn_ratelimited("%s: Warning: No 802.3ad response from the link partner for any adapters in the bond\n", + best->slave ? + best->slave->bond->dev->name : "NULL"); } best->is_active = 1; @@ -1796,8 +1797,6 @@ void bond_3ad_initiate_agg_selection(struct bonding *bond, int timeout) BOND_AD_INFO(bond).agg_select_timer = timeout; } -static u16 aggregator_identifier; - /** * bond_3ad_initialize - initialize a bond's 802.3ad parameters and structures * @bond: bonding struct to work on @@ -1811,7 +1810,7 @@ void bond_3ad_initialize(struct bonding *bond, u16 tick_resolution) if (!MAC_ADDRESS_EQUAL(&(BOND_AD_INFO(bond).system.sys_mac_addr), bond->dev->dev_addr)) { - aggregator_identifier = 0; + BOND_AD_INFO(bond).aggregator_identifier = 0; BOND_AD_INFO(bond).system.sys_priority = 0xFFFF; BOND_AD_INFO(bond).system.sys_mac_addr = *((struct mac_addr *)bond->dev->dev_addr); @@ -1841,16 +1840,16 @@ void bond_3ad_bind_slave(struct slave *slave) struct aggregator *aggregator; /* check that the slave has not been initialized yet. */ - if (SLAVE_AD_INFO(slave).port.slave != slave) { + if (SLAVE_AD_INFO(slave)->port.slave != slave) { /* port initialization */ - port = &(SLAVE_AD_INFO(slave).port); + port = &(SLAVE_AD_INFO(slave)->port); ad_initialize_port(port, bond->params.lacp_fast); __initialize_port_locks(slave); port->slave = slave; - port->actor_port_number = SLAVE_AD_INFO(slave).id; + port->actor_port_number = SLAVE_AD_INFO(slave)->id; /* key is determined according to the link speed, duplex and user key(which * is yet not supported) */ @@ -1875,12 +1874,12 @@ void bond_3ad_bind_slave(struct slave *slave) __disable_port(port); /* aggregator initialization */ - aggregator = &(SLAVE_AD_INFO(slave).aggregator); + aggregator = &(SLAVE_AD_INFO(slave)->aggregator); ad_initialize_agg(aggregator); aggregator->aggregator_mac_address = *((struct mac_addr *)bond->dev->dev_addr); - aggregator->aggregator_identifier = (++aggregator_identifier); + aggregator->aggregator_identifier = ++BOND_AD_INFO(bond).aggregator_identifier; aggregator->slave = slave; aggregator->is_active = 0; aggregator->num_of_ports = 0; @@ -1904,8 +1903,8 @@ void bond_3ad_unbind_slave(struct slave *slave) struct slave *slave_iter; struct list_head *iter; - aggregator = &(SLAVE_AD_INFO(slave).aggregator); - port = &(SLAVE_AD_INFO(slave).port); + aggregator = &(SLAVE_AD_INFO(slave)->aggregator); + port = &(SLAVE_AD_INFO(slave)->port); /* if slave is null, the whole port is not initialized */ if (!port->slave) { @@ -1933,7 +1932,7 @@ void bond_3ad_unbind_slave(struct slave *slave) (aggregator->lag_ports->next_port_in_aggregator)) { /* find new aggregator for the related port(s) */ bond_for_each_slave(bond, slave_iter, iter) { - new_aggregator = &(SLAVE_AD_INFO(slave_iter).aggregator); + new_aggregator = &(SLAVE_AD_INFO(slave_iter)->aggregator); /* if the new aggregator is empty, or it is * connected to our port only */ @@ -1950,7 +1949,7 @@ void bond_3ad_unbind_slave(struct slave *slave) * new aggregator */ if ((new_aggregator) && ((!new_aggregator->lag_ports) || ((new_aggregator->lag_ports == port) && !new_aggregator->lag_ports->next_port_in_aggregator))) { - pr_debug("Some port(s) related to LAG %d - replaceing with LAG %d\n", + pr_debug("Some port(s) related to LAG %d - replacing with LAG %d\n", aggregator->aggregator_identifier, new_aggregator->aggregator_identifier); @@ -2011,7 +2010,7 @@ void bond_3ad_unbind_slave(struct slave *slave) /* find the aggregator that this port is connected to */ bond_for_each_slave(bond, slave_iter, iter) { - temp_aggregator = &(SLAVE_AD_INFO(slave_iter).aggregator); + temp_aggregator = &(SLAVE_AD_INFO(slave_iter)->aggregator); prev_port = NULL; /* search the port in the aggregator's related ports */ for (temp_port = temp_aggregator->lag_ports; temp_port; @@ -2064,6 +2063,7 @@ void bond_3ad_state_machine_handler(struct work_struct *work) struct list_head *iter; struct slave *slave; struct port *port; + bool should_notify_rtnl = BOND_SLAVE_NOTIFY_LATER; read_lock(&bond->lock); rcu_read_lock(); @@ -2076,13 +2076,13 @@ void bond_3ad_state_machine_handler(struct work_struct *work) if (BOND_AD_INFO(bond).agg_select_timer && !(--BOND_AD_INFO(bond).agg_select_timer)) { slave = bond_first_slave_rcu(bond); - port = slave ? &(SLAVE_AD_INFO(slave).port) : NULL; + port = slave ? &(SLAVE_AD_INFO(slave)->port) : NULL; /* select the active aggregator for the bond */ if (port) { if (!port->slave) { - pr_warn("%s: Warning: bond's first port is uninitialized\n", - bond->dev->name); + pr_warn_ratelimited("%s: Warning: bond's first port is uninitialized\n", + bond->dev->name); goto re_arm; } @@ -2094,10 +2094,10 @@ void bond_3ad_state_machine_handler(struct work_struct *work) /* for each port run the state machines */ bond_for_each_slave_rcu(bond, slave, iter) { - port = &(SLAVE_AD_INFO(slave).port); + port = &(SLAVE_AD_INFO(slave)->port); if (!port->slave) { - pr_warn("%s: Warning: Found an uninitialized port\n", - bond->dev->name); + pr_warn_ratelimited("%s: Warning: Found an uninitialized port\n", + bond->dev->name); goto re_arm; } @@ -2121,8 +2121,19 @@ void bond_3ad_state_machine_handler(struct work_struct *work) } re_arm: + bond_for_each_slave_rcu(bond, slave, iter) { + if (slave->should_notify) { + should_notify_rtnl = BOND_SLAVE_NOTIFY_NOW; + break; + } + } rcu_read_unlock(); read_unlock(&bond->lock); + + if (should_notify_rtnl && rtnl_trylock()) { + bond_slave_state_notify(bond); + rtnl_unlock(); + } queue_delayed_work(bond->wq, &bond->ad_work, ad_delta_in_ticks); } @@ -2144,11 +2155,11 @@ static int bond_3ad_rx_indication(struct lacpdu *lacpdu, struct slave *slave, if (length >= sizeof(struct lacpdu)) { - port = &(SLAVE_AD_INFO(slave).port); + port = &(SLAVE_AD_INFO(slave)->port); if (!port->slave) { - pr_warn("%s: Warning: port of slave %s is uninitialized\n", - slave->dev->name, slave->bond->dev->name); + pr_warn_ratelimited("%s: Warning: port of slave %s is uninitialized\n", + slave->dev->name, slave->bond->dev->name); return ret; } @@ -2201,7 +2212,7 @@ void bond_3ad_adapter_speed_changed(struct slave *slave) { struct port *port; - port = &(SLAVE_AD_INFO(slave).port); + port = &(SLAVE_AD_INFO(slave)->port); /* if slave is null, the whole port is not initialized */ if (!port->slave) { @@ -2234,7 +2245,7 @@ void bond_3ad_adapter_duplex_changed(struct slave *slave) { struct port *port; - port = &(SLAVE_AD_INFO(slave).port); + port = &(SLAVE_AD_INFO(slave)->port); /* if slave is null, the whole port is not initialized */ if (!port->slave) { @@ -2268,7 +2279,7 @@ void bond_3ad_handle_link_change(struct slave *slave, char link) { struct port *port; - port = &(SLAVE_AD_INFO(slave).port); + port = &(SLAVE_AD_INFO(slave)->port); /* if slave is null, the whole port is not initialized */ if (!port->slave) { @@ -2300,9 +2311,9 @@ void bond_3ad_handle_link_change(struct slave *slave, char link) port->actor_oper_port_key = (port->actor_admin_port_key &= ~AD_SPEED_KEY_BITS); } - pr_debug("Port %d changed link status to %s", - port->actor_port_number, - (link == BOND_LINK_UP) ? "UP" : "DOWN"); + pr_debug("Port %d changed link status to %s\n", + port->actor_port_number, + link == BOND_LINK_UP ? "UP" : "DOWN"); /* there is no need to reselect a new aggregator, just signal the * state machines to reinitialize */ @@ -2336,7 +2347,7 @@ int bond_3ad_set_carrier(struct bonding *bond) ret = 0; goto out; } - active = __get_active_agg(&(SLAVE_AD_INFO(first_slave).aggregator)); + 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) { @@ -2373,24 +2384,23 @@ int __bond_3ad_get_active_agg_info(struct bonding *bond, struct port *port; bond_for_each_slave_rcu(bond, slave, iter) { - port = &(SLAVE_AD_INFO(slave).port); + port = &(SLAVE_AD_INFO(slave)->port); if (port->aggregator && port->aggregator->is_active) { aggregator = port->aggregator; break; } } - if (aggregator) { - ad_info->aggregator_id = aggregator->aggregator_identifier; - ad_info->ports = aggregator->num_of_ports; - ad_info->actor_key = aggregator->actor_oper_aggregator_key; - ad_info->partner_key = aggregator->partner_oper_aggregator_key; - memcpy(ad_info->partner_system, - aggregator->partner_system.mac_addr_value, ETH_ALEN); - return 0; - } + if (!aggregator) + return -1; - return -1; + ad_info->aggregator_id = aggregator->aggregator_identifier; + ad_info->ports = aggregator->num_of_ports; + ad_info->actor_key = aggregator->actor_oper_aggregator_key; + ad_info->partner_key = aggregator->partner_oper_aggregator_key; + ether_addr_copy(ad_info->partner_system, + aggregator->partner_system.mac_addr_value); + return 0; } /* Wrapper used to hold bond->lock so no slave manipulation can occur */ @@ -2430,22 +2440,22 @@ int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev) goto err_free; } - slave_agg_no = bond_xmit_hash(bond, skb, slaves_in_agg); + slave_agg_no = bond_xmit_hash(bond, skb) % slaves_in_agg; first_ok_slave = NULL; bond_for_each_slave_rcu(bond, slave, iter) { - agg = SLAVE_AD_INFO(slave).port.aggregator; + agg = SLAVE_AD_INFO(slave)->port.aggregator; if (!agg || agg->aggregator_identifier != agg_id) continue; if (slave_agg_no >= 0) { - if (!first_ok_slave && SLAVE_IS_OK(slave)) + if (!first_ok_slave && bond_slave_can_tx(slave)) first_ok_slave = slave; slave_agg_no--; continue; } - if (SLAVE_IS_OK(slave)) { + if (bond_slave_can_tx(slave)) { bond_dev_queue_xmit(bond, skb, slave->dev); goto out; } @@ -2469,7 +2479,7 @@ out: return NETDEV_TX_OK; err_free: /* no suitable interface, frame not sent */ - kfree_skb(skb); + dev_kfree_skb_any(skb); goto out; } @@ -2512,7 +2522,7 @@ void bond_3ad_update_lacp_rate(struct bonding *bond) lacp_fast = bond->params.lacp_fast; bond_for_each_slave(bond, slave, iter) { - port = &(SLAVE_AD_INFO(slave).port); + port = &(SLAVE_AD_INFO(slave)->port); __get_state_machine_lock(port); if (lacp_fast) port->actor_oper_port_state |= AD_STATE_LACP_TIMEOUT; diff --git a/drivers/net/bonding/bond_3ad.h b/drivers/net/bonding/bond_3ad.h index 13dc9d3c5e3..bb03b1df2f3 100644 --- a/drivers/net/bonding/bond_3ad.h +++ b/drivers/net/bonding/bond_3ad.h @@ -28,7 +28,7 @@ #include <linux/netdevice.h> #include <linux/if_ether.h> -// General definitions +/* General definitions */ #define PKT_TYPE_LACPDU cpu_to_be16(ETH_P_SLOW) #define AD_TIMER_INTERVAL 100 /*msec*/ @@ -47,54 +47,54 @@ enum { BOND_AD_COUNT = 2, }; -// rx machine states(43.4.11 in the 802.3ad standard) +/* rx machine states(43.4.11 in the 802.3ad standard) */ typedef enum { AD_RX_DUMMY, - AD_RX_INITIALIZE, // rx Machine - AD_RX_PORT_DISABLED, // rx Machine - AD_RX_LACP_DISABLED, // rx Machine - AD_RX_EXPIRED, // rx Machine - AD_RX_DEFAULTED, // rx Machine - AD_RX_CURRENT // rx Machine + AD_RX_INITIALIZE, /* rx Machine */ + AD_RX_PORT_DISABLED, /* rx Machine */ + AD_RX_LACP_DISABLED, /* rx Machine */ + AD_RX_EXPIRED, /* rx Machine */ + AD_RX_DEFAULTED, /* rx Machine */ + AD_RX_CURRENT /* rx Machine */ } rx_states_t; -// periodic machine states(43.4.12 in the 802.3ad standard) +/* periodic machine states(43.4.12 in the 802.3ad standard) */ typedef enum { AD_PERIODIC_DUMMY, - AD_NO_PERIODIC, // periodic machine - AD_FAST_PERIODIC, // periodic machine - AD_SLOW_PERIODIC, // periodic machine - AD_PERIODIC_TX // periodic machine + AD_NO_PERIODIC, /* periodic machine */ + AD_FAST_PERIODIC, /* periodic machine */ + AD_SLOW_PERIODIC, /* periodic machine */ + AD_PERIODIC_TX /* periodic machine */ } periodic_states_t; -// mux machine states(43.4.13 in the 802.3ad standard) +/* mux machine states(43.4.13 in the 802.3ad standard) */ typedef enum { AD_MUX_DUMMY, - AD_MUX_DETACHED, // mux machine - AD_MUX_WAITING, // mux machine - AD_MUX_ATTACHED, // mux machine - AD_MUX_COLLECTING_DISTRIBUTING // mux machine + AD_MUX_DETACHED, /* mux machine */ + AD_MUX_WAITING, /* mux machine */ + AD_MUX_ATTACHED, /* mux machine */ + AD_MUX_COLLECTING_DISTRIBUTING /* mux machine */ } mux_states_t; -// tx machine states(43.4.15 in the 802.3ad standard) +/* tx machine states(43.4.15 in the 802.3ad standard) */ typedef enum { AD_TX_DUMMY, - AD_TRANSMIT // tx Machine + AD_TRANSMIT /* tx Machine */ } tx_states_t; -// rx indication types +/* rx indication types */ typedef enum { - AD_TYPE_LACPDU = 1, // type lacpdu - AD_TYPE_MARKER // type marker + AD_TYPE_LACPDU = 1, /* type lacpdu */ + AD_TYPE_MARKER /* type marker */ } pdu_type_t; -// rx marker indication types +/* rx marker indication types */ typedef enum { - AD_MARKER_INFORMATION_SUBTYPE = 1, // marker imformation subtype - AD_MARKER_RESPONSE_SUBTYPE // marker response subtype + AD_MARKER_INFORMATION_SUBTYPE = 1, /* marker imformation subtype */ + AD_MARKER_RESPONSE_SUBTYPE /* marker response subtype */ } bond_marker_subtype_t; -// timers types(43.4.9 in the 802.3ad standard) +/* timers types(43.4.9 in the 802.3ad standard) */ typedef enum { AD_CURRENT_WHILE_TIMER, AD_ACTOR_CHURN_TIMER, @@ -105,35 +105,35 @@ typedef enum { #pragma pack(1) -// Link Aggregation Control Protocol(LACP) data unit structure(43.4.2.2 in the 802.3ad standard) +/* Link Aggregation Control Protocol(LACP) data unit structure(43.4.2.2 in the 802.3ad standard) */ typedef struct lacpdu { - u8 subtype; // = LACP(= 0x01) + u8 subtype; /* = LACP(= 0x01) */ u8 version_number; - u8 tlv_type_actor_info; // = actor information(type/length/value) - u8 actor_information_length; // = 20 + u8 tlv_type_actor_info; /* = actor information(type/length/value) */ + u8 actor_information_length; /* = 20 */ __be16 actor_system_priority; struct mac_addr actor_system; __be16 actor_key; __be16 actor_port_priority; __be16 actor_port; u8 actor_state; - u8 reserved_3_1[3]; // = 0 - u8 tlv_type_partner_info; // = partner information - u8 partner_information_length; // = 20 + u8 reserved_3_1[3]; /* = 0 */ + u8 tlv_type_partner_info; /* = partner information */ + u8 partner_information_length; /* = 20 */ __be16 partner_system_priority; struct mac_addr partner_system; __be16 partner_key; __be16 partner_port_priority; __be16 partner_port; u8 partner_state; - u8 reserved_3_2[3]; // = 0 - u8 tlv_type_collector_info; // = collector information - u8 collector_information_length; // = 16 + u8 reserved_3_2[3]; /* = 0 */ + u8 tlv_type_collector_info; /* = collector information */ + u8 collector_information_length;/* = 16 */ __be16 collector_max_delay; u8 reserved_12[12]; - u8 tlv_type_terminator; // = terminator - u8 terminator_length; // = 0 - u8 reserved_50[50]; // = 0 + u8 tlv_type_terminator; /* = terminator */ + u8 terminator_length; /* = 0 */ + u8 reserved_50[50]; /* = 0 */ } __packed lacpdu_t; typedef struct lacpdu_header { @@ -141,20 +141,20 @@ typedef struct lacpdu_header { struct lacpdu lacpdu; } __packed lacpdu_header_t; -// Marker Protocol Data Unit(PDU) structure(43.5.3.2 in the 802.3ad standard) +/* Marker Protocol Data Unit(PDU) structure(43.5.3.2 in the 802.3ad standard) */ typedef struct bond_marker { - u8 subtype; // = 0x02 (marker PDU) - u8 version_number; // = 0x01 - u8 tlv_type; // = 0x01 (marker information) - // = 0x02 (marker response information) - u8 marker_length; // = 0x16 - u16 requester_port; // The number assigned to the port by the requester - struct mac_addr requester_system; // The requester's system id - u32 requester_transaction_id; // The transaction id allocated by the requester, - u16 pad; // = 0 - u8 tlv_type_terminator; // = 0x00 - u8 terminator_length; // = 0x00 - u8 reserved_90[90]; // = 0 + u8 subtype; /* = 0x02 (marker PDU) */ + u8 version_number; /* = 0x01 */ + u8 tlv_type; /* = 0x01 (marker information) */ + /* = 0x02 (marker response information) */ + u8 marker_length; /* = 0x16 */ + u16 requester_port; /* The number assigned to the port by the requester */ + struct mac_addr requester_system; /* The requester's system id */ + u32 requester_transaction_id; /* The transaction id allocated by the requester, */ + u16 pad; /* = 0 */ + u8 tlv_type_terminator; /* = 0x00 */ + u8 terminator_length; /* = 0x00 */ + u8 reserved_90[90]; /* = 0 */ } __packed bond_marker_t; typedef struct bond_marker_header { @@ -173,7 +173,7 @@ struct port; #pragma pack(8) #endif -// aggregator structure(43.4.5 in the 802.3ad standard) +/* aggregator structure(43.4.5 in the 802.3ad standard) */ typedef struct aggregator { struct mac_addr aggregator_mac_address; u16 aggregator_identifier; @@ -183,12 +183,12 @@ typedef struct aggregator { struct mac_addr partner_system; u16 partner_system_priority; u16 partner_oper_aggregator_key; - u16 receive_state; // BOOLEAN - u16 transmit_state; // BOOLEAN + u16 receive_state; /* BOOLEAN */ + u16 transmit_state; /* BOOLEAN */ struct port *lag_ports; - // ****** PRIVATE PARAMETERS ****** - struct slave *slave; // pointer to the bond slave that this aggregator belongs to - u16 is_active; // BOOLEAN. Indicates if this aggregator is active + /* ****** PRIVATE PARAMETERS ****** */ + struct slave *slave; /* pointer to the bond slave that this aggregator belongs to */ + u16 is_active; /* BOOLEAN. Indicates if this aggregator is active */ u16 num_of_ports; } aggregator_t; @@ -201,12 +201,12 @@ struct port_params { u16 port_state; }; -// port structure(43.4.6 in the 802.3ad standard) +/* port structure(43.4.6 in the 802.3ad standard) */ typedef struct port { u16 actor_port_number; u16 actor_port_priority; - struct mac_addr actor_system; // This parameter is added here although it is not specified in the standard, just for simplification - u16 actor_system_priority; // This parameter is added here although it is not specified in the standard, just for simplification + struct mac_addr actor_system; /* This parameter is added here although it is not specified in the standard, just for simplification */ + u16 actor_system_priority; /* This parameter is added here although it is not specified in the standard, just for simplification */ u16 actor_port_aggregator_identifier; bool ntt; u16 actor_admin_port_key; @@ -219,24 +219,24 @@ typedef struct port { bool is_enabled; - // ****** PRIVATE PARAMETERS ****** - u16 sm_vars; // all state machines variables for this port - rx_states_t sm_rx_state; // state machine rx state - u16 sm_rx_timer_counter; // state machine rx timer counter - periodic_states_t sm_periodic_state;// state machine periodic state - u16 sm_periodic_timer_counter; // state machine periodic timer counter - mux_states_t sm_mux_state; // state machine mux state - u16 sm_mux_timer_counter; // state machine mux timer counter - tx_states_t sm_tx_state; // state machine tx state - u16 sm_tx_timer_counter; // state machine tx timer counter(allways on - enter to transmit state 3 time per second) - struct slave *slave; // pointer to the bond slave that this port belongs to - struct aggregator *aggregator; // pointer to an aggregator that this port related to - struct port *next_port_in_aggregator; // Next port on the linked list of the parent aggregator - u32 transaction_id; // continuous number for identification of Marker PDU's; - struct lacpdu lacpdu; // the lacpdu that will be sent for this port + /* ****** PRIVATE PARAMETERS ****** */ + u16 sm_vars; /* all state machines variables for this port */ + rx_states_t sm_rx_state; /* state machine rx state */ + u16 sm_rx_timer_counter; /* state machine rx timer counter */ + periodic_states_t sm_periodic_state; /* state machine periodic state */ + u16 sm_periodic_timer_counter; /* state machine periodic timer counter */ + mux_states_t sm_mux_state; /* state machine mux state */ + u16 sm_mux_timer_counter; /* state machine mux timer counter */ + tx_states_t sm_tx_state; /* state machine tx state */ + u16 sm_tx_timer_counter; /* state machine tx timer counter(allways on - enter to transmit state 3 time per second) */ + struct slave *slave; /* pointer to the bond slave that this port belongs to */ + struct aggregator *aggregator; /* pointer to an aggregator that this port related to */ + struct port *next_port_in_aggregator; /* Next port on the linked list of the parent aggregator */ + u32 transaction_id; /* continuous number for identification of Marker PDU's; */ + struct lacpdu lacpdu; /* the lacpdu that will be sent for this port */ } port_t; -// system structure +/* system structure */ struct ad_system { u16 sys_priority; struct mac_addr sys_mac_addr; @@ -246,26 +246,26 @@ struct ad_system { #pragma pack() #endif -// ================= AD Exported structures to the main bonding code ================== +/* ========== AD Exported structures to the main bonding code ========== */ #define BOND_AD_INFO(bond) ((bond)->ad_info) #define SLAVE_AD_INFO(slave) ((slave)->ad_info) struct ad_bond_info { - struct ad_system system; /* 802.3ad system structure */ - u32 agg_select_timer; // Timer to select aggregator after all adapter's hand shakes + struct ad_system system; /* 802.3ad system structure */ + u32 agg_select_timer; /* Timer to select aggregator after all adapter's hand shakes */ + u16 aggregator_identifier; }; struct ad_slave_info { - struct aggregator aggregator; // 802.3ad aggregator structure - struct port port; // 802.3ad port structure - spinlock_t state_machine_lock; /* mutex state machines vs. - incoming LACPDU */ + struct aggregator aggregator; /* 802.3ad aggregator structure */ + struct port port; /* 802.3ad port structure */ + spinlock_t state_machine_lock; /* mutex state machines vs. incoming LACPDU */ u16 id; }; -// ================= AD Exported functions to the main bonding code ================== +/* ========== AD Exported functions to the main bonding code ========== */ void bond_3ad_initialize(struct bonding *bond, u16 tick_resolution); -void bond_3ad_bind_slave(struct slave *slave); +void bond_3ad_bind_slave(struct slave *slave); void bond_3ad_unbind_slave(struct slave *slave); void bond_3ad_state_machine_handler(struct work_struct *); void bond_3ad_initiate_agg_selection(struct bonding *bond, int timeout); @@ -280,5 +280,5 @@ int bond_3ad_lacpdu_recv(const struct sk_buff *skb, struct bonding *bond, struct slave *slave); int bond_3ad_set_carrier(struct bonding *bond); void bond_3ad_update_lacp_rate(struct bonding *bond); -#endif //__BOND_3AD_H__ +#endif /* __BOND_3AD_H__ */ diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c index a2c47476804..76c0dade233 100644 --- a/drivers/net/bonding/bond_alb.c +++ b/drivers/net/bonding/bond_alb.c @@ -82,7 +82,8 @@ static inline struct arp_pkt *arp_pkt(const struct sk_buff *skb) } /* Forward declaration */ -static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[]); +static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[], + bool strict_match); static void rlb_purge_src_ip(struct bonding *bond, struct arp_pkt *arp); static void rlb_src_unlink(struct bonding *bond, u32 index); static void rlb_src_link(struct bonding *bond, u32 ip_src_hash, @@ -93,9 +94,8 @@ static inline u8 _simple_hash(const u8 *hash_start, int hash_size) int i; u8 hash = 0; - for (i = 0; i < hash_size; i++) { + for (i = 0; i < hash_size; i++) hash ^= hash_start[i]; - } return hash; } @@ -190,9 +190,8 @@ static int tlb_initialize(struct bonding *bond) bond_info->tx_hashtbl = new_hashtbl; - for (i = 0; i < TLB_HASH_TABLE_SIZE; i++) { + for (i = 0; i < TLB_HASH_TABLE_SIZE; i++) tlb_init_table_entry(&bond_info->tx_hashtbl[i], 0); - } _unlock_tx_hashtbl_bh(bond); @@ -230,7 +229,7 @@ static struct slave *tlb_get_least_loaded_slave(struct bonding *bond) /* Find the slave with the largest gap */ bond_for_each_slave_rcu(bond, slave, iter) { - if (SLAVE_IS_OK(slave)) { + if (bond_slave_can_tx(slave)) { long long gap = compute_gap(slave); if (max_gap < gap) { @@ -264,9 +263,8 @@ static struct slave *__tlb_choose_channel(struct bonding *bond, u32 hash_index, hash_table[hash_index].next = next_index; hash_table[hash_index].prev = TLB_NULL_INDEX; - if (next_index != TLB_NULL_INDEX) { + if (next_index != TLB_NULL_INDEX) hash_table[next_index].prev = hash_index; - } slave_info->head = hash_index; slave_info->load += @@ -274,9 +272,8 @@ static struct slave *__tlb_choose_channel(struct bonding *bond, u32 hash_index, } } - if (assigned_slave) { + if (assigned_slave) hash_table[hash_index].tx_bytes += skb_len; - } return assigned_slave; } @@ -329,7 +326,7 @@ static void rlb_update_entry_from_arp(struct bonding *bond, struct arp_pkt *arp) _lock_rx_hashtbl_bh(bond); - hash_index = _simple_hash((u8*)&(arp->ip_src), sizeof(arp->ip_src)); + hash_index = _simple_hash((u8 *)&(arp->ip_src), sizeof(arp->ip_src)); client_info = &(bond_info->rx_hashtbl[hash_index]); if ((client_info->assigned) && @@ -337,7 +334,7 @@ static void rlb_update_entry_from_arp(struct bonding *bond, struct arp_pkt *arp) (client_info->ip_dst == arp->ip_src) && (!ether_addr_equal_64bits(client_info->mac_dst, arp->mac_src))) { /* update the clients MAC address */ - memcpy(client_info->mac_dst, arp->mac_src, ETH_ALEN); + ether_addr_copy(client_info->mac_dst, arp->mac_src); client_info->ntt = 1; bond_info->rx_ntt = 1; } @@ -387,7 +384,7 @@ static struct slave *rlb_next_rx_slave(struct bonding *bond) bool found = false; bond_for_each_slave(bond, slave, iter) { - if (!SLAVE_IS_OK(slave)) + if (!bond_slave_can_tx(slave)) continue; if (!found) { if (!before || before->speed < slave->speed) @@ -420,7 +417,7 @@ static struct slave *__rlb_next_rx_slave(struct bonding *bond) bool found = false; bond_for_each_slave_rcu(bond, slave, iter) { - if (!SLAVE_IS_OK(slave)) + if (!bond_slave_can_tx(slave)) continue; if (!found) { if (!before || before->speed < slave->speed) @@ -451,9 +448,8 @@ static struct slave *__rlb_next_rx_slave(struct bonding *bond) */ static void rlb_teach_disabled_mac_on_primary(struct bonding *bond, u8 addr[]) { - if (!bond->curr_active_slave) { + if (!bond->curr_active_slave) return; - } if (!bond->alb_info.primary_is_promisc) { if (!dev_set_promiscuity(bond->curr_active_slave->dev, 1)) @@ -464,7 +460,7 @@ static void rlb_teach_disabled_mac_on_primary(struct bonding *bond, u8 addr[]) bond->alb_info.rlb_promisc_timeout_counter = 0; - alb_send_learning_packets(bond->curr_active_slave, addr); + alb_send_learning_packets(bond->curr_active_slave, addr, true); } /* slave being removed should not be active at this point @@ -513,9 +509,8 @@ static void rlb_clear_slave(struct bonding *bond, struct slave *slave) write_lock_bh(&bond->curr_slave_lock); - if (slave != bond->curr_active_slave) { + if (slave != bond->curr_active_slave) rlb_teach_disabled_mac_on_primary(bond, slave->dev->dev_addr); - } write_unlock_bh(&bond->curr_slave_lock); } @@ -524,9 +519,8 @@ static void rlb_update_client(struct rlb_client_info *client_info) { int i; - if (!client_info->slave) { + if (!client_info->slave) return; - } for (i = 0; i < RLB_ARP_BURST_SIZE; i++) { struct sk_buff *skb; @@ -574,9 +568,8 @@ static void rlb_update_rx_clients(struct bonding *bond) client_info = &(bond_info->rx_hashtbl[hash_index]); if (client_info->ntt) { rlb_update_client(client_info); - if (bond_info->rlb_update_retry_counter == 0) { + if (bond_info->rlb_update_retry_counter == 0) client_info->ntt = 0; - } } } @@ -610,10 +603,10 @@ static void rlb_req_update_slave_clients(struct bonding *bond, struct slave *sla } } - // update the team's flag only after the whole iteration + /* update the team's flag only after the whole iteration */ if (ntt) { bond_info->rx_ntt = 1; - //fasten the change + /* fasten the change */ bond_info->rlb_update_retry_counter = RLB_UPDATE_RETRY; } @@ -677,9 +670,9 @@ static struct slave *rlb_choose_channel(struct sk_buff *skb, struct bonding *bon /* the entry is already assigned to this client */ if (!ether_addr_equal_64bits(arp->mac_dst, mac_bcast)) { /* update mac address from arp */ - memcpy(client_info->mac_dst, arp->mac_dst, ETH_ALEN); + ether_addr_copy(client_info->mac_dst, arp->mac_dst); } - memcpy(client_info->mac_src, arp->mac_src, ETH_ALEN); + ether_addr_copy(client_info->mac_src, arp->mac_src); assigned_slave = client_info->slave; if (assigned_slave) { @@ -719,8 +712,8 @@ static struct slave *rlb_choose_channel(struct sk_buff *skb, struct bonding *bon * will be updated with clients actual unicast mac address * upon receiving an arp reply. */ - memcpy(client_info->mac_dst, arp->mac_dst, ETH_ALEN); - memcpy(client_info->mac_src, arp->mac_src, ETH_ALEN); + ether_addr_copy(client_info->mac_dst, arp->mac_dst); + ether_addr_copy(client_info->mac_src, arp->mac_src); client_info->slave = assigned_slave; if (!ether_addr_equal_64bits(client_info->mac_dst, mac_bcast)) { @@ -730,7 +723,7 @@ static struct slave *rlb_choose_channel(struct sk_buff *skb, struct bonding *bon client_info->ntt = 0; } - if (!vlan_get_tag(skb, &client_info->vlan_id)) + if (vlan_get_tag(skb, &client_info->vlan_id)) client_info->vlan_id = 0; if (!client_info->assigned) { @@ -762,7 +755,7 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond) /* Don't modify or load balance ARPs that do not originate locally * (e.g.,arrive via a bridge). */ - if (!bond_slave_has_mac_rcu(bond, arp->mac_src)) + if (!bond_slave_has_mac_rx(bond, arp->mac_src)) return NULL; if (arp->op_code == htons(ARPOP_REPLY)) { @@ -770,9 +763,8 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond) * rx channel */ tx_slave = rlb_choose_channel(skb, bond); - if (tx_slave) { - memcpy(arp->mac_src,tx_slave->dev->dev_addr, ETH_ALEN); - } + if (tx_slave) + ether_addr_copy(arp->mac_src, tx_slave->dev->dev_addr); pr_debug("Server sent ARP Reply packet\n"); } else if (arp->op_code == htons(ARPOP_REQUEST)) { /* Create an entry in the rx_hashtbl for this client as a @@ -824,9 +816,8 @@ static void rlb_rebalance(struct bonding *bond) } /* update the team's flag only after the whole iteration */ - if (ntt) { + if (ntt) bond_info->rx_ntt = 1; - } _unlock_rx_hashtbl_bh(bond); } @@ -923,7 +914,7 @@ static void rlb_src_link(struct bonding *bond, u32 ip_src_hash, u32 ip_dst_hash) static void rlb_purge_src_ip(struct bonding *bond, struct arp_pkt *arp) { struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond)); - u32 ip_src_hash = _simple_hash((u8*)&(arp->ip_src), sizeof(arp->ip_src)); + u32 ip_src_hash = _simple_hash((u8 *)&(arp->ip_src), sizeof(arp->ip_src)); u32 index; _lock_rx_hashtbl_bh(bond); @@ -957,9 +948,8 @@ static int rlb_initialize(struct bonding *bond) bond_info->rx_hashtbl_used_head = RLB_NULL_INDEX; - for (i = 0; i < RLB_HASH_TABLE_SIZE; i++) { + for (i = 0; i < RLB_HASH_TABLE_SIZE; i++) rlb_init_table_entry(bond_info->rx_hashtbl + i); - } _unlock_rx_hashtbl_bh(bond); @@ -1006,7 +996,7 @@ static void rlb_clear_vlan(struct bonding *bond, unsigned short vlan_id) /*********************** tlb/rlb shared functions *********************/ static void alb_send_lp_vid(struct slave *slave, u8 mac_addr[], - u16 vid) + __be16 vlan_proto, u16 vid) { struct learning_pkt pkt; struct sk_buff *skb; @@ -1014,9 +1004,9 @@ static void alb_send_lp_vid(struct slave *slave, u8 mac_addr[], char *data; memset(&pkt, 0, size); - memcpy(pkt.mac_dst, mac_addr, ETH_ALEN); - memcpy(pkt.mac_src, mac_addr, ETH_ALEN); - pkt.type = cpu_to_be16(ETH_P_LOOP); + ether_addr_copy(pkt.mac_dst, mac_addr); + ether_addr_copy(pkt.mac_src, mac_addr); + pkt.type = cpu_to_be16(ETH_P_LOOPBACK); skb = dev_alloc_skb(size); if (!skb) @@ -1032,7 +1022,7 @@ static void alb_send_lp_vid(struct slave *slave, u8 mac_addr[], skb->dev = slave->dev; if (vid) { - skb = vlan_put_tag(skb, htons(ETH_P_8021Q), vid); + skb = vlan_put_tag(skb, vlan_proto, vid); if (!skb) { pr_err("%s: Error: failed to insert VLAN tag\n", slave->bond->dev->name); @@ -1043,22 +1033,45 @@ static void alb_send_lp_vid(struct slave *slave, u8 mac_addr[], dev_queue_xmit(skb); } - -static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[]) +static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[], + bool strict_match) { struct bonding *bond = bond_get_bond_by_slave(slave); struct net_device *upper; struct list_head *iter; + struct bond_vlan_tag tags[BOND_MAX_VLAN_ENCAP]; /* send untagged */ - alb_send_lp_vid(slave, mac_addr, 0); + alb_send_lp_vid(slave, mac_addr, 0, 0); - /* loop through vlans and send one packet for each */ + /* loop through all devices and see if we need to send a packet + * for that device. + */ rcu_read_lock(); netdev_for_each_all_upper_dev_rcu(bond->dev, upper, iter) { - if (upper->priv_flags & IFF_802_1Q_VLAN) - alb_send_lp_vid(slave, mac_addr, - vlan_dev_vlan_id(upper)); + if (is_vlan_dev(upper) && vlan_get_encap_level(upper) == 0) { + if (strict_match && + ether_addr_equal_64bits(mac_addr, + upper->dev_addr)) { + alb_send_lp_vid(slave, mac_addr, + vlan_dev_vlan_proto(upper), + vlan_dev_vlan_id(upper)); + } else if (!strict_match) { + alb_send_lp_vid(slave, upper->dev_addr, + vlan_dev_vlan_proto(upper), + vlan_dev_vlan_id(upper)); + } + } + + /* If this is a macvlan device, then only send updates + * when strict_match is turned off. + */ + if (netif_is_macvlan(upper) && !strict_match) { + memset(tags, 0, sizeof(tags)); + bond_verify_device_path(bond->dev, upper, tags); + alb_send_lp_vid(slave, upper->dev_addr, + tags[0].vlan_proto, tags[0].vlan_id); + } } rcu_read_unlock(); } @@ -1068,7 +1081,7 @@ static int alb_set_slave_mac_addr(struct slave *slave, u8 addr[]) struct net_device *dev = slave->dev; struct sockaddr s_addr; - if (slave->bond->params.mode == BOND_MODE_TLB) { + if (BOND_MODE(slave->bond) == BOND_MODE_TLB) { memcpy(dev->dev_addr, addr, dev->addr_len); return 0; } @@ -1097,7 +1110,7 @@ static void alb_swap_mac_addr(struct slave *slave1, struct slave *slave2) { u8 tmp_mac_addr[ETH_ALEN]; - memcpy(tmp_mac_addr, slave1->dev->dev_addr, ETH_ALEN); + ether_addr_copy(tmp_mac_addr, slave1->dev->dev_addr); alb_set_slave_mac_addr(slave1, slave2->dev->dev_addr); alb_set_slave_mac_addr(slave2, tmp_mac_addr); @@ -1111,14 +1124,14 @@ static void alb_swap_mac_addr(struct slave *slave1, struct slave *slave2) static void alb_fasten_mac_swap(struct bonding *bond, struct slave *slave1, struct slave *slave2) { - int slaves_state_differ = (SLAVE_IS_OK(slave1) != SLAVE_IS_OK(slave2)); + int slaves_state_differ = (bond_slave_can_tx(slave1) != bond_slave_can_tx(slave2)); struct slave *disabled_slave = NULL; ASSERT_RTNL(); /* fasten the change in the switch */ - if (SLAVE_IS_OK(slave1)) { - alb_send_learning_packets(slave1, slave1->dev->dev_addr); + if (bond_slave_can_tx(slave1)) { + alb_send_learning_packets(slave1, slave1->dev->dev_addr, false); if (bond->alb_info.rlb_enabled) { /* inform the clients that the mac address * has changed @@ -1129,8 +1142,8 @@ static void alb_fasten_mac_swap(struct bonding *bond, struct slave *slave1, disabled_slave = slave1; } - if (SLAVE_IS_OK(slave2)) { - alb_send_learning_packets(slave2, slave2->dev->dev_addr); + if (bond_slave_can_tx(slave2)) { + alb_send_learning_packets(slave2, slave2->dev->dev_addr, false); if (bond->alb_info.rlb_enabled) { /* inform the clients that the mac address * has changed @@ -1254,9 +1267,9 @@ static int alb_handle_addr_collision_on_attach(struct bonding *bond, struct slav if (free_mac_slave) { alb_set_slave_mac_addr(slave, free_mac_slave->perm_hwaddr); - pr_warning("%s: Warning: the hw address of slave %s is in use by the bond; giving it the hw address of %s\n", - bond->dev->name, slave->dev->name, - free_mac_slave->dev->name); + pr_warn("%s: Warning: the hw address of slave %s is in use by the bond; giving it the hw address of %s\n", + bond->dev->name, slave->dev->name, + free_mac_slave->dev->name); } else if (has_bond_addr) { pr_err("%s: Error: the hw address of slave %s is in use by the bond; couldn't find a slave with a free hw address to give it (this should not have happened)\n", @@ -1294,12 +1307,12 @@ static int alb_set_mac_address(struct bonding *bond, void *addr) bond_for_each_slave(bond, slave, iter) { /* save net_device's current hw address */ - memcpy(tmp_addr, slave->dev->dev_addr, ETH_ALEN); + ether_addr_copy(tmp_addr, slave->dev->dev_addr); res = dev_set_mac_address(slave->dev, addr); /* restore net_device's hw address */ - memcpy(slave->dev->dev_addr, tmp_addr, ETH_ALEN); + ether_addr_copy(slave->dev->dev_addr, tmp_addr); if (res) goto unwind; @@ -1315,9 +1328,9 @@ unwind: bond_for_each_slave(bond, rollback_slave, iter) { if (rollback_slave == slave) break; - memcpy(tmp_addr, rollback_slave->dev->dev_addr, ETH_ALEN); + ether_addr_copy(tmp_addr, rollback_slave->dev->dev_addr); dev_set_mac_address(rollback_slave->dev, &sa); - memcpy(rollback_slave->dev->dev_addr, tmp_addr, ETH_ALEN); + ether_addr_copy(rollback_slave->dev->dev_addr, tmp_addr); } return res; @@ -1330,9 +1343,8 @@ int bond_alb_initialize(struct bonding *bond, int rlb_enabled) int res; res = tlb_initialize(bond); - if (res) { + if (res) return res; - } if (rlb_enabled) { bond->alb_info.rlb_enabled = 1; @@ -1355,9 +1367,79 @@ void bond_alb_deinitialize(struct bonding *bond) tlb_deinitialize(bond); - if (bond_info->rlb_enabled) { + if (bond_info->rlb_enabled) rlb_deinitialize(bond); +} + +static int bond_do_alb_xmit(struct sk_buff *skb, struct bonding *bond, + struct slave *tx_slave) +{ + struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond)); + struct ethhdr *eth_data = eth_hdr(skb); + + if (!tx_slave) { + /* unbalanced or unassigned, send through primary */ + tx_slave = rcu_dereference(bond->curr_active_slave); + if (bond->params.tlb_dynamic_lb) + bond_info->unbalanced_load += skb->len; + } + + if (tx_slave && bond_slave_can_tx(tx_slave)) { + if (tx_slave != rcu_dereference(bond->curr_active_slave)) { + ether_addr_copy(eth_data->h_source, + tx_slave->dev->dev_addr); + } + + bond_dev_queue_xmit(bond, skb, tx_slave->dev); + goto out; + } + + if (tx_slave && bond->params.tlb_dynamic_lb) { + _lock_tx_hashtbl(bond); + __tlb_clear_slave(bond, tx_slave, 0); + _unlock_tx_hashtbl(bond); } + + /* no suitable interface, frame not sent */ + dev_kfree_skb_any(skb); +out: + return NETDEV_TX_OK; +} + +int bond_tlb_xmit(struct sk_buff *skb, struct net_device *bond_dev) +{ + struct bonding *bond = netdev_priv(bond_dev); + struct ethhdr *eth_data; + struct slave *tx_slave = NULL; + u32 hash_index; + + skb_reset_mac_header(skb); + eth_data = eth_hdr(skb); + + /* Do not TX balance any multicast or broadcast */ + if (!is_multicast_ether_addr(eth_data->h_dest)) { + switch (skb->protocol) { + case htons(ETH_P_IP): + case htons(ETH_P_IPX): + /* In case of IPX, it will falback to L2 hash */ + case htons(ETH_P_IPV6): + hash_index = bond_xmit_hash(bond, skb); + if (bond->params.tlb_dynamic_lb) { + tx_slave = tlb_choose_channel(bond, + hash_index & 0xFF, + skb->len); + } else { + struct list_head *iter; + int idx = hash_index % bond->slave_cnt; + + bond_for_each_slave_rcu(bond, tx_slave, iter) + if (--idx < 0) + break; + } + break; + } + } + return bond_do_alb_xmit(skb, bond, tx_slave); } int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev) @@ -1368,7 +1450,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev) struct slave *tx_slave = NULL; static const __be32 ip_bcast = htonl(0xffffffff); int hash_size = 0; - int do_tx_balance = 1; + bool do_tx_balance = true; u32 hash_index = 0; const u8 *hash_start = NULL; struct ipv6hdr *ip6hdr; @@ -1383,7 +1465,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev) if (ether_addr_equal_64bits(eth_data->h_dest, mac_bcast) || (iph->daddr == ip_bcast) || (iph->protocol == IPPROTO_IGMP)) { - do_tx_balance = 0; + do_tx_balance = false; break; } hash_start = (char *)&(iph->daddr); @@ -1395,7 +1477,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev) * that here just in case. */ if (ether_addr_equal_64bits(eth_data->h_dest, mac_bcast)) { - do_tx_balance = 0; + do_tx_balance = false; break; } @@ -1403,7 +1485,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev) * broadcasts in IPv4. */ if (ether_addr_equal_64bits(eth_data->h_dest, mac_v6_allmcast)) { - do_tx_balance = 0; + do_tx_balance = false; break; } @@ -1413,7 +1495,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev) */ ip6hdr = ipv6_hdr(skb); if (ipv6_addr_any(&ip6hdr->saddr)) { - do_tx_balance = 0; + do_tx_balance = false; break; } @@ -1423,7 +1505,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev) case ETH_P_IPX: if (ipx_hdr(skb)->ipx_checksum != IPX_NO_CHECKSUM) { /* something is wrong with this packet */ - do_tx_balance = 0; + do_tx_balance = false; break; } @@ -1432,21 +1514,20 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev) * this family since it has an "ARP" like * mechanism */ - do_tx_balance = 0; + do_tx_balance = false; break; } - hash_start = (char*)eth_data->h_dest; + hash_start = (char *)eth_data->h_dest; hash_size = ETH_ALEN; break; case ETH_P_ARP: - do_tx_balance = 0; - if (bond_info->rlb_enabled) { + do_tx_balance = false; + if (bond_info->rlb_enabled) tx_slave = rlb_arp_xmit(skb, bond); - } break; default: - do_tx_balance = 0; + do_tx_balance = false; break; } @@ -1455,33 +1536,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev) tx_slave = tlb_choose_channel(bond, hash_index, skb->len); } - if (!tx_slave) { - /* unbalanced or unassigned, send through primary */ - tx_slave = rcu_dereference(bond->curr_active_slave); - bond_info->unbalanced_load += skb->len; - } - - if (tx_slave && SLAVE_IS_OK(tx_slave)) { - if (tx_slave != rcu_dereference(bond->curr_active_slave)) { - memcpy(eth_data->h_source, - tx_slave->dev->dev_addr, - ETH_ALEN); - } - - bond_dev_queue_xmit(bond, skb, tx_slave->dev); - goto out; - } else { - if (tx_slave) { - _lock_tx_hashtbl(bond); - __tlb_clear_slave(bond, tx_slave, 0); - _unlock_tx_hashtbl(bond); - } - } - - /* no suitable interface, frame not sent */ - kfree_skb(skb); -out: - return NETDEV_TX_OK; + return bond_do_alb_xmit(skb, bond, tx_slave); } void bond_alb_monitor(struct work_struct *work) @@ -1505,6 +1560,8 @@ void bond_alb_monitor(struct work_struct *work) /* send learning packets */ if (bond_info->lp_counter >= BOND_ALB_LP_TICKS(bond)) { + bool strict_match; + /* change of curr_active_slave involves swapping of mac addresses. * in order to avoid this swapping from happening while * sending the learning packets, the curr_slave_lock must be held for @@ -1512,8 +1569,17 @@ void bond_alb_monitor(struct work_struct *work) */ read_lock(&bond->curr_slave_lock); - bond_for_each_slave_rcu(bond, slave, iter) - alb_send_learning_packets(slave, slave->dev->dev_addr); + bond_for_each_slave_rcu(bond, slave, iter) { + /* If updating current_active, use all currently + * user mac addreses (!strict_match). Otherwise, only + * use mac of the slave device. + * In RLB mode, we always use strict matches. + */ + strict_match = (slave != bond->curr_active_slave || + bond_info->rlb_enabled); + alb_send_learning_packets(slave, slave->dev->dev_addr, + strict_match); + } read_unlock(&bond->curr_slave_lock); @@ -1577,11 +1643,10 @@ void bond_alb_monitor(struct work_struct *work) --bond_info->rlb_update_delay_counter; } else { rlb_update_rx_clients(bond); - if (bond_info->rlb_update_retry_counter) { + if (bond_info->rlb_update_retry_counter) --bond_info->rlb_update_retry_counter; - } else { + else bond_info->rx_ntt = 0; - } } } } @@ -1598,23 +1663,20 @@ int bond_alb_init_slave(struct bonding *bond, struct slave *slave) int res; res = alb_set_slave_mac_addr(slave, slave->perm_hwaddr); - if (res) { + if (res) return res; - } res = alb_handle_addr_collision_on_attach(bond, slave); - if (res) { + if (res) return res; - } tlb_init_slave(slave); /* order a rebalance ASAP */ bond->alb_info.tx_rebalance_counter = BOND_TLB_REBALANCE_TICKS; - if (bond->alb_info.rlb_enabled) { + if (bond->alb_info.rlb_enabled) bond->alb_info.rlb_rebalance = 1; - } return 0; } @@ -1645,9 +1707,8 @@ void bond_alb_handle_link_change(struct bonding *bond, struct slave *slave, char if (link == BOND_LINK_DOWN) { tlb_clear_slave(bond, slave, 0); - if (bond->alb_info.rlb_enabled) { + if (bond->alb_info.rlb_enabled) rlb_clear_slave(bond, slave); - } } else if (link == BOND_LINK_UP) { /* order a rebalance ASAP */ bond_info->tx_rebalance_counter = BOND_TLB_REBALANCE_TICKS; @@ -1719,18 +1780,18 @@ void bond_alb_handle_active_change(struct bonding *bond, struct slave *new_slave /* in TLB mode, the slave might flip down/up with the old dev_addr, * and thus filter bond->dev_addr's packets, so force bond's mac */ - if (bond->params.mode == BOND_MODE_TLB) { + if (BOND_MODE(bond) == BOND_MODE_TLB) { struct sockaddr sa; u8 tmp_addr[ETH_ALEN]; - memcpy(tmp_addr, new_slave->dev->dev_addr, ETH_ALEN); + ether_addr_copy(tmp_addr, new_slave->dev->dev_addr); memcpy(sa.sa_data, bond->dev->dev_addr, bond->dev->addr_len); sa.sa_family = bond->dev->type; /* we don't care if it can't change its mac, best effort */ dev_set_mac_address(new_slave->dev, &sa); - memcpy(new_slave->dev->dev_addr, tmp_addr, ETH_ALEN); + ether_addr_copy(new_slave->dev->dev_addr, tmp_addr); } /* curr_active_slave must be set before calling alb_swap_mac_addr */ @@ -1741,7 +1802,8 @@ void bond_alb_handle_active_change(struct bonding *bond, struct slave *new_slave } else { /* set the new_slave to the bond mac address */ alb_set_slave_mac_addr(new_slave, bond->dev->dev_addr); - alb_send_learning_packets(new_slave, bond->dev->dev_addr); + alb_send_learning_packets(new_slave, bond->dev->dev_addr, + false); } write_lock_bh(&bond->curr_slave_lock); @@ -1759,14 +1821,12 @@ int bond_alb_set_mac_address(struct net_device *bond_dev, void *addr) struct slave *swap_slave; int res; - if (!is_valid_ether_addr(sa->sa_data)) { + if (!is_valid_ether_addr(sa->sa_data)) return -EADDRNOTAVAIL; - } res = alb_set_mac_address(bond, addr); - if (res) { + if (res) return res; - } memcpy(bond_dev->dev_addr, sa->sa_data, bond_dev->addr_len); @@ -1774,9 +1834,8 @@ int bond_alb_set_mac_address(struct net_device *bond_dev, void *addr) * Otherwise we'll need to pass the new address to it and handle * duplications. */ - if (!bond->curr_active_slave) { + if (!bond->curr_active_slave) return 0; - } swap_slave = bond_slave_has_mac(bond, bond_dev->dev_addr); @@ -1787,7 +1846,8 @@ int bond_alb_set_mac_address(struct net_device *bond_dev, void *addr) alb_set_slave_mac_addr(bond->curr_active_slave, bond_dev->dev_addr); read_lock(&bond->lock); - alb_send_learning_packets(bond->curr_active_slave, bond_dev->dev_addr); + alb_send_learning_packets(bond->curr_active_slave, + bond_dev->dev_addr, false); if (bond->alb_info.rlb_enabled) { /* inform clients mac address has changed */ rlb_req_update_slave_clients(bond, bond->curr_active_slave); @@ -1800,8 +1860,7 @@ int bond_alb_set_mac_address(struct net_device *bond_dev, void *addr) void bond_alb_clear_vlan(struct bonding *bond, unsigned short vlan_id) { - if (bond->alb_info.rlb_enabled) { + if (bond->alb_info.rlb_enabled) rlb_clear_vlan(bond, vlan_id); - } } diff --git a/drivers/net/bonding/bond_alb.h b/drivers/net/bonding/bond_alb.h index e09dd4bfaff..5fc76c01636 100644 --- a/drivers/net/bonding/bond_alb.h +++ b/drivers/net/bonding/bond_alb.h @@ -175,6 +175,7 @@ void bond_alb_deinit_slave(struct bonding *bond, struct slave *slave); void bond_alb_handle_link_change(struct bonding *bond, struct slave *slave, char link); void bond_alb_handle_active_change(struct bonding *bond, struct slave *new_slave); int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev); +int bond_tlb_xmit(struct sk_buff *skb, struct net_device *bond_dev); void bond_alb_monitor(struct work_struct *); int bond_alb_set_mac_address(struct net_device *bond_dev, void *addr); void bond_alb_clear_vlan(struct bonding *bond, unsigned short vlan_id); diff --git a/drivers/net/bonding/bond_debugfs.c b/drivers/net/bonding/bond_debugfs.c index 5fc4c235147..658e761c456 100644 --- a/drivers/net/bonding/bond_debugfs.c +++ b/drivers/net/bonding/bond_debugfs.c @@ -23,7 +23,7 @@ static int bond_debug_rlb_hash_show(struct seq_file *m, void *v) struct rlb_client_info *client_info; u32 hash_index; - if (bond->params.mode != BOND_MODE_ALB) + if (BOND_MODE(bond) != BOND_MODE_ALB) return 0; seq_printf(m, "SourceIP DestinationIP " @@ -69,7 +69,7 @@ void bond_debug_register(struct bonding *bond) debugfs_create_dir(bond->dev->name, bonding_debug_root); if (!bond->debug_dir) { - pr_warning("%s: Warning: failed to register to debugfs\n", + pr_warn("%s: Warning: failed to register to debugfs\n", bond->dev->name); return; } @@ -98,9 +98,8 @@ void bond_debug_reregister(struct bonding *bond) if (d) { bond->debug_dir = d; } else { - pr_warning("%s: Warning: failed to reregister, " - "so just unregister old one\n", - bond->dev->name); + pr_warn("%s: Warning: failed to reregister, so just unregister old one\n", + bond->dev->name); bond_debug_unregister(bond); } } @@ -110,8 +109,7 @@ void bond_create_debugfs(void) bonding_debug_root = debugfs_create_dir("bonding", NULL); if (!bonding_debug_root) { - pr_warning("Warning: Cannot create bonding directory" - " in debugfs\n"); + pr_warn("Warning: Cannot create bonding directory in debugfs\n"); } } diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 3220b488dd1..701f86cd599 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -79,7 +79,6 @@ #include <net/pkt_sched.h> #include <linux/rculist.h> #include <net/flow_keys.h> -#include <linux/reciprocal_div.h> #include "bonding.h" #include "bond_3ad.h" #include "bond_alb.h" @@ -87,13 +86,11 @@ /*---------------------------- Module parameters ----------------------------*/ /* monitor all links that often (in milliseconds). <=0 disables monitoring */ -#define BOND_LINK_MON_INTERV 0 -#define BOND_LINK_ARP_INTERV 0 static int max_bonds = BOND_DEFAULT_MAX_BONDS; static int tx_queues = BOND_DEFAULT_TX_QUEUES; static int num_peer_notif = 1; -static int miimon = BOND_LINK_MON_INTERV; +static int miimon; static int updelay; static int downdelay; static int use_carrier = 1; @@ -104,7 +101,7 @@ static char *lacp_rate; static int min_links; static char *ad_select; static char *xmit_hash_policy; -static int arp_interval = BOND_LINK_ARP_INTERV; +static int arp_interval; static char *arp_ip_target[BOND_MAX_ARP_TARGETS]; static char *arp_validate; static char *arp_all_targets; @@ -209,67 +206,6 @@ static int bond_mode = BOND_MODE_ROUNDROBIN; static int xmit_hashtype = BOND_XMIT_POLICY_LAYER2; static int lacp_fast; -const struct bond_parm_tbl bond_lacp_tbl[] = { -{ "slow", AD_LACP_SLOW}, -{ "fast", AD_LACP_FAST}, -{ NULL, -1}, -}; - -const struct bond_parm_tbl bond_mode_tbl[] = { -{ "balance-rr", BOND_MODE_ROUNDROBIN}, -{ "active-backup", BOND_MODE_ACTIVEBACKUP}, -{ "balance-xor", BOND_MODE_XOR}, -{ "broadcast", BOND_MODE_BROADCAST}, -{ "802.3ad", BOND_MODE_8023AD}, -{ "balance-tlb", BOND_MODE_TLB}, -{ "balance-alb", BOND_MODE_ALB}, -{ NULL, -1}, -}; - -const struct bond_parm_tbl xmit_hashtype_tbl[] = { -{ "layer2", BOND_XMIT_POLICY_LAYER2}, -{ "layer3+4", BOND_XMIT_POLICY_LAYER34}, -{ "layer2+3", BOND_XMIT_POLICY_LAYER23}, -{ "encap2+3", BOND_XMIT_POLICY_ENCAP23}, -{ "encap3+4", BOND_XMIT_POLICY_ENCAP34}, -{ NULL, -1}, -}; - -const struct bond_parm_tbl arp_all_targets_tbl[] = { -{ "any", BOND_ARP_TARGETS_ANY}, -{ "all", BOND_ARP_TARGETS_ALL}, -{ NULL, -1}, -}; - -const struct bond_parm_tbl arp_validate_tbl[] = { -{ "none", BOND_ARP_VALIDATE_NONE}, -{ "active", BOND_ARP_VALIDATE_ACTIVE}, -{ "backup", BOND_ARP_VALIDATE_BACKUP}, -{ "all", BOND_ARP_VALIDATE_ALL}, -{ NULL, -1}, -}; - -const struct bond_parm_tbl fail_over_mac_tbl[] = { -{ "none", BOND_FOM_NONE}, -{ "active", BOND_FOM_ACTIVE}, -{ "follow", BOND_FOM_FOLLOW}, -{ NULL, -1}, -}; - -const struct bond_parm_tbl pri_reselect_tbl[] = { -{ "always", BOND_PRI_RESELECT_ALWAYS}, -{ "better", BOND_PRI_RESELECT_BETTER}, -{ "failure", BOND_PRI_RESELECT_FAILURE}, -{ NULL, -1}, -}; - -struct bond_parm_tbl ad_select_tbl[] = { -{ "stable", BOND_AD_STABLE}, -{ "bandwidth", BOND_AD_BANDWIDTH}, -{ "count", BOND_AD_COUNT}, -{ NULL, -1}, -}; - /*-------------------------- Forward declarations ---------------------------*/ static int bond_init(struct net_device *bond_dev); @@ -407,7 +343,7 @@ static int bond_set_carrier(struct bonding *bond) if (!bond_has_slaves(bond)) goto down; - if (bond->params.mode == BOND_MODE_8023AD) + if (BOND_MODE(bond) == BOND_MODE_8023AD) return bond_3ad_set_carrier(bond); bond_for_each_slave(bond, slave, iter) { @@ -561,7 +497,7 @@ static int bond_set_promiscuity(struct bonding *bond, int inc) struct list_head *iter; int err = 0; - if (USES_PRIMARY(bond->params.mode)) { + if (bond_uses_primary(bond)) { /* write lock already acquired */ if (bond->curr_active_slave) { err = dev_set_promiscuity(bond->curr_active_slave->dev, @@ -587,7 +523,7 @@ static int bond_set_allmulti(struct bonding *bond, int inc) struct list_head *iter; int err = 0; - if (USES_PRIMARY(bond->params.mode)) { + if (bond_uses_primary(bond)) { /* write lock already acquired */ if (bond->curr_active_slave) { err = dev_set_allmulti(bond->curr_active_slave->dev, @@ -638,7 +574,7 @@ static void bond_hw_addr_flush(struct net_device *bond_dev, dev_uc_unsync(slave_dev, bond_dev); dev_mc_unsync(slave_dev, bond_dev); - if (bond->params.mode == BOND_MODE_8023AD) { + if (BOND_MODE(bond) == BOND_MODE_8023AD) { /* del lacpdu mc addr from mc list */ u8 lacpdu_multicast[ETH_ALEN] = MULTICAST_LACPDU_ADDR; @@ -649,8 +585,8 @@ static void bond_hw_addr_flush(struct net_device *bond_dev, /*--------------------------- Active slave change ---------------------------*/ /* Update the hardware address list and promisc/allmulti for the new and - * old active slaves (if any). Modes that are !USES_PRIMARY keep all - * slaves up date at all times; only the USES_PRIMARY modes need to call + * old active slaves (if any). Modes that are not using primary keep all + * slaves up date at all times; only the modes that use primary need to call * this function to swap these settings during a failover. */ static void bond_hw_addr_swap(struct bonding *bond, struct slave *new_active, @@ -737,12 +673,12 @@ static void bond_do_fail_over_mac(struct bonding *bond, write_unlock_bh(&bond->curr_slave_lock); if (old_active) { - memcpy(tmp_mac, new_active->dev->dev_addr, ETH_ALEN); - memcpy(saddr.sa_data, old_active->dev->dev_addr, - ETH_ALEN); + ether_addr_copy(tmp_mac, new_active->dev->dev_addr); + ether_addr_copy(saddr.sa_data, + old_active->dev->dev_addr); saddr.sa_family = new_active->dev->type; } else { - memcpy(saddr.sa_data, bond->dev->dev_addr, ETH_ALEN); + ether_addr_copy(saddr.sa_data, bond->dev->dev_addr); saddr.sa_family = bond->dev->type; } @@ -756,7 +692,7 @@ static void bond_do_fail_over_mac(struct bonding *bond, if (!old_active) goto out; - memcpy(saddr.sa_data, tmp_mac, ETH_ALEN); + ether_addr_copy(saddr.sa_data, tmp_mac); saddr.sa_family = old_active->dev->type; rv = dev_set_mac_address(old_active->dev, &saddr); @@ -811,7 +747,7 @@ static struct slave *bond_find_best_slave(struct bonding *bond) bond_for_each_slave(bond, slave, iter) { if (slave->link == BOND_LINK_UP) return slave; - if (slave->link == BOND_LINK_BACK && IS_UP(slave->dev) && + if (slave->link == BOND_LINK_BACK && bond_slave_is_up(slave) && slave->delay < mintime) { mintime = slave->delay; bestslave = slave; @@ -862,11 +798,11 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active) return; if (new_active) { - new_active->jiffies = jiffies; + new_active->last_link_up = jiffies; if (new_active->link == BOND_LINK_BACK) { - if (USES_PRIMARY(bond->params.mode)) { - pr_info("%s: making interface %s the new active one %d ms earlier.\n", + if (bond_uses_primary(bond)) { + pr_info("%s: making interface %s the new active one %d ms earlier\n", bond->dev->name, new_active->dev->name, (bond->params.updelay - new_active->delay) * bond->params.miimon); } @@ -874,40 +810,44 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active) new_active->delay = 0; new_active->link = BOND_LINK_UP; - if (bond->params.mode == BOND_MODE_8023AD) + if (BOND_MODE(bond) == BOND_MODE_8023AD) bond_3ad_handle_link_change(new_active, BOND_LINK_UP); if (bond_is_lb(bond)) bond_alb_handle_link_change(bond, new_active, BOND_LINK_UP); } else { - if (USES_PRIMARY(bond->params.mode)) { - pr_info("%s: making interface %s the new active one.\n", + if (bond_uses_primary(bond)) { + pr_info("%s: making interface %s the new active one\n", bond->dev->name, new_active->dev->name); } } } - if (USES_PRIMARY(bond->params.mode)) + if (bond_uses_primary(bond)) bond_hw_addr_swap(bond, new_active, old_active); if (bond_is_lb(bond)) { bond_alb_handle_active_change(bond, new_active); if (old_active) - bond_set_slave_inactive_flags(old_active); + bond_set_slave_inactive_flags(old_active, + BOND_SLAVE_NOTIFY_NOW); if (new_active) - bond_set_slave_active_flags(new_active); + bond_set_slave_active_flags(new_active, + BOND_SLAVE_NOTIFY_NOW); } else { rcu_assign_pointer(bond->curr_active_slave, new_active); } - if (bond->params.mode == BOND_MODE_ACTIVEBACKUP) { + if (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP) { if (old_active) - bond_set_slave_inactive_flags(old_active); + bond_set_slave_inactive_flags(old_active, + BOND_SLAVE_NOTIFY_NOW); if (new_active) { bool should_notify_peers = false; - bond_set_slave_active_flags(new_active); + bond_set_slave_active_flags(new_active, + BOND_SLAVE_NOTIFY_NOW); if (bond->params.fail_over_mac) bond_do_fail_over_mac(bond, new_active, @@ -936,8 +876,8 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active) * resend only if bond is brought up with the affected * bonding modes and the retransmission is enabled */ if (netif_running(bond->dev) && (bond->params.resend_igmp > 0) && - ((USES_PRIMARY(bond->params.mode) && new_active) || - bond->params.mode == BOND_MODE_ROUNDROBIN)) { + ((bond_uses_primary(bond) && new_active) || + BOND_MODE(bond) == BOND_MODE_ROUNDROBIN)) { bond->igmp_retrans = bond->params.resend_igmp; queue_delayed_work(bond->wq, &bond->mcast_work, 1); } @@ -970,7 +910,7 @@ void bond_select_active_slave(struct bonding *bond) pr_info("%s: first active interface up!\n", bond->dev->name); } else { - pr_info("%s: now running without any active interface !\n", + pr_info("%s: now running without any active interface!\n", bond->dev->name); } } @@ -982,12 +922,12 @@ static inline int slave_enable_netpoll(struct slave *slave) struct netpoll *np; int err = 0; - np = kzalloc(sizeof(*np), GFP_ATOMIC); + np = kzalloc(sizeof(*np), GFP_KERNEL); err = -ENOMEM; if (!np) goto out; - err = __netpoll_setup(np, slave->dev, GFP_ATOMIC); + err = __netpoll_setup(np, slave->dev); if (err) { kfree(np); goto out; @@ -1006,14 +946,6 @@ static inline void slave_disable_netpoll(struct slave *slave) slave->np = NULL; __netpoll_free_async(np); } -static inline bool slave_dev_support_netpoll(struct net_device *slave_dev) -{ - if (slave_dev->priv_flags & IFF_DISABLE_NETPOLL) - return false; - if (!slave_dev->netdev_ops->ndo_poll_controller) - return false; - return true; -} static void bond_poll_controller(struct net_device *bond_dev) { @@ -1026,11 +958,11 @@ static void bond_netpoll_cleanup(struct net_device *bond_dev) struct slave *slave; bond_for_each_slave(bond, slave, iter) - if (IS_UP(slave->dev)) + if (bond_slave_is_up(slave)) slave_disable_netpoll(slave); } -static int bond_netpoll_setup(struct net_device *dev, struct netpoll_info *ni, gfp_t gfp) +static int bond_netpoll_setup(struct net_device *dev, struct netpoll_info *ni) { struct bonding *bond = netdev_priv(dev); struct list_head *iter; @@ -1093,10 +1025,14 @@ static netdev_features_t bond_fix_features(struct net_device *dev, NETIF_F_FRAGLIST | NETIF_F_ALL_TSO | \ NETIF_F_HIGHDMA | NETIF_F_LRO) +#define BOND_ENC_FEATURES (NETIF_F_ALL_CSUM | NETIF_F_SG | NETIF_F_RXCSUM |\ + NETIF_F_TSO | NETIF_F_GSO_UDP_TUNNEL) + static void bond_compute_features(struct bonding *bond) { unsigned int flags, dst_release_flag = IFF_XMIT_DST_RELEASE; netdev_features_t vlan_features = BOND_VLAN_FEATURES; + netdev_features_t enc_features = BOND_ENC_FEATURES; struct net_device *bond_dev = bond->dev; struct list_head *iter; struct slave *slave; @@ -1106,11 +1042,15 @@ static void bond_compute_features(struct bonding *bond) if (!bond_has_slaves(bond)) goto done; + vlan_features &= NETIF_F_ALL_FOR_ALL; bond_for_each_slave(bond, slave, iter) { vlan_features = netdev_increment_features(vlan_features, slave->dev->vlan_features, BOND_VLAN_FEATURES); + enc_features = netdev_increment_features(enc_features, + slave->dev->hw_enc_features, + BOND_ENC_FEATURES); dst_release_flag &= slave->dev->priv_flags; if (slave->dev->hard_header_len > max_hard_header_len) max_hard_header_len = slave->dev->hard_header_len; @@ -1121,6 +1061,7 @@ static void bond_compute_features(struct bonding *bond) done: bond_dev->vlan_features = vlan_features; + bond_dev->hw_enc_features = enc_features; bond_dev->hard_header_len = max_hard_header_len; bond_dev->gso_max_segs = gso_max_segs; netif_set_gso_max_size(bond_dev, gso_max_size); @@ -1152,7 +1093,7 @@ static bool bond_should_deliver_exact_match(struct sk_buff *skb, struct bonding *bond) { if (bond_is_slave_inactive(slave)) { - if (bond->params.mode == BOND_MODE_ALB && + if (BOND_MODE(bond) == BOND_MODE_ALB && skb->pkt_type != PACKET_BROADCAST && skb->pkt_type != PACKET_MULTICAST) return false; @@ -1179,9 +1120,6 @@ static rx_handler_result_t bond_handle_frame(struct sk_buff **pskb) slave = bond_slave_get_rcu(skb->dev); bond = slave->bond; - if (bond->params.arp_interval) - slave->dev->last_rx = jiffies; - recv_probe = ACCESS_ONCE(bond->recv_probe); if (recv_probe) { ret = recv_probe(skb, bond, slave); @@ -1197,7 +1135,7 @@ static rx_handler_result_t bond_handle_frame(struct sk_buff **pskb) skb->dev = bond->dev; - if (bond->params.mode == BOND_MODE_ALB && + if (BOND_MODE(bond) == BOND_MODE_ALB && bond->dev->priv_flags & IFF_BRIDGE_PORT && skb->pkt_type == PACKET_HOST) { @@ -1206,7 +1144,7 @@ static rx_handler_result_t bond_handle_frame(struct sk_buff **pskb) kfree_skb(skb); return RX_HANDLER_CONSUMED; } - memcpy(eth_hdr(skb)->h_dest, bond->dev->dev_addr, ETH_ALEN); + ether_addr_copy(eth_hdr(skb)->h_dest, bond->dev->dev_addr); } return ret; @@ -1234,6 +1172,35 @@ static void bond_upper_dev_unlink(struct net_device *bond_dev, rtmsg_ifinfo(RTM_NEWLINK, slave_dev, IFF_SLAVE, GFP_KERNEL); } +static struct slave *bond_alloc_slave(struct bonding *bond) +{ + struct slave *slave = NULL; + + slave = kzalloc(sizeof(struct slave), GFP_KERNEL); + if (!slave) + return NULL; + + if (BOND_MODE(bond) == BOND_MODE_8023AD) { + SLAVE_AD_INFO(slave) = kzalloc(sizeof(struct ad_slave_info), + GFP_KERNEL); + if (!SLAVE_AD_INFO(slave)) { + kfree(slave); + return NULL; + } + } + return slave; +} + +static void bond_free_slave(struct slave *slave) +{ + struct bonding *bond = bond_get_bond_by_slave(slave); + + if (BOND_MODE(bond) == BOND_MODE_8023AD) + kfree(SLAVE_AD_INFO(slave)); + + kfree(slave); +} + /* enslave device <slave> to bond device <master> */ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) { @@ -1247,16 +1214,21 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) if (!bond->params.use_carrier && slave_dev->ethtool_ops->get_link == NULL && slave_ops->ndo_do_ioctl == NULL) { - pr_warning("%s: Warning: no link monitoring support for %s\n", - bond_dev->name, slave_dev->name); + pr_warn("%s: Warning: no link monitoring support for %s\n", + bond_dev->name, slave_dev->name); } /* already enslaved */ if (slave_dev->flags & IFF_SLAVE) { - pr_debug("Error, Device was already enslaved\n"); + pr_debug("Error: Device was already enslaved\n"); return -EBUSY; } + if (bond_dev == slave_dev) { + pr_err("%s: cannot enslave bond to itself.\n", bond_dev->name); + return -EPERM; + } + /* vlan challenged mutual exclusion */ /* no need to lock since we're protected by rtnl_lock */ if (slave_dev->features & NETIF_F_VLAN_CHALLENGED) { @@ -1266,9 +1238,9 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) bond_dev->name, slave_dev->name, bond_dev->name); return -EPERM; } else { - pr_warning("%s: Warning: enslaved VLAN challenged slave %s. Adding VLANs will be blocked as long as %s is part of bond %s\n", - bond_dev->name, slave_dev->name, - slave_dev->name, bond_dev->name); + pr_warn("%s: Warning: enslaved VLAN challenged slave %s. Adding VLANs will be blocked as long as %s is part of bond %s\n", + bond_dev->name, slave_dev->name, + slave_dev->name, bond_dev->name); } } else { pr_debug("%s: ! NETIF_F_VLAN_CHALLENGED\n", slave_dev->name); @@ -1281,7 +1253,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) * enslaving it; the old ifenslave will not. */ if ((slave_dev->flags & IFF_UP)) { - pr_err("%s is up. This may be due to an out of date ifenslave.\n", + pr_err("%s is up - this may be due to an out of date ifenslave\n", slave_dev->name); res = -EPERM; goto err_undo_flags; @@ -1325,20 +1297,23 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) bond_dev); } } else if (bond_dev->type != slave_dev->type) { - pr_err("%s ether type (%d) is different from other slaves (%d), can not enslave it.\n", - slave_dev->name, - slave_dev->type, bond_dev->type); + pr_err("%s ether type (%d) is different from other slaves (%d), can not enslave it\n", + slave_dev->name, slave_dev->type, bond_dev->type); res = -EINVAL; goto err_undo_flags; } if (slave_ops->ndo_set_mac_address == NULL) { if (!bond_has_slaves(bond)) { - pr_warning("%s: Warning: The first slave device specified does not support setting the MAC address. Setting fail_over_mac to active.", - bond_dev->name); - bond->params.fail_over_mac = BOND_FOM_ACTIVE; + pr_warn("%s: Warning: The first slave device specified does not support setting the MAC address\n", + bond_dev->name); + if (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP) { + bond->params.fail_over_mac = BOND_FOM_ACTIVE; + pr_warn("%s: Setting fail_over_mac to active for active-backup mode\n", + bond_dev->name); + } } else if (bond->params.fail_over_mac != BOND_FOM_ACTIVE) { - pr_err("%s: Error: The slave device specified does not support setting the MAC address, but fail_over_mac is not set to active.\n", + pr_err("%s: Error: The slave device specified does not support setting the MAC address, but fail_over_mac is not set to active\n", bond_dev->name); res = -EOPNOTSUPP; goto err_undo_flags; @@ -1353,11 +1328,14 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) bond->dev->addr_assign_type == NET_ADDR_RANDOM) bond_set_dev_addr(bond->dev, slave_dev); - new_slave = kzalloc(sizeof(struct slave), GFP_KERNEL); + new_slave = bond_alloc_slave(bond); if (!new_slave) { res = -ENOMEM; goto err_undo_flags; } + + new_slave->bond = bond; + new_slave->dev = slave_dev; /* * Set the new_slave's queue_id to be zero. Queue ID mapping * is set via sysfs or module option if desired. @@ -1377,9 +1355,10 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) * that need it, and for restoring it upon release, and then * set it to the master's address */ - memcpy(new_slave->perm_hwaddr, slave_dev->dev_addr, ETH_ALEN); + ether_addr_copy(new_slave->perm_hwaddr, slave_dev->dev_addr); - if (!bond->params.fail_over_mac) { + if (!bond->params.fail_over_mac || + BOND_MODE(bond) != BOND_MODE_ACTIVEBACKUP) { /* * Set slave to master's mac address. The application already * set the master's mac address to that of the first slave @@ -1400,8 +1379,6 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) goto err_restore_mac; } - new_slave->bond = bond; - new_slave->dev = slave_dev; slave_dev->priv_flags |= IFF_BONDING; if (bond_is_lb(bond)) { @@ -1413,10 +1390,10 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) goto err_close; } - /* If the mode USES_PRIMARY, then the following is handled by + /* If the mode uses primary, then the following is handled by * bond_change_active_slave(). */ - if (!USES_PRIMARY(bond->params.mode)) { + if (!bond_uses_primary(bond)) { /* set promiscuity level to new slave */ if (bond_dev->flags & IFF_PROMISC) { res = dev_set_promiscuity(slave_dev, 1); @@ -1439,7 +1416,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) netif_addr_unlock_bh(bond_dev); } - if (bond->params.mode == BOND_MODE_8023AD) { + if (BOND_MODE(bond) == BOND_MODE_8023AD) { /* add lacpdu mc addr to mc list */ u8 lacpdu_multicast[ETH_ALEN] = MULTICAST_LACPDU_ADDR; @@ -1460,10 +1437,10 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) bond_update_speed_duplex(new_slave); - new_slave->last_arp_rx = jiffies - + new_slave->last_rx = jiffies - (msecs_to_jiffies(bond->params.arp_interval) + 1); for (i = 0; i < BOND_MAX_ARP_TARGETS; i++) - new_slave->target_last_arp_rx[i] = new_slave->last_arp_rx; + new_slave->target_last_arp_rx[i] = new_slave->last_rx; if (bond->params.miimon && !bond->params.use_carrier) { link_reporting = bond_check_dev_link(bond, slave_dev, 1); @@ -1478,12 +1455,12 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) * supported); thus, we don't need to change * the messages for netif_carrier. */ - pr_warning("%s: Warning: MII and ETHTOOL support not available for interface %s, and arp_interval/arp_ip_target module parameters not specified, thus bonding will not detect link failures! see bonding.txt for details.\n", - bond_dev->name, slave_dev->name); + pr_warn("%s: Warning: MII and ETHTOOL support not available for interface %s, and arp_interval/arp_ip_target module parameters not specified, thus bonding will not detect link failures! see bonding.txt for details\n", + bond_dev->name, slave_dev->name); } else if (link_reporting == -1) { /* unable get link status using mii/ethtool */ - pr_warning("%s: Warning: can't get link status from interface %s; the network driver associated with this interface does not support MII or ETHTOOL link status reporting, thus miimon has no effect on this interface.\n", - bond_dev->name, slave_dev->name); + pr_warn("%s: Warning: can't get link status from interface %s; the network driver associated with this interface does not support MII or ETHTOOL link status reporting, thus miimon has no effect on this interface\n", + bond_dev->name, slave_dev->name); } } @@ -1507,12 +1484,12 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) } if (new_slave->link != BOND_LINK_DOWN) - new_slave->jiffies = jiffies; + new_slave->last_link_up = jiffies; pr_debug("Initial state of slave_dev is BOND_LINK_%s\n", - new_slave->link == BOND_LINK_DOWN ? "DOWN" : - (new_slave->link == BOND_LINK_UP ? "UP" : "BACK")); + new_slave->link == BOND_LINK_DOWN ? "DOWN" : + (new_slave->link == BOND_LINK_UP ? "UP" : "BACK")); - if (USES_PRIMARY(bond->params.mode) && bond->params.primary[0]) { + if (bond_uses_primary(bond) && bond->params.primary[0]) { /* if there is a primary slave, remember it */ if (strcmp(bond->params.primary, new_slave->dev->name) == 0) { bond->primary_slave = new_slave; @@ -1520,26 +1497,27 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) } } - switch (bond->params.mode) { + switch (BOND_MODE(bond)) { case BOND_MODE_ACTIVEBACKUP: - bond_set_slave_inactive_flags(new_slave); + bond_set_slave_inactive_flags(new_slave, + BOND_SLAVE_NOTIFY_NOW); break; case BOND_MODE_8023AD: /* in 802.3ad mode, the internal mechanism * will activate the slaves in the selected * aggregator */ - bond_set_slave_inactive_flags(new_slave); + bond_set_slave_inactive_flags(new_slave, BOND_SLAVE_NOTIFY_NOW); /* if this is the first slave */ if (!prev_slave) { - SLAVE_AD_INFO(new_slave).id = 1; + SLAVE_AD_INFO(new_slave)->id = 1; /* Initialize AD with the number of times that the AD timer is called in 1 second * can be called only after the mac address of the bond is set */ bond_3ad_initialize(bond, 1000/AD_TIMER_INTERVAL); } else { - SLAVE_AD_INFO(new_slave).id = - SLAVE_AD_INFO(prev_slave).id + 1; + SLAVE_AD_INFO(new_slave)->id = + SLAVE_AD_INFO(prev_slave)->id + 1; } bond_3ad_bind_slave(new_slave); @@ -1547,7 +1525,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) case BOND_MODE_TLB: case BOND_MODE_ALB: bond_set_active_slave(new_slave); - bond_set_slave_inactive_flags(new_slave); + bond_set_slave_inactive_flags(new_slave, BOND_SLAVE_NOTIFY_NOW); break; default: pr_debug("This slave is always active in trunk mode\n"); @@ -1569,10 +1547,8 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) slave_dev->npinfo = bond->dev->npinfo; if (slave_dev->npinfo) { if (slave_enable_netpoll(new_slave)) { - read_unlock(&bond->lock); - pr_info("Error, %s: master_dev is using netpoll, " - "but new slave device does not support netpoll.\n", - bond_dev->name); + pr_info("Error, %s: master_dev is using netpoll, but new slave device does not support netpoll\n", + bond_dev->name); res = -EBUSY; goto err_detach; } @@ -1602,16 +1578,18 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) bond_compute_features(bond); bond_set_carrier(bond); - if (USES_PRIMARY(bond->params.mode)) { + if (bond_uses_primary(bond)) { + block_netpoll_tx(); write_lock_bh(&bond->curr_slave_lock); bond_select_active_slave(bond); write_unlock_bh(&bond->curr_slave_lock); + unblock_netpoll_tx(); } - pr_info("%s: enslaving %s as a%s interface with a%s link.\n", + pr_info("%s: Enslaving %s as %s interface with %s link\n", bond_dev->name, slave_dev->name, - bond_is_active_slave(new_slave) ? "n active" : " backup", - new_slave->link != BOND_LINK_DOWN ? "n up" : " down"); + bond_is_active_slave(new_slave) ? "an active" : "a backup", + new_slave->link != BOND_LINK_DOWN ? "an up" : "a down"); /* enslave is successful */ return 0; @@ -1624,17 +1602,19 @@ err_unregister: netdev_rx_handler_unregister(slave_dev); err_detach: - if (!USES_PRIMARY(bond->params.mode)) + if (!bond_uses_primary(bond)) bond_hw_addr_flush(bond_dev, slave_dev); vlan_vids_del_by_dev(slave_dev, bond_dev); if (bond->primary_slave == new_slave) bond->primary_slave = NULL; if (bond->curr_active_slave == new_slave) { + block_netpoll_tx(); write_lock_bh(&bond->curr_slave_lock); bond_change_active_slave(bond, NULL); bond_select_active_slave(bond); write_unlock_bh(&bond->curr_slave_lock); + unblock_netpoll_tx(); } slave_disable_netpoll(new_slave); @@ -1643,12 +1623,13 @@ err_close: dev_close(slave_dev); err_restore_mac: - if (!bond->params.fail_over_mac) { + if (!bond->params.fail_over_mac || + BOND_MODE(bond) != BOND_MODE_ACTIVEBACKUP) { /* XXX TODO - fom follow mode needs to change master's * MAC if this slave's MAC is in use by the bond, or at * least print a warning. */ - memcpy(addr.sa_data, new_slave->perm_hwaddr, ETH_ALEN); + ether_addr_copy(addr.sa_data, new_slave->perm_hwaddr); addr.sa_family = slave_dev->type; dev_set_mac_address(slave_dev, &addr); } @@ -1657,7 +1638,7 @@ err_restore_mtu: dev_set_mtu(slave_dev, new_slave->original_mtu); err_free: - kfree(new_slave); + bond_free_slave(new_slave); err_undo_flags: /* Enslave of first slave has failed and we need to fix master's mac */ @@ -1693,7 +1674,7 @@ static int __bond_release_one(struct net_device *bond_dev, /* slave is not a slave or master is not master of this slave */ if (!(slave_dev->flags & IFF_SLAVE) || !netdev_has_upper_dev(slave_dev, bond_dev)) { - pr_err("%s: Error: cannot release %s.\n", + pr_err("%s: Error: cannot release %s\n", bond_dev->name, slave_dev->name); return -EINVAL; } @@ -1709,9 +1690,6 @@ static int __bond_release_one(struct net_device *bond_dev, return -EINVAL; } - /* release the slave from its bond */ - bond->slave_cnt--; - bond_sysfs_slave_del(slave); bond_upper_dev_unlink(bond_dev, slave_dev); @@ -1722,12 +1700,12 @@ static int __bond_release_one(struct net_device *bond_dev, write_lock_bh(&bond->lock); /* Inform AD package of unbinding of slave. */ - if (bond->params.mode == BOND_MODE_8023AD) + if (BOND_MODE(bond) == BOND_MODE_8023AD) bond_3ad_unbind_slave(slave); write_unlock_bh(&bond->lock); - pr_info("%s: releasing %s interface %s\n", + pr_info("%s: Releasing %s interface %s\n", bond_dev->name, bond_is_active_slave(slave) ? "active" : "backup", slave_dev->name); @@ -1736,13 +1714,14 @@ static int __bond_release_one(struct net_device *bond_dev, bond->current_arp_slave = NULL; - if (!all && !bond->params.fail_over_mac) { + if (!all && (!bond->params.fail_over_mac || + BOND_MODE(bond) != BOND_MODE_ACTIVEBACKUP)) { if (ether_addr_equal_64bits(bond_dev->dev_addr, slave->perm_hwaddr) && bond_has_slaves(bond)) - pr_warn("%s: Warning: the permanent HWaddr of %s - %pM - is still in use by %s. Set the HWaddr of %s to a different address to avoid conflicts.\n", - bond_dev->name, slave_dev->name, - slave->perm_hwaddr, - bond_dev->name, slave_dev->name); + pr_warn("%s: Warning: the permanent HWaddr of %s - %pM - is still in use by %s - set the HWaddr of %s to a different address to avoid conflicts\n", + bond_dev->name, slave_dev->name, + slave->perm_hwaddr, + bond_dev->name, slave_dev->name); } if (bond->primary_slave == slave) @@ -1764,7 +1743,7 @@ static int __bond_release_one(struct net_device *bond_dev, } if (all) { - rcu_assign_pointer(bond->curr_active_slave, NULL); + RCU_INIT_POINTER(bond->curr_active_slave, NULL); } else if (oldcurrent == slave) { /* * Note that we hold RTNL over this sequence, so there @@ -1783,15 +1762,16 @@ static int __bond_release_one(struct net_device *bond_dev, eth_hw_addr_random(bond_dev); if (vlan_uses_dev(bond_dev)) { - pr_warning("%s: Warning: clearing HW address of %s while it still has VLANs.\n", - bond_dev->name, bond_dev->name); - pr_warning("%s: When re-adding slaves, make sure the bond's HW address matches its VLANs'.\n", - bond_dev->name); + pr_warn("%s: Warning: clearing HW address of %s while it still has VLANs\n", + bond_dev->name, bond_dev->name); + pr_warn("%s: When re-adding slaves, make sure the bond's HW address matches its VLANs\n", + bond_dev->name); } } unblock_netpoll_tx(); synchronize_rcu(); + bond->slave_cnt--; if (!bond_has_slaves(bond)) { call_netdevice_notifiers(NETDEV_CHANGEADDR, bond->dev); @@ -1801,16 +1781,16 @@ static int __bond_release_one(struct net_device *bond_dev, bond_compute_features(bond); if (!(bond_dev->features & NETIF_F_VLAN_CHALLENGED) && (old_features & NETIF_F_VLAN_CHALLENGED)) - pr_info("%s: last VLAN challenged slave %s left bond %s. VLAN blocking is removed\n", + pr_info("%s: last VLAN challenged slave %s left bond %s - VLAN blocking is removed\n", bond_dev->name, slave_dev->name, bond_dev->name); /* must do this from outside any spinlocks */ vlan_vids_del_by_dev(slave_dev, bond_dev); - /* If the mode USES_PRIMARY, then this cases was handled above by + /* If the mode uses primary, then this cases was handled above by * bond_change_active_slave(..., NULL) */ - if (!USES_PRIMARY(bond->params.mode)) { + if (!bond_uses_primary(bond)) { /* unset promiscuity level from slave * NOTE: The NETDEV_CHANGEADDR call above may change the value * of the IFF_PROMISC flag in the bond_dev, but we need the @@ -1833,9 +1813,10 @@ static int __bond_release_one(struct net_device *bond_dev, /* close slave before restoring its mac address */ dev_close(slave_dev); - if (bond->params.fail_over_mac != BOND_FOM_ACTIVE) { + if (bond->params.fail_over_mac != BOND_FOM_ACTIVE || + BOND_MODE(bond) != BOND_MODE_ACTIVEBACKUP) { /* restore original ("permanent") mac address */ - memcpy(addr.sa_data, slave->perm_hwaddr, ETH_ALEN); + ether_addr_copy(addr.sa_data, slave->perm_hwaddr); addr.sa_family = slave_dev->type; dev_set_mac_address(slave_dev, &addr); } @@ -1844,7 +1825,7 @@ static int __bond_release_one(struct net_device *bond_dev, slave_dev->priv_flags &= ~IFF_BONDING; - kfree(slave); + bond_free_slave(slave); return 0; /* deletion OK */ } @@ -1868,7 +1849,7 @@ static int bond_release_and_destroy(struct net_device *bond_dev, ret = bond_release(bond_dev, slave_dev); if (ret == 0 && !bond_has_slaves(bond)) { bond_dev->priv_flags |= IFF_DISABLE_NETPOLL; - pr_info("%s: destroying bond %s.\n", + pr_info("%s: Destroying bond %s\n", bond_dev->name, bond_dev->name); unregister_netdevice(bond_dev); } @@ -1879,12 +1860,10 @@ static int bond_info_query(struct net_device *bond_dev, struct ifbond *info) { struct bonding *bond = netdev_priv(bond_dev); - info->bond_mode = bond->params.mode; + info->bond_mode = BOND_MODE(bond); info->miimon = bond->params.miimon; - read_lock(&bond->lock); info->num_slaves = bond->slave_cnt; - read_unlock(&bond->lock); return 0; } @@ -1896,7 +1875,6 @@ static int bond_slave_info_query(struct net_device *bond_dev, struct ifslave *in int i = 0, res = -ENODEV; struct slave *slave; - read_lock(&bond->lock); bond_for_each_slave(bond, slave, iter) { if (i++ == (int)info->slave_id) { res = 0; @@ -1907,7 +1885,6 @@ static int bond_slave_info_query(struct net_device *bond_dev, struct ifslave *in break; } } - read_unlock(&bond->lock); return res; } @@ -1937,9 +1914,9 @@ static int bond_miimon_inspect(struct bonding *bond) slave->link = BOND_LINK_FAIL; slave->delay = bond->params.downdelay; if (slave->delay) { - pr_info("%s: link status down for %sinterface %s, disabling it in %d ms.\n", + pr_info("%s: link status down for %sinterface %s, disabling it in %d ms\n", bond->dev->name, - (bond->params.mode == + (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP) ? (bond_is_active_slave(slave) ? "active " : "backup ") : "", @@ -1953,8 +1930,8 @@ static int bond_miimon_inspect(struct bonding *bond) * recovered before downdelay expired */ slave->link = BOND_LINK_UP; - slave->jiffies = jiffies; - pr_info("%s: link status up again after %d ms for interface %s.\n", + slave->last_link_up = jiffies; + pr_info("%s: link status up again after %d ms for interface %s\n", bond->dev->name, (bond->params.downdelay - slave->delay) * bond->params.miimon, @@ -1979,7 +1956,7 @@ static int bond_miimon_inspect(struct bonding *bond) slave->delay = bond->params.updelay; if (slave->delay) { - pr_info("%s: link status up for interface %s, enabling it in %d ms.\n", + pr_info("%s: link status up for interface %s, enabling it in %d ms\n", bond->dev->name, slave->dev->name, ignore_updelay ? 0 : bond->params.updelay * @@ -1989,7 +1966,7 @@ static int bond_miimon_inspect(struct bonding *bond) case BOND_LINK_BACK: if (!link_state) { slave->link = BOND_LINK_DOWN; - pr_info("%s: link status down again after %d ms for interface %s.\n", + pr_info("%s: link status down again after %d ms for interface %s\n", bond->dev->name, (bond->params.updelay - slave->delay) * bond->params.miimon, @@ -2028,12 +2005,12 @@ static void bond_miimon_commit(struct bonding *bond) case BOND_LINK_UP: slave->link = BOND_LINK_UP; - slave->jiffies = jiffies; + slave->last_link_up = jiffies; - if (bond->params.mode == BOND_MODE_8023AD) { + if (BOND_MODE(bond) == BOND_MODE_8023AD) { /* prevent it from being the active one */ bond_set_backup_slave(slave); - } else if (bond->params.mode != BOND_MODE_ACTIVEBACKUP) { + } else if (BOND_MODE(bond) != BOND_MODE_ACTIVEBACKUP) { /* make it immediately active */ bond_set_active_slave(slave); } else if (slave != bond->primary_slave) { @@ -2041,13 +2018,13 @@ static void bond_miimon_commit(struct bonding *bond) bond_set_backup_slave(slave); } - pr_info("%s: link status definitely up for interface %s, %u Mbps %s duplex.\n", + pr_info("%s: link status definitely up for interface %s, %u Mbps %s duplex\n", bond->dev->name, slave->dev->name, slave->speed == SPEED_UNKNOWN ? 0 : slave->speed, slave->duplex ? "full" : "half"); /* notify ad that the link status has changed */ - if (bond->params.mode == BOND_MODE_8023AD) + if (BOND_MODE(bond) == BOND_MODE_8023AD) bond_3ad_handle_link_change(slave, BOND_LINK_UP); if (bond_is_lb(bond)) @@ -2066,14 +2043,15 @@ static void bond_miimon_commit(struct bonding *bond) slave->link = BOND_LINK_DOWN; - if (bond->params.mode == BOND_MODE_ACTIVEBACKUP || - bond->params.mode == BOND_MODE_8023AD) - bond_set_slave_inactive_flags(slave); + if (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP || + BOND_MODE(bond) == BOND_MODE_8023AD) + bond_set_slave_inactive_flags(slave, + BOND_SLAVE_NOTIFY_NOW); pr_info("%s: link status definitely down for interface %s, disabling it\n", bond->dev->name, slave->dev->name); - if (bond->params.mode == BOND_MODE_8023AD) + if (BOND_MODE(bond) == BOND_MODE_8023AD) bond_3ad_handle_link_change(slave, BOND_LINK_DOWN); @@ -2185,95 +2163,125 @@ static bool bond_has_this_ip(struct bonding *bond, __be32 ip) * switches in VLAN mode (especially if ports are configured as * "native" to a VLAN) might not pass non-tagged frames. */ -static void bond_arp_send(struct net_device *slave_dev, int arp_op, __be32 dest_ip, __be32 src_ip, unsigned short vlan_id) +static void bond_arp_send(struct net_device *slave_dev, int arp_op, + __be32 dest_ip, __be32 src_ip, + struct bond_vlan_tag *tags) { struct sk_buff *skb; + int i; - pr_debug("arp %d on slave %s: dst %pI4 src %pI4 vid %d\n", arp_op, - slave_dev->name, &dest_ip, &src_ip, vlan_id); + pr_debug("arp %d on slave %s: dst %pI4 src %pI4\n", + arp_op, slave_dev->name, &dest_ip, &src_ip); skb = arp_create(arp_op, ETH_P_ARP, dest_ip, slave_dev, src_ip, NULL, slave_dev->dev_addr, NULL); if (!skb) { - pr_err("ARP packet allocation failed\n"); + net_err_ratelimited("ARP packet allocation failed\n"); return; } - if (vlan_id) { - skb = vlan_put_tag(skb, htons(ETH_P_8021Q), vlan_id); + + /* Go through all the tags backwards and add them to the packet */ + for (i = BOND_MAX_VLAN_ENCAP - 1; i > 0; i--) { + if (!tags[i].vlan_id) + continue; + + pr_debug("inner tag: proto %X vid %X\n", + ntohs(tags[i].vlan_proto), tags[i].vlan_id); + skb = __vlan_put_tag(skb, tags[i].vlan_proto, + tags[i].vlan_id); + if (!skb) { + net_err_ratelimited("failed to insert inner VLAN tag\n"); + return; + } + } + /* Set the outer tag */ + if (tags[0].vlan_id) { + pr_debug("outer tag: proto %X vid %X\n", + ntohs(tags[0].vlan_proto), tags[0].vlan_id); + skb = vlan_put_tag(skb, tags[0].vlan_proto, tags[0].vlan_id); if (!skb) { - pr_err("failed to insert VLAN tag\n"); + net_err_ratelimited("failed to insert outer VLAN tag\n"); return; } } arp_xmit(skb); } +/* Validate the device path between the @start_dev and the @end_dev. + * The path is valid if the @end_dev is reachable through device + * stacking. + * When the path is validated, collect any vlan information in the + * path. + */ +bool bond_verify_device_path(struct net_device *start_dev, + struct net_device *end_dev, + struct bond_vlan_tag *tags) +{ + struct net_device *upper; + struct list_head *iter; + int idx; + + if (start_dev == end_dev) + return true; + + netdev_for_each_upper_dev_rcu(start_dev, upper, iter) { + if (bond_verify_device_path(upper, end_dev, tags)) { + if (is_vlan_dev(upper)) { + idx = vlan_get_encap_level(upper); + if (idx >= BOND_MAX_VLAN_ENCAP) + return false; + + tags[idx].vlan_proto = + vlan_dev_vlan_proto(upper); + tags[idx].vlan_id = vlan_dev_vlan_id(upper); + } + return true; + } + } + + return false; +} static void bond_arp_send_all(struct bonding *bond, struct slave *slave) { - struct net_device *upper, *vlan_upper; - struct list_head *iter, *vlan_iter; struct rtable *rt; + struct bond_vlan_tag tags[BOND_MAX_VLAN_ENCAP]; __be32 *targets = bond->params.arp_targets, addr; - int i, vlan_id; + int i; + bool ret; for (i = 0; i < BOND_MAX_ARP_TARGETS && targets[i]; i++) { pr_debug("basa: target %pI4\n", &targets[i]); + memset(tags, 0, sizeof(tags)); /* Find out through which dev should the packet go */ rt = ip_route_output(dev_net(bond->dev), targets[i], 0, RTO_ONLINK, 0); if (IS_ERR(rt)) { - pr_debug("%s: no route to arp_ip_target %pI4\n", - bond->dev->name, &targets[i]); + /* there's no route to target - try to send arp + * probe to generate any traffic (arp_validate=0) + */ + if (bond->params.arp_validate) + net_warn_ratelimited("%s: no route to arp_ip_target %pI4 and arp_validate is set\n", + bond->dev->name, + &targets[i]); + bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], + 0, tags); continue; } - vlan_id = 0; - /* bond device itself */ if (rt->dst.dev == bond->dev) goto found; rcu_read_lock(); - /* first we search only for vlan devices. for every vlan - * found we verify its upper dev list, searching for the - * rt->dst.dev. If found we save the tag of the vlan and - * proceed to send the packet. - * - * TODO: QinQ? - */ - netdev_for_each_all_upper_dev_rcu(bond->dev, vlan_upper, - vlan_iter) { - if (!is_vlan_dev(vlan_upper)) - continue; - netdev_for_each_all_upper_dev_rcu(vlan_upper, upper, - iter) { - if (upper == rt->dst.dev) { - vlan_id = vlan_dev_vlan_id(vlan_upper); - rcu_read_unlock(); - goto found; - } - } - } - - /* if the device we're looking for is not on top of any of - * our upper vlans, then just search for any dev that - * matches, and in case it's a vlan - save the id - */ - netdev_for_each_all_upper_dev_rcu(bond->dev, upper, iter) { - if (upper == rt->dst.dev) { - /* if it's a vlan - get its VID */ - if (is_vlan_dev(upper)) - vlan_id = vlan_dev_vlan_id(upper); - - rcu_read_unlock(); - goto found; - } - } + ret = bond_verify_device_path(bond->dev, rt->dst.dev, tags); rcu_read_unlock(); + if (ret) + goto found; + /* Not our device - skip */ pr_debug("%s: no path to arp_ip_target %pI4 via rt.dev %s\n", bond->dev->name, &targets[i], @@ -2286,7 +2294,7 @@ found: addr = bond_confirm_addr(rt->dst.dev, targets[i], 0); ip_rt_put(rt); bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], - addr, vlan_id); + addr, tags); } } @@ -2304,7 +2312,7 @@ static void bond_validate_arp(struct bonding *bond, struct slave *slave, __be32 pr_debug("bva: sip %pI4 not found in targets\n", &sip); return; } - slave->last_arp_rx = jiffies; + slave->last_rx = jiffies; slave->target_last_arp_rx[i] = jiffies; } @@ -2312,17 +2320,19 @@ int bond_arp_rcv(const struct sk_buff *skb, struct bonding *bond, struct slave *slave) { struct arphdr *arp = (struct arphdr *)skb->data; + struct slave *curr_active_slave; unsigned char *arp_ptr; __be32 sip, tip; - int alen; + int alen, is_arp = skb->protocol == __cpu_to_be16(ETH_P_ARP); - if (skb->protocol != __cpu_to_be16(ETH_P_ARP)) + if (!slave_do_arp_validate(bond, slave)) { + if ((slave_do_arp_validate_only(bond) && is_arp) || + !slave_do_arp_validate_only(bond)) + slave->last_rx = jiffies; return RX_HANDLER_ANOTHER; - - read_lock(&bond->lock); - - if (!slave_do_arp_validate(bond, slave)) - goto out_unlock; + } else if (!is_arp) { + return RX_HANDLER_ANOTHER; + } alen = arp_hdr_len(bond->dev); @@ -2356,6 +2366,8 @@ int bond_arp_rcv(const struct sk_buff *skb, struct bonding *bond, bond->params.arp_validate, slave_do_arp_validate(bond, slave), &sip, &tip); + curr_active_slave = rcu_dereference(bond->curr_active_slave); + /* * Backup slaves won't see the ARP reply, but do come through * here for each ARP probe (so we swap the sip/tip to validate @@ -2369,15 +2381,15 @@ int bond_arp_rcv(const struct sk_buff *skb, struct bonding *bond, * is done to avoid endless looping when we can't reach the * arp_ip_target and fool ourselves with our own arp requests. */ + if (bond_is_active_slave(slave)) bond_validate_arp(bond, slave, sip, tip); - else if (bond->curr_active_slave && - time_after(slave_last_rx(bond, bond->curr_active_slave), - bond->curr_active_slave->jiffies)) + else if (curr_active_slave && + time_after(slave_last_rx(bond, curr_active_slave), + curr_active_slave->last_link_up)) bond_validate_arp(bond, slave, tip, sip); out_unlock: - read_unlock(&bond->lock); if (arp != (struct arphdr *)skb->data) kfree(arp); return RX_HANDLER_ANOTHER; @@ -2410,7 +2422,7 @@ static void bond_loadbalance_arp_mon(struct work_struct *work) arp_work.work); struct slave *slave, *oldcurrent; struct list_head *iter; - int do_failover = 0; + int do_failover = 0, slave_state_changed = 0; if (!bond_has_slaves(bond)) goto re_arm; @@ -2420,9 +2432,9 @@ static void bond_loadbalance_arp_mon(struct work_struct *work) oldcurrent = ACCESS_ONCE(bond->curr_active_slave); /* see if any of the previous devices are up now (i.e. they have * xmt and rcv traffic). the curr_active_slave does not come into - * the picture unless it is null. also, slave->jiffies is not needed - * here because we send an arp on each slave and give a slave as - * long as it needs to get the tx/rx within the delta. + * the picture unless it is null. also, slave->last_link_up is not + * needed here because we send an arp on each slave and give a slave + * as long as it needs to get the tx/rx within the delta. * TODO: what about up/down delay in arp mode? it wasn't here before * so it can wait */ @@ -2431,10 +2443,10 @@ static void bond_loadbalance_arp_mon(struct work_struct *work) if (slave->link != BOND_LINK_UP) { if (bond_time_in_interval(bond, trans_start, 1) && - bond_time_in_interval(bond, slave->dev->last_rx, 1)) { + bond_time_in_interval(bond, slave->last_rx, 1)) { slave->link = BOND_LINK_UP; - bond_set_active_slave(slave); + slave_state_changed = 1; /* primary_slave has no meaning in round-robin * mode. the window of a slave being up and @@ -2442,7 +2454,7 @@ static void bond_loadbalance_arp_mon(struct work_struct *work) * is closed. */ if (!oldcurrent) { - pr_info("%s: link status definitely up for interface %s, ", + pr_info("%s: link status definitely up for interface %s\n", bond->dev->name, slave->dev->name); do_failover = 1; @@ -2460,17 +2472,16 @@ static void bond_loadbalance_arp_mon(struct work_struct *work) * if we don't know our ip yet */ if (!bond_time_in_interval(bond, trans_start, 2) || - !bond_time_in_interval(bond, slave->dev->last_rx, 2)) { + !bond_time_in_interval(bond, slave->last_rx, 2)) { slave->link = BOND_LINK_DOWN; - bond_set_backup_slave(slave); + slave_state_changed = 1; if (slave->link_failure_count < UINT_MAX) slave->link_failure_count++; - pr_info("%s: interface %s is now down.\n", - bond->dev->name, - slave->dev->name); + pr_info("%s: interface %s is now down\n", + bond->dev->name, slave->dev->name); if (slave == oldcurrent) do_failover = 1; @@ -2484,25 +2495,30 @@ static void bond_loadbalance_arp_mon(struct work_struct *work) * do - all replies will be rx'ed on same link causing slaves * to be unstable during low/no traffic periods */ - if (IS_UP(slave->dev)) + if (bond_slave_is_up(slave)) bond_arp_send_all(bond, slave); } rcu_read_unlock(); - if (do_failover) { - /* the bond_select_active_slave must hold RTNL - * and curr_slave_lock for write. - */ + if (do_failover || slave_state_changed) { if (!rtnl_trylock()) goto re_arm; - block_netpoll_tx(); - write_lock_bh(&bond->curr_slave_lock); - bond_select_active_slave(bond); + if (slave_state_changed) { + bond_slave_state_change(bond); + } else if (do_failover) { + /* the bond_select_active_slave must hold RTNL + * and curr_slave_lock for write. + */ + block_netpoll_tx(); + write_lock_bh(&bond->curr_slave_lock); - write_unlock_bh(&bond->curr_slave_lock); - unblock_netpoll_tx(); + bond_select_active_slave(bond); + + write_unlock_bh(&bond->curr_slave_lock); + unblock_netpoll_tx(); + } rtnl_unlock(); } @@ -2544,7 +2560,7 @@ static int bond_ab_arp_inspect(struct bonding *bond) * active. This avoids bouncing, as the last receive * times need a full ARP monitor cycle to be updated. */ - if (bond_time_in_interval(bond, slave->jiffies, 2)) + if (bond_time_in_interval(bond, slave->last_link_up, 2)) continue; /* @@ -2610,11 +2626,12 @@ static void bond_ab_arp_commit(struct bonding *bond) slave->link = BOND_LINK_UP; if (bond->current_arp_slave) { bond_set_slave_inactive_flags( - bond->current_arp_slave); + bond->current_arp_slave, + BOND_SLAVE_NOTIFY_NOW); bond->current_arp_slave = NULL; } - pr_info("%s: link status definitely up for interface %s.\n", + pr_info("%s: link status definitely up for interface %s\n", bond->dev->name, slave->dev->name); if (!bond->curr_active_slave || @@ -2630,7 +2647,8 @@ static void bond_ab_arp_commit(struct bonding *bond) slave->link_failure_count++; slave->link = BOND_LINK_DOWN; - bond_set_slave_inactive_flags(slave); + bond_set_slave_inactive_flags(slave, + BOND_SLAVE_NOTIFY_NOW); pr_info("%s: link status definitely down for interface %s, disabling it\n", bond->dev->name, slave->dev->name); @@ -2666,28 +2684,25 @@ do_failover: * * Called with rcu_read_lock hold. */ -static void bond_ab_arp_probe(struct bonding *bond) +static bool bond_ab_arp_probe(struct bonding *bond) { struct slave *slave, *before = NULL, *new_slave = NULL, - *curr_arp_slave = rcu_dereference(bond->current_arp_slave); + *curr_arp_slave = rcu_dereference(bond->current_arp_slave), + *curr_active_slave = rcu_dereference(bond->curr_active_slave); struct list_head *iter; bool found = false; + bool should_notify_rtnl = BOND_SLAVE_NOTIFY_LATER; - read_lock(&bond->curr_slave_lock); - - if (curr_arp_slave && bond->curr_active_slave) + if (curr_arp_slave && curr_active_slave) pr_info("PROBE: c_arp %s && cas %s BAD\n", curr_arp_slave->dev->name, - bond->curr_active_slave->dev->name); + curr_active_slave->dev->name); - if (bond->curr_active_slave) { - bond_arp_send_all(bond, bond->curr_active_slave); - read_unlock(&bond->curr_slave_lock); - return; + if (curr_active_slave) { + bond_arp_send_all(bond, curr_active_slave); + return should_notify_rtnl; } - read_unlock(&bond->curr_slave_lock); - /* if we don't have a curr_active_slave, search for the next available * backup slave from the current_arp_slave and make it the candidate * for becoming the curr_active_slave @@ -2696,16 +2711,16 @@ static void bond_ab_arp_probe(struct bonding *bond) if (!curr_arp_slave) { curr_arp_slave = bond_first_slave_rcu(bond); if (!curr_arp_slave) - return; + return should_notify_rtnl; } - bond_set_slave_inactive_flags(curr_arp_slave); + bond_set_slave_inactive_flags(curr_arp_slave, BOND_SLAVE_NOTIFY_LATER); bond_for_each_slave_rcu(bond, slave, iter) { - if (!found && !before && IS_UP(slave->dev)) + if (!found && !before && bond_slave_is_up(slave)) before = slave; - if (found && !new_slave && IS_UP(slave->dev)) + if (found && !new_slave && bond_slave_is_up(slave)) new_slave = slave; /* if the link state is up at this point, we * mark it down - this can happen if we have @@ -2714,14 +2729,15 @@ static void bond_ab_arp_probe(struct bonding *bond) * one the current slave so it is still marked * up when it is actually down */ - if (!IS_UP(slave->dev) && slave->link == BOND_LINK_UP) { + if (!bond_slave_is_up(slave) && slave->link == BOND_LINK_UP) { slave->link = BOND_LINK_DOWN; if (slave->link_failure_count < UINT_MAX) slave->link_failure_count++; - bond_set_slave_inactive_flags(slave); + bond_set_slave_inactive_flags(slave, + BOND_SLAVE_NOTIFY_LATER); - pr_info("%s: backup interface %s is now down.\n", + pr_info("%s: backup interface %s is now down\n", bond->dev->name, slave->dev->name); } if (slave == curr_arp_slave) @@ -2732,13 +2748,22 @@ static void bond_ab_arp_probe(struct bonding *bond) new_slave = before; if (!new_slave) - return; + goto check_state; new_slave->link = BOND_LINK_BACK; - bond_set_slave_active_flags(new_slave); + bond_set_slave_active_flags(new_slave, BOND_SLAVE_NOTIFY_LATER); bond_arp_send_all(bond, new_slave); - new_slave->jiffies = jiffies; + new_slave->last_link_up = jiffies; rcu_assign_pointer(bond->current_arp_slave, new_slave); + +check_state: + bond_for_each_slave_rcu(bond, slave, iter) { + if (slave->should_notify) { + should_notify_rtnl = BOND_SLAVE_NOTIFY_NOW; + break; + } + } + return should_notify_rtnl; } static void bond_activebackup_arp_mon(struct work_struct *work) @@ -2746,6 +2771,7 @@ static void bond_activebackup_arp_mon(struct work_struct *work) struct bonding *bond = container_of(work, struct bonding, arp_work.work); bool should_notify_peers = false; + bool should_notify_rtnl = false; int delta_in_ticks; delta_in_ticks = msecs_to_jiffies(bond->params.arp_interval); @@ -2773,17 +2799,23 @@ static void bond_activebackup_arp_mon(struct work_struct *work) rcu_read_lock(); } - bond_ab_arp_probe(bond); + should_notify_rtnl = bond_ab_arp_probe(bond); rcu_read_unlock(); re_arm: if (bond->params.arp_interval) queue_delayed_work(bond->wq, &bond->arp_work, delta_in_ticks); - if (should_notify_peers) { + if (should_notify_peers || should_notify_rtnl) { if (!rtnl_trylock()) return; - call_netdevice_notifiers(NETDEV_NOTIFY_PEERS, bond->dev); + + if (should_notify_peers) + call_netdevice_notifiers(NETDEV_NOTIFY_PEERS, + bond->dev); + if (should_notify_rtnl) + bond_slave_state_notify(bond); + rtnl_unlock(); } } @@ -2860,7 +2892,7 @@ static int bond_slave_netdev_event(unsigned long event, bond_update_speed_duplex(slave); - if (bond->params.mode == BOND_MODE_8023AD) { + if (BOND_MODE(bond) == BOND_MODE_8023AD) { if (old_speed != slave->speed) bond_3ad_adapter_speed_changed(slave); if (old_duplex != slave->duplex) @@ -2888,7 +2920,7 @@ static int bond_slave_netdev_event(unsigned long event, break; case NETDEV_CHANGENAME: /* we don't care if we don't have primary set */ - if (!USES_PRIMARY(bond->params.mode) || + if (!bond_uses_primary(bond) || !bond->params.primary[0]) break; @@ -2902,12 +2934,15 @@ static int bond_slave_netdev_event(unsigned long event, break; } - pr_info("%s: Primary slave changed to %s, reselecting active slave.\n", - bond->dev->name, bond->primary_slave ? slave_dev->name : - "none"); + pr_info("%s: Primary slave changed to %s, reselecting active slave\n", + bond->dev->name, + bond->primary_slave ? slave_dev->name : "none"); + + block_netpoll_tx(); write_lock_bh(&bond->curr_slave_lock); bond_select_active_slave(bond); write_unlock_bh(&bond->curr_slave_lock); + unblock_netpoll_tx(); break; case NETDEV_FEAT_CHANGE: bond_compute_features(bond); @@ -2937,8 +2972,7 @@ static int bond_netdev_event(struct notifier_block *this, struct net_device *event_dev = netdev_notifier_info_to_dev(ptr); pr_debug("event_dev: %s, event: %lx\n", - event_dev ? event_dev->name : "None", - event); + event_dev ? event_dev->name : "None", event); if (!(event_dev->priv_flags & IFF_BONDING)) return NOTIFY_DONE; @@ -2987,7 +3021,7 @@ static bool bond_flow_dissect(struct bonding *bond, struct sk_buff *skb, fk->ports = 0; noff = skb_network_offset(skb); if (skb->protocol == htons(ETH_P_IP)) { - if (!pskb_may_pull(skb, noff + sizeof(*iph))) + if (unlikely(!pskb_may_pull(skb, noff + sizeof(*iph)))) return false; iph = ip_hdr(skb); fk->src = iph->saddr; @@ -2996,7 +3030,7 @@ static bool bond_flow_dissect(struct bonding *bond, struct sk_buff *skb, if (!ip_is_fragment(iph)) proto = iph->protocol; } else if (skb->protocol == htons(ETH_P_IPV6)) { - if (!pskb_may_pull(skb, noff + sizeof(*iph6))) + if (unlikely(!pskb_may_pull(skb, noff + sizeof(*iph6)))) return false; iph6 = ipv6_hdr(skb); fk->src = (__force __be32)ipv6_addr_hash(&iph6->saddr); @@ -3016,20 +3050,18 @@ static bool bond_flow_dissect(struct bonding *bond, struct sk_buff *skb, * bond_xmit_hash - generate a hash value based on the xmit policy * @bond: bonding device * @skb: buffer to use for headers - * @count: modulo value * * This function will extract the necessary headers from the skb buffer and use * them to generate a hash based on the xmit_policy set in the bonding device - * which will be reduced modulo count before returning. */ -int bond_xmit_hash(struct bonding *bond, struct sk_buff *skb, int count) +u32 bond_xmit_hash(struct bonding *bond, struct sk_buff *skb) { struct flow_keys flow; u32 hash; if (bond->params.xmit_policy == BOND_XMIT_POLICY_LAYER2 || !bond_flow_dissect(bond, skb, &flow)) - return bond_eth_hash(skb) % count; + return bond_eth_hash(skb); if (bond->params.xmit_policy == BOND_XMIT_POLICY_LAYER23 || bond->params.xmit_policy == BOND_XMIT_POLICY_ENCAP23) @@ -3040,7 +3072,7 @@ int bond_xmit_hash(struct bonding *bond, struct sk_buff *skb, int count) hash ^= (hash >> 16); hash ^= (hash >> 8); - return hash % count; + return hash; } /*-------------------------- Device entry points ----------------------------*/ @@ -3051,7 +3083,7 @@ static void bond_work_init_all(struct bonding *bond) bond_resend_igmp_join_requests_delayed); INIT_DELAYED_WORK(&bond->alb_work, bond_alb_monitor); INIT_DELAYED_WORK(&bond->mii_work, bond_mii_monitor); - if (bond->params.mode == BOND_MODE_ACTIVEBACKUP) + if (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP) INIT_DELAYED_WORK(&bond->arp_work, bond_activebackup_arp_mon); else INIT_DELAYED_WORK(&bond->arp_work, bond_loadbalance_arp_mon); @@ -3078,11 +3110,13 @@ static int bond_open(struct net_device *bond_dev) if (bond_has_slaves(bond)) { read_lock(&bond->curr_slave_lock); bond_for_each_slave(bond, slave, iter) { - if ((bond->params.mode == BOND_MODE_ACTIVEBACKUP) + if (bond_uses_primary(bond) && (slave != bond->curr_active_slave)) { - bond_set_slave_inactive_flags(slave); + bond_set_slave_inactive_flags(slave, + BOND_SLAVE_NOTIFY_NOW); } else { - bond_set_slave_active_flags(slave); + bond_set_slave_active_flags(slave, + BOND_SLAVE_NOTIFY_NOW); } } read_unlock(&bond->curr_slave_lock); @@ -3095,9 +3129,10 @@ static int bond_open(struct net_device *bond_dev) /* bond_alb_initialize must be called before the timer * is started. */ - if (bond_alb_initialize(bond, (bond->params.mode == BOND_MODE_ALB))) + if (bond_alb_initialize(bond, (BOND_MODE(bond) == BOND_MODE_ALB))) return -ENOMEM; - queue_delayed_work(bond->wq, &bond->alb_work, 0); + if (bond->params.tlb_dynamic_lb) + queue_delayed_work(bond->wq, &bond->alb_work, 0); } if (bond->params.miimon) /* link check interval, in milliseconds. */ @@ -3105,11 +3140,10 @@ static int bond_open(struct net_device *bond_dev) if (bond->params.arp_interval) { /* arp interval, in milliseconds. */ queue_delayed_work(bond->wq, &bond->arp_work, 0); - if (bond->params.arp_validate) - bond->recv_probe = bond_arp_rcv; + bond->recv_probe = bond_arp_rcv; } - if (bond->params.mode == BOND_MODE_8023AD) { + if (BOND_MODE(bond) == BOND_MODE_8023AD) { queue_delayed_work(bond->wq, &bond->ad_work, 0); /* register to receive LACPDUs */ bond->recv_probe = bond_3ad_lacpdu_recv; @@ -3187,6 +3221,7 @@ static int bond_do_ioctl(struct net_device *bond_dev, struct ifreq *ifr, int cmd struct ifslave k_sinfo; struct ifslave __user *u_sinfo = NULL; struct mii_ioctl_data *mii = NULL; + struct bond_opt_value newval; struct net *net; int res = 0; @@ -3282,7 +3317,8 @@ static int bond_do_ioctl(struct net_device *bond_dev, struct ifreq *ifr, int cmd break; case BOND_CHANGE_ACTIVE_OLD: case SIOCBONDCHANGEACTIVE: - res = bond_option_active_slave_set(bond, slave_dev); + bond_opt_initstr(&newval, slave_dev->name); + res = __bond_opt_set(bond, BOND_OPT_ACTIVE_SLAVE, &newval); break; default: res = -EOPNOTSUPP; @@ -3312,7 +3348,7 @@ static void bond_set_rx_mode(struct net_device *bond_dev) rcu_read_lock(); - if (USES_PRIMARY(bond->params.mode)) { + if (bond_uses_primary(bond)) { slave = rcu_dereference(bond->curr_active_slave); if (slave) { dev_uc_sync(slave->dev, bond_dev); @@ -3391,8 +3427,8 @@ static int bond_change_mtu(struct net_device *bond_dev, int new_mtu) struct list_head *iter; int res = 0; - pr_debug("bond=%p, name=%s, new_mtu=%d\n", bond, - (bond_dev ? bond_dev->name : "None"), new_mtu); + pr_debug("bond=%p, name=%s, new_mtu=%d\n", + bond, bond_dev ? bond_dev->name : "None", new_mtu); /* Can't hold bond->lock with bh disabled here since * some base drivers panic. On the other hand we can't @@ -3411,8 +3447,7 @@ static int bond_change_mtu(struct net_device *bond_dev, int new_mtu) bond_for_each_slave(bond, slave, iter) { pr_debug("s %p c_m %p\n", - slave, - slave->dev->netdev_ops->ndo_change_mtu); + slave, slave->dev->netdev_ops->ndo_change_mtu); res = dev_set_mtu(slave->dev, new_mtu); @@ -3467,7 +3502,7 @@ static int bond_set_mac_address(struct net_device *bond_dev, void *addr) struct list_head *iter; int res = 0; - if (bond->params.mode == BOND_MODE_ALB) + if (BOND_MODE(bond) == BOND_MODE_ALB) return bond_alb_set_mac_address(bond_dev, addr); @@ -3477,7 +3512,8 @@ static int bond_set_mac_address(struct net_device *bond_dev, void *addr) /* If fail_over_mac is enabled, do nothing and return success. * Returning an error causes ifenslave to fail. */ - if (bond->params.fail_over_mac) + if (bond->params.fail_over_mac && + BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP) return 0; if (!is_valid_ether_addr(sa->sa_data)) @@ -3499,15 +3535,7 @@ static int bond_set_mac_address(struct net_device *bond_dev, void *addr) */ bond_for_each_slave(bond, slave, iter) { - const struct net_device_ops *slave_ops = slave->dev->netdev_ops; pr_debug("slave %p %s\n", slave, slave->dev->name); - - if (slave_ops->ndo_set_mac_address == NULL) { - res = -EOPNOTSUPP; - pr_debug("EOPNOTSUPP %s\n", slave->dev->name); - goto unwind; - } - res = dev_set_mac_address(slave->dev, addr); if (res) { /* TODO: consider downing the slave @@ -3565,7 +3593,7 @@ static void bond_xmit_slave_id(struct bonding *bond, struct sk_buff *skb, int sl /* Here we start from the slave with slave_id */ bond_for_each_slave_rcu(bond, slave, iter) { if (--i < 0) { - if (slave_can_tx(slave)) { + if (bond_slave_can_tx(slave)) { bond_dev_queue_xmit(bond, skb, slave->dev); return; } @@ -3577,13 +3605,13 @@ static void bond_xmit_slave_id(struct bonding *bond, struct sk_buff *skb, int sl bond_for_each_slave_rcu(bond, slave, iter) { if (--i < 0) break; - if (slave_can_tx(slave)) { + if (bond_slave_can_tx(slave)) { bond_dev_queue_xmit(bond, skb, slave->dev); return; } } /* no slave that can tx has been found */ - kfree_skb(skb); + dev_kfree_skb_any(skb); } /** @@ -3596,8 +3624,9 @@ static void bond_xmit_slave_id(struct bonding *bond, struct sk_buff *skb, int sl */ static u32 bond_rr_gen_slave_id(struct bonding *bond) { - int packets_per_slave = bond->params.packets_per_slave; u32 slave_id; + struct reciprocal_value reciprocal_packets_per_slave; + int packets_per_slave = bond->params.packets_per_slave; switch (packets_per_slave) { case 0: @@ -3607,8 +3636,10 @@ static u32 bond_rr_gen_slave_id(struct bonding *bond) slave_id = bond->rr_tx_counter; break; default: + reciprocal_packets_per_slave = + bond->params.reciprocal_packets_per_slave; slave_id = reciprocal_divide(bond->rr_tx_counter, - packets_per_slave); + reciprocal_packets_per_slave); break; } bond->rr_tx_counter++; @@ -3631,7 +3662,7 @@ static int bond_xmit_roundrobin(struct sk_buff *skb, struct net_device *bond_dev */ if (iph->protocol == IPPROTO_IGMP && skb->protocol == htons(ETH_P_IP)) { slave = rcu_dereference(bond->curr_active_slave); - if (slave && slave_can_tx(slave)) + if (slave && bond_slave_can_tx(slave)) bond_dev_queue_xmit(bond, skb, slave->dev); else bond_xmit_slave_id(bond, skb, 0); @@ -3656,7 +3687,7 @@ static int bond_xmit_activebackup(struct sk_buff *skb, struct net_device *bond_d if (slave) bond_dev_queue_xmit(bond, skb, slave->dev); else - kfree_skb(skb); + dev_kfree_skb_any(skb); return NETDEV_TX_OK; } @@ -3669,7 +3700,7 @@ static int bond_xmit_xor(struct sk_buff *skb, struct net_device *bond_dev) { struct bonding *bond = netdev_priv(bond_dev); - bond_xmit_slave_id(bond, skb, bond_xmit_hash(bond, skb, bond->slave_cnt)); + bond_xmit_slave_id(bond, skb, bond_xmit_hash(bond, skb) % bond->slave_cnt); return NETDEV_TX_OK; } @@ -3684,22 +3715,22 @@ static int bond_xmit_broadcast(struct sk_buff *skb, struct net_device *bond_dev) bond_for_each_slave_rcu(bond, slave, iter) { if (bond_is_last_slave(bond, slave)) break; - if (IS_UP(slave->dev) && slave->link == BOND_LINK_UP) { + if (bond_slave_is_up(slave) && slave->link == BOND_LINK_UP) { struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC); if (!skb2) { - pr_err("%s: Error: bond_xmit_broadcast(): skb_clone() failed\n", - bond_dev->name); + net_err_ratelimited("%s: Error: %s: skb_clone() failed\n", + bond_dev->name, __func__); continue; } /* bond_dev_queue_xmit always returns 0 */ bond_dev_queue_xmit(bond, skb2, slave->dev); } } - if (slave && IS_UP(slave->dev) && slave->link == BOND_LINK_UP) + if (slave && bond_slave_is_up(slave) && slave->link == BOND_LINK_UP) bond_dev_queue_xmit(bond, skb, slave->dev); else - kfree_skb(skb); + dev_kfree_skb_any(skb); return NETDEV_TX_OK; } @@ -3721,7 +3752,7 @@ static inline int bond_slave_override(struct bonding *bond, /* Find out if any slaves have the same mapping as this skb. */ bond_for_each_slave_rcu(bond, slave, iter) { if (slave->queue_id == skb->queue_mapping) { - if (slave_can_tx(slave)) { + if (bond_slave_can_tx(slave)) { bond_dev_queue_xmit(bond, skb, slave->dev); return 0; } @@ -3735,7 +3766,7 @@ static inline int bond_slave_override(struct bonding *bond, static u16 bond_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, select_queue_fallback_t fallback) { /* * This helper function exists to help dev_pick_tx get the correct @@ -3762,12 +3793,11 @@ static netdev_tx_t __bond_start_xmit(struct sk_buff *skb, struct net_device *dev { struct bonding *bond = netdev_priv(dev); - if (TX_QUEUE_OVERRIDE(bond->params.mode)) { - if (!bond_slave_override(bond, skb)) - return NETDEV_TX_OK; - } + if (bond_should_override_tx_queue(bond) && + !bond_slave_override(bond, skb)) + return NETDEV_TX_OK; - switch (bond->params.mode) { + switch (BOND_MODE(bond)) { case BOND_MODE_ROUNDROBIN: return bond_xmit_roundrobin(skb, dev); case BOND_MODE_ACTIVEBACKUP: @@ -3779,14 +3809,15 @@ static netdev_tx_t __bond_start_xmit(struct sk_buff *skb, struct net_device *dev case BOND_MODE_8023AD: return bond_3ad_xmit_xor(skb, dev); case BOND_MODE_ALB: - case BOND_MODE_TLB: return bond_alb_xmit(skb, dev); + case BOND_MODE_TLB: + return bond_tlb_xmit(skb, dev); default: /* Should never happen, mode already checked */ pr_err("%s: Error: Unknown bonding mode %d\n", - dev->name, bond->params.mode); + dev->name, BOND_MODE(bond)); WARN_ON_ONCE(1); - kfree_skb(skb); + dev_kfree_skb_any(skb); return NETDEV_TX_OK; } } @@ -3800,14 +3831,14 @@ static netdev_tx_t bond_start_xmit(struct sk_buff *skb, struct net_device *dev) * If we risk deadlock from transmitting this in the * netpoll path, tell netpoll to queue the frame for later tx */ - if (is_netpoll_tx_blocked(dev)) + if (unlikely(is_netpoll_tx_blocked(dev))) return NETDEV_TX_BUSY; rcu_read_lock(); if (bond_has_slaves(bond)) ret = __bond_start_xmit(skb, dev); else - kfree_skb(skb); + dev_kfree_skb_any(skb); rcu_read_unlock(); return ret; @@ -3824,14 +3855,14 @@ static int bond_ethtool_get_settings(struct net_device *bond_dev, ecmd->duplex = DUPLEX_UNKNOWN; ecmd->port = PORT_OTHER; - /* Since SLAVE_IS_OK returns false for all inactive or down slaves, we + /* Since bond_slave_can_tx returns false for all inactive or down slaves, we * do not need to check mode. Though link speed might not represent * the true receive or transmit bandwidth (not all modes are symmetric) * this is an accurate maximum. */ read_lock(&bond->lock); bond_for_each_slave(bond, slave, iter) { - if (SLAVE_IS_OK(slave)) { + if (bond_slave_can_tx(slave)) { if (slave->speed != SPEED_UNKNOWN) speed += slave->speed; if (ecmd->duplex == DUPLEX_UNKNOWN && @@ -3883,7 +3914,6 @@ static const struct net_device_ops bond_netdev_ops = { #endif .ndo_add_slave = bond_enslave, .ndo_del_slave = bond_release, - .ndo_get_slave = bond_get_slave, .ndo_fix_features = bond_fix_features, }; @@ -3923,7 +3953,7 @@ void bond_setup(struct net_device *bond_dev) /* Initialize the device options */ bond_dev->tx_queue_len = 0; bond_dev->flags |= IFF_MASTER|IFF_MULTICAST; - bond_dev->priv_flags |= IFF_BONDING; + bond_dev->priv_flags |= IFF_BONDING | IFF_UNICAST_FLT; bond_dev->priv_flags &= ~(IFF_XMIT_DST_RELEASE | IFF_TX_SKB_SHARING); /* At first, we block adding VLANs. That's the only way to @@ -3944,12 +3974,16 @@ void bond_setup(struct net_device *bond_dev) * capable */ + /* Don't allow bond devices to change network namespaces. */ + bond_dev->features |= NETIF_F_NETNS_LOCAL; + bond_dev->hw_features = BOND_VLAN_FEATURES | NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_FILTER; bond_dev->hw_features &= ~(NETIF_F_ALL_CSUM & ~NETIF_F_HW_CSUM); + bond_dev->hw_features |= NETIF_F_GSO_UDP_TUNNEL; bond_dev->features |= bond_dev->hw_features; } @@ -3968,7 +4002,7 @@ static void bond_uninit(struct net_device *bond_dev) /* Release the bonded slaves */ bond_for_each_slave(bond, slave, iter) __bond_release_one(bond_dev, slave->dev, true); - pr_info("%s: released all slaves\n", bond_dev->name); + pr_info("%s: Released all slaves\n", bond_dev->name); list_del(&bond->bond_list); @@ -3977,83 +4011,42 @@ static void bond_uninit(struct net_device *bond_dev) /*------------------------- Module initialization ---------------------------*/ -int bond_parm_tbl_lookup(int mode, const struct bond_parm_tbl *tbl) -{ - int i; - - for (i = 0; tbl[i].modename; i++) - if (mode == tbl[i].mode) - return tbl[i].mode; - - return -1; -} - -static int bond_parm_tbl_lookup_name(const char *modename, - const struct bond_parm_tbl *tbl) -{ - int i; - - for (i = 0; tbl[i].modename; i++) - if (strcmp(modename, tbl[i].modename) == 0) - return tbl[i].mode; - - return -1; -} - -/* - * Convert string input module parms. Accept either the - * number of the mode or its string name. A bit complicated because - * some mode names are substrings of other names, and calls from sysfs - * may have whitespace in the name (trailing newlines, for example). - */ -int bond_parse_parm(const char *buf, const struct bond_parm_tbl *tbl) -{ - int modeint; - char *p, modestr[BOND_MAX_MODENAME_LEN + 1]; - - for (p = (char *)buf; *p; p++) - if (!(isdigit(*p) || isspace(*p))) - break; - - if (*p && sscanf(buf, "%20s", modestr) != 0) - return bond_parm_tbl_lookup_name(modestr, tbl); - else if (sscanf(buf, "%d", &modeint) != 0) - return bond_parm_tbl_lookup(modeint, tbl); - - return -1; -} - static int bond_check_params(struct bond_params *params) { int arp_validate_value, fail_over_mac_value, primary_reselect_value, i; + struct bond_opt_value newval; + const struct bond_opt_value *valptr; int arp_all_targets_value; /* * Convert string parameters. */ if (mode) { - bond_mode = bond_parse_parm(mode, bond_mode_tbl); - if (bond_mode == -1) { - pr_err("Error: Invalid bonding mode \"%s\"\n", - mode == NULL ? "NULL" : mode); + bond_opt_initstr(&newval, mode); + valptr = bond_opt_parse(bond_opt_get(BOND_OPT_MODE), &newval); + if (!valptr) { + pr_err("Error: Invalid bonding mode \"%s\"\n", mode); return -EINVAL; } + bond_mode = valptr->value; } if (xmit_hash_policy) { if ((bond_mode != BOND_MODE_XOR) && - (bond_mode != BOND_MODE_8023AD)) { + (bond_mode != BOND_MODE_8023AD) && + (bond_mode != BOND_MODE_TLB)) { pr_info("xmit_hash_policy param is irrelevant in mode %s\n", - bond_mode_name(bond_mode)); + bond_mode_name(bond_mode)); } else { - xmit_hashtype = bond_parse_parm(xmit_hash_policy, - xmit_hashtype_tbl); - if (xmit_hashtype == -1) { + bond_opt_initstr(&newval, xmit_hash_policy); + valptr = bond_opt_parse(bond_opt_get(BOND_OPT_XMIT_HASH), + &newval); + if (!valptr) { pr_err("Error: Invalid xmit_hash_policy \"%s\"\n", - xmit_hash_policy == NULL ? "NULL" : xmit_hash_policy); return -EINVAL; } + xmit_hashtype = valptr->value; } } @@ -4062,97 +4055,98 @@ static int bond_check_params(struct bond_params *params) pr_info("lacp_rate param is irrelevant in mode %s\n", bond_mode_name(bond_mode)); } else { - lacp_fast = bond_parse_parm(lacp_rate, bond_lacp_tbl); - if (lacp_fast == -1) { + bond_opt_initstr(&newval, lacp_rate); + valptr = bond_opt_parse(bond_opt_get(BOND_OPT_LACP_RATE), + &newval); + if (!valptr) { pr_err("Error: Invalid lacp rate \"%s\"\n", - lacp_rate == NULL ? "NULL" : lacp_rate); + lacp_rate); return -EINVAL; } + lacp_fast = valptr->value; } } if (ad_select) { - params->ad_select = bond_parse_parm(ad_select, ad_select_tbl); - if (params->ad_select == -1) { - pr_err("Error: Invalid ad_select \"%s\"\n", - ad_select == NULL ? "NULL" : ad_select); + bond_opt_initstr(&newval, ad_select); + valptr = bond_opt_parse(bond_opt_get(BOND_OPT_AD_SELECT), + &newval); + if (!valptr) { + pr_err("Error: Invalid ad_select \"%s\"\n", ad_select); return -EINVAL; } - - if (bond_mode != BOND_MODE_8023AD) { - pr_warning("ad_select param only affects 802.3ad mode\n"); - } + params->ad_select = valptr->value; + if (bond_mode != BOND_MODE_8023AD) + pr_warn("ad_select param only affects 802.3ad mode\n"); } else { params->ad_select = BOND_AD_STABLE; } if (max_bonds < 0) { - pr_warning("Warning: max_bonds (%d) not in range %d-%d, so it was reset to BOND_DEFAULT_MAX_BONDS (%d)\n", - max_bonds, 0, INT_MAX, BOND_DEFAULT_MAX_BONDS); + pr_warn("Warning: max_bonds (%d) not in range %d-%d, so it was reset to BOND_DEFAULT_MAX_BONDS (%d)\n", + max_bonds, 0, INT_MAX, BOND_DEFAULT_MAX_BONDS); max_bonds = BOND_DEFAULT_MAX_BONDS; } if (miimon < 0) { - pr_warning("Warning: miimon module parameter (%d), not in range 0-%d, so it was reset to %d\n", - miimon, INT_MAX, BOND_LINK_MON_INTERV); - miimon = BOND_LINK_MON_INTERV; + pr_warn("Warning: miimon module parameter (%d), not in range 0-%d, so it was reset to 0\n", + miimon, INT_MAX); + miimon = 0; } if (updelay < 0) { - pr_warning("Warning: updelay module parameter (%d), not in range 0-%d, so it was reset to 0\n", - updelay, INT_MAX); + pr_warn("Warning: updelay module parameter (%d), not in range 0-%d, so it was reset to 0\n", + updelay, INT_MAX); updelay = 0; } if (downdelay < 0) { - pr_warning("Warning: downdelay module parameter (%d), not in range 0-%d, so it was reset to 0\n", - downdelay, INT_MAX); + pr_warn("Warning: downdelay module parameter (%d), not in range 0-%d, so it was reset to 0\n", + downdelay, INT_MAX); downdelay = 0; } if ((use_carrier != 0) && (use_carrier != 1)) { - pr_warning("Warning: use_carrier module parameter (%d), not of valid value (0/1), so it was set to 1\n", - use_carrier); + pr_warn("Warning: use_carrier module parameter (%d), not of valid value (0/1), so it was set to 1\n", + use_carrier); use_carrier = 1; } if (num_peer_notif < 0 || num_peer_notif > 255) { - pr_warning("Warning: num_grat_arp/num_unsol_na (%d) not in range 0-255 so it was reset to 1\n", - num_peer_notif); + pr_warn("Warning: num_grat_arp/num_unsol_na (%d) not in range 0-255 so it was reset to 1\n", + num_peer_notif); num_peer_notif = 1; } /* reset values for 802.3ad/TLB/ALB */ - if (BOND_NO_USES_ARP(bond_mode)) { + if (!bond_mode_uses_arp(bond_mode)) { if (!miimon) { - pr_warning("Warning: miimon must be specified, otherwise bonding will not detect link failure, speed and duplex which are essential for 802.3ad operation\n"); - pr_warning("Forcing miimon to 100msec\n"); + pr_warn("Warning: miimon must be specified, otherwise bonding will not detect link failure, speed and duplex which are essential for 802.3ad operation\n"); + pr_warn("Forcing miimon to 100msec\n"); miimon = BOND_DEFAULT_MIIMON; } } if (tx_queues < 1 || tx_queues > 255) { - pr_warning("Warning: tx_queues (%d) should be between " - "1 and 255, resetting to %d\n", - tx_queues, BOND_DEFAULT_TX_QUEUES); + pr_warn("Warning: tx_queues (%d) should be between 1 and 255, resetting to %d\n", + tx_queues, BOND_DEFAULT_TX_QUEUES); tx_queues = BOND_DEFAULT_TX_QUEUES; } if ((all_slaves_active != 0) && (all_slaves_active != 1)) { - pr_warning("Warning: all_slaves_active module parameter (%d), " - "not of valid value (0/1), so it was set to " - "0\n", all_slaves_active); + pr_warn("Warning: all_slaves_active module parameter (%d), not of valid value (0/1), so it was set to 0\n", + all_slaves_active); all_slaves_active = 0; } if (resend_igmp < 0 || resend_igmp > 255) { - pr_warning("Warning: resend_igmp (%d) should be between " - "0 and 255, resetting to %d\n", - resend_igmp, BOND_DEFAULT_RESEND_IGMP); + pr_warn("Warning: resend_igmp (%d) should be between 0 and 255, resetting to %d\n", + resend_igmp, BOND_DEFAULT_RESEND_IGMP); resend_igmp = BOND_DEFAULT_RESEND_IGMP; } - if (packets_per_slave < 0 || packets_per_slave > USHRT_MAX) { + bond_opt_initval(&newval, packets_per_slave); + if (!bond_opt_parse(bond_opt_get(BOND_OPT_PACKETS_PER_SLAVE), &newval)) { pr_warn("Warning: packets_per_slave (%d) should be between 0 and %u resetting to 1\n", packets_per_slave, USHRT_MAX); packets_per_slave = 1; @@ -4168,38 +4162,37 @@ static int bond_check_params(struct bond_params *params) /* just warn the user the up/down delay will have * no effect since miimon is zero... */ - pr_warning("Warning: miimon module parameter not set and updelay (%d) or downdelay (%d) module parameter is set; updelay and downdelay have no effect unless miimon is set\n", - updelay, downdelay); + pr_warn("Warning: miimon module parameter not set and updelay (%d) or downdelay (%d) module parameter is set; updelay and downdelay have no effect unless miimon is set\n", + updelay, downdelay); } } else { /* don't allow arp monitoring */ if (arp_interval) { - pr_warning("Warning: miimon (%d) and arp_interval (%d) can't be used simultaneously, disabling ARP monitoring\n", - miimon, arp_interval); + pr_warn("Warning: miimon (%d) and arp_interval (%d) can't be used simultaneously, disabling ARP monitoring\n", + miimon, arp_interval); arp_interval = 0; } if ((updelay % miimon) != 0) { - pr_warning("Warning: updelay (%d) is not a multiple of miimon (%d), updelay rounded to %d ms\n", - updelay, miimon, - (updelay / miimon) * miimon); + pr_warn("Warning: updelay (%d) is not a multiple of miimon (%d), updelay rounded to %d ms\n", + updelay, miimon, (updelay / miimon) * miimon); } updelay /= miimon; if ((downdelay % miimon) != 0) { - pr_warning("Warning: downdelay (%d) is not a multiple of miimon (%d), downdelay rounded to %d ms\n", - downdelay, miimon, - (downdelay / miimon) * miimon); + pr_warn("Warning: downdelay (%d) is not a multiple of miimon (%d), downdelay rounded to %d ms\n", + downdelay, miimon, + (downdelay / miimon) * miimon); } downdelay /= miimon; } if (arp_interval < 0) { - pr_warning("Warning: arp_interval module parameter (%d) , not in range 0-%d, so it was reset to %d\n", - arp_interval, INT_MAX, BOND_LINK_ARP_INTERV); - arp_interval = BOND_LINK_ARP_INTERV; + pr_warn("Warning: arp_interval module parameter (%d), not in range 0-%d, so it was reset to 0\n", + arp_interval, INT_MAX); + arp_interval = 0; } for (arp_ip_count = 0, i = 0; @@ -4208,118 +4201,121 @@ static int bond_check_params(struct bond_params *params) catch mistakes */ __be32 ip; if (!in4_pton(arp_ip_target[i], -1, (u8 *)&ip, -1, NULL) || - IS_IP_TARGET_UNUSABLE_ADDRESS(ip)) { - pr_warning("Warning: bad arp_ip_target module parameter (%s), ARP monitoring will not be performed\n", - arp_ip_target[i]); + !bond_is_ip_target_ok(ip)) { + pr_warn("Warning: bad arp_ip_target module parameter (%s), ARP monitoring will not be performed\n", + arp_ip_target[i]); arp_interval = 0; } else { if (bond_get_targets_ip(arp_target, ip) == -1) arp_target[arp_ip_count++] = ip; else - pr_warning("Warning: duplicate address %pI4 in arp_ip_target, skipping\n", - &ip); + pr_warn("Warning: duplicate address %pI4 in arp_ip_target, skipping\n", + &ip); } } if (arp_interval && !arp_ip_count) { /* don't allow arping if no arp_ip_target given... */ - pr_warning("Warning: arp_interval module parameter (%d) specified without providing an arp_ip_target parameter, arp_interval was reset to 0\n", - arp_interval); + pr_warn("Warning: arp_interval module parameter (%d) specified without providing an arp_ip_target parameter, arp_interval was reset to 0\n", + arp_interval); arp_interval = 0; } if (arp_validate) { - if (bond_mode != BOND_MODE_ACTIVEBACKUP) { - pr_err("arp_validate only supported in active-backup mode\n"); - return -EINVAL; - } if (!arp_interval) { pr_err("arp_validate requires arp_interval\n"); return -EINVAL; } - arp_validate_value = bond_parse_parm(arp_validate, - arp_validate_tbl); - if (arp_validate_value == -1) { + bond_opt_initstr(&newval, arp_validate); + valptr = bond_opt_parse(bond_opt_get(BOND_OPT_ARP_VALIDATE), + &newval); + if (!valptr) { pr_err("Error: invalid arp_validate \"%s\"\n", - arp_validate == NULL ? "NULL" : arp_validate); + arp_validate); return -EINVAL; } - } else + arp_validate_value = valptr->value; + } else { arp_validate_value = 0; + } arp_all_targets_value = 0; if (arp_all_targets) { - arp_all_targets_value = bond_parse_parm(arp_all_targets, - arp_all_targets_tbl); - - if (arp_all_targets_value == -1) { + bond_opt_initstr(&newval, arp_all_targets); + valptr = bond_opt_parse(bond_opt_get(BOND_OPT_ARP_ALL_TARGETS), + &newval); + if (!valptr) { pr_err("Error: invalid arp_all_targets_value \"%s\"\n", arp_all_targets); arp_all_targets_value = 0; + } else { + arp_all_targets_value = valptr->value; } } if (miimon) { pr_info("MII link monitoring set to %d ms\n", miimon); } else if (arp_interval) { + valptr = bond_opt_get_val(BOND_OPT_ARP_VALIDATE, + arp_validate_value); pr_info("ARP monitoring set to %d ms, validate %s, with %d target(s):", - arp_interval, - arp_validate_tbl[arp_validate_value].modename, - arp_ip_count); + arp_interval, valptr->string, arp_ip_count); for (i = 0; i < arp_ip_count; i++) - pr_info(" %s", arp_ip_target[i]); + pr_cont(" %s", arp_ip_target[i]); - pr_info("\n"); + pr_cont("\n"); } else if (max_bonds) { /* miimon and arp_interval not set, we need one so things * work as expected, see bonding.txt for details */ - pr_debug("Warning: either miimon or arp_interval and arp_ip_target module parameters must be specified, otherwise bonding will not detect link failures! see bonding.txt for details.\n"); + pr_debug("Warning: either miimon or arp_interval and arp_ip_target module parameters must be specified, otherwise bonding will not detect link failures! see bonding.txt for details\n"); } - if (primary && !USES_PRIMARY(bond_mode)) { + if (primary && !bond_mode_uses_primary(bond_mode)) { /* currently, using a primary only makes sense * in active backup, TLB or ALB modes */ - pr_warning("Warning: %s primary device specified but has no effect in %s mode\n", - primary, bond_mode_name(bond_mode)); + pr_warn("Warning: %s primary device specified but has no effect in %s mode\n", + primary, bond_mode_name(bond_mode)); primary = NULL; } if (primary && primary_reselect) { - primary_reselect_value = bond_parse_parm(primary_reselect, - pri_reselect_tbl); - if (primary_reselect_value == -1) { + bond_opt_initstr(&newval, primary_reselect); + valptr = bond_opt_parse(bond_opt_get(BOND_OPT_PRIMARY_RESELECT), + &newval); + if (!valptr) { pr_err("Error: Invalid primary_reselect \"%s\"\n", - primary_reselect == - NULL ? "NULL" : primary_reselect); + primary_reselect); return -EINVAL; } + primary_reselect_value = valptr->value; } else { primary_reselect_value = BOND_PRI_RESELECT_ALWAYS; } if (fail_over_mac) { - fail_over_mac_value = bond_parse_parm(fail_over_mac, - fail_over_mac_tbl); - if (fail_over_mac_value == -1) { + bond_opt_initstr(&newval, fail_over_mac); + valptr = bond_opt_parse(bond_opt_get(BOND_OPT_FAIL_OVER_MAC), + &newval); + if (!valptr) { pr_err("Error: invalid fail_over_mac \"%s\"\n", - arp_validate == NULL ? "NULL" : arp_validate); + fail_over_mac); return -EINVAL; } - + fail_over_mac_value = valptr->value; if (bond_mode != BOND_MODE_ACTIVEBACKUP) - pr_warning("Warning: fail_over_mac only affects active-backup mode.\n"); + pr_warn("Warning: fail_over_mac only affects active-backup mode\n"); } else { fail_over_mac_value = BOND_FOM_NONE; } if (lp_interval == 0) { - pr_warning("Warning: ip_interval must be between 1 and %d, so it was reset to %d\n", - INT_MAX, BOND_ALB_DEFAULT_LP_INTERVAL); + pr_warn("Warning: ip_interval must be between 1 and %d, so it was reset to %d\n", + INT_MAX, BOND_ALB_DEFAULT_LP_INTERVAL); lp_interval = BOND_ALB_DEFAULT_LP_INTERVAL; } @@ -4343,10 +4339,19 @@ static int bond_check_params(struct bond_params *params) params->resend_igmp = resend_igmp; params->min_links = min_links; params->lp_interval = lp_interval; - if (packets_per_slave > 1) - params->packets_per_slave = reciprocal_value(packets_per_slave); - else - params->packets_per_slave = packets_per_slave; + params->packets_per_slave = packets_per_slave; + params->tlb_dynamic_lb = 1; /* Default value */ + if (packets_per_slave > 0) { + params->reciprocal_packets_per_slave = + reciprocal_value(packets_per_slave); + } else { + /* reciprocal_packets_per_slave is unused if + * packets_per_slave is 0 or 1, just initialize it + */ + params->reciprocal_packets_per_slave = + (struct reciprocal_value) { 0 }; + } + if (primary) { strncpy(params->primary, primary, IFNAMSIZ); params->primary[IFNAMSIZ - 1] = 0; @@ -4524,6 +4529,7 @@ static int __init bonding_init(void) out: return res; err: + bond_destroy_debugfs(); bond_netlink_fini(); err_link: unregister_pernet_subsys(&bond_net_ops); diff --git a/drivers/net/bonding/bond_netlink.c b/drivers/net/bonding/bond_netlink.c index 21c648854a8..5ab3c1847e6 100644 --- a/drivers/net/bonding/bond_netlink.c +++ b/drivers/net/bonding/bond_netlink.c @@ -19,35 +19,49 @@ #include <linux/if_ether.h> #include <net/netlink.h> #include <net/rtnetlink.h> -#include <linux/reciprocal_div.h> #include "bonding.h" -int bond_get_slave(struct net_device *slave_dev, struct sk_buff *skb) +static size_t bond_get_slave_size(const struct net_device *bond_dev, + const struct net_device *slave_dev) +{ + return nla_total_size(sizeof(u8)) + /* IFLA_BOND_SLAVE_STATE */ + nla_total_size(sizeof(u8)) + /* IFLA_BOND_SLAVE_MII_STATUS */ + nla_total_size(sizeof(u32)) + /* IFLA_BOND_SLAVE_LINK_FAILURE_COUNT */ + nla_total_size(MAX_ADDR_LEN) + /* IFLA_BOND_SLAVE_PERM_HWADDR */ + nla_total_size(sizeof(u16)) + /* IFLA_BOND_SLAVE_QUEUE_ID */ + nla_total_size(sizeof(u16)) + /* IFLA_BOND_SLAVE_AD_AGGREGATOR_ID */ + 0; +} + +static int bond_fill_slave_info(struct sk_buff *skb, + const struct net_device *bond_dev, + const struct net_device *slave_dev) { struct slave *slave = bond_slave_get_rtnl(slave_dev); - const struct aggregator *agg; - if (nla_put_u8(skb, IFLA_SLAVE_STATE, bond_slave_state(slave))) + if (nla_put_u8(skb, IFLA_BOND_SLAVE_STATE, bond_slave_state(slave))) goto nla_put_failure; - if (nla_put_u8(skb, IFLA_SLAVE_MII_STATUS, slave->link)) + if (nla_put_u8(skb, IFLA_BOND_SLAVE_MII_STATUS, slave->link)) goto nla_put_failure; - if (nla_put_u32(skb, IFLA_SLAVE_LINK_FAILURE_COUNT, + if (nla_put_u32(skb, IFLA_BOND_SLAVE_LINK_FAILURE_COUNT, slave->link_failure_count)) goto nla_put_failure; - if (nla_put(skb, IFLA_SLAVE_PERM_HWADDR, + if (nla_put(skb, IFLA_BOND_SLAVE_PERM_HWADDR, slave_dev->addr_len, slave->perm_hwaddr)) goto nla_put_failure; - if (nla_put_u16(skb, IFLA_SLAVE_QUEUE_ID, slave->queue_id)) + if (nla_put_u16(skb, IFLA_BOND_SLAVE_QUEUE_ID, slave->queue_id)) goto nla_put_failure; - if (slave->bond->params.mode == BOND_MODE_8023AD) { - agg = SLAVE_AD_INFO(slave).port.aggregator; + if (BOND_MODE(slave->bond) == BOND_MODE_8023AD) { + const struct aggregator *agg; + + agg = SLAVE_AD_INFO(slave)->port.aggregator; if (agg) - if (nla_put_u16(skb, IFLA_SLAVE_AD_AGGREGATOR_ID, + if (nla_put_u16(skb, IFLA_BOND_SLAVE_AD_AGGREGATOR_ID, agg->aggregator_identifier)) goto nla_put_failure; } @@ -99,6 +113,7 @@ static int bond_changelink(struct net_device *bond_dev, struct nlattr *tb[], struct nlattr *data[]) { struct bonding *bond = netdev_priv(bond_dev); + struct bond_opt_value newval; int miimon = 0; int err; @@ -108,51 +123,57 @@ static int bond_changelink(struct net_device *bond_dev, if (data[IFLA_BOND_MODE]) { int mode = nla_get_u8(data[IFLA_BOND_MODE]); - err = bond_option_mode_set(bond, mode); + bond_opt_initval(&newval, mode); + err = __bond_opt_set(bond, BOND_OPT_MODE, &newval); if (err) return err; } if (data[IFLA_BOND_ACTIVE_SLAVE]) { int ifindex = nla_get_u32(data[IFLA_BOND_ACTIVE_SLAVE]); struct net_device *slave_dev; + char *active_slave = ""; - if (ifindex == 0) { - slave_dev = NULL; - } else { + if (ifindex != 0) { slave_dev = __dev_get_by_index(dev_net(bond_dev), ifindex); if (!slave_dev) return -ENODEV; + active_slave = slave_dev->name; } - err = bond_option_active_slave_set(bond, slave_dev); + bond_opt_initstr(&newval, active_slave); + err = __bond_opt_set(bond, BOND_OPT_ACTIVE_SLAVE, &newval); if (err) return err; } if (data[IFLA_BOND_MIIMON]) { miimon = nla_get_u32(data[IFLA_BOND_MIIMON]); - err = bond_option_miimon_set(bond, miimon); + bond_opt_initval(&newval, miimon); + err = __bond_opt_set(bond, BOND_OPT_MIIMON, &newval); if (err) return err; } if (data[IFLA_BOND_UPDELAY]) { int updelay = nla_get_u32(data[IFLA_BOND_UPDELAY]); - err = bond_option_updelay_set(bond, updelay); + bond_opt_initval(&newval, updelay); + err = __bond_opt_set(bond, BOND_OPT_UPDELAY, &newval); if (err) return err; } if (data[IFLA_BOND_DOWNDELAY]) { int downdelay = nla_get_u32(data[IFLA_BOND_DOWNDELAY]); - err = bond_option_downdelay_set(bond, downdelay); + bond_opt_initval(&newval, downdelay); + err = __bond_opt_set(bond, BOND_OPT_DOWNDELAY, &newval); if (err) return err; } if (data[IFLA_BOND_USE_CARRIER]) { int use_carrier = nla_get_u8(data[IFLA_BOND_USE_CARRIER]); - err = bond_option_use_carrier_set(bond, use_carrier); + bond_opt_initval(&newval, use_carrier); + err = __bond_opt_set(bond, BOND_OPT_USE_CARRIER, &newval); if (err) return err; } @@ -160,26 +181,34 @@ static int bond_changelink(struct net_device *bond_dev, int arp_interval = nla_get_u32(data[IFLA_BOND_ARP_INTERVAL]); if (arp_interval && miimon) { - pr_err("%s: ARP monitoring cannot be used with MII monitoring.\n", + pr_err("%s: ARP monitoring cannot be used with MII monitoring\n", bond->dev->name); return -EINVAL; } - err = bond_option_arp_interval_set(bond, arp_interval); + bond_opt_initval(&newval, arp_interval); + err = __bond_opt_set(bond, BOND_OPT_ARP_INTERVAL, &newval); if (err) return err; } if (data[IFLA_BOND_ARP_IP_TARGET]) { - __be32 targets[BOND_MAX_ARP_TARGETS] = { 0, }; struct nlattr *attr; int i = 0, rem; + bond_option_arp_ip_targets_clear(bond); nla_for_each_nested(attr, data[IFLA_BOND_ARP_IP_TARGET], rem) { __be32 target = nla_get_be32(attr); - targets[i++] = target; - } - err = bond_option_arp_ip_targets_set(bond, targets, i); + bond_opt_initval(&newval, (__force u64)target); + err = __bond_opt_set(bond, BOND_OPT_ARP_TARGETS, + &newval); + if (err) + break; + i++; + } + if (i == 0 && bond->params.arp_interval) + pr_warn("%s: Removing last arp target with arp_interval on\n", + bond->dev->name); if (err) return err; } @@ -187,12 +216,13 @@ static int bond_changelink(struct net_device *bond_dev, int arp_validate = nla_get_u32(data[IFLA_BOND_ARP_VALIDATE]); if (arp_validate && miimon) { - pr_err("%s: ARP validating cannot be used with MII monitoring.\n", + pr_err("%s: ARP validating cannot be used with MII monitoring\n", bond->dev->name); return -EINVAL; } - err = bond_option_arp_validate_set(bond, arp_validate); + bond_opt_initval(&newval, arp_validate); + err = __bond_opt_set(bond, BOND_OPT_ARP_VALIDATE, &newval); if (err) return err; } @@ -200,7 +230,8 @@ static int bond_changelink(struct net_device *bond_dev, int arp_all_targets = nla_get_u32(data[IFLA_BOND_ARP_ALL_TARGETS]); - err = bond_option_arp_all_targets_set(bond, arp_all_targets); + bond_opt_initval(&newval, arp_all_targets); + err = __bond_opt_set(bond, BOND_OPT_ARP_ALL_TARGETS, &newval); if (err) return err; } @@ -213,7 +244,8 @@ static int bond_changelink(struct net_device *bond_dev, if (dev) primary = dev->name; - err = bond_option_primary_set(bond, primary); + bond_opt_initstr(&newval, primary); + err = __bond_opt_set(bond, BOND_OPT_PRIMARY, &newval); if (err) return err; } @@ -221,7 +253,8 @@ static int bond_changelink(struct net_device *bond_dev, int primary_reselect = nla_get_u8(data[IFLA_BOND_PRIMARY_RESELECT]); - err = bond_option_primary_reselect_set(bond, primary_reselect); + bond_opt_initval(&newval, primary_reselect); + err = __bond_opt_set(bond, BOND_OPT_PRIMARY_RESELECT, &newval); if (err) return err; } @@ -229,7 +262,8 @@ static int bond_changelink(struct net_device *bond_dev, int fail_over_mac = nla_get_u8(data[IFLA_BOND_FAIL_OVER_MAC]); - err = bond_option_fail_over_mac_set(bond, fail_over_mac); + bond_opt_initval(&newval, fail_over_mac); + err = __bond_opt_set(bond, BOND_OPT_FAIL_OVER_MAC, &newval); if (err) return err; } @@ -237,7 +271,8 @@ static int bond_changelink(struct net_device *bond_dev, int xmit_hash_policy = nla_get_u8(data[IFLA_BOND_XMIT_HASH_POLICY]); - err = bond_option_xmit_hash_policy_set(bond, xmit_hash_policy); + bond_opt_initval(&newval, xmit_hash_policy); + err = __bond_opt_set(bond, BOND_OPT_XMIT_HASH, &newval); if (err) return err; } @@ -245,7 +280,8 @@ static int bond_changelink(struct net_device *bond_dev, int resend_igmp = nla_get_u32(data[IFLA_BOND_RESEND_IGMP]); - err = bond_option_resend_igmp_set(bond, resend_igmp); + bond_opt_initval(&newval, resend_igmp); + err = __bond_opt_set(bond, BOND_OPT_RESEND_IGMP, &newval); if (err) return err; } @@ -253,7 +289,8 @@ static int bond_changelink(struct net_device *bond_dev, int num_peer_notif = nla_get_u8(data[IFLA_BOND_NUM_PEER_NOTIF]); - err = bond_option_num_peer_notif_set(bond, num_peer_notif); + bond_opt_initval(&newval, num_peer_notif); + err = __bond_opt_set(bond, BOND_OPT_NUM_PEER_NOTIF, &newval); if (err) return err; } @@ -261,8 +298,8 @@ static int bond_changelink(struct net_device *bond_dev, int all_slaves_active = nla_get_u8(data[IFLA_BOND_ALL_SLAVES_ACTIVE]); - err = bond_option_all_slaves_active_set(bond, - all_slaves_active); + bond_opt_initval(&newval, all_slaves_active); + err = __bond_opt_set(bond, BOND_OPT_ALL_SLAVES_ACTIVE, &newval); if (err) return err; } @@ -270,7 +307,8 @@ static int bond_changelink(struct net_device *bond_dev, int min_links = nla_get_u32(data[IFLA_BOND_MIN_LINKS]); - err = bond_option_min_links_set(bond, min_links); + bond_opt_initval(&newval, min_links); + err = __bond_opt_set(bond, BOND_OPT_MINLINKS, &newval); if (err) return err; } @@ -278,7 +316,8 @@ static int bond_changelink(struct net_device *bond_dev, int lp_interval = nla_get_u32(data[IFLA_BOND_LP_INTERVAL]); - err = bond_option_lp_interval_set(bond, lp_interval); + bond_opt_initval(&newval, lp_interval); + err = __bond_opt_set(bond, BOND_OPT_LP_INTERVAL, &newval); if (err) return err; } @@ -286,8 +325,8 @@ static int bond_changelink(struct net_device *bond_dev, int packets_per_slave = nla_get_u32(data[IFLA_BOND_PACKETS_PER_SLAVE]); - err = bond_option_packets_per_slave_set(bond, - packets_per_slave); + bond_opt_initval(&newval, packets_per_slave); + err = __bond_opt_set(bond, BOND_OPT_PACKETS_PER_SLAVE, &newval); if (err) return err; } @@ -295,7 +334,8 @@ static int bond_changelink(struct net_device *bond_dev, int lacp_rate = nla_get_u8(data[IFLA_BOND_AD_LACP_RATE]); - err = bond_option_lacp_rate_set(bond, lacp_rate); + bond_opt_initval(&newval, lacp_rate); + err = __bond_opt_set(bond, BOND_OPT_LACP_RATE, &newval); if (err) return err; } @@ -303,7 +343,8 @@ static int bond_changelink(struct net_device *bond_dev, int ad_select = nla_get_u8(data[IFLA_BOND_AD_SELECT]); - err = bond_option_ad_select_set(bond, ad_select); + bond_opt_initval(&newval, ad_select); + err = __bond_opt_set(bond, BOND_OPT_AD_SELECT, &newval); if (err) return err; } @@ -366,7 +407,7 @@ static int bond_fill_info(struct sk_buff *skb, unsigned int packets_per_slave; int i, targets_added; - if (nla_put_u8(skb, IFLA_BOND_MODE, bond->params.mode)) + if (nla_put_u8(skb, IFLA_BOND_MODE, BOND_MODE(bond))) goto nla_put_failure; if (slave_dev && @@ -452,9 +493,6 @@ static int bond_fill_info(struct sk_buff *skb, goto nla_put_failure; packets_per_slave = bond->params.packets_per_slave; - if (packets_per_slave > 1) - packets_per_slave = reciprocal_value(packets_per_slave); - if (nla_put_u32(skb, IFLA_BOND_PACKETS_PER_SLAVE, packets_per_slave)) goto nla_put_failure; @@ -467,7 +505,7 @@ static int bond_fill_info(struct sk_buff *skb, bond->params.ad_select)) goto nla_put_failure; - if (bond->params.mode == BOND_MODE_8023AD) { + if (BOND_MODE(bond) == BOND_MODE_8023AD) { struct ad_info info; if (!bond_3ad_get_active_agg_info(bond, &info)) { @@ -518,6 +556,8 @@ struct rtnl_link_ops bond_link_ops __read_mostly = { .get_num_tx_queues = bond_get_num_tx_queues, .get_num_rx_queues = bond_get_num_tx_queues, /* Use the same number as for TX queues */ + .get_slave_size = bond_get_slave_size, + .fill_slave_info = bond_fill_slave_info, }; int __init bond_netlink_init(void) diff --git a/drivers/net/bonding/bond_options.c b/drivers/net/bonding/bond_options.c index 945a6668da8..540e0167bf2 100644 --- a/drivers/net/bonding/bond_options.c +++ b/drivers/net/bonding/bond_options.c @@ -14,52 +14,687 @@ #include <linux/errno.h> #include <linux/if.h> #include <linux/netdevice.h> -#include <linux/rwlock.h> +#include <linux/spinlock.h> #include <linux/rcupdate.h> -#include <linux/reciprocal_div.h> +#include <linux/ctype.h> +#include <linux/inet.h> #include "bonding.h" -int bond_option_mode_set(struct bonding *bond, int mode) +static int bond_option_active_slave_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_miimon_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_updelay_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_downdelay_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_use_carrier_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_arp_interval_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_arp_ip_target_add(struct bonding *bond, __be32 target); +static int bond_option_arp_ip_target_rem(struct bonding *bond, __be32 target); +static int bond_option_arp_ip_targets_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_arp_validate_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_arp_all_targets_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_primary_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_primary_reselect_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_fail_over_mac_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_xmit_hash_policy_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_resend_igmp_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_num_peer_notif_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_all_slaves_active_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_min_links_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_lp_interval_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_pps_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_lacp_rate_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_ad_select_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_queue_id_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_mode_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_slaves_set(struct bonding *bond, + const struct bond_opt_value *newval); +static int bond_option_tlb_dynamic_lb_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}, + { "active-backup", BOND_MODE_ACTIVEBACKUP, 0}, + { "balance-xor", BOND_MODE_XOR, 0}, + { "broadcast", BOND_MODE_BROADCAST, 0}, + { "802.3ad", BOND_MODE_8023AD, 0}, + { "balance-tlb", BOND_MODE_TLB, 0}, + { "balance-alb", BOND_MODE_ALB, 0}, + { NULL, -1, 0}, +}; + +static const struct bond_opt_value bond_pps_tbl[] = { + { "default", 1, BOND_VALFLAG_DEFAULT}, + { "maxval", USHRT_MAX, BOND_VALFLAG_MAX}, + { NULL, -1, 0}, +}; + +static const struct bond_opt_value bond_xmit_hashtype_tbl[] = { + { "layer2", BOND_XMIT_POLICY_LAYER2, BOND_VALFLAG_DEFAULT}, + { "layer3+4", BOND_XMIT_POLICY_LAYER34, 0}, + { "layer2+3", BOND_XMIT_POLICY_LAYER23, 0}, + { "encap2+3", BOND_XMIT_POLICY_ENCAP23, 0}, + { "encap3+4", BOND_XMIT_POLICY_ENCAP34, 0}, + { NULL, -1, 0}, +}; + +static const struct bond_opt_value bond_arp_validate_tbl[] = { + { "none", BOND_ARP_VALIDATE_NONE, BOND_VALFLAG_DEFAULT}, + { "active", BOND_ARP_VALIDATE_ACTIVE, 0}, + { "backup", BOND_ARP_VALIDATE_BACKUP, 0}, + { "all", BOND_ARP_VALIDATE_ALL, 0}, + { "filter", BOND_ARP_FILTER, 0}, + { "filter_active", BOND_ARP_FILTER_ACTIVE, 0}, + { "filter_backup", BOND_ARP_FILTER_BACKUP, 0}, + { NULL, -1, 0}, +}; + +static const struct bond_opt_value bond_arp_all_targets_tbl[] = { + { "any", BOND_ARP_TARGETS_ANY, BOND_VALFLAG_DEFAULT}, + { "all", BOND_ARP_TARGETS_ALL, 0}, + { NULL, -1, 0}, +}; + +static const struct bond_opt_value bond_fail_over_mac_tbl[] = { + { "none", BOND_FOM_NONE, BOND_VALFLAG_DEFAULT}, + { "active", BOND_FOM_ACTIVE, 0}, + { "follow", BOND_FOM_FOLLOW, 0}, + { NULL, -1, 0}, +}; + +static const struct bond_opt_value bond_intmax_tbl[] = { + { "off", 0, BOND_VALFLAG_DEFAULT}, + { "maxval", INT_MAX, BOND_VALFLAG_MAX}, + { NULL, -1, 0} +}; + +static const struct bond_opt_value bond_lacp_rate_tbl[] = { + { "slow", AD_LACP_SLOW, 0}, + { "fast", AD_LACP_FAST, 0}, + { NULL, -1, 0}, +}; + +static const struct bond_opt_value bond_ad_select_tbl[] = { + { "stable", BOND_AD_STABLE, BOND_VALFLAG_DEFAULT}, + { "bandwidth", BOND_AD_BANDWIDTH, 0}, + { "count", BOND_AD_COUNT, 0}, + { NULL, -1, 0}, +}; + +static const struct bond_opt_value bond_num_peer_notif_tbl[] = { + { "off", 0, 0}, + { "maxval", 255, BOND_VALFLAG_MAX}, + { "default", 1, BOND_VALFLAG_DEFAULT}, + { NULL, -1, 0} +}; + +static const struct bond_opt_value bond_primary_reselect_tbl[] = { + { "always", BOND_PRI_RESELECT_ALWAYS, BOND_VALFLAG_DEFAULT}, + { "better", BOND_PRI_RESELECT_BETTER, 0}, + { "failure", BOND_PRI_RESELECT_FAILURE, 0}, + { NULL, -1}, +}; + +static const struct bond_opt_value bond_use_carrier_tbl[] = { + { "off", 0, 0}, + { "on", 1, BOND_VALFLAG_DEFAULT}, + { NULL, -1, 0} +}; + +static const struct bond_opt_value bond_all_slaves_active_tbl[] = { + { "off", 0, BOND_VALFLAG_DEFAULT}, + { "on", 1, 0}, + { NULL, -1, 0} +}; + +static const struct bond_opt_value bond_resend_igmp_tbl[] = { + { "off", 0, 0}, + { "maxval", 255, BOND_VALFLAG_MAX}, + { "default", 1, BOND_VALFLAG_DEFAULT}, + { NULL, -1, 0} +}; + +static const struct bond_opt_value bond_lp_interval_tbl[] = { + { "minval", 1, BOND_VALFLAG_MIN | BOND_VALFLAG_DEFAULT}, + { "maxval", INT_MAX, BOND_VALFLAG_MAX}, + { NULL, -1, 0}, +}; + +static const struct bond_opt_value bond_tlb_dynamic_lb_tbl[] = { + { "off", 0, 0}, + { "on", 1, BOND_VALFLAG_DEFAULT}, + { NULL, -1, 0} +}; + +static const struct bond_option bond_opts[] = { + [BOND_OPT_MODE] = { + .id = BOND_OPT_MODE, + .name = "mode", + .desc = "bond device mode", + .flags = BOND_OPTFLAG_NOSLAVES | BOND_OPTFLAG_IFDOWN, + .values = bond_mode_tbl, + .set = bond_option_mode_set + }, + [BOND_OPT_PACKETS_PER_SLAVE] = { + .id = BOND_OPT_PACKETS_PER_SLAVE, + .name = "packets_per_slave", + .desc = "Packets to send per slave in RR mode", + .unsuppmodes = BOND_MODE_ALL_EX(BIT(BOND_MODE_ROUNDROBIN)), + .values = bond_pps_tbl, + .set = bond_option_pps_set + }, + [BOND_OPT_XMIT_HASH] = { + .id = BOND_OPT_XMIT_HASH, + .name = "xmit_hash_policy", + .desc = "balance-xor, 802.3ad, and tlb hashing method", + .values = bond_xmit_hashtype_tbl, + .set = bond_option_xmit_hash_policy_set + }, + [BOND_OPT_ARP_VALIDATE] = { + .id = BOND_OPT_ARP_VALIDATE, + .name = "arp_validate", + .desc = "validate src/dst of ARP probes", + .unsuppmodes = BIT(BOND_MODE_8023AD) | BIT(BOND_MODE_TLB) | + BIT(BOND_MODE_ALB), + .values = bond_arp_validate_tbl, + .set = bond_option_arp_validate_set + }, + [BOND_OPT_ARP_ALL_TARGETS] = { + .id = BOND_OPT_ARP_ALL_TARGETS, + .name = "arp_all_targets", + .desc = "fail on any/all arp targets timeout", + .values = bond_arp_all_targets_tbl, + .set = bond_option_arp_all_targets_set + }, + [BOND_OPT_FAIL_OVER_MAC] = { + .id = BOND_OPT_FAIL_OVER_MAC, + .name = "fail_over_mac", + .desc = "For active-backup, do not set all slaves to the same MAC", + .flags = BOND_OPTFLAG_NOSLAVES, + .values = bond_fail_over_mac_tbl, + .set = bond_option_fail_over_mac_set + }, + [BOND_OPT_ARP_INTERVAL] = { + .id = BOND_OPT_ARP_INTERVAL, + .name = "arp_interval", + .desc = "arp interval in milliseconds", + .unsuppmodes = BIT(BOND_MODE_8023AD) | BIT(BOND_MODE_TLB) | + BIT(BOND_MODE_ALB), + .values = bond_intmax_tbl, + .set = bond_option_arp_interval_set + }, + [BOND_OPT_ARP_TARGETS] = { + .id = BOND_OPT_ARP_TARGETS, + .name = "arp_ip_target", + .desc = "arp targets in n.n.n.n form", + .flags = BOND_OPTFLAG_RAWVAL, + .set = bond_option_arp_ip_targets_set + }, + [BOND_OPT_DOWNDELAY] = { + .id = BOND_OPT_DOWNDELAY, + .name = "downdelay", + .desc = "Delay before considering link down, in milliseconds", + .values = bond_intmax_tbl, + .set = bond_option_downdelay_set + }, + [BOND_OPT_UPDELAY] = { + .id = BOND_OPT_UPDELAY, + .name = "updelay", + .desc = "Delay before considering link up, in milliseconds", + .values = bond_intmax_tbl, + .set = bond_option_updelay_set + }, + [BOND_OPT_LACP_RATE] = { + .id = BOND_OPT_LACP_RATE, + .name = "lacp_rate", + .desc = "LACPDU tx rate to request from 802.3ad partner", + .flags = BOND_OPTFLAG_IFDOWN, + .unsuppmodes = BOND_MODE_ALL_EX(BIT(BOND_MODE_8023AD)), + .values = bond_lacp_rate_tbl, + .set = bond_option_lacp_rate_set + }, + [BOND_OPT_MINLINKS] = { + .id = BOND_OPT_MINLINKS, + .name = "min_links", + .desc = "Minimum number of available links before turning on carrier", + .values = bond_intmax_tbl, + .set = bond_option_min_links_set + }, + [BOND_OPT_AD_SELECT] = { + .id = BOND_OPT_AD_SELECT, + .name = "ad_select", + .desc = "803.ad aggregation selection logic", + .flags = BOND_OPTFLAG_IFDOWN, + .values = bond_ad_select_tbl, + .set = bond_option_ad_select_set + }, + [BOND_OPT_NUM_PEER_NOTIF] = { + .id = BOND_OPT_NUM_PEER_NOTIF, + .name = "num_unsol_na", + .desc = "Number of peer notifications to send on failover event", + .values = bond_num_peer_notif_tbl, + .set = bond_option_num_peer_notif_set + }, + [BOND_OPT_MIIMON] = { + .id = BOND_OPT_MIIMON, + .name = "miimon", + .desc = "Link check interval in milliseconds", + .values = bond_intmax_tbl, + .set = bond_option_miimon_set + }, + [BOND_OPT_PRIMARY] = { + .id = BOND_OPT_PRIMARY, + .name = "primary", + .desc = "Primary network device to use", + .flags = BOND_OPTFLAG_RAWVAL, + .unsuppmodes = BOND_MODE_ALL_EX(BIT(BOND_MODE_ACTIVEBACKUP) | + BIT(BOND_MODE_TLB) | + BIT(BOND_MODE_ALB)), + .set = bond_option_primary_set + }, + [BOND_OPT_PRIMARY_RESELECT] = { + .id = BOND_OPT_PRIMARY_RESELECT, + .name = "primary_reselect", + .desc = "Reselect primary slave once it comes up", + .values = bond_primary_reselect_tbl, + .set = bond_option_primary_reselect_set + }, + [BOND_OPT_USE_CARRIER] = { + .id = BOND_OPT_USE_CARRIER, + .name = "use_carrier", + .desc = "Use netif_carrier_ok (vs MII ioctls) in miimon", + .values = bond_use_carrier_tbl, + .set = bond_option_use_carrier_set + }, + [BOND_OPT_ACTIVE_SLAVE] = { + .id = BOND_OPT_ACTIVE_SLAVE, + .name = "active_slave", + .desc = "Currently active slave", + .flags = BOND_OPTFLAG_RAWVAL, + .unsuppmodes = BOND_MODE_ALL_EX(BIT(BOND_MODE_ACTIVEBACKUP) | + BIT(BOND_MODE_TLB) | + BIT(BOND_MODE_ALB)), + .set = bond_option_active_slave_set + }, + [BOND_OPT_QUEUE_ID] = { + .id = BOND_OPT_QUEUE_ID, + .name = "queue_id", + .desc = "Set queue id of a slave", + .flags = BOND_OPTFLAG_RAWVAL, + .set = bond_option_queue_id_set + }, + [BOND_OPT_ALL_SLAVES_ACTIVE] = { + .id = BOND_OPT_ALL_SLAVES_ACTIVE, + .name = "all_slaves_active", + .desc = "Keep all frames received on an interface by setting active flag for all slaves", + .values = bond_all_slaves_active_tbl, + .set = bond_option_all_slaves_active_set + }, + [BOND_OPT_RESEND_IGMP] = { + .id = BOND_OPT_RESEND_IGMP, + .name = "resend_igmp", + .desc = "Number of IGMP membership reports to send on link failure", + .values = bond_resend_igmp_tbl, + .set = bond_option_resend_igmp_set + }, + [BOND_OPT_LP_INTERVAL] = { + .id = BOND_OPT_LP_INTERVAL, + .name = "lp_interval", + .desc = "The number of seconds between instances where the bonding driver sends learning packets to each slave's peer switch", + .values = bond_lp_interval_tbl, + .set = bond_option_lp_interval_set + }, + [BOND_OPT_SLAVES] = { + .id = BOND_OPT_SLAVES, + .name = "slaves", + .desc = "Slave membership management", + .flags = BOND_OPTFLAG_RAWVAL, + .set = bond_option_slaves_set + }, + [BOND_OPT_TLB_DYNAMIC_LB] = { + .id = BOND_OPT_TLB_DYNAMIC_LB, + .name = "tlb_dynamic_lb", + .desc = "Enable dynamic flow shuffling", + .unsuppmodes = BOND_MODE_ALL_EX(BIT(BOND_MODE_TLB)), + .values = bond_tlb_dynamic_lb_tbl, + .flags = BOND_OPTFLAG_IFDOWN, + .set = bond_option_tlb_dynamic_lb_set, + }, + { } +}; + +/* Searches for an option by name */ +const struct bond_option *bond_opt_get_by_name(const char *name) { - if (bond_parm_tbl_lookup(mode, bond_mode_tbl) < 0) { - pr_err("%s: Ignoring invalid mode value %d.\n", - bond->dev->name, mode); - return -EINVAL; + const struct bond_option *opt; + int option; + + for (option = 0; option < BOND_OPT_LAST; option++) { + opt = bond_opt_get(option); + if (opt && !strcmp(opt->name, name)) + return opt; } - if (bond->dev->flags & IFF_UP) { - pr_err("%s: unable to update mode because interface is up.\n", - bond->dev->name); - return -EPERM; + return NULL; +} + +/* Searches for a value in opt's values[] table */ +const struct bond_opt_value *bond_opt_get_val(unsigned int option, u64 val) +{ + const struct bond_option *opt; + int i; + + opt = bond_opt_get(option); + if (WARN_ON(!opt)) + return NULL; + for (i = 0; opt->values && opt->values[i].string; i++) + if (opt->values[i].value == val) + return &opt->values[i]; + + return NULL; +} + +/* Searches for a value in opt's values[] table which matches the flagmask */ +static const struct bond_opt_value *bond_opt_get_flags(const struct bond_option *opt, + u32 flagmask) +{ + int i; + + for (i = 0; opt->values && opt->values[i].string; i++) + if (opt->values[i].flags & flagmask) + return &opt->values[i]; + + return NULL; +} + +/* If maxval is missing then there's no range to check. In case minval is + * missing then it's considered to be 0. + */ +static bool bond_opt_check_range(const struct bond_option *opt, u64 val) +{ + const struct bond_opt_value *minval, *maxval; + + minval = bond_opt_get_flags(opt, BOND_VALFLAG_MIN); + maxval = bond_opt_get_flags(opt, BOND_VALFLAG_MAX); + if (!maxval || (minval && val < minval->value) || val > maxval->value) + return false; + + return true; +} + +/** + * bond_opt_parse - parse option value + * @opt: the option to parse against + * @val: value to parse + * + * This function tries to extract the value from @val and check if it's + * a possible match for the option and returns NULL if a match isn't found, + * or the struct_opt_value that matched. It also strips the new line from + * @val->string if it's present. + */ +const struct bond_opt_value *bond_opt_parse(const struct bond_option *opt, + struct bond_opt_value *val) +{ + char *p, valstr[BOND_OPT_MAX_NAMELEN + 1] = { 0, }; + const struct bond_opt_value *tbl; + const struct bond_opt_value *ret = NULL; + bool checkval; + int i, rv; + + /* No parsing if the option wants a raw val */ + if (opt->flags & BOND_OPTFLAG_RAWVAL) + return val; + + tbl = opt->values; + if (!tbl) + goto out; + + /* ULLONG_MAX is used to bypass string processing */ + checkval = val->value != ULLONG_MAX; + if (!checkval) { + if (!val->string) + goto out; + p = strchr(val->string, '\n'); + if (p) + *p = '\0'; + for (p = val->string; *p; p++) + if (!(isdigit(*p) || isspace(*p))) + break; + /* The following code extracts the string to match or the value + * and sets checkval appropriately + */ + if (*p) { + rv = sscanf(val->string, "%32s", valstr); + } else { + rv = sscanf(val->string, "%llu", &val->value); + checkval = true; + } + if (!rv) + goto out; } - if (bond_has_slaves(bond)) { - pr_err("%s: unable to update mode because bond has slaves.\n", - bond->dev->name); - return -EPERM; + for (i = 0; tbl[i].string; i++) { + /* Check for exact match */ + if (checkval) { + if (val->value == tbl[i].value) + ret = &tbl[i]; + } else { + if (!strcmp(valstr, "default") && + (tbl[i].flags & BOND_VALFLAG_DEFAULT)) + ret = &tbl[i]; + + if (!strcmp(valstr, tbl[i].string)) + ret = &tbl[i]; + } + /* Found an exact match */ + if (ret) + goto out; + } + /* Possible range match */ + if (checkval && bond_opt_check_range(opt, val->value)) + ret = val; +out: + return ret; +} + +/* Check opt's dependencies against bond mode and currently set options */ +static int bond_opt_check_deps(struct bonding *bond, + const struct bond_option *opt) +{ + struct bond_params *params = &bond->params; + + if (test_bit(params->mode, &opt->unsuppmodes)) + return -EACCES; + if ((opt->flags & BOND_OPTFLAG_NOSLAVES) && bond_has_slaves(bond)) + return -ENOTEMPTY; + if ((opt->flags & BOND_OPTFLAG_IFDOWN) && (bond->dev->flags & IFF_UP)) + return -EBUSY; + + return 0; +} + +static void bond_opt_dep_print(struct bonding *bond, + const struct bond_option *opt) +{ + const struct bond_opt_value *modeval; + struct bond_params *params; + + params = &bond->params; + modeval = bond_opt_get_val(BOND_OPT_MODE, params->mode); + if (test_bit(params->mode, &opt->unsuppmodes)) + pr_err("%s: option %s: mode dependency failed, not supported in mode %s(%llu)\n", + bond->dev->name, opt->name, + modeval->string, modeval->value); +} + +static void bond_opt_error_interpret(struct bonding *bond, + const struct bond_option *opt, + int error, const struct bond_opt_value *val) +{ + const struct bond_opt_value *minval, *maxval; + char *p; + + switch (error) { + case -EINVAL: + if (val) { + if (val->string) { + /* sometimes RAWVAL opts may have new lines */ + p = strchr(val->string, '\n'); + if (p) + *p = '\0'; + pr_err("%s: option %s: invalid value (%s)\n", + bond->dev->name, opt->name, val->string); + } else { + pr_err("%s: option %s: invalid value (%llu)\n", + bond->dev->name, opt->name, val->value); + } + } + minval = bond_opt_get_flags(opt, BOND_VALFLAG_MIN); + maxval = bond_opt_get_flags(opt, BOND_VALFLAG_MAX); + if (!maxval) + break; + pr_err("%s: option %s: allowed values %llu - %llu\n", + bond->dev->name, opt->name, minval ? minval->value : 0, + maxval->value); + break; + case -EACCES: + bond_opt_dep_print(bond, opt); + break; + case -ENOTEMPTY: + pr_err("%s: option %s: unable to set because the bond device has slaves\n", + bond->dev->name, opt->name); + break; + case -EBUSY: + pr_err("%s: option %s: unable to set because the bond device is up\n", + bond->dev->name, opt->name); + break; + default: + break; } +} + +/** + * __bond_opt_set - set a bonding option + * @bond: target bond device + * @option: option to set + * @val: value to set it to + * + * This function is used to change the bond's option value, it can be + * used for both enabling/changing an option and for disabling it. RTNL lock + * must be obtained before calling this function. + */ +int __bond_opt_set(struct bonding *bond, + unsigned int option, struct bond_opt_value *val) +{ + const struct bond_opt_value *retval = NULL; + const struct bond_option *opt; + int ret = -ENOENT; + + ASSERT_RTNL(); - if (BOND_NO_USES_ARP(mode) && bond->params.arp_interval) { + opt = bond_opt_get(option); + if (WARN_ON(!val) || WARN_ON(!opt)) + goto out; + ret = bond_opt_check_deps(bond, opt); + if (ret) + goto out; + retval = bond_opt_parse(opt, val); + if (!retval) { + ret = -EINVAL; + goto out; + } + ret = opt->set(bond, retval); +out: + if (ret) + bond_opt_error_interpret(bond, opt, ret, val); + + return ret; +} + +/** + * bond_opt_tryset_rtnl - try to acquire rtnl and call __bond_opt_set + * @bond: target bond device + * @option: option to set + * @buf: value to set it to + * + * This function tries to acquire RTNL without blocking and if successful + * calls __bond_opt_set. It is mainly used for sysfs option manipulation. + */ +int bond_opt_tryset_rtnl(struct bonding *bond, unsigned int option, char *buf) +{ + struct bond_opt_value optval; + int ret; + + if (!rtnl_trylock()) + return restart_syscall(); + bond_opt_initstr(&optval, buf); + ret = __bond_opt_set(bond, option, &optval); + rtnl_unlock(); + + return ret; +} + +/** + * bond_opt_get - get a pointer to an option + * @option: option for which to return a pointer + * + * This function checks if option is valid and if so returns a pointer + * to its entry in the bond_opts[] option array. + */ +const struct bond_option *bond_opt_get(unsigned int option) +{ + if (!BOND_OPT_VALID(option)) + return NULL; + + return &bond_opts[option]; +} + +int bond_option_mode_set(struct bonding *bond, const struct bond_opt_value *newval) +{ + if (!bond_mode_uses_arp(newval->value) && bond->params.arp_interval) { pr_info("%s: %s mode is incompatible with arp monitoring, start mii monitoring\n", - bond->dev->name, bond_mode_tbl[mode].modename); + bond->dev->name, newval->string); /* disable arp monitoring */ bond->params.arp_interval = 0; /* set miimon to default value */ bond->params.miimon = BOND_DEFAULT_MIIMON; - pr_info("%s: Setting MII monitoring interval to %d.\n", + pr_info("%s: Setting MII monitoring interval to %d\n", bond->dev->name, bond->params.miimon); } /* don't cache arp_validate between modes */ bond->params.arp_validate = BOND_ARP_VALIDATE_NONE; - bond->params.mode = mode; + bond->params.mode = newval->value; + return 0; } static struct net_device *__bond_option_active_slave_get(struct bonding *bond, struct slave *slave) { - return USES_PRIMARY(bond->params.mode) && slave ? slave->dev : NULL; + return bond_uses_primary(bond) && slave ? slave->dev : NULL; } struct net_device *bond_option_active_slave_get_rcu(struct bonding *bond) @@ -74,39 +709,43 @@ struct net_device *bond_option_active_slave_get(struct bonding *bond) return __bond_option_active_slave_get(bond, bond->curr_active_slave); } -int bond_option_active_slave_set(struct bonding *bond, - struct net_device *slave_dev) +static int bond_option_active_slave_set(struct bonding *bond, + const struct bond_opt_value *newval) { + char ifname[IFNAMSIZ] = { 0, }; + struct net_device *slave_dev; int ret = 0; + sscanf(newval->string, "%15s", ifname); /* IFNAMSIZ */ + if (!strlen(ifname) || newval->string[0] == '\n') { + slave_dev = NULL; + } else { + slave_dev = __dev_get_by_name(dev_net(bond->dev), ifname); + if (!slave_dev) + return -ENODEV; + } + if (slave_dev) { if (!netif_is_bond_slave(slave_dev)) { - pr_err("Device %s is not bonding slave.\n", + pr_err("Device %s is not bonding slave\n", slave_dev->name); return -EINVAL; } if (bond->dev != netdev_master_upper_dev_get(slave_dev)) { - pr_err("%s: Device %s is not our slave.\n", + pr_err("%s: Device %s is not our slave\n", bond->dev->name, slave_dev->name); return -EINVAL; } } - if (!USES_PRIMARY(bond->params.mode)) { - pr_err("%s: Unable to change active slave; %s is in mode %d\n", - bond->dev->name, bond->dev->name, bond->params.mode); - return -EINVAL; - } - block_netpoll_tx(); write_lock_bh(&bond->curr_slave_lock); /* check to see if we are clearing active */ if (!slave_dev) { - pr_info("%s: Clearing current active slave.\n", - bond->dev->name); - rcu_assign_pointer(bond->curr_active_slave, NULL); + pr_info("%s: Clearing current active slave\n", bond->dev->name); + RCU_INIT_POINTER(bond->curr_active_slave, NULL); bond_select_active_slave(bond); } else { struct slave *old_active = bond->curr_active_slave; @@ -116,16 +755,16 @@ int bond_option_active_slave_set(struct bonding *bond, if (new_active == old_active) { /* do nothing */ - pr_info("%s: %s is already the current active slave.\n", + pr_info("%s: %s is already the current active slave\n", bond->dev->name, new_active->dev->name); } else { if (old_active && (new_active->link == BOND_LINK_UP) && - IS_UP(new_active->dev)) { - pr_info("%s: Setting %s as active slave.\n", + bond_slave_is_up(new_active)) { + pr_info("%s: Setting %s as active slave\n", bond->dev->name, new_active->dev->name); bond_change_active_slave(bond, new_active); } else { - pr_err("%s: Could not set %s as active slave; either %s is down or the link is down.\n", + pr_err("%s: Could not set %s as active slave; either %s is down or the link is down\n", bond->dev->name, new_active->dev->name, new_active->dev->name); ret = -EINVAL; @@ -135,29 +774,30 @@ int bond_option_active_slave_set(struct bonding *bond, write_unlock_bh(&bond->curr_slave_lock); unblock_netpoll_tx(); + return ret; } -int bond_option_miimon_set(struct bonding *bond, int miimon) +/* There are two tricky bits here. First, if MII monitoring is activated, then + * we must disable ARP monitoring. Second, if the timer isn't running, we must + * start it. + */ +static int bond_option_miimon_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if (miimon < 0) { - pr_err("%s: Invalid miimon value %d not in range %d-%d; rejected.\n", - bond->dev->name, miimon, 0, INT_MAX); - return -EINVAL; - } - pr_info("%s: Setting MII monitoring interval to %d.\n", - bond->dev->name, miimon); - bond->params.miimon = miimon; + pr_info("%s: Setting MII monitoring interval to %llu\n", + bond->dev->name, newval->value); + bond->params.miimon = newval->value; if (bond->params.updelay) - pr_info("%s: Note: Updating updelay (to %d) since it is a multiple of the miimon value.\n", + pr_info("%s: Note: Updating updelay (to %d) since it is a multiple of the miimon value\n", bond->dev->name, bond->params.updelay * bond->params.miimon); if (bond->params.downdelay) - pr_info("%s: Note: Updating downdelay (to %d) since it is a multiple of the miimon value.\n", + pr_info("%s: Note: Updating downdelay (to %d) since it is a multiple of the miimon value\n", bond->dev->name, bond->params.downdelay * bond->params.miimon); - if (miimon && bond->params.arp_interval) { - pr_info("%s: MII monitoring cannot be used with ARP monitoring. Disabling ARP monitoring...\n", + if (newval->value && bond->params.arp_interval) { + pr_info("%s: MII monitoring cannot be used with ARP monitoring - disabling ARP monitoring...\n", bond->dev->name); bond->params.arp_interval = 0; if (bond->params.arp_validate) @@ -169,111 +809,97 @@ int bond_option_miimon_set(struct bonding *bond, int miimon) * timer will get fired off when the open function * is called. */ - if (!miimon) { + if (!newval->value) { cancel_delayed_work_sync(&bond->mii_work); } else { cancel_delayed_work_sync(&bond->arp_work); queue_delayed_work(bond->wq, &bond->mii_work, 0); } } + return 0; } -int bond_option_updelay_set(struct bonding *bond, int updelay) +/* Set up and down delays. These must be multiples of the + * MII monitoring value, and are stored internally as the multiplier. + * Thus, we must translate to MS for the real world. + */ +static int bond_option_updelay_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if (!(bond->params.miimon)) { + int value = newval->value; + + if (!bond->params.miimon) { pr_err("%s: Unable to set up delay as MII monitoring is disabled\n", bond->dev->name); return -EPERM; } - - if (updelay < 0) { - pr_err("%s: Invalid up delay value %d not in range %d-%d; rejected.\n", - bond->dev->name, updelay, 0, INT_MAX); - return -EINVAL; - } else { - if ((updelay % bond->params.miimon) != 0) { - pr_warn("%s: Warning: up delay (%d) is not a multiple of miimon (%d), updelay rounded to %d ms\n", - bond->dev->name, updelay, - bond->params.miimon, - (updelay / bond->params.miimon) * - bond->params.miimon); - } - bond->params.updelay = updelay / bond->params.miimon; - pr_info("%s: Setting up delay to %d.\n", - bond->dev->name, - bond->params.updelay * bond->params.miimon); + if ((value % bond->params.miimon) != 0) { + pr_warn("%s: Warning: up delay (%d) is not a multiple of miimon (%d), updelay rounded to %d ms\n", + bond->dev->name, value, + bond->params.miimon, + (value / bond->params.miimon) * + bond->params.miimon); } + bond->params.updelay = value / bond->params.miimon; + pr_info("%s: Setting up delay to %d\n", + bond->dev->name, bond->params.updelay * bond->params.miimon); return 0; } -int bond_option_downdelay_set(struct bonding *bond, int downdelay) +static int bond_option_downdelay_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if (!(bond->params.miimon)) { + int value = newval->value; + + if (!bond->params.miimon) { pr_err("%s: Unable to set down delay as MII monitoring is disabled\n", bond->dev->name); return -EPERM; } - - if (downdelay < 0) { - pr_err("%s: Invalid down delay value %d not in range %d-%d; rejected.\n", - bond->dev->name, downdelay, 0, INT_MAX); - return -EINVAL; - } else { - if ((downdelay % bond->params.miimon) != 0) { - pr_warn("%s: Warning: down delay (%d) is not a multiple of miimon (%d), delay rounded to %d ms\n", - bond->dev->name, downdelay, - bond->params.miimon, - (downdelay / bond->params.miimon) * - bond->params.miimon); - } - bond->params.downdelay = downdelay / bond->params.miimon; - pr_info("%s: Setting down delay to %d.\n", - bond->dev->name, - bond->params.downdelay * bond->params.miimon); + if ((value % bond->params.miimon) != 0) { + pr_warn("%s: Warning: down delay (%d) is not a multiple of miimon (%d), delay rounded to %d ms\n", + bond->dev->name, value, + bond->params.miimon, + (value / bond->params.miimon) * + bond->params.miimon); } + bond->params.downdelay = value / bond->params.miimon; + pr_info("%s: Setting down delay to %d\n", + bond->dev->name, bond->params.downdelay * bond->params.miimon); return 0; } -int bond_option_use_carrier_set(struct bonding *bond, int use_carrier) +static int bond_option_use_carrier_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if ((use_carrier == 0) || (use_carrier == 1)) { - bond->params.use_carrier = use_carrier; - pr_info("%s: Setting use_carrier to %d.\n", - bond->dev->name, use_carrier); - } else { - pr_info("%s: Ignoring invalid use_carrier value %d.\n", - bond->dev->name, use_carrier); - } + pr_info("%s: Setting use_carrier to %llu\n", + bond->dev->name, newval->value); + bond->params.use_carrier = newval->value; return 0; } -int bond_option_arp_interval_set(struct bonding *bond, int arp_interval) +/* There are two tricky bits here. First, if ARP monitoring is activated, then + * we must disable MII monitoring. Second, if the ARP timer isn't running, + * we must start it. + */ +static int bond_option_arp_interval_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if (arp_interval < 0) { - pr_err("%s: Invalid arp_interval value %d not in range 0-%d; rejected.\n", - bond->dev->name, arp_interval, INT_MAX); - return -EINVAL; - } - if (BOND_NO_USES_ARP(bond->params.mode)) { - pr_info("%s: ARP monitoring cannot be used with ALB/TLB/802.3ad. Only MII monitoring is supported on %s.\n", - bond->dev->name, bond->dev->name); - return -EINVAL; - } - pr_info("%s: Setting ARP monitoring interval to %d.\n", - bond->dev->name, arp_interval); - bond->params.arp_interval = arp_interval; - if (arp_interval) { + pr_info("%s: Setting ARP monitoring interval to %llu\n", + bond->dev->name, newval->value); + bond->params.arp_interval = newval->value; + if (newval->value) { if (bond->params.miimon) { - pr_info("%s: ARP monitoring cannot be used with MII monitoring. %s Disabling MII monitoring.\n", + pr_info("%s: ARP monitoring cannot be used with MII monitoring. %s Disabling MII monitoring\n", bond->dev->name, bond->dev->name); bond->params.miimon = 0; } if (!bond->params.arp_targets[0]) - pr_info("%s: ARP monitoring has been set up, but no ARP targets have been specified.\n", + pr_info("%s: ARP monitoring has been set up, but no ARP targets have been specified\n", bond->dev->name); } if (bond->dev->flags & IFF_UP) { @@ -282,14 +908,13 @@ int bond_option_arp_interval_set(struct bonding *bond, int arp_interval) * timer will get fired off when the open function * is called. */ - if (!arp_interval) { + if (!newval->value) { if (bond->params.arp_validate) bond->recv_probe = NULL; cancel_delayed_work_sync(&bond->arp_work); } else { /* arp_validate can be set only in active-backup mode */ - if (bond->params.arp_validate) - bond->recv_probe = bond_arp_rcv; + bond->recv_probe = bond_arp_rcv; cancel_delayed_work_sync(&bond->mii_work); queue_delayed_work(bond->wq, &bond->arp_work, 0); } @@ -318,7 +943,7 @@ static int _bond_option_arp_ip_target_add(struct bonding *bond, __be32 target) __be32 *targets = bond->params.arp_targets; int ind; - if (IS_IP_TARGET_UNUSABLE_ADDRESS(target)) { + if (!bond_is_ip_target_ok(target)) { pr_err("%s: invalid ARP target %pI4 specified for addition\n", bond->dev->name, &target); return -EINVAL; @@ -332,19 +957,18 @@ static int _bond_option_arp_ip_target_add(struct bonding *bond, __be32 target) ind = bond_get_targets_ip(targets, 0); /* first free slot */ if (ind == -1) { - pr_err("%s: ARP target table is full!\n", - bond->dev->name); + pr_err("%s: ARP target table is full!\n", bond->dev->name); return -EINVAL; } - pr_info("%s: adding ARP target %pI4.\n", bond->dev->name, &target); + pr_info("%s: Adding ARP target %pI4\n", bond->dev->name, &target); _bond_options_arp_ip_target_set(bond, ind, target, jiffies); return 0; } -int bond_option_arp_ip_target_add(struct bonding *bond, __be32 target) +static int bond_option_arp_ip_target_add(struct bonding *bond, __be32 target) { int ret; @@ -356,7 +980,7 @@ int bond_option_arp_ip_target_add(struct bonding *bond, __be32 target) return ret; } -int bond_option_arp_ip_target_rem(struct bonding *bond, __be32 target) +static int bond_option_arp_ip_target_rem(struct bonding *bond, __be32 target) { __be32 *targets = bond->params.arp_targets; struct list_head *iter; @@ -364,7 +988,7 @@ int bond_option_arp_ip_target_rem(struct bonding *bond, __be32 target) unsigned long *targets_rx; int ind, i; - if (IS_IP_TARGET_UNUSABLE_ADDRESS(target)) { + if (!bond_is_ip_target_ok(target)) { pr_err("%s: invalid ARP target %pI4 specified for removal\n", bond->dev->name, &target); return -EINVAL; @@ -372,17 +996,16 @@ int bond_option_arp_ip_target_rem(struct bonding *bond, __be32 target) ind = bond_get_targets_ip(targets, target); if (ind == -1) { - pr_err("%s: unable to remove nonexistent ARP target %pI4.\n", + pr_err("%s: unable to remove nonexistent ARP target %pI4\n", bond->dev->name, &target); return -EINVAL; } if (ind == 0 && !targets[1] && bond->params.arp_interval) - pr_warn("%s: removing last arp target with arp_interval on\n", + pr_warn("%s: Removing last arp target with arp_interval on\n", bond->dev->name); - pr_info("%s: removing ARP target %pI4.\n", bond->dev->name, - &target); + pr_info("%s: Removing ARP target %pI4\n", bond->dev->name, &target); /* not to race with bond_arp_rcv */ write_lock_bh(&bond->lock); @@ -402,99 +1025,88 @@ int bond_option_arp_ip_target_rem(struct bonding *bond, __be32 target) return 0; } -int bond_option_arp_ip_targets_set(struct bonding *bond, __be32 *targets, - int count) +void bond_option_arp_ip_targets_clear(struct bonding *bond) { - int i, ret = 0; + int i; /* not to race with bond_arp_rcv */ write_lock_bh(&bond->lock); - - /* clear table */ for (i = 0; i < BOND_MAX_ARP_TARGETS; i++) _bond_options_arp_ip_target_set(bond, i, 0, 0); - - if (count == 0 && bond->params.arp_interval) - pr_warn("%s: removing last arp target with arp_interval on\n", - bond->dev->name); - - for (i = 0; i < count; i++) { - ret = _bond_option_arp_ip_target_add(bond, targets[i]); - if (ret) - break; - } - write_unlock_bh(&bond->lock); - return ret; } -int bond_option_arp_validate_set(struct bonding *bond, int arp_validate) +static int bond_option_arp_ip_targets_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if (bond_parm_tbl_lookup(arp_validate, arp_validate_tbl) < 0) { - pr_err("%s: Ignoring invalid arp_validate value %d.\n", - bond->dev->name, arp_validate); - return -EINVAL; + int ret = -EPERM; + __be32 target; + + if (newval->string) { + if (!in4_pton(newval->string+1, -1, (u8 *)&target, -1, NULL)) { + pr_err("%s: invalid ARP target %pI4 specified\n", + bond->dev->name, &target); + return ret; + } + if (newval->string[0] == '+') + ret = bond_option_arp_ip_target_add(bond, target); + else if (newval->string[0] == '-') + ret = bond_option_arp_ip_target_rem(bond, target); + else + pr_err("no command found in arp_ip_targets file for bond %s - use +<addr> or -<addr>\n", + bond->dev->name); + } else { + target = newval->value; + ret = bond_option_arp_ip_target_add(bond, target); } - if (bond->params.mode != BOND_MODE_ACTIVEBACKUP) { - pr_err("%s: arp_validate only supported in active-backup mode.\n", - bond->dev->name); - return -EINVAL; - } + return ret; +} - pr_info("%s: setting arp_validate to %s (%d).\n", - bond->dev->name, arp_validate_tbl[arp_validate].modename, - arp_validate); +static int bond_option_arp_validate_set(struct bonding *bond, + const struct bond_opt_value *newval) +{ + pr_info("%s: Setting arp_validate to %s (%llu)\n", + bond->dev->name, newval->string, newval->value); if (bond->dev->flags & IFF_UP) { - if (!arp_validate) + if (!newval->value) bond->recv_probe = NULL; else if (bond->params.arp_interval) bond->recv_probe = bond_arp_rcv; } - bond->params.arp_validate = arp_validate; + bond->params.arp_validate = newval->value; return 0; } -int bond_option_arp_all_targets_set(struct bonding *bond, int arp_all_targets) +static int bond_option_arp_all_targets_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if (bond_parm_tbl_lookup(arp_all_targets, arp_all_targets_tbl) < 0) { - pr_err("%s: Ignoring invalid arp_all_targets value %d.\n", - bond->dev->name, arp_all_targets); - return -EINVAL; - } - - pr_info("%s: setting arp_all_targets to %s (%d).\n", - bond->dev->name, arp_all_targets_tbl[arp_all_targets].modename, - arp_all_targets); - - bond->params.arp_all_targets = arp_all_targets; + pr_info("%s: Setting arp_all_targets to %s (%llu)\n", + bond->dev->name, newval->string, newval->value); + bond->params.arp_all_targets = newval->value; return 0; } -int bond_option_primary_set(struct bonding *bond, const char *primary) +static int bond_option_primary_set(struct bonding *bond, + const struct bond_opt_value *newval) { + char *p, *primary = newval->string; struct list_head *iter; struct slave *slave; - int err = 0; block_netpoll_tx(); read_lock(&bond->lock); write_lock_bh(&bond->curr_slave_lock); - if (!USES_PRIMARY(bond->params.mode)) { - pr_err("%s: Unable to set primary slave; %s is in mode %d\n", - bond->dev->name, bond->dev->name, bond->params.mode); - err = -EINVAL; - goto out; - } - + p = strchr(primary, '\n'); + if (p) + *p = '\0'; /* check to see if we are clearing primary */ if (!strlen(primary)) { - pr_info("%s: Setting primary slave to None.\n", - bond->dev->name); + pr_info("%s: Setting primary slave to None\n", bond->dev->name); bond->primary_slave = NULL; memset(bond->params.primary, 0, sizeof(bond->params.primary)); bond_select_active_slave(bond); @@ -503,7 +1115,7 @@ int bond_option_primary_set(struct bonding *bond, const char *primary) bond_for_each_slave(bond, slave, iter) { if (strncmp(slave->dev->name, primary, IFNAMSIZ) == 0) { - pr_info("%s: Setting %s as primary slave.\n", + pr_info("%s: Setting %s as primary slave\n", bond->dev->name, slave->dev->name); bond->primary_slave = slave; strcpy(bond->params.primary, slave->dev->name); @@ -512,10 +1124,15 @@ int bond_option_primary_set(struct bonding *bond, const char *primary) } } + if (bond->primary_slave) { + pr_info("%s: Setting primary slave to None\n", bond->dev->name); + bond->primary_slave = NULL; + bond_select_active_slave(bond); + } strncpy(bond->params.primary, primary, IFNAMSIZ); bond->params.primary[IFNAMSIZ - 1] = 0; - pr_info("%s: Recording %s as primary, but it has not been enslaved to %s yet.\n", + pr_info("%s: Recording %s as primary, but it has not been enslaved to %s yet\n", bond->dev->name, primary, bond->dev->name); out: @@ -523,21 +1140,15 @@ out: read_unlock(&bond->lock); unblock_netpoll_tx(); - return err; + return 0; } -int bond_option_primary_reselect_set(struct bonding *bond, int primary_reselect) +static int bond_option_primary_reselect_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if (bond_parm_tbl_lookup(primary_reselect, pri_reselect_tbl) < 0) { - pr_err("%s: Ignoring invalid primary_reselect value %d.\n", - bond->dev->name, primary_reselect); - return -EINVAL; - } - - bond->params.primary_reselect = primary_reselect; - pr_info("%s: setting primary_reselect to %s (%d).\n", - bond->dev->name, pri_reselect_tbl[primary_reselect].modename, - primary_reselect); + pr_info("%s: Setting primary_reselect to %s (%llu)\n", + bond->dev->name, newval->string, newval->value); + bond->params.primary_reselect = newval->value; block_netpoll_tx(); write_lock_bh(&bond->curr_slave_lock); @@ -548,85 +1159,56 @@ int bond_option_primary_reselect_set(struct bonding *bond, int primary_reselect) return 0; } -int bond_option_fail_over_mac_set(struct bonding *bond, int fail_over_mac) +static int bond_option_fail_over_mac_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if (bond_parm_tbl_lookup(fail_over_mac, fail_over_mac_tbl) < 0) { - pr_err("%s: Ignoring invalid fail_over_mac value %d.\n", - bond->dev->name, fail_over_mac); - return -EINVAL; - } - - if (bond_has_slaves(bond)) { - pr_err("%s: Can't alter fail_over_mac with slaves in bond.\n", - bond->dev->name); - return -EPERM; - } - - bond->params.fail_over_mac = fail_over_mac; - pr_info("%s: Setting fail_over_mac to %s (%d).\n", - bond->dev->name, fail_over_mac_tbl[fail_over_mac].modename, - fail_over_mac); + pr_info("%s: Setting fail_over_mac to %s (%llu)\n", + bond->dev->name, newval->string, newval->value); + bond->params.fail_over_mac = newval->value; return 0; } -int bond_option_xmit_hash_policy_set(struct bonding *bond, int xmit_hash_policy) +static int bond_option_xmit_hash_policy_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if (bond_parm_tbl_lookup(xmit_hash_policy, xmit_hashtype_tbl) < 0) { - pr_err("%s: Ignoring invalid xmit_hash_policy value %d.\n", - bond->dev->name, xmit_hash_policy); - return -EINVAL; - } - - bond->params.xmit_policy = xmit_hash_policy; - pr_info("%s: setting xmit hash policy to %s (%d).\n", - bond->dev->name, - xmit_hashtype_tbl[xmit_hash_policy].modename, xmit_hash_policy); + pr_info("%s: Setting xmit hash policy to %s (%llu)\n", + bond->dev->name, newval->string, newval->value); + bond->params.xmit_policy = newval->value; return 0; } -int bond_option_resend_igmp_set(struct bonding *bond, int resend_igmp) +static int bond_option_resend_igmp_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if (resend_igmp < 0 || resend_igmp > 255) { - pr_err("%s: Invalid resend_igmp value %d not in range 0-255; rejected.\n", - bond->dev->name, resend_igmp); - return -EINVAL; - } - - bond->params.resend_igmp = resend_igmp; - pr_info("%s: Setting resend_igmp to %d.\n", - bond->dev->name, resend_igmp); + pr_info("%s: Setting resend_igmp to %llu\n", + bond->dev->name, newval->value); + bond->params.resend_igmp = newval->value; return 0; } -int bond_option_num_peer_notif_set(struct bonding *bond, int num_peer_notif) +static int bond_option_num_peer_notif_set(struct bonding *bond, + const struct bond_opt_value *newval) { - bond->params.num_peer_notif = num_peer_notif; + bond->params.num_peer_notif = newval->value; + return 0; } -int bond_option_all_slaves_active_set(struct bonding *bond, - int all_slaves_active) +static int bond_option_all_slaves_active_set(struct bonding *bond, + const struct bond_opt_value *newval) { struct list_head *iter; struct slave *slave; - if (all_slaves_active == bond->params.all_slaves_active) + if (newval->value == bond->params.all_slaves_active) return 0; - - if ((all_slaves_active == 0) || (all_slaves_active == 1)) { - bond->params.all_slaves_active = all_slaves_active; - } else { - pr_info("%s: Ignoring invalid all_slaves_active value %d.\n", - bond->dev->name, all_slaves_active); - return -EINVAL; - } - + bond->params.all_slaves_active = newval->value; bond_for_each_slave(bond, slave, iter) { if (!bond_is_active_slave(slave)) { - if (all_slaves_active) + if (newval->value) slave->inactive = 0; else slave->inactive = 1; @@ -636,97 +1218,177 @@ int bond_option_all_slaves_active_set(struct bonding *bond, return 0; } -int bond_option_min_links_set(struct bonding *bond, int min_links) +static int bond_option_min_links_set(struct bonding *bond, + const struct bond_opt_value *newval) { - pr_info("%s: Setting min links value to %u\n", - bond->dev->name, min_links); - bond->params.min_links = min_links; + pr_info("%s: Setting min links value to %llu\n", + bond->dev->name, newval->value); + bond->params.min_links = newval->value; return 0; } -int bond_option_lp_interval_set(struct bonding *bond, int lp_interval) +static int bond_option_lp_interval_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if (lp_interval <= 0) { - pr_err("%s: lp_interval must be between 1 and %d\n", - bond->dev->name, INT_MAX); - return -EINVAL; - } - - bond->params.lp_interval = lp_interval; + bond->params.lp_interval = newval->value; return 0; } -int bond_option_packets_per_slave_set(struct bonding *bond, - int packets_per_slave) +static int bond_option_pps_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if (packets_per_slave < 0 || packets_per_slave > USHRT_MAX) { - pr_err("%s: packets_per_slave must be between 0 and %u\n", - bond->dev->name, USHRT_MAX); - return -EINVAL; + bond->params.packets_per_slave = newval->value; + if (newval->value > 0) { + bond->params.reciprocal_packets_per_slave = + reciprocal_value(newval->value); + } else { + /* reciprocal_packets_per_slave is unused if + * packets_per_slave is 0 or 1, just initialize it + */ + bond->params.reciprocal_packets_per_slave = + (struct reciprocal_value) { 0 }; } - if (bond->params.mode != BOND_MODE_ROUNDROBIN) - pr_warn("%s: Warning: packets_per_slave has effect only in balance-rr mode\n", - bond->dev->name); + return 0; +} - if (packets_per_slave > 1) - bond->params.packets_per_slave = - reciprocal_value(packets_per_slave); - else - bond->params.packets_per_slave = packets_per_slave; +static int bond_option_lacp_rate_set(struct bonding *bond, + const struct bond_opt_value *newval) +{ + pr_info("%s: Setting LACP rate to %s (%llu)\n", + bond->dev->name, newval->string, newval->value); + bond->params.lacp_fast = newval->value; + bond_3ad_update_lacp_rate(bond); return 0; } -int bond_option_lacp_rate_set(struct bonding *bond, int lacp_rate) +static int bond_option_ad_select_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if (bond_parm_tbl_lookup(lacp_rate, bond_lacp_tbl) < 0) { - pr_err("%s: Ignoring invalid LACP rate value %d.\n", - bond->dev->name, lacp_rate); - return -EINVAL; - } + pr_info("%s: Setting ad_select to %s (%llu)\n", + bond->dev->name, newval->string, newval->value); + bond->params.ad_select = newval->value; - if (bond->dev->flags & IFF_UP) { - pr_err("%s: Unable to update LACP rate because interface is up.\n", - bond->dev->name); - return -EPERM; - } + return 0; +} - if (bond->params.mode != BOND_MODE_8023AD) { - pr_err("%s: Unable to update LACP rate because bond is not in 802.3ad mode.\n", - bond->dev->name); - return -EPERM; +static int bond_option_queue_id_set(struct bonding *bond, + const struct bond_opt_value *newval) +{ + struct slave *slave, *update_slave; + struct net_device *sdev; + struct list_head *iter; + char *delim; + int ret = 0; + u16 qid; + + /* delim will point to queue id if successful */ + delim = strchr(newval->string, ':'); + if (!delim) + goto err_no_cmd; + + /* Terminate string that points to device name and bump it + * up one, so we can read the queue id there. + */ + *delim = '\0'; + if (sscanf(++delim, "%hd\n", &qid) != 1) + goto err_no_cmd; + + /* Check buffer length, valid ifname and queue id */ + if (!dev_valid_name(newval->string) || + qid > bond->dev->real_num_tx_queues) + goto err_no_cmd; + + /* Get the pointer to that interface if it exists */ + sdev = __dev_get_by_name(dev_net(bond->dev), newval->string); + if (!sdev) + goto err_no_cmd; + + /* Search for thes slave and check for duplicate qids */ + update_slave = NULL; + bond_for_each_slave(bond, slave, iter) { + if (sdev == slave->dev) + /* We don't need to check the matching + * slave for dups, since we're overwriting it + */ + update_slave = slave; + else if (qid && qid == slave->queue_id) { + goto err_no_cmd; + } } - bond->params.lacp_fast = lacp_rate; - bond_3ad_update_lacp_rate(bond); - pr_info("%s: Setting LACP rate to %s (%d).\n", - bond->dev->name, bond_lacp_tbl[lacp_rate].modename, - lacp_rate); + if (!update_slave) + goto err_no_cmd; + + /* Actually set the qids for the slave */ + update_slave->queue_id = qid; + +out: + return ret; + +err_no_cmd: + pr_info("invalid input for queue_id set for %s\n", bond->dev->name); + ret = -EPERM; + goto out; - return 0; } -int bond_option_ad_select_set(struct bonding *bond, int ad_select) +static int bond_option_slaves_set(struct bonding *bond, + const struct bond_opt_value *newval) { - if (bond_parm_tbl_lookup(ad_select, ad_select_tbl) < 0) { - pr_err("%s: Ignoring invalid ad_select value %d.\n", - bond->dev->name, ad_select); - return -EINVAL; + char command[IFNAMSIZ + 1] = { 0, }; + struct net_device *dev; + char *ifname; + int ret; + + sscanf(newval->string, "%16s", command); /* IFNAMSIZ*/ + ifname = command + 1; + if ((strlen(command) <= 1) || + !dev_valid_name(ifname)) + goto err_no_cmd; + + dev = __dev_get_by_name(dev_net(bond->dev), ifname); + if (!dev) { + pr_info("%s: interface %s does not exist!\n", + bond->dev->name, ifname); + ret = -ENODEV; + goto out; } - if (bond->dev->flags & IFF_UP) { - pr_err("%s: Unable to update ad_select because interface is up.\n", - bond->dev->name); - return -EPERM; + switch (command[0]) { + case '+': + pr_info("%s: Adding slave %s\n", bond->dev->name, dev->name); + ret = bond_enslave(bond->dev, dev); + break; + + case '-': + pr_info("%s: Removing slave %s\n", bond->dev->name, dev->name); + ret = bond_release(bond->dev, dev); + break; + + default: + goto err_no_cmd; } - bond->params.ad_select = ad_select; - pr_info("%s: Setting ad_select to %s (%d).\n", - bond->dev->name, ad_select_tbl[ad_select].modename, - ad_select); +out: + return ret; + +err_no_cmd: + pr_err("no command found in slaves file for bond %s - use +ifname or -ifname\n", + bond->dev->name); + ret = -EPERM; + goto out; +} + +static int bond_option_tlb_dynamic_lb_set(struct bonding *bond, + const struct bond_opt_value *newval) +{ + pr_info("%s: Setting dynamic-lb to %s (%llu)\n", + bond->dev->name, newval->string, newval->value); + bond->params.tlb_dynamic_lb = newval->value; return 0; } diff --git a/drivers/net/bonding/bond_options.h b/drivers/net/bonding/bond_options.h new file mode 100644 index 00000000000..17ded5b2917 --- /dev/null +++ b/drivers/net/bonding/bond_options.h @@ -0,0 +1,130 @@ +/* + * drivers/net/bond/bond_options.h - bonding options + * Copyright (c) 2013 Nikolay Aleksandrov <nikolay@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef _BOND_OPTIONS_H +#define _BOND_OPTIONS_H + +#define BOND_OPT_MAX_NAMELEN 32 +#define BOND_OPT_VALID(opt) ((opt) < BOND_OPT_LAST) +#define BOND_MODE_ALL_EX(x) (~(x)) + +/* Option flags: + * BOND_OPTFLAG_NOSLAVES - check if the bond device is empty before setting + * BOND_OPTFLAG_IFDOWN - check if the bond device is down before setting + * BOND_OPTFLAG_RAWVAL - the option parses the value itself + */ +enum { + BOND_OPTFLAG_NOSLAVES = BIT(0), + BOND_OPTFLAG_IFDOWN = BIT(1), + BOND_OPTFLAG_RAWVAL = BIT(2) +}; + +/* Value type flags: + * BOND_VALFLAG_DEFAULT - mark the value as default + * BOND_VALFLAG_(MIN|MAX) - mark the value as min/max + */ +enum { + BOND_VALFLAG_DEFAULT = BIT(0), + BOND_VALFLAG_MIN = BIT(1), + BOND_VALFLAG_MAX = BIT(2) +}; + +/* Option IDs, their bit positions correspond to their IDs */ +enum { + BOND_OPT_MODE, + BOND_OPT_PACKETS_PER_SLAVE, + BOND_OPT_XMIT_HASH, + BOND_OPT_ARP_VALIDATE, + BOND_OPT_ARP_ALL_TARGETS, + BOND_OPT_FAIL_OVER_MAC, + BOND_OPT_ARP_INTERVAL, + BOND_OPT_ARP_TARGETS, + BOND_OPT_DOWNDELAY, + BOND_OPT_UPDELAY, + BOND_OPT_LACP_RATE, + BOND_OPT_MINLINKS, + BOND_OPT_AD_SELECT, + BOND_OPT_NUM_PEER_NOTIF, + BOND_OPT_MIIMON, + BOND_OPT_PRIMARY, + BOND_OPT_PRIMARY_RESELECT, + BOND_OPT_USE_CARRIER, + BOND_OPT_ACTIVE_SLAVE, + BOND_OPT_QUEUE_ID, + BOND_OPT_ALL_SLAVES_ACTIVE, + BOND_OPT_RESEND_IGMP, + BOND_OPT_LP_INTERVAL, + BOND_OPT_SLAVES, + BOND_OPT_TLB_DYNAMIC_LB, + BOND_OPT_LAST +}; + +/* This structure is used for storing option values and for passing option + * values when changing an option. The logic when used as an arg is as follows: + * - if string != NULL -> parse it, if the opt is RAW type then return it, else + * return the parse result + * - if string == NULL -> parse value + */ +struct bond_opt_value { + char *string; + u64 value; + u32 flags; +}; + +struct bonding; + +struct bond_option { + int id; + const char *name; + const char *desc; + u32 flags; + + /* unsuppmodes is used to denote modes in which the option isn't + * supported. + */ + unsigned long unsuppmodes; + /* supported values which this option can have, can be a subset of + * BOND_OPTVAL_RANGE's value range + */ + const struct bond_opt_value *values; + + int (*set)(struct bonding *bond, const struct bond_opt_value *val); +}; + +int __bond_opt_set(struct bonding *bond, unsigned int option, + struct bond_opt_value *val); +int bond_opt_tryset_rtnl(struct bonding *bond, unsigned int option, char *buf); + +const struct bond_opt_value *bond_opt_parse(const struct bond_option *opt, + struct bond_opt_value *val); +const struct bond_option *bond_opt_get(unsigned int option); +const struct bond_option *bond_opt_get_by_name(const char *name); +const struct bond_opt_value *bond_opt_get_val(unsigned int option, u64 val); + +/* This helper is used to initialize a bond_opt_value structure for parameter + * passing. There should be either a valid string or value, but not both. + * When value is ULLONG_MAX then string will be used. + */ +static inline void __bond_opt_init(struct bond_opt_value *optval, + char *string, u64 value) +{ + memset(optval, 0, sizeof(*optval)); + optval->value = ULLONG_MAX; + if (value == ULLONG_MAX) + optval->string = string; + else + optval->value = value; +} +#define bond_opt_initval(optval, value) __bond_opt_init(optval, NULL, value) +#define bond_opt_initstr(optval, str) __bond_opt_init(optval, str, ULLONG_MAX) + +void bond_option_arp_ip_targets_clear(struct bonding *bond); + +#endif /* _BOND_OPTIONS_H */ diff --git a/drivers/net/bonding/bond_procfs.c b/drivers/net/bonding/bond_procfs.c index 8515b344258..b215b479bb3 100644 --- a/drivers/net/bonding/bond_procfs.c +++ b/drivers/net/bonding/bond_procfs.c @@ -65,37 +65,42 @@ static void bond_info_seq_stop(struct seq_file *seq, void *v) static void bond_info_show_master(struct seq_file *seq) { struct bonding *bond = seq->private; + const struct bond_opt_value *optval; struct slave *curr; int i; - read_lock(&bond->curr_slave_lock); - curr = bond->curr_active_slave; - read_unlock(&bond->curr_slave_lock); + curr = rcu_dereference(bond->curr_active_slave); seq_printf(seq, "Bonding Mode: %s", - bond_mode_name(bond->params.mode)); + bond_mode_name(BOND_MODE(bond))); - if (bond->params.mode == BOND_MODE_ACTIVEBACKUP && - bond->params.fail_over_mac) - seq_printf(seq, " (fail_over_mac %s)", - fail_over_mac_tbl[bond->params.fail_over_mac].modename); + if (BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP && + bond->params.fail_over_mac) { + optval = bond_opt_get_val(BOND_OPT_FAIL_OVER_MAC, + bond->params.fail_over_mac); + seq_printf(seq, " (fail_over_mac %s)", optval->string); + } seq_printf(seq, "\n"); - if (bond->params.mode == BOND_MODE_XOR || - bond->params.mode == BOND_MODE_8023AD) { + if (BOND_MODE(bond) == BOND_MODE_XOR || + BOND_MODE(bond) == BOND_MODE_8023AD) { + optval = bond_opt_get_val(BOND_OPT_XMIT_HASH, + bond->params.xmit_policy); seq_printf(seq, "Transmit Hash Policy: %s (%d)\n", - xmit_hashtype_tbl[bond->params.xmit_policy].modename, - bond->params.xmit_policy); + optval->string, bond->params.xmit_policy); } - if (USES_PRIMARY(bond->params.mode)) { + if (bond_uses_primary(bond)) { seq_printf(seq, "Primary Slave: %s", (bond->primary_slave) ? bond->primary_slave->dev->name : "None"); - if (bond->primary_slave) + if (bond->primary_slave) { + optval = bond_opt_get_val(BOND_OPT_PRIMARY_RESELECT, + bond->params.primary_reselect); seq_printf(seq, " (primary_reselect %s)", - pri_reselect_tbl[bond->params.primary_reselect].modename); + optval->string); + } seq_printf(seq, "\nCurrently Active Slave: %s\n", (curr) ? curr->dev->name : "None"); @@ -129,15 +134,17 @@ static void bond_info_show_master(struct seq_file *seq) seq_printf(seq, "\n"); } - if (bond->params.mode == BOND_MODE_8023AD) { + if (BOND_MODE(bond) == BOND_MODE_8023AD) { struct ad_info ad_info; seq_puts(seq, "\n802.3ad info\n"); seq_printf(seq, "LACP rate: %s\n", (bond->params.lacp_fast) ? "fast" : "slow"); seq_printf(seq, "Min links: %d\n", bond->params.min_links); + optval = bond_opt_get_val(BOND_OPT_AD_SELECT, + bond->params.ad_select); seq_printf(seq, "Aggregator selection policy (ad_select): %s\n", - ad_select_tbl[bond->params.ad_select].modename); + optval->string); if (__bond_3ad_get_active_agg_info(bond, &ad_info)) { seq_printf(seq, "bond %s has no active aggregator\n", @@ -181,9 +188,9 @@ static void bond_info_show_slave(struct seq_file *seq, seq_printf(seq, "Permanent HW addr: %pM\n", slave->perm_hwaddr); - if (bond->params.mode == BOND_MODE_8023AD) { + if (BOND_MODE(bond) == BOND_MODE_8023AD) { const struct aggregator *agg - = SLAVE_AD_INFO(slave).port.aggregator; + = SLAVE_AD_INFO(slave)->port.aggregator; if (agg) seq_printf(seq, "Aggregator ID: %d\n", @@ -245,8 +252,8 @@ void bond_create_proc_entry(struct bonding *bond) S_IRUGO, bn->proc_dir, &bond_info_fops, bond); if (bond->proc_entry == NULL) - pr_warning("Warning: Cannot create /proc/net/%s/%s\n", - DRV_NAME, bond_dev->name); + pr_warn("Warning: Cannot create /proc/net/%s/%s\n", + DRV_NAME, bond_dev->name); else memcpy(bond->proc_file_name, bond_dev->name, IFNAMSIZ); } @@ -272,8 +279,8 @@ void __net_init bond_create_proc_dir(struct bond_net *bn) if (!bn->proc_dir) { bn->proc_dir = proc_mkdir(DRV_NAME, bn->net->proc_net); if (!bn->proc_dir) - pr_warning("Warning: cannot create /proc/net/%s\n", - DRV_NAME); + pr_warn("Warning: Cannot create /proc/net/%s\n", + DRV_NAME); } } diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c index 011f163c2c6..daed52f68ce 100644 --- a/drivers/net/bonding/bond_sysfs.c +++ b/drivers/net/bonding/bond_sysfs.c @@ -39,15 +39,13 @@ #include <net/net_namespace.h> #include <net/netns/generic.h> #include <linux/nsproxy.h> -#include <linux/reciprocal_div.h> #include "bonding.h" #define to_dev(obj) container_of(obj, struct device, kobj) #define to_bond(cd) ((struct bonding *)(netdev_priv(to_net_dev(cd)))) -/* - * "show" function for the bond_masters attribute. +/* "show" function for the bond_masters attribute. * The class parameter is ignored. */ static ssize_t bonding_show_bonds(struct class *cls, @@ -89,14 +87,12 @@ static struct net_device *bond_get_by_name(struct bond_net *bn, const char *ifna return NULL; } -/* - * "store" function for the bond_masters attribute. This is what +/* "store" function for the bond_masters attribute. This is what * creates and deletes entire bonds. * * The class parameter is ignored. * */ - static ssize_t bonding_store_bonds(struct class *cls, struct class_attribute *attr, const char *buffer, size_t count) @@ -118,9 +114,9 @@ static ssize_t bonding_store_bonds(struct class *cls, rv = bond_create(bn->net, ifname); if (rv) { if (rv == -EEXIST) - pr_info("%s already exists.\n", ifname); + pr_info("%s already exists\n", ifname); else - pr_info("%s creation failed.\n", ifname); + pr_info("%s creation failed\n", ifname); res = rv; } } else if (command[0] == '-') { @@ -145,7 +141,7 @@ static ssize_t bonding_store_bonds(struct class *cls, return res; err_no_cmd: - pr_err("no command found in bonding_masters. Use +ifname or -ifname.\n"); + pr_err("no command found in bonding_masters - use +ifname or -ifname\n"); return -EPERM; } @@ -159,9 +155,26 @@ static const struct class_attribute class_attr_bonding_masters = { .store = bonding_store_bonds, }; -/* - * Show the slaves in the current bond. - */ +/* Generic "store" method for bonding sysfs option setting */ +static ssize_t bonding_sysfs_store_option(struct device *d, + struct device_attribute *attr, + const char *buffer, size_t count) +{ + struct bonding *bond = to_bond(d); + const struct bond_option *opt; + int ret; + + opt = bond_opt_get_by_name(attr->attr.name); + if (WARN_ON(!opt)) + return -ENOENT; + ret = bond_opt_tryset_rtnl(bond, opt->id, (char *)buffer); + if (!ret) + ret = count; + + return ret; +} + +/* Show the slaves in the current bond. */ static ssize_t bonding_show_slaves(struct device *d, struct device_attribute *attr, char *buf) { @@ -191,290 +204,87 @@ static ssize_t bonding_show_slaves(struct device *d, return res; } - -/* - * Set the slaves in the current bond. - * This is supposed to be only thin wrapper for bond_enslave and bond_release. - * All hard work should be done there. - */ -static ssize_t bonding_store_slaves(struct device *d, - struct device_attribute *attr, - const char *buffer, size_t count) -{ - char command[IFNAMSIZ + 1] = { 0, }; - char *ifname; - int res, ret = count; - struct net_device *dev; - struct bonding *bond = to_bond(d); - - if (!rtnl_trylock()) - return restart_syscall(); - - sscanf(buffer, "%16s", command); /* IFNAMSIZ*/ - ifname = command + 1; - if ((strlen(command) <= 1) || - !dev_valid_name(ifname)) - goto err_no_cmd; - - dev = __dev_get_by_name(dev_net(bond->dev), ifname); - if (!dev) { - pr_info("%s: Interface %s does not exist!\n", - bond->dev->name, ifname); - ret = -ENODEV; - goto out; - } - - switch (command[0]) { - case '+': - pr_info("%s: Adding slave %s.\n", bond->dev->name, dev->name); - res = bond_enslave(bond->dev, dev); - break; - - case '-': - pr_info("%s: Removing slave %s.\n", bond->dev->name, dev->name); - res = bond_release(bond->dev, dev); - break; - - default: - goto err_no_cmd; - } - - if (res) - ret = res; - goto out; - -err_no_cmd: - pr_err("no command found in slaves file for bond %s. Use +ifname or -ifname.\n", - bond->dev->name); - ret = -EPERM; - -out: - rtnl_unlock(); - return ret; -} - static DEVICE_ATTR(slaves, S_IRUGO | S_IWUSR, bonding_show_slaves, - bonding_store_slaves); + bonding_sysfs_store_option); -/* - * Show and set the bonding mode. The bond interface must be down to - * change the mode. - */ +/* Show the bonding mode. */ static ssize_t bonding_show_mode(struct device *d, struct device_attribute *attr, char *buf) { struct bonding *bond = to_bond(d); + const struct bond_opt_value *val; - return sprintf(buf, "%s %d\n", - bond_mode_tbl[bond->params.mode].modename, - bond->params.mode); -} - -static ssize_t bonding_store_mode(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - int new_value, ret; - struct bonding *bond = to_bond(d); - - new_value = bond_parse_parm(buf, bond_mode_tbl); - if (new_value < 0) { - pr_err("%s: Ignoring invalid mode value %.*s.\n", - bond->dev->name, (int)strlen(buf) - 1, buf); - return -EINVAL; - } - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_mode_set(bond, new_value); - if (!ret) { - pr_info("%s: setting mode to %s (%d).\n", - bond->dev->name, bond_mode_tbl[new_value].modename, - new_value); - ret = count; - } + val = bond_opt_get_val(BOND_OPT_MODE, BOND_MODE(bond)); - rtnl_unlock(); - return ret; + return sprintf(buf, "%s %d\n", val->string, BOND_MODE(bond)); } static DEVICE_ATTR(mode, S_IRUGO | S_IWUSR, - bonding_show_mode, bonding_store_mode); + bonding_show_mode, bonding_sysfs_store_option); -/* - * Show and set the bonding transmit hash method. - */ +/* Show the bonding transmit hash method. */ static ssize_t bonding_show_xmit_hash(struct device *d, struct device_attribute *attr, char *buf) { struct bonding *bond = to_bond(d); + const struct bond_opt_value *val; - return sprintf(buf, "%s %d\n", - xmit_hashtype_tbl[bond->params.xmit_policy].modename, - bond->params.xmit_policy); -} - -static ssize_t bonding_store_xmit_hash(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - int new_value, ret; - struct bonding *bond = to_bond(d); - - new_value = bond_parse_parm(buf, xmit_hashtype_tbl); - if (new_value < 0) { - pr_err("%s: Ignoring invalid xmit hash policy value %.*s.\n", - bond->dev->name, - (int)strlen(buf) - 1, buf); - return -EINVAL; - } - - if (!rtnl_trylock()) - return restart_syscall(); + val = bond_opt_get_val(BOND_OPT_XMIT_HASH, bond->params.xmit_policy); - ret = bond_option_xmit_hash_policy_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; + return sprintf(buf, "%s %d\n", val->string, bond->params.xmit_policy); } static DEVICE_ATTR(xmit_hash_policy, S_IRUGO | S_IWUSR, - bonding_show_xmit_hash, bonding_store_xmit_hash); + bonding_show_xmit_hash, bonding_sysfs_store_option); -/* - * Show and set arp_validate. - */ +/* Show arp_validate. */ static ssize_t bonding_show_arp_validate(struct device *d, struct device_attribute *attr, char *buf) { struct bonding *bond = to_bond(d); + const struct bond_opt_value *val; - return sprintf(buf, "%s %d\n", - arp_validate_tbl[bond->params.arp_validate].modename, - bond->params.arp_validate); -} - -static ssize_t bonding_store_arp_validate(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct bonding *bond = to_bond(d); - int new_value, ret; - - new_value = bond_parse_parm(buf, arp_validate_tbl); - if (new_value < 0) { - pr_err("%s: Ignoring invalid arp_validate value %s\n", - bond->dev->name, buf); - return -EINVAL; - } - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_arp_validate_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); + val = bond_opt_get_val(BOND_OPT_ARP_VALIDATE, + bond->params.arp_validate); - return ret; + return sprintf(buf, "%s %d\n", val->string, bond->params.arp_validate); } - static DEVICE_ATTR(arp_validate, S_IRUGO | S_IWUSR, bonding_show_arp_validate, - bonding_store_arp_validate); -/* - * Show and set arp_all_targets. - */ + bonding_sysfs_store_option); + +/* Show arp_all_targets. */ static ssize_t bonding_show_arp_all_targets(struct device *d, struct device_attribute *attr, char *buf) { struct bonding *bond = to_bond(d); - int value = bond->params.arp_all_targets; - - return sprintf(buf, "%s %d\n", arp_all_targets_tbl[value].modename, - value); -} + const struct bond_opt_value *val; -static ssize_t bonding_store_arp_all_targets(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct bonding *bond = to_bond(d); - int new_value, ret; - - new_value = bond_parse_parm(buf, arp_all_targets_tbl); - if (new_value < 0) { - pr_err("%s: Ignoring invalid arp_all_targets value %s\n", - bond->dev->name, buf); - return -EINVAL; - } - - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_arp_all_targets_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - - return ret; + val = bond_opt_get_val(BOND_OPT_ARP_ALL_TARGETS, + bond->params.arp_all_targets); + return sprintf(buf, "%s %d\n", + val->string, bond->params.arp_all_targets); } - static DEVICE_ATTR(arp_all_targets, S_IRUGO | S_IWUSR, - bonding_show_arp_all_targets, bonding_store_arp_all_targets); + bonding_show_arp_all_targets, bonding_sysfs_store_option); -/* - * Show and store fail_over_mac. User only allowed to change the - * value when there are no slaves. - */ +/* Show fail_over_mac. */ static ssize_t bonding_show_fail_over_mac(struct device *d, struct device_attribute *attr, char *buf) { struct bonding *bond = to_bond(d); + const struct bond_opt_value *val; - return sprintf(buf, "%s %d\n", - fail_over_mac_tbl[bond->params.fail_over_mac].modename, - bond->params.fail_over_mac); -} - -static ssize_t bonding_store_fail_over_mac(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - int new_value, ret; - struct bonding *bond = to_bond(d); - - new_value = bond_parse_parm(buf, fail_over_mac_tbl); - if (new_value < 0) { - pr_err("%s: Ignoring invalid fail_over_mac value %s.\n", - bond->dev->name, buf); - return -EINVAL; - } - - if (!rtnl_trylock()) - return restart_syscall(); + val = bond_opt_get_val(BOND_OPT_FAIL_OVER_MAC, + bond->params.fail_over_mac); - ret = bond_option_fail_over_mac_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; + return sprintf(buf, "%s %d\n", val->string, bond->params.fail_over_mac); } - static DEVICE_ATTR(fail_over_mac, S_IRUGO | S_IWUSR, - bonding_show_fail_over_mac, bonding_store_fail_over_mac); + bonding_show_fail_over_mac, bonding_sysfs_store_option); -/* - * Show and set the arp timer interval. There are two tricky bits - * here. First, if ARP monitoring is activated, then we must disable - * MII monitoring. Second, if the ARP timer isn't running, we must - * start it. - */ +/* Show the arp timer interval. */ static ssize_t bonding_show_arp_interval(struct device *d, struct device_attribute *attr, char *buf) @@ -483,42 +293,16 @@ static ssize_t bonding_show_arp_interval(struct device *d, return sprintf(buf, "%d\n", bond->params.arp_interval); } - -static ssize_t bonding_store_arp_interval(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct bonding *bond = to_bond(d); - int new_value, ret; - - if (sscanf(buf, "%d", &new_value) != 1) { - pr_err("%s: no arp_interval value specified.\n", - bond->dev->name); - return -EINVAL; - } - - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_arp_interval_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; -} static DEVICE_ATTR(arp_interval, S_IRUGO | S_IWUSR, - bonding_show_arp_interval, bonding_store_arp_interval); + bonding_show_arp_interval, bonding_sysfs_store_option); -/* - * Show and set the arp targets. - */ +/* Show the arp targets. */ static ssize_t bonding_show_arp_targets(struct device *d, struct device_attribute *attr, char *buf) { - int i, res = 0; struct bonding *bond = to_bond(d); + int i, res = 0; for (i = 0; i < BOND_MAX_ARP_TARGETS; i++) { if (bond->params.arp_targets[i]) @@ -527,47 +311,13 @@ static ssize_t bonding_show_arp_targets(struct device *d, } if (res) buf[res-1] = '\n'; /* eat the leftover space */ - return res; -} - -static ssize_t bonding_store_arp_targets(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct bonding *bond = to_bond(d); - __be32 target; - int ret = -EPERM; - if (!in4_pton(buf + 1, -1, (u8 *)&target, -1, NULL)) { - pr_err("%s: invalid ARP target %pI4 specified\n", - bond->dev->name, &target); - return -EPERM; - } - - if (!rtnl_trylock()) - return restart_syscall(); - - if (buf[0] == '+') - ret = bond_option_arp_ip_target_add(bond, target); - else if (buf[0] == '-') - ret = bond_option_arp_ip_target_rem(bond, target); - else - pr_err("no command found in arp_ip_targets file for bond %s. Use +<addr> or -<addr>.\n", - bond->dev->name); - - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; + return res; } -static DEVICE_ATTR(arp_ip_target, S_IRUGO | S_IWUSR , bonding_show_arp_targets, bonding_store_arp_targets); +static DEVICE_ATTR(arp_ip_target, S_IRUGO | S_IWUSR, + bonding_show_arp_targets, bonding_sysfs_store_option); -/* - * Show and set the up and down delays. These must be multiples of the - * MII monitoring value, and are stored internally as the multiplier. - * Thus, we must translate to MS for the real world. - */ +/* Show the up and down delays. */ static ssize_t bonding_show_downdelay(struct device *d, struct device_attribute *attr, char *buf) @@ -576,31 +326,8 @@ static ssize_t bonding_show_downdelay(struct device *d, return sprintf(buf, "%d\n", bond->params.downdelay * bond->params.miimon); } - -static ssize_t bonding_store_downdelay(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - int new_value, ret; - struct bonding *bond = to_bond(d); - - if (sscanf(buf, "%d", &new_value) != 1) { - pr_err("%s: no down delay value specified.\n", bond->dev->name); - return -EINVAL; - } - - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_downdelay_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; -} static DEVICE_ATTR(downdelay, S_IRUGO | S_IWUSR, - bonding_show_downdelay, bonding_store_downdelay); + bonding_show_downdelay, bonding_sysfs_store_option); static ssize_t bonding_show_updelay(struct device *d, struct device_attribute *attr, @@ -611,74 +338,23 @@ static ssize_t bonding_show_updelay(struct device *d, return sprintf(buf, "%d\n", bond->params.updelay * bond->params.miimon); } - -static ssize_t bonding_store_updelay(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - int new_value, ret; - struct bonding *bond = to_bond(d); - - if (sscanf(buf, "%d", &new_value) != 1) { - pr_err("%s: no up delay value specified.\n", - bond->dev->name); - return -EINVAL; - } - - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_updelay_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; -} static DEVICE_ATTR(updelay, S_IRUGO | S_IWUSR, - bonding_show_updelay, bonding_store_updelay); + bonding_show_updelay, bonding_sysfs_store_option); -/* - * Show and set the LACP interval. Interface must be down, and the mode - * must be set to 802.3ad mode. - */ +/* Show the LACP interval. */ static ssize_t bonding_show_lacp(struct device *d, struct device_attribute *attr, char *buf) { struct bonding *bond = to_bond(d); + const struct bond_opt_value *val; - return sprintf(buf, "%s %d\n", - bond_lacp_tbl[bond->params.lacp_fast].modename, - bond->params.lacp_fast); -} - -static ssize_t bonding_store_lacp(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct bonding *bond = to_bond(d); - int new_value, ret; - - new_value = bond_parse_parm(buf, bond_lacp_tbl); - if (new_value < 0) { - pr_err("%s: Ignoring invalid LACP rate value %.*s.\n", - bond->dev->name, (int)strlen(buf) - 1, buf); - return -EINVAL; - } + val = bond_opt_get_val(BOND_OPT_LACP_RATE, bond->params.lacp_fast); - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_lacp_rate_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; + return sprintf(buf, "%s %d\n", val->string, bond->params.lacp_fast); } static DEVICE_ATTR(lacp_rate, S_IRUGO | S_IWUSR, - bonding_show_lacp, bonding_store_lacp); + bonding_show_lacp, bonding_sysfs_store_option); static ssize_t bonding_show_min_links(struct device *d, struct device_attribute *attr, @@ -686,79 +362,26 @@ static ssize_t bonding_show_min_links(struct device *d, { struct bonding *bond = to_bond(d); - return sprintf(buf, "%d\n", bond->params.min_links); -} - -static ssize_t bonding_store_min_links(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct bonding *bond = to_bond(d); - int ret; - unsigned int new_value; - - ret = kstrtouint(buf, 0, &new_value); - if (ret < 0) { - pr_err("%s: Ignoring invalid min links value %s.\n", - bond->dev->name, buf); - return ret; - } - - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_min_links_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; + return sprintf(buf, "%u\n", bond->params.min_links); } static DEVICE_ATTR(min_links, S_IRUGO | S_IWUSR, - bonding_show_min_links, bonding_store_min_links); + bonding_show_min_links, bonding_sysfs_store_option); static ssize_t bonding_show_ad_select(struct device *d, struct device_attribute *attr, char *buf) { struct bonding *bond = to_bond(d); + const struct bond_opt_value *val; - return sprintf(buf, "%s %d\n", - ad_select_tbl[bond->params.ad_select].modename, - bond->params.ad_select); -} - - -static ssize_t bonding_store_ad_select(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - int new_value, ret; - struct bonding *bond = to_bond(d); - - new_value = bond_parse_parm(buf, ad_select_tbl); - if (new_value < 0) { - pr_err("%s: Ignoring invalid ad_select value %.*s.\n", - bond->dev->name, (int)strlen(buf) - 1, buf); - return -EINVAL; - } - - if (!rtnl_trylock()) - return restart_syscall(); + val = bond_opt_get_val(BOND_OPT_AD_SELECT, bond->params.ad_select); - ret = bond_option_ad_select_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; + return sprintf(buf, "%s %d\n", val->string, bond->params.ad_select); } static DEVICE_ATTR(ad_select, S_IRUGO | S_IWUSR, - bonding_show_ad_select, bonding_store_ad_select); + bonding_show_ad_select, bonding_sysfs_store_option); -/* - * Show and set the number of peer notifications to send after a failover event. - */ +/* Show and set the number of peer notifications to send after a failover event. */ static ssize_t bonding_show_num_peer_notif(struct device *d, struct device_attribute *attr, char *buf) @@ -772,24 +395,12 @@ static ssize_t bonding_store_num_peer_notif(struct device *d, const char *buf, size_t count) { struct bonding *bond = to_bond(d); - u8 new_value; int ret; - ret = kstrtou8(buf, 10, &new_value); - if (ret) { - pr_err("%s: invalid value %s specified.\n", - bond->dev->name, buf); - return ret; - } - - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_num_peer_notif_set(bond, new_value); + ret = bond_opt_tryset_rtnl(bond, BOND_OPT_NUM_PEER_NOTIF, (char *)buf); if (!ret) ret = count; - rtnl_unlock(); return ret; } static DEVICE_ATTR(num_grat_arp, S_IRUGO | S_IWUSR, @@ -797,12 +408,7 @@ static DEVICE_ATTR(num_grat_arp, S_IRUGO | S_IWUSR, static DEVICE_ATTR(num_unsol_na, S_IRUGO | S_IWUSR, bonding_show_num_peer_notif, bonding_store_num_peer_notif); -/* - * Show and set the MII monitor interval. There are two tricky bits - * here. First, if MII monitoring is activated, then we must disable - * ARP monitoring. Second, if the timer isn't running, we must - * start it. - */ +/* Show the MII monitor interval. */ static ssize_t bonding_show_miimon(struct device *d, struct device_attribute *attr, char *buf) @@ -811,40 +417,10 @@ static ssize_t bonding_show_miimon(struct device *d, return sprintf(buf, "%d\n", bond->params.miimon); } - -static ssize_t bonding_store_miimon(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - int new_value, ret; - struct bonding *bond = to_bond(d); - - if (sscanf(buf, "%d", &new_value) != 1) { - pr_err("%s: no miimon value specified.\n", - bond->dev->name); - return -EINVAL; - } - - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_miimon_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; -} static DEVICE_ATTR(miimon, S_IRUGO | S_IWUSR, - bonding_show_miimon, bonding_store_miimon); + bonding_show_miimon, bonding_sysfs_store_option); -/* - * Show and set the primary slave. The store function is much - * simpler than bonding_store_slaves function because it only needs to - * handle one interface name. - * The bond must be a mode that supports a primary for this be - * set. - */ +/* Show the primary slave. */ static ssize_t bonding_show_primary(struct device *d, struct device_attribute *attr, char *buf) @@ -857,78 +433,27 @@ static ssize_t bonding_show_primary(struct device *d, return count; } - -static ssize_t bonding_store_primary(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct bonding *bond = to_bond(d); - char ifname[IFNAMSIZ]; - int ret; - - sscanf(buf, "%15s", ifname); /* IFNAMSIZ */ - if (ifname[0] == '\n') - ifname[0] = '\0'; - - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_primary_set(bond, ifname); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; -} static DEVICE_ATTR(primary, S_IRUGO | S_IWUSR, - bonding_show_primary, bonding_store_primary); + bonding_show_primary, bonding_sysfs_store_option); -/* - * Show and set the primary_reselect flag. - */ +/* Show the primary_reselect flag. */ static ssize_t bonding_show_primary_reselect(struct device *d, struct device_attribute *attr, char *buf) { struct bonding *bond = to_bond(d); + const struct bond_opt_value *val; - return sprintf(buf, "%s %d\n", - pri_reselect_tbl[bond->params.primary_reselect].modename, - bond->params.primary_reselect); -} - -static ssize_t bonding_store_primary_reselect(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - int new_value, ret; - struct bonding *bond = to_bond(d); - - new_value = bond_parse_parm(buf, pri_reselect_tbl); - if (new_value < 0) { - pr_err("%s: Ignoring invalid primary_reselect value %.*s.\n", - bond->dev->name, - (int) strlen(buf) - 1, buf); - return -EINVAL; - } - - if (!rtnl_trylock()) - return restart_syscall(); + val = bond_opt_get_val(BOND_OPT_PRIMARY_RESELECT, + bond->params.primary_reselect); - ret = bond_option_primary_reselect_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; + return sprintf(buf, "%s %d\n", + val->string, bond->params.primary_reselect); } static DEVICE_ATTR(primary_reselect, S_IRUGO | S_IWUSR, - bonding_show_primary_reselect, - bonding_store_primary_reselect); + bonding_show_primary_reselect, bonding_sysfs_store_option); -/* - * Show and set the use_carrier flag. - */ +/* Show the use_carrier flag. */ static ssize_t bonding_show_carrier(struct device *d, struct device_attribute *attr, char *buf) @@ -937,37 +462,11 @@ static ssize_t bonding_show_carrier(struct device *d, return sprintf(buf, "%d\n", bond->params.use_carrier); } - -static ssize_t bonding_store_carrier(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - int new_value, ret; - struct bonding *bond = to_bond(d); - - if (sscanf(buf, "%d", &new_value) != 1) { - pr_err("%s: no use_carrier value specified.\n", - bond->dev->name); - return -EINVAL; - } - - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_use_carrier_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; -} static DEVICE_ATTR(use_carrier, S_IRUGO | S_IWUSR, - bonding_show_carrier, bonding_store_carrier); + bonding_show_carrier, bonding_sysfs_store_option); -/* - * Show and set currently active_slave. - */ +/* Show currently active_slave. */ static ssize_t bonding_show_active_slave(struct device *d, struct device_attribute *attr, char *buf) @@ -984,47 +483,10 @@ static ssize_t bonding_show_active_slave(struct device *d, return count; } - -static ssize_t bonding_store_active_slave(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - int ret; - struct bonding *bond = to_bond(d); - char ifname[IFNAMSIZ]; - struct net_device *dev; - - if (!rtnl_trylock()) - return restart_syscall(); - - sscanf(buf, "%15s", ifname); /* IFNAMSIZ */ - if (!strlen(ifname) || buf[0] == '\n') { - dev = NULL; - } else { - dev = __dev_get_by_name(dev_net(bond->dev), ifname); - if (!dev) { - ret = -ENODEV; - goto out; - } - } - - ret = bond_option_active_slave_set(bond, dev); - if (!ret) - ret = count; - - out: - rtnl_unlock(); - - return ret; - -} static DEVICE_ATTR(active_slave, S_IRUGO | S_IWUSR, - bonding_show_active_slave, bonding_store_active_slave); - + bonding_show_active_slave, bonding_sysfs_store_option); -/* - * Show link status of the bond interface. - */ +/* Show link status of the bond interface. */ static ssize_t bonding_show_mii_status(struct device *d, struct device_attribute *attr, char *buf) @@ -1035,9 +497,7 @@ static ssize_t bonding_show_mii_status(struct device *d, } static DEVICE_ATTR(mii_status, S_IRUGO, bonding_show_mii_status, NULL); -/* - * Show current 802.3ad aggregator ID. - */ +/* Show current 802.3ad aggregator ID. */ static ssize_t bonding_show_ad_aggregator(struct device *d, struct device_attribute *attr, char *buf) @@ -1045,7 +505,7 @@ static ssize_t bonding_show_ad_aggregator(struct device *d, int count = 0; struct bonding *bond = to_bond(d); - if (bond->params.mode == BOND_MODE_8023AD) { + if (BOND_MODE(bond) == BOND_MODE_8023AD) { struct ad_info ad_info; count = sprintf(buf, "%d\n", bond_3ad_get_active_agg_info(bond, &ad_info) @@ -1057,9 +517,7 @@ static ssize_t bonding_show_ad_aggregator(struct device *d, static DEVICE_ATTR(ad_aggregator, S_IRUGO, bonding_show_ad_aggregator, NULL); -/* - * Show number of active 802.3ad ports. - */ +/* Show number of active 802.3ad ports. */ static ssize_t bonding_show_ad_num_ports(struct device *d, struct device_attribute *attr, char *buf) @@ -1067,7 +525,7 @@ static ssize_t bonding_show_ad_num_ports(struct device *d, int count = 0; struct bonding *bond = to_bond(d); - if (bond->params.mode == BOND_MODE_8023AD) { + if (BOND_MODE(bond) == BOND_MODE_8023AD) { struct ad_info ad_info; count = sprintf(buf, "%d\n", bond_3ad_get_active_agg_info(bond, &ad_info) @@ -1079,9 +537,7 @@ static ssize_t bonding_show_ad_num_ports(struct device *d, static DEVICE_ATTR(ad_num_ports, S_IRUGO, bonding_show_ad_num_ports, NULL); -/* - * Show current 802.3ad actor key. - */ +/* Show current 802.3ad actor key. */ static ssize_t bonding_show_ad_actor_key(struct device *d, struct device_attribute *attr, char *buf) @@ -1089,7 +545,7 @@ static ssize_t bonding_show_ad_actor_key(struct device *d, int count = 0; struct bonding *bond = to_bond(d); - if (bond->params.mode == BOND_MODE_8023AD) { + if (BOND_MODE(bond) == BOND_MODE_8023AD) { struct ad_info ad_info; count = sprintf(buf, "%d\n", bond_3ad_get_active_agg_info(bond, &ad_info) @@ -1101,9 +557,7 @@ static ssize_t bonding_show_ad_actor_key(struct device *d, static DEVICE_ATTR(ad_actor_key, S_IRUGO, bonding_show_ad_actor_key, NULL); -/* - * Show current 802.3ad partner key. - */ +/* Show current 802.3ad partner key. */ static ssize_t bonding_show_ad_partner_key(struct device *d, struct device_attribute *attr, char *buf) @@ -1111,7 +565,7 @@ static ssize_t bonding_show_ad_partner_key(struct device *d, int count = 0; struct bonding *bond = to_bond(d); - if (bond->params.mode == BOND_MODE_8023AD) { + if (BOND_MODE(bond) == BOND_MODE_8023AD) { struct ad_info ad_info; count = sprintf(buf, "%d\n", bond_3ad_get_active_agg_info(bond, &ad_info) @@ -1123,9 +577,7 @@ static ssize_t bonding_show_ad_partner_key(struct device *d, static DEVICE_ATTR(ad_partner_key, S_IRUGO, bonding_show_ad_partner_key, NULL); -/* - * Show current 802.3ad partner mac. - */ +/* Show current 802.3ad partner mac. */ static ssize_t bonding_show_ad_partner_mac(struct device *d, struct device_attribute *attr, char *buf) @@ -1133,7 +585,7 @@ static ssize_t bonding_show_ad_partner_mac(struct device *d, int count = 0; struct bonding *bond = to_bond(d); - if (bond->params.mode == BOND_MODE_8023AD) { + if (BOND_MODE(bond) == BOND_MODE_8023AD) { struct ad_info ad_info; if (!bond_3ad_get_active_agg_info(bond, &ad_info)) count = sprintf(buf, "%pM\n", ad_info.partner_system); @@ -1143,9 +595,7 @@ static ssize_t bonding_show_ad_partner_mac(struct device *d, } static DEVICE_ATTR(ad_partner_mac, S_IRUGO, bonding_show_ad_partner_mac, NULL); -/* - * Show the queue_ids of the slaves in the current bond. - */ +/* Show the queue_ids of the slaves in the current bond. */ static ssize_t bonding_show_queue_id(struct device *d, struct device_attribute *attr, char *buf) @@ -1176,88 +626,11 @@ static ssize_t bonding_show_queue_id(struct device *d, return res; } - -/* - * Set the queue_ids of the slaves in the current bond. The bond - * interface must be enslaved for this to work. - */ -static ssize_t bonding_store_queue_id(struct device *d, - struct device_attribute *attr, - const char *buffer, size_t count) -{ - struct slave *slave, *update_slave; - struct bonding *bond = to_bond(d); - struct list_head *iter; - u16 qid; - int ret = count; - char *delim; - struct net_device *sdev = NULL; - - if (!rtnl_trylock()) - return restart_syscall(); - - /* delim will point to queue id if successful */ - delim = strchr(buffer, ':'); - if (!delim) - goto err_no_cmd; - - /* - * Terminate string that points to device name and bump it - * up one, so we can read the queue id there. - */ - *delim = '\0'; - if (sscanf(++delim, "%hd\n", &qid) != 1) - goto err_no_cmd; - - /* Check buffer length, valid ifname and queue id */ - if (strlen(buffer) > IFNAMSIZ || - !dev_valid_name(buffer) || - qid > bond->dev->real_num_tx_queues) - goto err_no_cmd; - - /* Get the pointer to that interface if it exists */ - sdev = __dev_get_by_name(dev_net(bond->dev), buffer); - if (!sdev) - goto err_no_cmd; - - /* Search for thes slave and check for duplicate qids */ - update_slave = NULL; - bond_for_each_slave(bond, slave, iter) { - if (sdev == slave->dev) - /* - * We don't need to check the matching - * slave for dups, since we're overwriting it - */ - update_slave = slave; - else if (qid && qid == slave->queue_id) { - goto err_no_cmd; - } - } - - if (!update_slave) - goto err_no_cmd; - - /* Actually set the qids for the slave */ - update_slave->queue_id = qid; - -out: - rtnl_unlock(); - return ret; - -err_no_cmd: - pr_info("invalid input for queue_id set for %s.\n", - bond->dev->name); - ret = -EPERM; - goto out; -} - static DEVICE_ATTR(queue_id, S_IRUGO | S_IWUSR, bonding_show_queue_id, - bonding_store_queue_id); + bonding_sysfs_store_option); -/* - * Show and set the all_slaves_active flag. - */ +/* Show the all_slaves_active flag. */ static ssize_t bonding_show_slaves_active(struct device *d, struct device_attribute *attr, char *buf) @@ -1266,36 +639,10 @@ static ssize_t bonding_show_slaves_active(struct device *d, return sprintf(buf, "%d\n", bond->params.all_slaves_active); } - -static ssize_t bonding_store_slaves_active(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct bonding *bond = to_bond(d); - int new_value, ret; - - if (sscanf(buf, "%d", &new_value) != 1) { - pr_err("%s: no all_slaves_active value specified.\n", - bond->dev->name); - return -EINVAL; - } - - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_all_slaves_active_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; -} static DEVICE_ATTR(all_slaves_active, S_IRUGO | S_IWUSR, - bonding_show_slaves_active, bonding_store_slaves_active); + bonding_show_slaves_active, bonding_sysfs_store_option); -/* - * Show and set the number of IGMP membership reports to send on link failure - */ +/* Show the number of IGMP membership reports to send on link failure */ static ssize_t bonding_show_resend_igmp(struct device *d, struct device_attribute *attr, char *buf) @@ -1304,33 +651,8 @@ static ssize_t bonding_show_resend_igmp(struct device *d, return sprintf(buf, "%d\n", bond->params.resend_igmp); } - -static ssize_t bonding_store_resend_igmp(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - int new_value, ret = count; - struct bonding *bond = to_bond(d); - - if (sscanf(buf, "%d", &new_value) != 1) { - pr_err("%s: no resend_igmp value specified.\n", - bond->dev->name); - return -EINVAL; - } - - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_resend_igmp_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; -} - static DEVICE_ATTR(resend_igmp, S_IRUGO | S_IWUSR, - bonding_show_resend_igmp, bonding_store_resend_igmp); + bonding_show_resend_igmp, bonding_sysfs_store_option); static ssize_t bonding_show_lp_interval(struct device *d, @@ -1338,35 +660,21 @@ static ssize_t bonding_show_lp_interval(struct device *d, char *buf) { struct bonding *bond = to_bond(d); + return sprintf(buf, "%d\n", bond->params.lp_interval); } +static DEVICE_ATTR(lp_interval, S_IRUGO | S_IWUSR, + bonding_show_lp_interval, bonding_sysfs_store_option); -static ssize_t bonding_store_lp_interval(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) +static ssize_t bonding_show_tlb_dynamic_lb(struct device *d, + struct device_attribute *attr, + char *buf) { struct bonding *bond = to_bond(d); - int new_value, ret; - - if (sscanf(buf, "%d", &new_value) != 1) { - pr_err("%s: no lp interval value specified.\n", - bond->dev->name); - return -EINVAL; - } - - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_lp_interval_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; + return sprintf(buf, "%d\n", bond->params.tlb_dynamic_lb); } - -static DEVICE_ATTR(lp_interval, S_IRUGO | S_IWUSR, - bonding_show_lp_interval, bonding_store_lp_interval); +static DEVICE_ATTR(tlb_dynamic_lb, S_IRUGO | S_IWUSR, + bonding_show_tlb_dynamic_lb, bonding_sysfs_store_option); static ssize_t bonding_show_packets_per_slave(struct device *d, struct device_attribute *attr, @@ -1375,39 +683,10 @@ static ssize_t bonding_show_packets_per_slave(struct device *d, struct bonding *bond = to_bond(d); unsigned int packets_per_slave = bond->params.packets_per_slave; - if (packets_per_slave > 1) - packets_per_slave = reciprocal_value(packets_per_slave); - return sprintf(buf, "%u\n", packets_per_slave); } - -static ssize_t bonding_store_packets_per_slave(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct bonding *bond = to_bond(d); - int new_value, ret; - - if (sscanf(buf, "%d", &new_value) != 1) { - pr_err("%s: no packets_per_slave value specified.\n", - bond->dev->name); - return -EINVAL; - } - - if (!rtnl_trylock()) - return restart_syscall(); - - ret = bond_option_packets_per_slave_set(bond, new_value); - if (!ret) - ret = count; - - rtnl_unlock(); - return ret; -} - static DEVICE_ATTR(packets_per_slave, S_IRUGO | S_IWUSR, - bonding_show_packets_per_slave, - bonding_store_packets_per_slave); + bonding_show_packets_per_slave, bonding_sysfs_store_option); static struct attribute *per_bond_attrs[] = { &dev_attr_slaves.attr, @@ -1441,6 +720,7 @@ static struct attribute *per_bond_attrs[] = { &dev_attr_min_links.attr, &dev_attr_lp_interval.attr, &dev_attr_packets_per_slave.attr, + &dev_attr_tlb_dynamic_lb.attr, NULL, }; @@ -1449,8 +729,7 @@ static struct attribute_group bonding_group = { .attrs = per_bond_attrs, }; -/* - * Initialize sysfs. This sets up the bonding_masters file in +/* Initialize sysfs. This sets up the bonding_masters file in * /sys/class/net. */ int bond_create_sysfs(struct bond_net *bn) @@ -1462,8 +741,7 @@ int bond_create_sysfs(struct bond_net *bn) ret = netdev_class_create_file_ns(&bn->class_attr_bonding_masters, bn->net); - /* - * Permit multiple loads of the module by ignoring failures to + /* Permit multiple loads of the module by ignoring failures to * create the bonding_masters sysfs file. Bonding devices * created by second or subsequent loads of the module will * not be listed in, or controllable by, bonding_masters, but @@ -1477,7 +755,7 @@ int bond_create_sysfs(struct bond_net *bn) /* Is someone being kinky and naming a device bonding_master? */ if (__dev_get_by_name(bn->net, class_attr_bonding_masters.attr.name)) - pr_err("network device named %s already exists in sysfs", + pr_err("network device named %s already exists in sysfs\n", class_attr_bonding_masters.attr.name); ret = 0; } @@ -1486,16 +764,13 @@ int bond_create_sysfs(struct bond_net *bn) } -/* - * Remove /sys/class/net/bonding_masters. - */ +/* Remove /sys/class/net/bonding_masters. */ void bond_destroy_sysfs(struct bond_net *bn) { netdev_class_remove_file_ns(&bn->class_attr_bonding_masters, bn->net); } -/* - * Initialize sysfs for each bond. This sets up and registers +/* Initialize sysfs for each bond. This sets up and registers * the 'bondctl' directory for each individual bond under /sys/class/net. */ void bond_prepare_sysfs_group(struct bonding *bond) diff --git a/drivers/net/bonding/bond_sysfs_slave.c b/drivers/net/bonding/bond_sysfs_slave.c index 7cb97de3cf6..198677f58ce 100644 --- a/drivers/net/bonding/bond_sysfs_slave.c +++ b/drivers/net/bonding/bond_sysfs_slave.c @@ -69,8 +69,8 @@ static ssize_t ad_aggregator_id_show(struct slave *slave, char *buf) { const struct aggregator *agg; - if (slave->bond->params.mode == BOND_MODE_8023AD) { - agg = SLAVE_AD_INFO(slave).port.aggregator; + if (BOND_MODE(slave->bond) == BOND_MODE_8023AD) { + agg = SLAVE_AD_INFO(slave)->port.aggregator; if (agg) return sprintf(buf, "%d\n", agg->aggregator_identifier); @@ -102,7 +102,7 @@ static ssize_t slave_show(struct kobject *kobj, return slave_attr->show(slave, buf); } -const struct sysfs_ops slave_sysfs_ops = { +static const struct sysfs_ops slave_sysfs_ops = { .show = slave_show, }; @@ -118,7 +118,7 @@ int bond_sysfs_slave_add(struct slave *slave) int err; err = kobject_init_and_add(&slave->kobj, &slave_ktype, - &(slave->dev->dev.kobj), "slave"); + &(slave->dev->dev.kobj), "bonding_slave"); if (err) return err; diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h index 8a935f8f2b3..0b4d9cde0b0 100644 --- a/drivers/net/bonding/bonding.h +++ b/drivers/net/bonding/bonding.h @@ -23,8 +23,11 @@ #include <linux/netpoll.h> #include <linux/inetdevice.h> #include <linux/etherdevice.h> +#include <linux/reciprocal_div.h> + #include "bond_3ad.h" #include "bond_alb.h" +#include "bond_options.h" #define DRV_VERSION "3.7.1" #define DRV_RELDATE "April 27, 2011" @@ -33,46 +36,11 @@ #define bond_version DRV_DESCRIPTION ": v" DRV_VERSION " (" DRV_RELDATE ")\n" +#define BOND_MAX_VLAN_ENCAP 2 #define BOND_MAX_ARP_TARGETS 16 #define BOND_DEFAULT_MIIMON 100 -#define IS_UP(dev) \ - ((((dev)->flags & IFF_UP) == IFF_UP) && \ - netif_running(dev) && \ - netif_carrier_ok(dev)) - -/* - * Checks whether slave is ready for transmit. - */ -#define SLAVE_IS_OK(slave) \ - (((slave)->dev->flags & IFF_UP) && \ - netif_running((slave)->dev) && \ - ((slave)->link == BOND_LINK_UP) && \ - bond_is_active_slave(slave)) - - -#define USES_PRIMARY(mode) \ - (((mode) == BOND_MODE_ACTIVEBACKUP) || \ - ((mode) == BOND_MODE_TLB) || \ - ((mode) == BOND_MODE_ALB)) - -#define BOND_NO_USES_ARP(mode) \ - (((mode) == BOND_MODE_8023AD) || \ - ((mode) == BOND_MODE_TLB) || \ - ((mode) == BOND_MODE_ALB)) - -#define TX_QUEUE_OVERRIDE(mode) \ - (((mode) == BOND_MODE_ACTIVEBACKUP) || \ - ((mode) == BOND_MODE_ROUNDROBIN)) - -#define BOND_MODE_IS_LB(mode) \ - (((mode) == BOND_MODE_TLB) || \ - ((mode) == BOND_MODE_ALB)) - -#define IS_IP_TARGET_UNUSABLE_ADDRESS(a) \ - ((htonl(INADDR_BROADCAST) == a) || \ - ipv4_is_zeronet(a)) /* * Less bad way to call ioctl from within the kernel; this needs to be * done some other way to get the call out of interrupt context. @@ -86,6 +54,8 @@ set_fs(fs); \ res; }) +#define BOND_MODE(bond) ((bond)->params.mode) + /* slave list primitives */ #define bond_slave_list(bond) (&(bond)->dev->adj_list.lower) @@ -171,6 +141,8 @@ struct bond_params { int resend_igmp; int lp_interval; int packets_per_slave; + int tlb_dynamic_lb; + struct reciprocal_value reciprocal_packets_per_slave; }; struct bond_parm_tbl { @@ -178,27 +150,27 @@ struct bond_parm_tbl { int mode; }; -#define BOND_MAX_MODENAME_LEN 20 - struct slave { struct net_device *dev; /* first - useful for panic debug */ struct bonding *bond; /* our master */ int delay; - unsigned long jiffies; - unsigned long last_arp_rx; + /* all three in jiffies */ + unsigned long last_link_up; + unsigned long last_rx; unsigned long target_last_arp_rx[BOND_MAX_ARP_TARGETS]; s8 link; /* one of BOND_LINK_XXXX */ s8 new_link; u8 backup:1, /* indicates backup slave. Value corresponds with BOND_STATE_ACTIVE and BOND_STATE_BACKUP */ - inactive:1; /* indicates inactive slave */ + inactive:1, /* indicates inactive slave */ + should_notify:1; /* indicateds whether the state changed */ u8 duplex; u32 original_mtu; u32 link_failure_count; u32 speed; u16 queue_id; u8 perm_hwaddr[ETH_ALEN]; - struct ad_slave_info ad_info; /* HUGE - better to dynamically alloc */ + struct ad_slave_info *ad_info; struct tlb_slave_info tlb_info; #ifdef CONFIG_NET_POLL_CONTROLLER struct netpoll *np; @@ -260,6 +232,11 @@ struct bonding { #define bond_slave_get_rtnl(dev) \ ((struct slave *) rtnl_dereference(dev->rx_handler_data)) +struct bond_vlan_tag { + __be16 vlan_proto; + unsigned short vlan_id; +}; + /** * Returns NULL if the net_device does not belong to any of the bond's slaves * @@ -273,21 +250,48 @@ static inline struct slave *bond_get_slave_by_dev(struct bonding *bond, static inline struct bonding *bond_get_bond_by_slave(struct slave *slave) { - if (!slave || !slave->bond) - return NULL; return slave->bond; } +static inline bool bond_should_override_tx_queue(struct bonding *bond) +{ + return BOND_MODE(bond) == BOND_MODE_ACTIVEBACKUP || + BOND_MODE(bond) == BOND_MODE_ROUNDROBIN; +} + static inline bool bond_is_lb(const struct bonding *bond) { - return BOND_MODE_IS_LB(bond->params.mode); + return BOND_MODE(bond) == BOND_MODE_TLB || + BOND_MODE(bond) == BOND_MODE_ALB; +} + +static inline bool bond_mode_uses_arp(int mode) +{ + return mode != BOND_MODE_8023AD && mode != BOND_MODE_TLB && + mode != BOND_MODE_ALB; +} + +static inline bool bond_mode_uses_primary(int mode) +{ + return mode == BOND_MODE_ACTIVEBACKUP || mode == BOND_MODE_TLB || + mode == BOND_MODE_ALB; +} + +static inline bool bond_uses_primary(struct bonding *bond) +{ + return bond_mode_uses_primary(BOND_MODE(bond)); +} + +static inline bool bond_slave_is_up(struct slave *slave) +{ + return netif_running(slave->dev) && netif_carrier_ok(slave->dev); } static inline void bond_set_active_slave(struct slave *slave) { if (slave->backup) { slave->backup = 0; - rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_KERNEL); + rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_ATOMIC); } } @@ -295,7 +299,51 @@ static inline void bond_set_backup_slave(struct slave *slave) { if (!slave->backup) { slave->backup = 1; - rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_KERNEL); + rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_ATOMIC); + } +} + +static inline void bond_set_slave_state(struct slave *slave, + int slave_state, bool notify) +{ + if (slave->backup == slave_state) + return; + + slave->backup = slave_state; + if (notify) { + rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_ATOMIC); + slave->should_notify = 0; + } else { + if (slave->should_notify) + slave->should_notify = 0; + else + slave->should_notify = 1; + } +} + +static inline void bond_slave_state_change(struct bonding *bond) +{ + struct list_head *iter; + struct slave *tmp; + + bond_for_each_slave(bond, tmp, iter) { + if (tmp->link == BOND_LINK_UP) + bond_set_active_slave(tmp); + else if (tmp->link == BOND_LINK_DOWN) + bond_set_backup_slave(tmp); + } +} + +static inline void bond_slave_state_notify(struct bonding *bond) +{ + struct list_head *iter; + struct slave *tmp; + + bond_for_each_slave(bond, tmp, iter) { + if (tmp->should_notify) { + rtmsg_ifinfo(RTM_NEWLINK, tmp->dev, 0, GFP_ATOMIC); + tmp->should_notify = 0; + } } } @@ -309,6 +357,12 @@ static inline bool bond_is_active_slave(struct slave *slave) return !bond_slave_state(slave); } +static inline bool bond_slave_can_tx(struct slave *slave) +{ + return bond_slave_is_up(slave) && slave->link == BOND_LINK_UP && + bond_is_active_slave(slave); +} + #define BOND_PRI_RESELECT_ALWAYS 0 #define BOND_PRI_RESELECT_BETTER 1 #define BOND_PRI_RESELECT_FAILURE 2 @@ -325,6 +379,14 @@ static inline bool bond_is_active_slave(struct slave *slave) #define BOND_ARP_VALIDATE_BACKUP (1 << BOND_STATE_BACKUP) #define BOND_ARP_VALIDATE_ALL (BOND_ARP_VALIDATE_ACTIVE | \ BOND_ARP_VALIDATE_BACKUP) +#define BOND_ARP_FILTER (BOND_ARP_VALIDATE_ALL + 1) +#define BOND_ARP_FILTER_ACTIVE (BOND_ARP_VALIDATE_ACTIVE | \ + BOND_ARP_FILTER) +#define BOND_ARP_FILTER_BACKUP (BOND_ARP_VALIDATE_BACKUP | \ + BOND_ARP_FILTER) + +#define BOND_SLAVE_NOTIFY_NOW true +#define BOND_SLAVE_NOTIFY_LATER false static inline int slave_do_arp_validate(struct bonding *bond, struct slave *slave) @@ -332,6 +394,16 @@ static inline int slave_do_arp_validate(struct bonding *bond, return bond->params.arp_validate & (1 << bond_slave_state(slave)); } +static inline int slave_do_arp_validate_only(struct bonding *bond) +{ + return bond->params.arp_validate & BOND_ARP_FILTER; +} + +static inline int bond_is_ip_target_ok(__be32 addr) +{ + return !ipv4_is_lbcast(addr) && !ipv4_is_zeronet(addr); +} + /* Get the oldest arp which we've received on this slave for bond's * arp_targets. */ @@ -351,14 +423,10 @@ static inline unsigned long slave_oldest_target_arp_rx(struct bonding *bond, static inline unsigned long slave_last_rx(struct bonding *bond, struct slave *slave) { - if (slave_do_arp_validate(bond, slave)) { - if (bond->params.arp_all_targets == BOND_ARP_TARGETS_ALL) - return slave_oldest_target_arp_rx(bond, slave); - else - return slave->last_arp_rx; - } + if (bond->params.arp_all_targets == BOND_ARP_TARGETS_ALL) + return slave_oldest_target_arp_rx(bond, slave); - return slave->dev->last_rx; + return slave->last_rx; } #ifdef CONFIG_NET_POLL_CONTROLLER @@ -377,17 +445,19 @@ static inline void bond_netpoll_send_skb(const struct slave *slave, } #endif -static inline void bond_set_slave_inactive_flags(struct slave *slave) +static inline void bond_set_slave_inactive_flags(struct slave *slave, + bool notify) { if (!bond_is_lb(slave->bond)) - bond_set_backup_slave(slave); + bond_set_slave_state(slave, BOND_STATE_BACKUP, notify); if (!slave->bond->params.all_slaves_active) slave->inactive = 1; } -static inline void bond_set_slave_active_flags(struct slave *slave) +static inline void bond_set_slave_active_flags(struct slave *slave, + bool notify) { - bond_set_active_slave(slave); + bond_set_slave_state(slave, BOND_STATE_ACTIVE, notify); slave->inactive = 0; } @@ -411,16 +481,14 @@ 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 bond_net { + struct net *net; /* Associated network namespace */ + struct list_head dev_list; +#ifdef CONFIG_PROC_FS + struct proc_dir_entry *proc_dir; +#endif + struct class_attribute class_attr_bonding_masters; +}; int bond_arp_rcv(const struct sk_buff *skb, struct bonding *bond, struct slave *slave); void bond_dev_queue_xmit(struct bonding *bond, struct sk_buff *skb, struct net_device *slave_dev); @@ -432,10 +500,7 @@ int bond_sysfs_slave_add(struct slave *slave); void bond_sysfs_slave_del(struct slave *slave); int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev); int bond_release(struct net_device *bond_dev, struct net_device *slave_dev); -int bond_get_slave(struct net_device *slave_dev, struct sk_buff *skb); -int bond_xmit_hash(struct bonding *bond, struct sk_buff *skb, int count); -int bond_parse_parm(const char *mode_arg, const struct bond_parm_tbl *tbl); -int bond_parm_tbl_lookup(int mode, const struct bond_parm_tbl *tbl); +u32 bond_xmit_hash(struct bonding *bond, struct sk_buff *skb); void bond_select_active_slave(struct bonding *bond); void bond_change_active_slave(struct bonding *bond, struct slave *new_active); void bond_create_debugfs(void); @@ -448,47 +513,12 @@ void bond_setup(struct net_device *bond_dev); unsigned int bond_get_num_tx_queues(void); int bond_netlink_init(void); void bond_netlink_fini(void); -int bond_option_mode_set(struct bonding *bond, int mode); -int bond_option_active_slave_set(struct bonding *bond, struct net_device *slave_dev); -int bond_option_miimon_set(struct bonding *bond, int miimon); -int bond_option_updelay_set(struct bonding *bond, int updelay); -int bond_option_downdelay_set(struct bonding *bond, int downdelay); -int bond_option_use_carrier_set(struct bonding *bond, int use_carrier); -int bond_option_arp_interval_set(struct bonding *bond, int arp_interval); -int bond_option_arp_ip_targets_set(struct bonding *bond, __be32 *targets, - int count); -int bond_option_arp_ip_target_add(struct bonding *bond, __be32 target); -int bond_option_arp_ip_target_rem(struct bonding *bond, __be32 target); -int bond_option_arp_validate_set(struct bonding *bond, int arp_validate); -int bond_option_arp_all_targets_set(struct bonding *bond, int arp_all_targets); -int bond_option_primary_set(struct bonding *bond, const char *primary); -int bond_option_primary_reselect_set(struct bonding *bond, - int primary_reselect); -int bond_option_fail_over_mac_set(struct bonding *bond, int fail_over_mac); -int bond_option_xmit_hash_policy_set(struct bonding *bond, - int xmit_hash_policy); -int bond_option_resend_igmp_set(struct bonding *bond, int resend_igmp); -int bond_option_num_peer_notif_set(struct bonding *bond, int num_peer_notif); -int bond_option_all_slaves_active_set(struct bonding *bond, - int all_slaves_active); -int bond_option_min_links_set(struct bonding *bond, int min_links); -int bond_option_lp_interval_set(struct bonding *bond, int min_links); -int bond_option_packets_per_slave_set(struct bonding *bond, - int packets_per_slave); -int bond_option_lacp_rate_set(struct bonding *bond, int lacp_rate); -int bond_option_ad_select_set(struct bonding *bond, int ad_select); struct net_device *bond_option_active_slave_get_rcu(struct bonding *bond); struct net_device *bond_option_active_slave_get(struct bonding *bond); const char *bond_slave_link_status(s8 link); - -struct bond_net { - struct net * net; /* Associated network namespace */ - struct list_head dev_list; -#ifdef CONFIG_PROC_FS - struct proc_dir_entry * proc_dir; -#endif - struct class_attribute class_attr_bonding_masters; -}; +bool bond_verify_device_path(struct net_device *start_dev, + struct net_device *end_dev, + struct bond_vlan_tag *tags); #ifdef CONFIG_PROC_FS void bond_create_proc_entry(struct bonding *bond); @@ -540,6 +570,27 @@ static inline struct slave *bond_slave_has_mac_rcu(struct bonding *bond, return NULL; } +/* Caller must hold rcu_read_lock() for read */ +static inline bool bond_slave_has_mac_rx(struct bonding *bond, const u8 *mac) +{ + struct list_head *iter; + struct slave *tmp; + struct netdev_hw_addr *ha; + + bond_for_each_slave_rcu(bond, tmp, iter) + if (ether_addr_equal_64bits(mac, tmp->dev->dev_addr)) + return true; + + if (netdev_uc_empty(bond->dev)) + return false; + + netdev_for_each_uc_addr(ha, bond->dev) + if (ether_addr_equal_64bits(mac, ha->addr)) + return true; + + return false; +} + /* Check if the ip is present in arp ip list, or first free slot if ip == 0 * Returns -1 if not found, index if found */ @@ -559,7 +610,6 @@ static inline int bond_get_targets_ip(__be32 *targets, __be32 ip) /* exported from bond_main.c */ extern int bond_net_id; extern const struct bond_parm_tbl bond_lacp_tbl[]; -extern const struct bond_parm_tbl bond_mode_tbl[]; extern const struct bond_parm_tbl xmit_hashtype_tbl[]; extern const struct bond_parm_tbl arp_validate_tbl[]; extern const struct bond_parm_tbl arp_all_targets_tbl[]; |