diff options
Diffstat (limited to 'drivers/net/bonding')
| -rw-r--r-- | drivers/net/bonding/Makefile | 3 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_3ad.c | 667 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_3ad.h | 59 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_alb.c | 37 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_ipv6.c | 216 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_main.c | 479 | ||||
| -rw-r--r-- | drivers/net/bonding/bond_sysfs.c | 160 | ||||
| -rw-r--r-- | drivers/net/bonding/bonding.h | 68 |
8 files changed, 1046 insertions, 643 deletions
diff --git a/drivers/net/bonding/Makefile b/drivers/net/bonding/Makefile index 5cdae2bc055..6f9c6faef24 100644 --- a/drivers/net/bonding/Makefile +++ b/drivers/net/bonding/Makefile @@ -6,3 +6,6 @@ obj-$(CONFIG_BONDING) += bonding.o bonding-objs := bond_main.o bond_3ad.o bond_alb.o bond_sysfs.o +ipv6-$(subst m,y,$(CONFIG_IPV6)) += bond_ipv6.o +bonding-objs += $(ipv6-y) + diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c index 6106660a4a4..8c2e5ab51f0 100644 --- a/drivers/net/bonding/bond_3ad.c +++ b/drivers/net/bonding/bond_3ad.c @@ -20,13 +20,12 @@ * */ -//#define BONDING_DEBUG 1 - #include <linux/skbuff.h> #include <linux/if_ether.h> #include <linux/netdevice.h> #include <linux/spinlock.h> #include <linux/ethtool.h> +#include <linux/etherdevice.h> #include <linux/if_bonding.h> #include <linux/pkt_sched.h> #include <net/net_namespace.h> @@ -96,33 +95,7 @@ static struct mac_addr null_mac_addr = {{0, 0, 0, 0, 0, 0}}; static u16 ad_ticks_per_sec; static const int ad_delta_in_ticks = (AD_TIMER_INTERVAL * HZ) / 1000; -// ================= 3AD api to bonding and kernel code ================== -static u16 __get_link_speed(struct port *port); -static u8 __get_duplex(struct port *port); -static inline void __initialize_port_locks(struct port *port); -//conversions -static u16 __ad_timer_to_ticks(u16 timer_type, u16 Par); - - -// ================= ad code helper functions ================== -//needed by ad_rx_machine(...) -static void __record_pdu(struct lacpdu *lacpdu, struct port *port); -static void __record_default(struct port *port); -static void __update_selected(struct lacpdu *lacpdu, struct port *port); -static void __update_default_selected(struct port *port); -static void __choose_matched(struct lacpdu *lacpdu, struct port *port); -static void __update_ntt(struct lacpdu *lacpdu, struct port *port); - -//needed for ad_mux_machine(..) -static void __attach_bond_to_agg(struct port *port); -static void __detach_bond_from_agg(struct port *port); -static int __agg_ports_are_ready(struct aggregator *aggregator); -static void __set_agg_ports_ready(struct aggregator *aggregator, int val); - -//needed for ad_agg_selection_logic(...) -static u32 __get_agg_bandwidth(struct aggregator *aggregator); -static struct aggregator *__get_active_agg(struct aggregator *aggregator); - +static const u8 lacpdu_mcast_addr[ETH_ALEN] = MULTICAST_LACPDU_ADDR; // ================= main 802.3ad protocol functions ================== static int ad_lacpdu_send(struct port *port); @@ -136,7 +109,6 @@ static void ad_agg_selection_logic(struct aggregator *aggregator); static void ad_clear_agg(struct aggregator *aggregator); static void ad_initialize_agg(struct aggregator *aggregator); static void ad_initialize_port(struct port *port, int lacp_fast); -static void ad_initialize_lacpdu(struct lacpdu *Lacpdu); static void ad_enable_collecting_distributing(struct port *port); static void ad_disable_collecting_distributing(struct port *port); static void ad_marker_info_received(struct bond_marker *marker_info, struct port *port); @@ -236,6 +208,17 @@ static inline struct aggregator *__get_next_agg(struct aggregator *aggregator) return &(SLAVE_AD_INFO(slave->next).aggregator); } +/* + * __agg_has_partner + * + * Return nonzero if aggregator has a partner (denoted by a non-zero ether + * address for the partner). Return 0 if not. + */ +static inline int __agg_has_partner(struct aggregator *agg) +{ + return !is_zero_ether_addr(agg->partner_system.mac_addr_value); +} + /** * __disable_port - disable the port's slave * @port: the port we're looking at @@ -274,14 +257,14 @@ static inline int __port_is_enabled(struct port *port) * __get_agg_selection_mode - get the aggregator selection mode * @port: the port we're looking at * - * Get the aggregator selection mode. Can be %BANDWIDTH or %COUNT. + * Get the aggregator selection mode. Can be %STABLE, %BANDWIDTH or %COUNT. */ static inline u32 __get_agg_selection_mode(struct port *port) { struct bonding *bond = __get_bond_by_port(port); if (bond == NULL) { - return AD_BANDWIDTH; + return BOND_AD_STABLE; } return BOND_AD_INFO(bond).agg_select_mode; @@ -369,7 +352,7 @@ static u16 __get_link_speed(struct port *port) } } - dprintk("Port %d Received link speed %d update from adapter\n", port->actor_port_number, speed); + pr_debug("Port %d Received link speed %d update from adapter\n", port->actor_port_number, speed); return speed; } @@ -395,12 +378,12 @@ static u8 __get_duplex(struct port *port) switch (slave->duplex) { case DUPLEX_FULL: retval=0x1; - dprintk("Port %d Received status full duplex update from adapter\n", port->actor_port_number); + pr_debug("Port %d Received status full duplex update from adapter\n", port->actor_port_number); break; case DUPLEX_HALF: default: retval=0x0; - dprintk("Port %d Received status NOT full duplex update from adapter\n", port->actor_port_number); + pr_debug("Port %d Received status NOT full duplex update from adapter\n", port->actor_port_number); break; } } @@ -473,33 +456,25 @@ static u16 __ad_timer_to_ticks(u16 timer_type, u16 par) */ static void __record_pdu(struct lacpdu *lacpdu, struct port *port) { - // validate lacpdu and port if (lacpdu && port) { + struct port_params *partner = &port->partner_oper; + // record the new parameter values for the partner operational - port->partner_oper_port_number = ntohs(lacpdu->actor_port); - port->partner_oper_port_priority = ntohs(lacpdu->actor_port_priority); - port->partner_oper_system = lacpdu->actor_system; - port->partner_oper_system_priority = ntohs(lacpdu->actor_system_priority); - port->partner_oper_key = ntohs(lacpdu->actor_key); - // zero partener's lase states - port->partner_oper_port_state = 0; - port->partner_oper_port_state |= (lacpdu->actor_state & AD_STATE_LACP_ACTIVITY); - port->partner_oper_port_state |= (lacpdu->actor_state & AD_STATE_LACP_TIMEOUT); - port->partner_oper_port_state |= (lacpdu->actor_state & AD_STATE_AGGREGATION); - port->partner_oper_port_state |= (lacpdu->actor_state & AD_STATE_SYNCHRONIZATION); - port->partner_oper_port_state |= (lacpdu->actor_state & AD_STATE_COLLECTING); - port->partner_oper_port_state |= (lacpdu->actor_state & AD_STATE_DISTRIBUTING); - port->partner_oper_port_state |= (lacpdu->actor_state & AD_STATE_DEFAULTED); - port->partner_oper_port_state |= (lacpdu->actor_state & AD_STATE_EXPIRED); + partner->port_number = ntohs(lacpdu->actor_port); + partner->port_priority = ntohs(lacpdu->actor_port_priority); + partner->system = lacpdu->actor_system; + partner->system_priority = ntohs(lacpdu->actor_system_priority); + partner->key = ntohs(lacpdu->actor_key); + partner->port_state = lacpdu->actor_state; // set actor_oper_port_state.defaulted to FALSE port->actor_oper_port_state &= ~AD_STATE_DEFAULTED; // set the partner sync. to on if the partner is sync. and the port is matched if ((port->sm_vars & AD_PORT_MATCHED) && (lacpdu->actor_state & AD_STATE_SYNCHRONIZATION)) { - port->partner_oper_port_state |= AD_STATE_SYNCHRONIZATION; + partner->port_state |= AD_STATE_SYNCHRONIZATION; } else { - port->partner_oper_port_state &= ~AD_STATE_SYNCHRONIZATION; + partner->port_state &= ~AD_STATE_SYNCHRONIZATION; } } } @@ -514,15 +489,10 @@ static void __record_pdu(struct lacpdu *lacpdu, struct port *port) */ static void __record_default(struct port *port) { - // validate the port if (port) { // record the partner admin parameters - port->partner_oper_port_number = port->partner_admin_port_number; - port->partner_oper_port_priority = port->partner_admin_port_priority; - port->partner_oper_system = port->partner_admin_system; - port->partner_oper_system_priority = port->partner_admin_system_priority; - port->partner_oper_key = port->partner_admin_key; - port->partner_oper_port_state = port->partner_admin_port_state; + memcpy(&port->partner_oper, &port->partner_admin, + sizeof(struct port_params)); // set actor_oper_port_state.defaulted to true port->actor_oper_port_state |= AD_STATE_DEFAULTED; @@ -544,16 +514,16 @@ static void __record_default(struct port *port) */ static void __update_selected(struct lacpdu *lacpdu, struct port *port) { - // validate lacpdu and port if (lacpdu && port) { + const struct port_params *partner = &port->partner_oper; + // check if any parameter is different - if ((ntohs(lacpdu->actor_port) != port->partner_oper_port_number) || - (ntohs(lacpdu->actor_port_priority) != port->partner_oper_port_priority) || - MAC_ADDRESS_COMPARE(&(lacpdu->actor_system), &(port->partner_oper_system)) || - (ntohs(lacpdu->actor_system_priority) != port->partner_oper_system_priority) || - (ntohs(lacpdu->actor_key) != port->partner_oper_key) || - ((lacpdu->actor_state & AD_STATE_AGGREGATION) != (port->partner_oper_port_state & AD_STATE_AGGREGATION)) - ) { + if (ntohs(lacpdu->actor_port) != partner->port_number + || ntohs(lacpdu->actor_port_priority) != partner->port_priority + || MAC_ADDRESS_COMPARE(&lacpdu->actor_system, &partner->system) + || ntohs(lacpdu->actor_system_priority) != partner->system_priority + || ntohs(lacpdu->actor_key) != partner->key + || (lacpdu->actor_state & AD_STATE_AGGREGATION) != (partner->port_state & AD_STATE_AGGREGATION)) { // update the state machine Selected variable port->sm_vars &= ~AD_PORT_SELECTED; } @@ -574,16 +544,18 @@ static void __update_selected(struct lacpdu *lacpdu, struct port *port) */ static void __update_default_selected(struct port *port) { - // validate the port if (port) { + const struct port_params *admin = &port->partner_admin; + const struct port_params *oper = &port->partner_oper; + // check if any parameter is different - if ((port->partner_admin_port_number != port->partner_oper_port_number) || - (port->partner_admin_port_priority != port->partner_oper_port_priority) || - MAC_ADDRESS_COMPARE(&(port->partner_admin_system), &(port->partner_oper_system)) || - (port->partner_admin_system_priority != port->partner_oper_system_priority) || - (port->partner_admin_key != port->partner_oper_key) || - ((port->partner_admin_port_state & AD_STATE_AGGREGATION) != (port->partner_oper_port_state & AD_STATE_AGGREGATION)) - ) { + if (admin->port_number != oper->port_number + || admin->port_priority != oper->port_priority + || MAC_ADDRESS_COMPARE(&admin->system, &oper->system) + || admin->system_priority != oper->system_priority + || admin->key != oper->key + || (admin->port_state & AD_STATE_AGGREGATION) + != (oper->port_state & AD_STATE_AGGREGATION)) { // update the state machine Selected variable port->sm_vars &= ~AD_PORT_SELECTED; } @@ -658,8 +630,8 @@ static void __update_ntt(struct lacpdu *lacpdu, struct port *port) ((lacpdu->partner_state & AD_STATE_SYNCHRONIZATION) != (port->actor_oper_port_state & AD_STATE_SYNCHRONIZATION)) || ((lacpdu->partner_state & AD_STATE_AGGREGATION) != (port->actor_oper_port_state & AD_STATE_AGGREGATION)) ) { - // set ntt to be TRUE - port->ntt = 1; + + port->ntt = true; } } } @@ -798,6 +770,7 @@ static struct aggregator *__get_active_agg(struct aggregator *aggregator) static inline void __update_lacpdu_from_port(struct port *port) { struct lacpdu *lacpdu = &port->lacpdu; + const struct port_params *partner = &port->partner_oper; /* update current actual Actor parameters */ /* lacpdu->subtype initialized @@ -818,12 +791,12 @@ static inline void __update_lacpdu_from_port(struct port *port) * lacpdu->partner_information_length initialized */ - lacpdu->partner_system_priority = htons(port->partner_oper_system_priority); - lacpdu->partner_system = port->partner_oper_system; - lacpdu->partner_key = htons(port->partner_oper_key); - lacpdu->partner_port_priority = htons(port->partner_oper_port_priority); - lacpdu->partner_port = htons(port->partner_oper_port_number); - lacpdu->partner_state = port->partner_oper_port_state; + lacpdu->partner_system_priority = htons(partner->system_priority); + lacpdu->partner_system = partner->system; + lacpdu->partner_key = htons(partner->key); + lacpdu->partner_port_priority = htons(partner->port_priority); + lacpdu->partner_port = htons(partner->port_number); + lacpdu->partner_state = partner->port_state; /* lacpdu->reserved_3_2 initialized * lacpdu->tlv_type_collector_info initialized @@ -853,7 +826,6 @@ static int ad_lacpdu_send(struct port *port) struct sk_buff *skb; struct lacpdu_header *lacpdu_header; int length = sizeof(struct lacpdu_header); - struct mac_addr lacpdu_multicast_address = AD_MULTICAST_LACPDU_ADDR; skb = dev_alloc_skb(length); if (!skb) { @@ -868,11 +840,11 @@ static int ad_lacpdu_send(struct port *port) lacpdu_header = (struct lacpdu_header *)skb_put(skb, length); - lacpdu_header->ad_header.destination_address = lacpdu_multicast_address; - /* Note: source addres is set to be the member's PERMANENT address, because we use it - to identify loopback lacpdus in receive. */ - lacpdu_header->ad_header.source_address = *((struct mac_addr *)(slave->perm_hwaddr)); - lacpdu_header->ad_header.length_type = PKT_TYPE_LACPDU; + memcpy(lacpdu_header->hdr.h_dest, lacpdu_mcast_addr, ETH_ALEN); + /* Note: source addres 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); + lacpdu_header->hdr.h_proto = PKT_TYPE_LACPDU; lacpdu_header->lacpdu = port->lacpdu; // struct copy @@ -895,7 +867,6 @@ static int ad_marker_send(struct port *port, struct bond_marker *marker) struct sk_buff *skb; struct bond_marker_header *marker_header; int length = sizeof(struct bond_marker_header); - struct mac_addr lacpdu_multicast_address = AD_MULTICAST_LACPDU_ADDR; skb = dev_alloc_skb(length + 16); if (!skb) { @@ -911,11 +882,11 @@ static int ad_marker_send(struct port *port, struct bond_marker *marker) marker_header = (struct bond_marker_header *)skb_put(skb, length); - marker_header->ad_header.destination_address = lacpdu_multicast_address; - /* Note: source addres is set to be the member's PERMANENT address, because we use it - to identify loopback MARKERs in receive. */ - marker_header->ad_header.source_address = *((struct mac_addr *)(slave->perm_hwaddr)); - marker_header->ad_header.length_type = PKT_TYPE_LACPDU; + memcpy(marker_header->hdr.h_dest, lacpdu_mcast_addr, ETH_ALEN); + /* Note: source addres 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); + marker_header->hdr.h_proto = PKT_TYPE_LACPDU; marker_header->marker = *marker; // struct copy @@ -972,7 +943,7 @@ static void ad_mux_machine(struct port *port) break; case AD_MUX_ATTACHED: // check also if agg_select_timer expired(so the edable port will take place only after this timer) - if ((port->sm_vars & AD_PORT_SELECTED) && (port->partner_oper_port_state & AD_STATE_SYNCHRONIZATION) && !__check_agg_selection_timer(port)) { + if ((port->sm_vars & AD_PORT_SELECTED) && (port->partner_oper.port_state & AD_STATE_SYNCHRONIZATION) && !__check_agg_selection_timer(port)) { port->sm_mux_state = AD_MUX_COLLECTING_DISTRIBUTING;// next state } else if (!(port->sm_vars & AD_PORT_SELECTED) || (port->sm_vars & AD_PORT_STANDBY)) { // if UNSELECTED or STANDBY port->sm_vars &= ~AD_PORT_READY_N; @@ -984,7 +955,7 @@ static void ad_mux_machine(struct port *port) break; case AD_MUX_COLLECTING_DISTRIBUTING: if (!(port->sm_vars & AD_PORT_SELECTED) || (port->sm_vars & AD_PORT_STANDBY) || - !(port->partner_oper_port_state & AD_STATE_SYNCHRONIZATION) + !(port->partner_oper.port_state & AD_STATE_SYNCHRONIZATION) ) { port->sm_mux_state = AD_MUX_ATTACHED;// next state @@ -1007,7 +978,7 @@ static void ad_mux_machine(struct port *port) // check if the state machine was changed if (port->sm_mux_state != last_state) { - dprintk("Mux Machine: Port=%d, Last State=%d, Curr State=%d\n", port->actor_port_number, last_state, port->sm_mux_state); + pr_debug("Mux Machine: Port=%d, Last State=%d, Curr State=%d\n", port->actor_port_number, last_state, port->sm_mux_state); switch (port->sm_mux_state) { case AD_MUX_DETACHED: __detach_bond_from_agg(port); @@ -1015,7 +986,7 @@ static void ad_mux_machine(struct port *port) ad_disable_collecting_distributing(port); port->actor_oper_port_state &= ~AD_STATE_COLLECTING; port->actor_oper_port_state &= ~AD_STATE_DISTRIBUTING; - port->ntt = 1; + port->ntt = true; break; case AD_MUX_WAITING: port->sm_mux_timer_counter = __ad_timer_to_ticks(AD_WAIT_WHILE_TIMER, 0); @@ -1026,13 +997,13 @@ static void ad_mux_machine(struct port *port) port->actor_oper_port_state &= ~AD_STATE_COLLECTING; port->actor_oper_port_state &= ~AD_STATE_DISTRIBUTING; ad_disable_collecting_distributing(port); - port->ntt = 1; + port->ntt = true; break; case AD_MUX_COLLECTING_DISTRIBUTING: port->actor_oper_port_state |= AD_STATE_COLLECTING; port->actor_oper_port_state |= AD_STATE_DISTRIBUTING; ad_enable_collecting_distributing(port); - port->ntt = 1; + port->ntt = true; break; default: //to silence the compiler break; @@ -1106,7 +1077,7 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port) // check if the State machine was changed or new lacpdu arrived if ((port->sm_rx_state != last_state) || (lacpdu)) { - dprintk("Rx Machine: Port=%d, Last State=%d, Curr State=%d\n", port->actor_port_number, last_state, port->sm_rx_state); + pr_debug("Rx Machine: Port=%d, Last State=%d, Curr State=%d\n", port->actor_port_number, last_state, port->sm_rx_state); switch (port->sm_rx_state) { case AD_RX_INITIALIZE: if (!(port->actor_oper_port_key & AD_DUPLEX_KEY_BITS)) { @@ -1128,7 +1099,7 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port) case AD_RX_LACP_DISABLED: port->sm_vars &= ~AD_PORT_SELECTED; __record_default(port); - port->partner_oper_port_state &= ~AD_STATE_AGGREGATION; + port->partner_oper.port_state &= ~AD_STATE_AGGREGATION; port->sm_vars |= AD_PORT_MATCHED; port->actor_oper_port_state &= ~AD_STATE_EXPIRED; break; @@ -1136,9 +1107,9 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port) //Reset of the Synchronization flag. (Standard 43.4.12) //This reset cause to disable this port in the COLLECTING_DISTRIBUTING state of the //mux machine in case of EXPIRED even if LINK_DOWN didn't arrive for the port. - port->partner_oper_port_state &= ~AD_STATE_SYNCHRONIZATION; + port->partner_oper.port_state &= ~AD_STATE_SYNCHRONIZATION; port->sm_vars &= ~AD_PORT_MATCHED; - port->partner_oper_port_state |= AD_SHORT_TIMEOUT; + port->partner_oper.port_state |= AD_SHORT_TIMEOUT; port->sm_rx_timer_counter = __ad_timer_to_ticks(AD_CURRENT_WHILE_TIMER, (u16)(AD_SHORT_TIMEOUT)); port->actor_oper_port_state |= AD_STATE_EXPIRED; break; @@ -1191,11 +1162,13 @@ static void ad_tx_machine(struct port *port) // check if there is something to send if (port->ntt && (port->sm_vars & AD_PORT_LACP_ENABLED)) { __update_lacpdu_from_port(port); - // send the lacpdu + if (ad_lacpdu_send(port) >= 0) { - dprintk("Sent LACPDU on port %d\n", port->actor_port_number); - // mark ntt as false, so it will not be sent again until demanded - port->ntt = 0; + pr_debug("Sent LACPDU on port %d\n", port->actor_port_number); + + /* mark ntt as false, so it will not be sent again until + demanded */ + port->ntt = false; } } // restart tx timer(to verify that we will not exceed AD_MAX_TX_IN_SECOND @@ -1218,7 +1191,7 @@ static void ad_periodic_machine(struct port *port) // check if port was reinitialized if (((port->sm_vars & AD_PORT_BEGIN) || !(port->sm_vars & AD_PORT_LACP_ENABLED) || !port->is_enabled) || - (!(port->actor_oper_port_state & AD_STATE_LACP_ACTIVITY) && !(port->partner_oper_port_state & AD_STATE_LACP_ACTIVITY)) + (!(port->actor_oper_port_state & AD_STATE_LACP_ACTIVITY) && !(port->partner_oper.port_state & AD_STATE_LACP_ACTIVITY)) ) { port->sm_periodic_state = AD_NO_PERIODIC; // next state } @@ -1232,12 +1205,12 @@ static void ad_periodic_machine(struct port *port) // If not expired, check if there is some new timeout parameter from the partner state switch (port->sm_periodic_state) { case AD_FAST_PERIODIC: - if (!(port->partner_oper_port_state & AD_STATE_LACP_TIMEOUT)) { + if (!(port->partner_oper.port_state & AD_STATE_LACP_TIMEOUT)) { port->sm_periodic_state = AD_SLOW_PERIODIC; // next state } break; case AD_SLOW_PERIODIC: - if ((port->partner_oper_port_state & AD_STATE_LACP_TIMEOUT)) { + if ((port->partner_oper.port_state & AD_STATE_LACP_TIMEOUT)) { // stop current timer port->sm_periodic_timer_counter = 0; port->sm_periodic_state = AD_PERIODIC_TX; // next state @@ -1253,7 +1226,7 @@ static void ad_periodic_machine(struct port *port) port->sm_periodic_state = AD_FAST_PERIODIC; // next state break; case AD_PERIODIC_TX: - if (!(port->partner_oper_port_state & AD_STATE_LACP_TIMEOUT)) { + if (!(port->partner_oper.port_state & AD_STATE_LACP_TIMEOUT)) { port->sm_periodic_state = AD_SLOW_PERIODIC; // next state } else { port->sm_periodic_state = AD_FAST_PERIODIC; // next state @@ -1266,7 +1239,7 @@ static void ad_periodic_machine(struct port *port) // check if the state machine was changed if (port->sm_periodic_state != last_state) { - dprintk("Periodic Machine: Port=%d, Last State=%d, Curr State=%d\n", port->actor_port_number, last_state, port->sm_periodic_state); + pr_debug("Periodic Machine: Port=%d, Last State=%d, Curr State=%d\n", port->actor_port_number, last_state, port->sm_periodic_state); switch (port->sm_periodic_state) { case AD_NO_PERIODIC: port->sm_periodic_timer_counter = 0; // zero timer @@ -1278,7 +1251,7 @@ static void ad_periodic_machine(struct port *port) port->sm_periodic_timer_counter = __ad_timer_to_ticks(AD_PERIODIC_TIMER, (u16)(AD_SLOW_PERIODIC_TIME))-1; // decrement 1 tick we lost in the PERIODIC_TX cycle break; case AD_PERIODIC_TX: - port->ntt = 1; + port->ntt = true; break; default: //to silence the compiler break; @@ -1323,7 +1296,7 @@ static void ad_port_selection_logic(struct port *port) port->next_port_in_aggregator=NULL; port->actor_port_aggregator_identifier=0; - dprintk("Port %d left LAG %d\n", port->actor_port_number, temp_aggregator->aggregator_identifier); + pr_debug("Port %d left LAG %d\n", port->actor_port_number, temp_aggregator->aggregator_identifier); // if the aggregator is empty, clear its parameters, and set it ready to be attached if (!temp_aggregator->lag_ports) { ad_clear_agg(temp_aggregator); @@ -1352,11 +1325,11 @@ static void ad_port_selection_logic(struct port *port) } // check if current aggregator suits us if (((aggregator->actor_oper_aggregator_key == port->actor_oper_port_key) && // if all parameters match AND - !MAC_ADDRESS_COMPARE(&(aggregator->partner_system), &(port->partner_oper_system)) && - (aggregator->partner_system_priority == port->partner_oper_system_priority) && - (aggregator->partner_oper_aggregator_key == port->partner_oper_key) + !MAC_ADDRESS_COMPARE(&(aggregator->partner_system), &(port->partner_oper.system)) && + (aggregator->partner_system_priority == port->partner_oper.system_priority) && + (aggregator->partner_oper_aggregator_key == port->partner_oper.key) ) && - ((MAC_ADDRESS_COMPARE(&(port->partner_oper_system), &(null_mac_addr)) && // partner answers + ((MAC_ADDRESS_COMPARE(&(port->partner_oper.system), &(null_mac_addr)) && // partner answers !aggregator->is_individual) // but is not individual OR ) ) { @@ -1366,7 +1339,7 @@ static void ad_port_selection_logic(struct port *port) port->next_port_in_aggregator=aggregator->lag_ports; port->aggregator->num_of_ports++; aggregator->lag_ports=port; - dprintk("Port %d joined LAG %d(existing LAG)\n", port->actor_port_number, port->aggregator->aggregator_identifier); + pr_debug("Port %d joined LAG %d(existing LAG)\n", port->actor_port_number, port->aggregator->aggregator_identifier); // mark this port as selected port->sm_vars |= AD_PORT_SELECTED; @@ -1385,16 +1358,16 @@ static void ad_port_selection_logic(struct port *port) // update the new aggregator's parameters // if port was responsed from the end-user if (port->actor_oper_port_key & AD_DUPLEX_KEY_BITS) {// if port is full duplex - port->aggregator->is_individual = 0; + port->aggregator->is_individual = false; } else { - port->aggregator->is_individual = 1; + port->aggregator->is_individual = true; } port->aggregator->actor_admin_aggregator_key = port->actor_admin_port_key; port->aggregator->actor_oper_aggregator_key = port->actor_oper_port_key; - port->aggregator->partner_system=port->partner_oper_system; - port->aggregator->partner_system_priority = port->partner_oper_system_priority; - port->aggregator->partner_oper_aggregator_key = port->partner_oper_key; + port->aggregator->partner_system=port->partner_oper.system; + port->aggregator->partner_system_priority = port->partner_oper.system_priority; + port->aggregator->partner_oper_aggregator_key = port->partner_oper.key; port->aggregator->receive_state = 1; port->aggregator->transmit_state = 1; port->aggregator->lag_ports = port; @@ -1403,7 +1376,7 @@ static void ad_port_selection_logic(struct port *port) // mark this port as selected port->sm_vars |= AD_PORT_SELECTED; - dprintk("Port %d joined LAG %d(new LAG)\n", port->actor_port_number, port->aggregator->aggregator_identifier); + pr_debug("Port %d joined LAG %d(new LAG)\n", port->actor_port_number, port->aggregator->aggregator_identifier); } else { printk(KERN_ERR DRV_NAME ": %s: Port %d (on %s) did not find a suitable aggregator\n", port->slave->dev->master->name, @@ -1414,9 +1387,82 @@ static void ad_port_selection_logic(struct port *port) // else set ready=FALSE in all aggregator's ports __set_agg_ports_ready(port->aggregator, __agg_ports_are_ready(port->aggregator)); - if (!__check_agg_selection_timer(port) && (aggregator = __get_first_agg(port))) { - ad_agg_selection_logic(aggregator); + aggregator = __get_first_agg(port); + ad_agg_selection_logic(aggregator); +} + +/* + * Decide if "agg" is a better choice for the new active aggregator that + * the current best, according to the ad_select policy. + */ +static struct aggregator *ad_agg_selection_test(struct aggregator *best, + struct aggregator *curr) +{ + /* + * 0. If no best, select current. + * + * 1. If the current agg is not individual, and the best is + * individual, select current. + * + * 2. If current agg is individual and the best is not, keep best. + * + * 3. Therefore, current and best are both individual or both not + * individual, so: + * + * 3a. If current agg partner replied, and best agg partner did not, + * select current. + * + * 3b. If current agg partner did not reply and best agg partner + * did reply, keep best. + * + * 4. Therefore, current and best both have partner replies or + * both do not, so perform selection policy: + * + * BOND_AD_COUNT: Select by count of ports. If count is equal, + * select by bandwidth. + * + * BOND_AD_STABLE, BOND_AD_BANDWIDTH: Select by bandwidth. + */ + if (!best) + return curr; + + if (!curr->is_individual && best->is_individual) + return curr; + + if (curr->is_individual && !best->is_individual) + return best; + + if (__agg_has_partner(curr) && !__agg_has_partner(best)) + return curr; + + if (!__agg_has_partner(curr) && __agg_has_partner(best)) + return best; + + switch (__get_agg_selection_mode(curr->lag_ports)) { + case BOND_AD_COUNT: + if (curr->num_of_ports > best->num_of_ports) + return curr; + + if (curr->num_of_ports < best->num_of_ports) + return best; + + /*FALLTHROUGH*/ + case BOND_AD_STABLE: + case BOND_AD_BANDWIDTH: + if (__get_agg_bandwidth(curr) > __get_agg_bandwidth(best)) + return curr; + + break; + + default: + printk(KERN_WARNING DRV_NAME + ": %s: Impossible agg select mode %d\n", + curr->slave->dev->master->name, + __get_agg_selection_mode(curr->lag_ports)); + break; } + + return best; } /** @@ -1424,156 +1470,138 @@ static void ad_port_selection_logic(struct port *port) * @aggregator: the aggregator we're looking at * * It is assumed that only one aggregator may be selected for a team. - * The logic of this function is to select (at first time) the aggregator with - * the most ports attached to it, and to reselect the active aggregator only if - * the previous aggregator has no more ports related to it. + * + * The logic of this function is to select the aggregator according to + * the ad_select policy: + * + * BOND_AD_STABLE: select the aggregator with the most ports attached to + * it, and to reselect the active aggregator only if the previous + * aggregator has no more ports related to it. + * + * BOND_AD_BANDWIDTH: select the aggregator with the highest total + * bandwidth, and reselect whenever a link state change takes place or the + * set of slaves in the bond changes. + * + * BOND_AD_COUNT: select the aggregator with largest number of ports + * (slaves), and reselect whenever a link state change takes place or the + * set of slaves in the bond changes. * * FIXME: this function MUST be called with the first agg in the bond, or * __get_active_agg() won't work correctly. This function should be better * called with the bond itself, and retrieve the first agg from it. */ -static void ad_agg_selection_logic(struct aggregator *aggregator) +static void ad_agg_selection_logic(struct aggregator *agg) { - struct aggregator *best_aggregator = NULL, *active_aggregator = NULL; - struct aggregator *last_active_aggregator = NULL, *origin_aggregator; + struct aggregator *best, *active, *origin; struct port *port; - u16 num_of_aggs=0; - origin_aggregator = aggregator; + origin = agg; - //get current active aggregator - last_active_aggregator = __get_active_agg(aggregator); + active = __get_active_agg(agg); + best = active; - // search for the aggregator with the most ports attached to it. do { - // count how many candidate lag's we have - if (aggregator->lag_ports) { - num_of_aggs++; - } - if (aggregator->is_active && !aggregator->is_individual && // if current aggregator is the active aggregator - MAC_ADDRESS_COMPARE(&(aggregator->partner_system), &(null_mac_addr))) { // and partner answers to 802.3ad PDUs - if (aggregator->num_of_ports) { // if any ports attached to the current aggregator - best_aggregator=NULL; // disregard the best aggregator that was chosen by now - break; // stop the selection of other aggregator if there are any ports attached to this active aggregator - } else { // no ports attached to this active aggregator - aggregator->is_active = 0; // mark this aggregator as not active anymore - } - } - if (aggregator->num_of_ports) { // if any ports attached - if (best_aggregator) { // if there is a candidte aggregator - //The reasons for choosing new best aggregator: - // 1. if current agg is NOT individual and the best agg chosen so far is individual OR - // current and best aggs are both individual or both not individual, AND - // 2a. current agg partner reply but best agg partner do not reply OR - // 2b. current agg partner reply OR current agg partner do not reply AND best agg partner also do not reply AND - // current has more ports/bandwidth, or same amount of ports but current has faster ports, THEN - // current agg become best agg so far - - //if current agg is NOT individual and the best agg chosen so far is individual change best_aggregator - if (!aggregator->is_individual && best_aggregator->is_individual) { - best_aggregator=aggregator; - } - // current and best aggs are both individual or both not individual - else if ((aggregator->is_individual && best_aggregator->is_individual) || - (!aggregator->is_individual && !best_aggregator->is_individual)) { - // current and best aggs are both individual or both not individual AND - // current agg partner reply but best agg partner do not reply - if ((MAC_ADDRESS_COMPARE(&(aggregator->partner_system), &(null_mac_addr)) && - !MAC_ADDRESS_COMPARE(&(best_aggregator->partner_system), &(null_mac_addr)))) { - best_aggregator=aggregator; - } - // current agg partner reply OR current agg partner do not reply AND best agg partner also do not reply - else if (! (!MAC_ADDRESS_COMPARE(&(aggregator->partner_system), &(null_mac_addr)) && - MAC_ADDRESS_COMPARE(&(best_aggregator->partner_system), &(null_mac_addr)))) { - if ((__get_agg_selection_mode(aggregator->lag_ports) == AD_BANDWIDTH)&& - (__get_agg_bandwidth(aggregator) > __get_agg_bandwidth(best_aggregator))) { - best_aggregator=aggregator; - } else if (__get_agg_selection_mode(aggregator->lag_ports) == AD_COUNT) { - if (((aggregator->num_of_ports > best_aggregator->num_of_ports) && - (aggregator->actor_oper_aggregator_key & AD_SPEED_KEY_BITS))|| - ((aggregator->num_of_ports == best_aggregator->num_of_ports) && - ((u16)(aggregator->actor_oper_aggregator_key & AD_SPEED_KEY_BITS) > - (u16)(best_aggregator->actor_oper_aggregator_key & AD_SPEED_KEY_BITS)))) { - best_aggregator=aggregator; - } - } - } - } - } else { - best_aggregator=aggregator; + agg->is_active = 0; + + if (agg->num_of_ports) + best = ad_agg_selection_test(best, agg); + + } while ((agg = __get_next_agg(agg))); + + if (best && + __get_agg_selection_mode(best->lag_ports) == BOND_AD_STABLE) { + /* + * For the STABLE policy, don't replace the old active + * aggregator if it's still active (it has an answering + * partner) or if both the best and active don't have an + * answering partner. + */ + if (active && active->lag_ports && + active->lag_ports->is_enabled && + (__agg_has_partner(active) || + (!__agg_has_partner(active) && !__agg_has_partner(best)))) { + if (!(!active->actor_oper_aggregator_key && + best->actor_oper_aggregator_key)) { + best = NULL; + active->is_active = 1; } } - aggregator->is_active = 0; // mark all aggregators as not active anymore - } while ((aggregator = __get_next_agg(aggregator))); - - // if we have new aggregator selected, don't replace the old aggregator if it has an answering partner, - // or if both old aggregator and new aggregator don't have answering partner - if (best_aggregator) { - if (last_active_aggregator && last_active_aggregator->lag_ports && last_active_aggregator->lag_ports->is_enabled && - (MAC_ADDRESS_COMPARE(&(last_active_aggregator->partner_system), &(null_mac_addr)) || // partner answers OR - (!MAC_ADDRESS_COMPARE(&(last_active_aggregator->partner_system), &(null_mac_addr)) && // both old and new - !MAC_ADDRESS_COMPARE(&(best_aggregator->partner_system), &(null_mac_addr)))) // partner do not answer - ) { - // if new aggregator has link, and old aggregator does not, replace old aggregator.(do nothing) - // -> don't replace otherwise. - if (!(!last_active_aggregator->actor_oper_aggregator_key && best_aggregator->actor_oper_aggregator_key)) { - best_aggregator=NULL; - last_active_aggregator->is_active = 1; // don't replace good old aggregator + } - } - } + if (best && (best == active)) { + best = NULL; + active->is_active = 1; } // if there is new best aggregator, activate it - if (best_aggregator) { - for (aggregator = __get_first_agg(best_aggregator->lag_ports); - aggregator; - aggregator = __get_next_agg(aggregator)) { - - dprintk("Agg=%d; Ports=%d; a key=%d; p key=%d; Indiv=%d; Active=%d\n", - aggregator->aggregator_identifier, aggregator->num_of_ports, - aggregator->actor_oper_aggregator_key, aggregator->partner_oper_aggregator_key, - aggregator->is_individual, aggregator->is_active); + if (best) { + pr_debug("best Agg=%d; P=%d; a k=%d; p k=%d; Ind=%d; Act=%d\n", + best->aggregator_identifier, best->num_of_ports, + best->actor_oper_aggregator_key, + best->partner_oper_aggregator_key, + best->is_individual, best->is_active); + pr_debug("best ports %p slave %p %s\n", + best->lag_ports, best->slave, + best->slave ? best->slave->dev->name : "NULL"); + + for (agg = __get_first_agg(best->lag_ports); agg; + agg = __get_next_agg(agg)) { + + pr_debug("Agg=%d; P=%d; a k=%d; p k=%d; Ind=%d; Act=%d\n", + agg->aggregator_identifier, agg->num_of_ports, + agg->actor_oper_aggregator_key, + agg->partner_oper_aggregator_key, + agg->is_individual, agg->is_active); } // check if any partner replys - if (best_aggregator->is_individual) { - printk(KERN_WARNING DRV_NAME ": %s: Warning: No 802.3ad response from " - "the link partner for any adapters in the bond\n", - best_aggregator->slave->dev->master->name); + if (best->is_individual) { + printk(KERN_WARNING DRV_NAME ": %s: Warning: No 802.3ad" + " response from the link partner for any" + " adapters in the bond\n", + best->slave->dev->master->name); } - // check if there are more than one aggregator - if (num_of_aggs > 1) { - dprintk("Warning: More than one Link Aggregation Group was " - "found in the bond. Only one group will function in the bond\n"); - } - - best_aggregator->is_active = 1; - dprintk("LAG %d choosed as the active LAG\n", best_aggregator->aggregator_identifier); - dprintk("Agg=%d; Ports=%d; a key=%d; p key=%d; Indiv=%d; Active=%d\n", - best_aggregator->aggregator_identifier, best_aggregator->num_of_ports, - best_aggregator->actor_oper_aggregator_key, best_aggregator->partner_oper_aggregator_key, - best_aggregator->is_individual, best_aggregator->is_active); + best->is_active = 1; + pr_debug("LAG %d chosen as the active LAG\n", + best->aggregator_identifier); + pr_debug("Agg=%d; P=%d; a k=%d; p k=%d; Ind=%d; Act=%d\n", + best->aggregator_identifier, best->num_of_ports, + best->actor_oper_aggregator_key, + best->partner_oper_aggregator_key, + best->is_individual, best->is_active); // disable the ports that were related to the former active_aggregator - if (last_active_aggregator) { - for (port=last_active_aggregator->lag_ports; port; port=port->next_port_in_aggregator) { + if (active) { + for (port = active->lag_ports; port; + port = port->next_port_in_aggregator) { __disable_port(port); } } } - // if the selected aggregator is of join individuals(partner_system is NULL), enable their ports - active_aggregator = __get_active_agg(origin_aggregator); + /* + * if the selected aggregator is of join individuals + * (partner_system is NULL), enable their ports + */ + active = __get_active_agg(origin); - if (active_aggregator) { - if (!MAC_ADDRESS_COMPARE(&(active_aggregator->partner_system), &(null_mac_addr))) { - for (port=active_aggregator->lag_ports; port; port=port->next_port_in_aggregator) { + if (active) { + if (!__agg_has_partner(active)) { + for (port = active->lag_ports; port; + port = port->next_port_in_aggregator) { __enable_port(port); } } } + + if (origin->slave) { + struct bonding *bond; + + bond = bond_get_bond_by_slave(origin->slave); + if (bond) + bond_3ad_set_carrier(bond); + } } /** @@ -1584,7 +1612,7 @@ static void ad_agg_selection_logic(struct aggregator *aggregator) static void ad_clear_agg(struct aggregator *aggregator) { if (aggregator) { - aggregator->is_individual = 0; + aggregator->is_individual = false; aggregator->actor_admin_aggregator_key = 0; aggregator->actor_oper_aggregator_key = 0; aggregator->partner_system = null_mac_addr; @@ -1595,7 +1623,7 @@ static void ad_clear_agg(struct aggregator *aggregator) aggregator->lag_ports = NULL; aggregator->is_active = 0; aggregator->num_of_ports = 0; - dprintk("LAG %d was cleared\n", aggregator->aggregator_identifier); + pr_debug("LAG %d was cleared\n", aggregator->aggregator_identifier); } } @@ -1623,13 +1651,32 @@ static void ad_initialize_agg(struct aggregator *aggregator) */ static void ad_initialize_port(struct port *port, int lacp_fast) { + static const struct port_params tmpl = { + .system_priority = 0xffff, + .key = 1, + .port_number = 1, + .port_priority = 0xff, + .port_state = 1, + }; + static const struct lacpdu lacpdu = { + .subtype = 0x01, + .version_number = 0x01, + .tlv_type_actor_info = 0x01, + .actor_information_length = 0x14, + .tlv_type_partner_info = 0x02, + .partner_information_length = 0x14, + .tlv_type_collector_info = 0x03, + .collector_information_length = 0x10, + .collector_max_delay = htons(AD_COLLECTOR_MAX_DELAY), + }; + if (port) { port->actor_port_number = 1; port->actor_port_priority = 0xff; port->actor_system = null_mac_addr; port->actor_system_priority = 0xffff; port->actor_port_aggregator_identifier = 0; - port->ntt = 0; + port->ntt = false; port->actor_admin_port_key = 1; port->actor_oper_port_key = 1; port->actor_admin_port_state = AD_STATE_AGGREGATION | AD_STATE_LACP_ACTIVITY; @@ -1639,19 +1686,10 @@ static void ad_initialize_port(struct port *port, int lacp_fast) port->actor_oper_port_state |= AD_STATE_LACP_TIMEOUT; } - port->partner_admin_system = null_mac_addr; - port->partner_oper_system = null_mac_addr; - port->partner_admin_system_priority = 0xffff; - port->partner_oper_system_priority = 0xffff; - port->partner_admin_key = 1; - port->partner_oper_key = 1; - port->partner_admin_port_number = 1; - port->partner_oper_port_number = 1; - port->partner_admin_port_priority = 0xff; - port->partner_oper_port_priority = 0xff; - port->partner_admin_port_state = 1; - port->partner_oper_port_state = 1; - port->is_enabled = 1; + memcpy(&port->partner_admin, &tmpl, sizeof(tmpl)); + memcpy(&port->partner_oper, &tmpl, sizeof(tmpl)); + + port->is_enabled = true; // ****** private parameters ****** port->sm_vars = 0x3; port->sm_rx_state = 0; @@ -1667,7 +1705,7 @@ static void ad_initialize_port(struct port *port, int lacp_fast) port->next_port_in_aggregator = NULL; port->transaction_id = 0; - ad_initialize_lacpdu(&(port->lacpdu)); + memcpy(&port->lacpdu, &lacpdu, sizeof(lacpdu)); } } @@ -1680,7 +1718,7 @@ static void ad_initialize_port(struct port *port, int lacp_fast) static void ad_enable_collecting_distributing(struct port *port) { if (port->aggregator->is_active) { - dprintk("Enabling port %d(LAG %d)\n", port->actor_port_number, port->aggregator->aggregator_identifier); + pr_debug("Enabling port %d(LAG %d)\n", port->actor_port_number, port->aggregator->aggregator_identifier); __enable_port(port); } } @@ -1693,7 +1731,7 @@ static void ad_enable_collecting_distributing(struct port *port) static void ad_disable_collecting_distributing(struct port *port) { if (port->aggregator && MAC_ADDRESS_COMPARE(&(port->aggregator->partner_system), &(null_mac_addr))) { - dprintk("Disabling port %d(LAG %d)\n", port->actor_port_number, port->aggregator->aggregator_identifier); + pr_debug("Disabling port %d(LAG %d)\n", port->actor_port_number, port->aggregator->aggregator_identifier); __disable_port(port); } } @@ -1731,7 +1769,7 @@ static void ad_marker_info_send(struct port *port) // send the marker information if (ad_marker_send(port, &marker) >= 0) { - dprintk("Sent Marker Information on port %d\n", port->actor_port_number); + pr_debug("Sent Marker Information on port %d\n", port->actor_port_number); } } #endif @@ -1755,7 +1793,7 @@ static void ad_marker_info_received(struct bond_marker *marker_info, // send the marker response if (ad_marker_send(port, &marker) >= 0) { - dprintk("Sent Marker Response on port %d\n", port->actor_port_number); + pr_debug("Sent Marker Response on port %d\n", port->actor_port_number); } } @@ -1776,53 +1814,6 @@ static void ad_marker_response_received(struct bond_marker *marker, // DO NOTHING, SINCE WE DECIDED NOT TO IMPLEMENT THIS FEATURE FOR NOW } -/** - * ad_initialize_lacpdu - initialize a given lacpdu structure - * @lacpdu: lacpdu structure to initialize - * - */ -static void ad_initialize_lacpdu(struct lacpdu *lacpdu) -{ - u16 index; - - // initialize lacpdu data - lacpdu->subtype = 0x01; - lacpdu->version_number = 0x01; - lacpdu->tlv_type_actor_info = 0x01; - lacpdu->actor_information_length = 0x14; - // lacpdu->actor_system_priority updated on send - // lacpdu->actor_system updated on send - // lacpdu->actor_key updated on send - // lacpdu->actor_port_priority updated on send - // lacpdu->actor_port updated on send - // lacpdu->actor_state updated on send - lacpdu->tlv_type_partner_info = 0x02; - lacpdu->partner_information_length = 0x14; - for (index=0; index<=2; index++) { - lacpdu->reserved_3_1[index]=0; - } - // lacpdu->partner_system_priority updated on send - // lacpdu->partner_system updated on send - // lacpdu->partner_key updated on send - // lacpdu->partner_port_priority updated on send - // lacpdu->partner_port updated on send - // lacpdu->partner_state updated on send - for (index=0; index<=2; index++) { - lacpdu->reserved_3_2[index]=0; - } - lacpdu->tlv_type_collector_info = 0x03; - lacpdu->collector_information_length= 0x10; - lacpdu->collector_max_delay = htons(AD_COLLECTOR_MAX_DELAY); - for (index=0; index<=11; index++) { - lacpdu->reserved_12[index]=0; - } - lacpdu->tlv_type_terminator = 0x00; - lacpdu->terminator_length = 0; - for (index=0; index<=49; index++) { - lacpdu->reserved_50[index]=0; - } -} - ////////////////////////////////////////////////////////////////////////////////////// // ================= AD exported functions to the main bonding code ================== ////////////////////////////////////////////////////////////////////////////////////// @@ -1830,6 +1821,19 @@ static void ad_initialize_lacpdu(struct lacpdu *lacpdu) // Check aggregators status in team every T seconds #define AD_AGGREGATOR_SELECTION_TIMER 8 +/* + * bond_3ad_initiate_agg_selection(struct bonding *bond) + * + * Set the aggregation selection timer, to initiate an agg selection in + * the very near future. Called during first initialization, and during + * any down to up transitions of the bond. + */ +void bond_3ad_initiate_agg_selection(struct bonding *bond, int timeout) +{ + BOND_AD_INFO(bond).agg_select_timer = timeout; + BOND_AD_INFO(bond).agg_select_mode = bond->params.ad_select; +} + static u16 aggregator_identifier; /** @@ -1854,9 +1858,9 @@ void bond_3ad_initialize(struct bonding *bond, u16 tick_resolution, int lacp_fas // initialize how many times this module is called in one second(should be about every 100ms) ad_ticks_per_sec = tick_resolution; - // initialize the aggregator selection timer(to activate an aggregation selection after initialize) - BOND_AD_INFO(bond).agg_select_timer = (AD_AGGREGATOR_SELECTION_TIMER * ad_ticks_per_sec); - BOND_AD_INFO(bond).agg_select_mode = AD_BANDWIDTH; + bond_3ad_initiate_agg_selection(bond, + AD_AGGREGATOR_SELECTION_TIMER * + ad_ticks_per_sec); } } @@ -1956,7 +1960,7 @@ void bond_3ad_unbind_slave(struct slave *slave) return; } - dprintk("Unbinding Link Aggregation Group %d\n", aggregator->aggregator_identifier); + pr_debug("Unbinding Link Aggregation Group %d\n", aggregator->aggregator_identifier); /* Tell the partner that this port is not suitable for aggregation */ port->actor_oper_port_state &= ~AD_STATE_AGGREGATION; @@ -1980,7 +1984,7 @@ void bond_3ad_unbind_slave(struct slave *slave) // if new aggregator found, copy the aggregator's parameters // and connect the related lag_ports to the new aggregator if ((new_aggregator) && ((!new_aggregator->lag_ports) || ((new_aggregator->lag_ports == port) && !new_aggregator->lag_ports->next_port_in_aggregator))) { - dprintk("Some port(s) related to LAG %d - replaceing with LAG %d\n", aggregator->aggregator_identifier, new_aggregator->aggregator_identifier); + pr_debug("Some port(s) related to LAG %d - replaceing with LAG %d\n", aggregator->aggregator_identifier, new_aggregator->aggregator_identifier); if ((new_aggregator->lag_ports == port) && new_aggregator->is_active) { printk(KERN_INFO DRV_NAME ": %s: Removing an active aggregator\n", @@ -2031,7 +2035,7 @@ void bond_3ad_unbind_slave(struct slave *slave) } } - dprintk("Unbinding port %d\n", port->actor_port_number); + pr_debug("Unbinding port %d\n", port->actor_port_number); // find the aggregator that this port is connected to temp_aggregator = __get_first_agg(port); for (; temp_aggregator; temp_aggregator = __get_next_agg(temp_aggregator)) { @@ -2162,7 +2166,7 @@ static void bond_3ad_rx_indication(struct lacpdu *lacpdu, struct slave *slave, u switch (lacpdu->subtype) { case AD_TYPE_LACPDU: - dprintk("Received LACPDU on port %d\n", port->actor_port_number); + pr_debug("Received LACPDU on port %d\n", port->actor_port_number); ad_rx_machine(lacpdu, port); break; @@ -2171,17 +2175,17 @@ static void bond_3ad_rx_indication(struct lacpdu *lacpdu, struct slave *slave, u switch (((struct bond_marker *)lacpdu)->tlv_type) { case AD_MARKER_INFORMATION_SUBTYPE: - dprintk("Received Marker Information on port %d\n", port->actor_port_number); + pr_debug("Received Marker Information on port %d\n", port->actor_port_number); ad_marker_info_received((struct bond_marker *)lacpdu, port); break; case AD_MARKER_RESPONSE_SUBTYPE: - dprintk("Received Marker Response on port %d\n", port->actor_port_number); + pr_debug("Received Marker Response on port %d\n", port->actor_port_number); ad_marker_response_received((struct bond_marker *)lacpdu, port); break; default: - dprintk("Received an unknown Marker subtype on slot %d\n", port->actor_port_number); + pr_debug("Received an unknown Marker subtype on slot %d\n", port->actor_port_number); } } } @@ -2209,7 +2213,7 @@ void bond_3ad_adapter_speed_changed(struct slave *slave) port->actor_admin_port_key &= ~AD_SPEED_KEY_BITS; port->actor_oper_port_key=port->actor_admin_port_key |= (__get_link_speed(port) << 1); - dprintk("Port %d changed speed\n", port->actor_port_number); + pr_debug("Port %d changed speed\n", port->actor_port_number); // there is no need to reselect a new aggregator, just signal the // state machines to reinitialize port->sm_vars |= AD_PORT_BEGIN; @@ -2237,7 +2241,7 @@ void bond_3ad_adapter_duplex_changed(struct slave *slave) port->actor_admin_port_key &= ~AD_DUPLEX_KEY_BITS; port->actor_oper_port_key=port->actor_admin_port_key |= __get_duplex(port); - dprintk("Port %d changed duplex\n", port->actor_port_number); + pr_debug("Port %d changed duplex\n", port->actor_port_number); // there is no need to reselect a new aggregator, just signal the // state machines to reinitialize port->sm_vars |= AD_PORT_BEGIN; @@ -2267,14 +2271,14 @@ void bond_3ad_handle_link_change(struct slave *slave, char link) // on link down we are zeroing duplex and speed since some of the adaptors(ce1000.lan) report full duplex/speed instead of N/A(duplex) / 0(speed) // on link up we are forcing recheck on the duplex and speed since some of he adaptors(ce1000.lan) report if (link == BOND_LINK_UP) { - port->is_enabled = 1; + port->is_enabled = true; port->actor_admin_port_key &= ~AD_DUPLEX_KEY_BITS; port->actor_oper_port_key=port->actor_admin_port_key |= __get_duplex(port); port->actor_admin_port_key &= ~AD_SPEED_KEY_BITS; port->actor_oper_port_key=port->actor_admin_port_key |= (__get_link_speed(port) << 1); } else { /* link has failed */ - port->is_enabled = 0; + port->is_enabled = false; port->actor_admin_port_key &= ~AD_DUPLEX_KEY_BITS; port->actor_oper_port_key= (port->actor_admin_port_key &= ~AD_SPEED_KEY_BITS); } @@ -2346,7 +2350,7 @@ int bond_3ad_get_active_agg_info(struct bonding *bond, struct ad_info *ad_info) int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev) { struct slave *slave, *start_at; - struct bonding *bond = dev->priv; + struct bonding *bond = netdev_priv(dev); int slave_agg_no; int slaves_in_agg; int agg_id; @@ -2426,7 +2430,7 @@ out: int bond_3ad_lacpdu_recv(struct sk_buff *skb, struct net_device *dev, struct packet_type* ptype, struct net_device *orig_dev) { - struct bonding *bond = dev->priv; + struct bonding *bond = netdev_priv(dev); struct slave *slave = NULL; int ret = NET_RX_DROP; @@ -2437,7 +2441,8 @@ int bond_3ad_lacpdu_recv(struct sk_buff *skb, struct net_device *dev, struct pac goto out; read_lock(&bond->lock); - slave = bond_get_slave_by_dev((struct bonding *)dev->priv, orig_dev); + slave = bond_get_slave_by_dev((struct bonding *)netdev_priv(dev), + orig_dev); if (!slave) goto out_unlock; diff --git a/drivers/net/bonding/bond_3ad.h b/drivers/net/bonding/bond_3ad.h index b5ee45f6d55..8a83eb283c2 100644 --- a/drivers/net/bonding/bond_3ad.h +++ b/drivers/net/bonding/bond_3ad.h @@ -33,7 +33,6 @@ #define AD_TIMER_INTERVAL 100 /*msec*/ #define MULTICAST_LACPDU_ADDR {0x01, 0x80, 0xC2, 0x00, 0x00, 0x02} -#define AD_MULTICAST_LACPDU_ADDR {MULTICAST_LACPDU_ADDR} #define AD_LACP_SLOW 0 #define AD_LACP_FAST 1 @@ -42,10 +41,11 @@ typedef struct mac_addr { u8 mac_addr_value[ETH_ALEN]; } mac_addr_t; -typedef enum { - AD_BANDWIDTH = 0, - AD_COUNT -} agg_selection_t; +enum { + BOND_AD_STABLE = 0, + BOND_AD_BANDWIDTH = 1, + BOND_AD_COUNT = 2, +}; // rx machine states(43.4.11 in the 802.3ad standard) typedef enum { @@ -105,12 +105,6 @@ typedef enum { #pragma pack(1) -typedef struct ad_header { - struct mac_addr destination_address; - struct mac_addr source_address; - __be16 length_type; -} ad_header_t; - // Link Aggregation Control Protocol(LACP) data unit structure(43.4.2.2 in the 802.3ad standard) typedef struct lacpdu { u8 subtype; // = LACP(= 0x01) @@ -143,7 +137,7 @@ typedef struct lacpdu { } lacpdu_t; typedef struct lacpdu_header { - struct ad_header ad_header; + struct ethhdr hdr; struct lacpdu lacpdu; } lacpdu_header_t; @@ -164,7 +158,7 @@ typedef struct bond_marker { } bond_marker_t; typedef struct bond_marker_header { - struct ad_header ad_header; + struct ethhdr hdr; struct bond_marker marker; } bond_marker_header_t; @@ -183,7 +177,7 @@ struct port; typedef struct aggregator { struct mac_addr aggregator_mac_address; u16 aggregator_identifier; - u16 is_individual; // BOOLEAN + bool is_individual; u16 actor_admin_aggregator_key; u16 actor_oper_aggregator_key; struct mac_addr partner_system; @@ -198,6 +192,15 @@ typedef struct aggregator { u16 num_of_ports; } aggregator_t; +struct port_params { + struct mac_addr system; + u16 system_priority; + u16 key; + u16 port_number; + u16 port_priority; + u16 port_state; +}; + // port structure(43.4.6 in the 802.3ad standard) typedef struct port { u16 actor_port_number; @@ -205,24 +208,17 @@ typedef struct port { 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; - u16 ntt; // BOOLEAN + bool ntt; u16 actor_admin_port_key; u16 actor_oper_port_key; u8 actor_admin_port_state; u8 actor_oper_port_state; - struct mac_addr partner_admin_system; - struct mac_addr partner_oper_system; - u16 partner_admin_system_priority; - u16 partner_oper_system_priority; - u16 partner_admin_key; - u16 partner_oper_key; - u16 partner_admin_port_number; - u16 partner_oper_port_number; - u16 partner_admin_port_priority; - u16 partner_oper_port_priority; - u8 partner_admin_port_state; - u8 partner_oper_port_state; - u16 is_enabled; // BOOLEAN + + struct port_params partner_admin; + struct port_params partner_oper; + + 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 @@ -241,10 +237,10 @@ typedef struct port { } port_t; // system structure -typedef struct ad_system { +struct ad_system { u16 sys_priority; struct mac_addr sys_mac_addr; -} ad_system_t; +}; #ifdef __ia64__ #pragma pack() @@ -255,7 +251,7 @@ typedef struct ad_system { #define SLAVE_AD_INFO(slave) ((slave)->ad_info) struct ad_bond_info { - ad_system_t system; // 802.3ad system structure + struct ad_system system; /* 802.3ad system structure */ u32 agg_select_timer; // Timer to select aggregator after all adapter's hand shakes u32 agg_select_mode; // Mode of selection of active aggregator(bandwidth/count) int lacp_fast; /* whether fast periodic tx should be @@ -277,6 +273,7 @@ void bond_3ad_initialize(struct bonding *bond, u16 tick_resolution, int lacp_fas int 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); void bond_3ad_adapter_speed_changed(struct slave *slave); void bond_3ad_adapter_duplex_changed(struct slave *slave); void bond_3ad_handle_link_change(struct slave *slave, char link); diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c index 87437c78847..27fb7f5c21c 100644 --- a/drivers/net/bonding/bond_alb.c +++ b/drivers/net/bonding/bond_alb.c @@ -20,8 +20,6 @@ * */ -//#define BONDING_DEBUG 1 - #include <linux/skbuff.h> #include <linux/netdevice.h> #include <linux/etherdevice.h> @@ -346,30 +344,37 @@ static void rlb_update_entry_from_arp(struct bonding *bond, struct arp_pkt *arp) static int rlb_arp_recv(struct sk_buff *skb, struct net_device *bond_dev, struct packet_type *ptype, struct net_device *orig_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond; struct arp_pkt *arp = (struct arp_pkt *)skb->data; int res = NET_RX_DROP; if (dev_net(bond_dev) != &init_net) goto out; - if (!(bond_dev->flags & IFF_MASTER)) + while (bond_dev->priv_flags & IFF_802_1Q_VLAN) + bond_dev = vlan_dev_real_dev(bond_dev); + + if (!(bond_dev->priv_flags & IFF_BONDING) || + !(bond_dev->flags & IFF_MASTER)) goto out; if (!arp) { - dprintk("Packet has no ARP data\n"); + pr_debug("Packet has no ARP data\n"); goto out; } if (skb->len < sizeof(struct arp_pkt)) { - dprintk("Packet is too small to be an ARP\n"); + pr_debug("Packet is too small to be an ARP\n"); goto out; } if (arp->op_code == htons(ARPOP_REPLY)) { /* update rx hash table for this ARP */ + printk("rar: update orig %s bond_dev %s\n", orig_dev->name, + bond_dev->name); + bond = netdev_priv(bond_dev); rlb_update_entry_from_arp(bond, arp); - dprintk("Server received an ARP Reply from client\n"); + pr_debug("Server received an ARP Reply from client\n"); } res = NET_RX_SUCCESS; @@ -723,7 +728,7 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond) if (tx_slave) { memcpy(arp->mac_src,tx_slave->dev->dev_addr, ETH_ALEN); } - dprintk("Server sent ARP Reply packet\n"); + 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 * place holder. @@ -743,7 +748,7 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond) * updated with their assigned mac. */ rlb_req_update_subnet_clients(bond, arp->ip_src); - dprintk("Server sent ARP Request packet\n"); + pr_debug("Server sent ARP Request packet\n"); } return tx_slave; @@ -818,7 +823,7 @@ static int rlb_initialize(struct bonding *bond) /*initialize packet type*/ pk_type->type = __constant_htons(ETH_P_ARP); - pk_type->dev = bond->dev; + pk_type->dev = NULL; pk_type->func = rlb_arp_recv; /* register to receive ARPs */ @@ -1211,11 +1216,6 @@ static int alb_set_mac_address(struct bonding *bond, void *addr) } bond_for_each_slave(bond, slave, i) { - if (slave->dev->set_mac_address == NULL) { - res = -EOPNOTSUPP; - goto unwind; - } - /* save net_device's current hw address */ memcpy(tmp_addr, slave->dev->dev_addr, ETH_ALEN); @@ -1224,9 +1224,8 @@ static int alb_set_mac_address(struct bonding *bond, void *addr) /* restore net_device's hw address */ memcpy(slave->dev->dev_addr, tmp_addr, ETH_ALEN); - if (res) { + if (res) goto unwind; - } } return 0; @@ -1285,7 +1284,7 @@ void bond_alb_deinitialize(struct bonding *bond) int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct ethhdr *eth_data; struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond)); struct slave *tx_slave = NULL; @@ -1706,7 +1705,7 @@ void bond_alb_handle_active_change(struct bonding *bond, struct slave *new_slave */ int bond_alb_set_mac_address(struct net_device *bond_dev, void *addr) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct sockaddr *sa = addr; struct slave *slave, *swap_slave; int res; diff --git a/drivers/net/bonding/bond_ipv6.c b/drivers/net/bonding/bond_ipv6.c new file mode 100644 index 00000000000..0d73bf5ac5a --- /dev/null +++ b/drivers/net/bonding/bond_ipv6.c @@ -0,0 +1,216 @@ +/* + * Copyright(c) 2008 Hewlett-Packard Development Company, L.P. + * + * 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. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write to the Free Software Foundation, Inc., + * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * The full GNU General Public License is included in this distribution in the + * file called LICENSE. + * + */ + +#include <linux/types.h> +#include <linux/if_vlan.h> +#include <net/ipv6.h> +#include <net/ndisc.h> +#include <net/addrconf.h> +#include "bonding.h" + +/* + * Assign bond->master_ipv6 to the next IPv6 address in the list, or + * zero it out if there are none. + */ +static void bond_glean_dev_ipv6(struct net_device *dev, struct in6_addr *addr) +{ + struct inet6_dev *idev; + struct inet6_ifaddr *ifa; + + if (!dev) + return; + + idev = in6_dev_get(dev); + if (!idev) + return; + + read_lock_bh(&idev->lock); + ifa = idev->addr_list; + if (ifa) + ipv6_addr_copy(addr, &ifa->addr); + else + ipv6_addr_set(addr, 0, 0, 0, 0); + + read_unlock_bh(&idev->lock); + + in6_dev_put(idev); +} + +static void bond_na_send(struct net_device *slave_dev, + struct in6_addr *daddr, + int router, + unsigned short vlan_id) +{ + struct in6_addr mcaddr; + struct icmp6hdr icmp6h = { + .icmp6_type = NDISC_NEIGHBOUR_ADVERTISEMENT, + }; + struct sk_buff *skb; + + icmp6h.icmp6_router = router; + icmp6h.icmp6_solicited = 0; + icmp6h.icmp6_override = 1; + + addrconf_addr_solict_mult(daddr, &mcaddr); + + pr_debug("ipv6 na on slave %s: dest %pI6, src %pI6\n", + slave_dev->name, &mcaddr, daddr); + + skb = ndisc_build_skb(slave_dev, &mcaddr, daddr, &icmp6h, daddr, + ND_OPT_TARGET_LL_ADDR); + + if (!skb) { + printk(KERN_ERR DRV_NAME ": NA packet allocation failed\n"); + return; + } + + if (vlan_id) { + skb = vlan_put_tag(skb, vlan_id); + if (!skb) { + printk(KERN_ERR DRV_NAME ": failed to insert VLAN tag\n"); + return; + } + } + + ndisc_send_skb(skb, slave_dev, NULL, &mcaddr, daddr, &icmp6h); +} + +/* + * Kick out an unsolicited Neighbor Advertisement for an IPv6 address on + * the bonding master. This will help the switch learn our address + * if in active-backup mode. + * + * Caller must hold curr_slave_lock for read or better + */ +void bond_send_unsolicited_na(struct bonding *bond) +{ + struct slave *slave = bond->curr_active_slave; + struct vlan_entry *vlan; + struct inet6_dev *idev; + int is_router; + + pr_debug("bond_send_unsol_na: bond %s slave %s\n", bond->dev->name, + slave ? slave->dev->name : "NULL"); + + if (!slave || !bond->send_unsol_na || + test_bit(__LINK_STATE_LINKWATCH_PENDING, &slave->dev->state)) + return; + + bond->send_unsol_na--; + + idev = in6_dev_get(bond->dev); + if (!idev) + return; + + is_router = !!idev->cnf.forwarding; + + in6_dev_put(idev); + + if (!ipv6_addr_any(&bond->master_ipv6)) + bond_na_send(slave->dev, &bond->master_ipv6, is_router, 0); + + list_for_each_entry(vlan, &bond->vlan_list, vlan_list) { + if (!ipv6_addr_any(&vlan->vlan_ipv6)) { + bond_na_send(slave->dev, &vlan->vlan_ipv6, is_router, + vlan->vlan_id); + } + } +} + +/* + * bond_inet6addr_event: handle inet6addr notifier chain events. + * + * We keep track of device IPv6 addresses primarily to use as source + * addresses in NS probes. + * + * We track one IPv6 for the main device (if it has one). + */ +static int bond_inet6addr_event(struct notifier_block *this, + unsigned long event, + void *ptr) +{ + struct inet6_ifaddr *ifa = ptr; + struct net_device *vlan_dev, *event_dev = ifa->idev->dev; + struct bonding *bond; + struct vlan_entry *vlan; + + if (dev_net(event_dev) != &init_net) + return NOTIFY_DONE; + + list_for_each_entry(bond, &bond_dev_list, bond_list) { + if (bond->dev == event_dev) { + switch (event) { + case NETDEV_UP: + if (ipv6_addr_any(&bond->master_ipv6)) + ipv6_addr_copy(&bond->master_ipv6, + &ifa->addr); + return NOTIFY_OK; + case NETDEV_DOWN: + if (ipv6_addr_equal(&bond->master_ipv6, + &ifa->addr)) + bond_glean_dev_ipv6(bond->dev, + &bond->master_ipv6); + return NOTIFY_OK; + default: + return NOTIFY_DONE; + } + } + + list_for_each_entry(vlan, &bond->vlan_list, vlan_list) { + vlan_dev = vlan_group_get_device(bond->vlgrp, + vlan->vlan_id); + if (vlan_dev == event_dev) { + switch (event) { + case NETDEV_UP: + if (ipv6_addr_any(&vlan->vlan_ipv6)) + ipv6_addr_copy(&vlan->vlan_ipv6, + &ifa->addr); + return NOTIFY_OK; + case NETDEV_DOWN: + if (ipv6_addr_equal(&vlan->vlan_ipv6, + &ifa->addr)) + bond_glean_dev_ipv6(vlan_dev, + &vlan->vlan_ipv6); + return NOTIFY_OK; + default: + return NOTIFY_DONE; + } + } + } + } + return NOTIFY_DONE; +} + +static struct notifier_block bond_inet6addr_notifier = { + .notifier_call = bond_inet6addr_event, +}; + +void bond_register_ipv6_notifier(void) +{ + register_inet6addr_notifier(&bond_inet6addr_notifier); +} + +void bond_unregister_ipv6_notifier(void) +{ + unregister_inet6addr_notifier(&bond_inet6addr_notifier); +} + diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index a3efba59eee..9fb388388fb 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -31,8 +31,6 @@ * */ -//#define BONDING_DEBUG 1 - #include <linux/kernel.h> #include <linux/module.h> #include <linux/types.h> @@ -89,6 +87,7 @@ static int max_bonds = BOND_DEFAULT_MAX_BONDS; static int num_grat_arp = 1; +static int num_unsol_na = 1; static int miimon = BOND_LINK_MON_INTERV; static int updelay = 0; static int downdelay = 0; @@ -96,6 +95,7 @@ static int use_carrier = 1; static char *mode = NULL; static char *primary = NULL; static char *lacp_rate = NULL; +static char *ad_select = NULL; static char *xmit_hash_policy = NULL; static int arp_interval = BOND_LINK_ARP_INTERV; static char *arp_ip_target[BOND_MAX_ARP_TARGETS] = { NULL, }; @@ -107,6 +107,8 @@ module_param(max_bonds, int, 0); MODULE_PARM_DESC(max_bonds, "Max number of bonded devices"); module_param(num_grat_arp, int, 0644); MODULE_PARM_DESC(num_grat_arp, "Number of gratuitous ARP packets to send on failover event"); +module_param(num_unsol_na, int, 0644); +MODULE_PARM_DESC(num_unsol_na, "Number of unsolicited IPv6 Neighbor Advertisements packets to send on failover event"); module_param(miimon, int, 0); MODULE_PARM_DESC(miimon, "Link check interval in milliseconds"); module_param(updelay, int, 0); @@ -127,6 +129,8 @@ MODULE_PARM_DESC(primary, "Primary network device to use"); module_param(lacp_rate, charp, 0); MODULE_PARM_DESC(lacp_rate, "LACPDU tx rate to request from 802.3ad partner " "(slow/fast)"); +module_param(ad_select, charp, 0); +MODULE_PARM_DESC(ad_select, "803.ad aggregation selection logic: stable (0, default), bandwidth (1), count (2)"); module_param(xmit_hash_policy, charp, 0); MODULE_PARM_DESC(xmit_hash_policy, "XOR hashing method: 0 for layer 2 (default)" ", 1 for layer 3+4"); @@ -150,7 +154,6 @@ LIST_HEAD(bond_dev_list); static struct proc_dir_entry *bond_proc_dir = NULL; #endif -extern struct rw_semaphore bonding_rwsem; static __be32 arp_target[BOND_MAX_ARP_TARGETS] = { 0, } ; static int arp_ip_count = 0; static int bond_mode = BOND_MODE_ROUNDROBIN; @@ -158,13 +161,13 @@ static int xmit_hashtype= BOND_XMIT_POLICY_LAYER2; static int lacp_fast = 0; -struct bond_parm_tbl bond_lacp_tbl[] = { +const struct bond_parm_tbl bond_lacp_tbl[] = { { "slow", AD_LACP_SLOW}, { "fast", AD_LACP_FAST}, { NULL, -1}, }; -struct bond_parm_tbl bond_mode_tbl[] = { +const struct bond_parm_tbl bond_mode_tbl[] = { { "balance-rr", BOND_MODE_ROUNDROBIN}, { "active-backup", BOND_MODE_ACTIVEBACKUP}, { "balance-xor", BOND_MODE_XOR}, @@ -175,14 +178,14 @@ struct bond_parm_tbl bond_mode_tbl[] = { { NULL, -1}, }; -struct bond_parm_tbl xmit_hashtype_tbl[] = { +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}, { NULL, -1}, }; -struct bond_parm_tbl arp_validate_tbl[] = { +const struct bond_parm_tbl arp_validate_tbl[] = { { "none", BOND_ARP_VALIDATE_NONE}, { "active", BOND_ARP_VALIDATE_ACTIVE}, { "backup", BOND_ARP_VALIDATE_BACKUP}, @@ -190,13 +193,20 @@ struct bond_parm_tbl arp_validate_tbl[] = { { NULL, -1}, }; -struct bond_parm_tbl fail_over_mac_tbl[] = { +const struct bond_parm_tbl fail_over_mac_tbl[] = { { "none", BOND_FOM_NONE}, { "active", BOND_FOM_ACTIVE}, { "follow", BOND_FOM_FOLLOW}, { 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 void bond_send_gratuitous_arp(struct bonding *bond); @@ -206,24 +216,20 @@ static void bond_deinit(struct net_device *bond_dev); static const char *bond_mode_name(int mode) { - switch (mode) { - case BOND_MODE_ROUNDROBIN : - return "load balancing (round-robin)"; - case BOND_MODE_ACTIVEBACKUP : - return "fault-tolerance (active-backup)"; - case BOND_MODE_XOR : - return "load balancing (xor)"; - case BOND_MODE_BROADCAST : - return "fault-tolerance (broadcast)"; - case BOND_MODE_8023AD: - return "IEEE 802.3ad Dynamic link aggregation"; - case BOND_MODE_TLB: - return "transmit load balancing"; - case BOND_MODE_ALB: - return "adaptive load balancing"; - default: + static const char *names[] = { + [BOND_MODE_ROUNDROBIN] = "load balancing (round-robin)", + [BOND_MODE_ACTIVEBACKUP] = "fault-tolerance (active-backup)", + [BOND_MODE_XOR] = "load balancing (xor)", + [BOND_MODE_BROADCAST] = "fault-tolerance (broadcast)", + [BOND_MODE_8023AD]= "IEEE 802.3ad Dynamic link aggregation", + [BOND_MODE_TLB] = "transmit load balancing", + [BOND_MODE_ALB] = "adaptive load balancing", + }; + + if (mode < 0 || mode > BOND_MODE_ALB) return "unknown"; - } + + return names[mode]; } /*---------------------------------- VLAN -----------------------------------*/ @@ -239,17 +245,16 @@ static int bond_add_vlan(struct bonding *bond, unsigned short vlan_id) { struct vlan_entry *vlan; - dprintk("bond: %s, vlan id %d\n", + pr_debug("bond: %s, vlan id %d\n", (bond ? bond->dev->name: "None"), vlan_id); - vlan = kmalloc(sizeof(struct vlan_entry), GFP_KERNEL); + vlan = kzalloc(sizeof(struct vlan_entry), GFP_KERNEL); if (!vlan) { return -ENOMEM; } INIT_LIST_HEAD(&vlan->vlan_list); vlan->vlan_id = vlan_id; - vlan->vlan_ip = 0; write_lock_bh(&bond->lock); @@ -257,7 +262,7 @@ static int bond_add_vlan(struct bonding *bond, unsigned short vlan_id) write_unlock_bh(&bond->lock); - dprintk("added VLAN ID %d on bond %s\n", vlan_id, bond->dev->name); + pr_debug("added VLAN ID %d on bond %s\n", vlan_id, bond->dev->name); return 0; } @@ -274,7 +279,7 @@ static int bond_del_vlan(struct bonding *bond, unsigned short vlan_id) struct vlan_entry *vlan; int res = -ENODEV; - dprintk("bond: %s, vlan id %d\n", bond->dev->name, vlan_id); + pr_debug("bond: %s, vlan id %d\n", bond->dev->name, vlan_id); write_lock_bh(&bond->lock); @@ -282,12 +287,10 @@ static int bond_del_vlan(struct bonding *bond, unsigned short vlan_id) if (vlan->vlan_id == vlan_id) { list_del(&vlan->vlan_list); - if ((bond->params.mode == BOND_MODE_TLB) || - (bond->params.mode == BOND_MODE_ALB)) { + if (bond_is_lb(bond)) bond_alb_clear_vlan(bond, vlan_id); - } - dprintk("removed VLAN ID %d from bond %s\n", vlan_id, + pr_debug("removed VLAN ID %d from bond %s\n", vlan_id, bond->dev->name); kfree(vlan); @@ -307,7 +310,7 @@ static int bond_del_vlan(struct bonding *bond, unsigned short vlan_id) } } - dprintk("couldn't find VLAN ID %d in bond %s\n", vlan_id, + pr_debug("couldn't find VLAN ID %d in bond %s\n", vlan_id, bond->dev->name); out: @@ -331,13 +334,13 @@ static int bond_has_challenged_slaves(struct bonding *bond) bond_for_each_slave(bond, slave, i) { if (slave->dev->features & NETIF_F_VLAN_CHALLENGED) { - dprintk("found VLAN challenged slave - %s\n", + pr_debug("found VLAN challenged slave - %s\n", slave->dev->name); return 1; } } - dprintk("no VLAN challenged slaves found\n"); + pr_debug("no VLAN challenged slaves found\n"); return 0; } @@ -442,7 +445,7 @@ int bond_dev_queue_xmit(struct bonding *bond, struct sk_buff *skb, struct net_de */ static void bond_vlan_rx_register(struct net_device *bond_dev, struct vlan_group *grp) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct slave *slave; int i; @@ -450,10 +453,11 @@ static void bond_vlan_rx_register(struct net_device *bond_dev, struct vlan_group bond_for_each_slave(bond, slave, i) { struct net_device *slave_dev = slave->dev; + const struct net_device_ops *slave_ops = slave_dev->netdev_ops; if ((slave_dev->features & NETIF_F_HW_VLAN_RX) && - slave_dev->vlan_rx_register) { - slave_dev->vlan_rx_register(slave_dev, grp); + slave_ops->ndo_vlan_rx_register) { + slave_ops->ndo_vlan_rx_register(slave_dev, grp); } } } @@ -465,16 +469,17 @@ static void bond_vlan_rx_register(struct net_device *bond_dev, struct vlan_group */ static void bond_vlan_rx_add_vid(struct net_device *bond_dev, uint16_t vid) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct slave *slave; int i, res; bond_for_each_slave(bond, slave, i) { struct net_device *slave_dev = slave->dev; + const struct net_device_ops *slave_ops = slave_dev->netdev_ops; if ((slave_dev->features & NETIF_F_HW_VLAN_FILTER) && - slave_dev->vlan_rx_add_vid) { - slave_dev->vlan_rx_add_vid(slave_dev, vid); + slave_ops->ndo_vlan_rx_add_vid) { + slave_ops->ndo_vlan_rx_add_vid(slave_dev, vid); } } @@ -493,21 +498,22 @@ static void bond_vlan_rx_add_vid(struct net_device *bond_dev, uint16_t vid) */ static void bond_vlan_rx_kill_vid(struct net_device *bond_dev, uint16_t vid) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct slave *slave; struct net_device *vlan_dev; int i, res; bond_for_each_slave(bond, slave, i) { struct net_device *slave_dev = slave->dev; + const struct net_device_ops *slave_ops = slave_dev->netdev_ops; if ((slave_dev->features & NETIF_F_HW_VLAN_FILTER) && - slave_dev->vlan_rx_kill_vid) { + slave_ops->ndo_vlan_rx_kill_vid) { /* Save and then restore vlan_dev in the grp array, * since the slave's driver might clear it. */ vlan_dev = vlan_group_get_device(bond->vlgrp, vid); - slave_dev->vlan_rx_kill_vid(slave_dev, vid); + slave_ops->ndo_vlan_rx_kill_vid(slave_dev, vid); vlan_group_set_device(bond->vlgrp, vid, vlan_dev); } } @@ -523,26 +529,23 @@ static void bond_vlan_rx_kill_vid(struct net_device *bond_dev, uint16_t vid) static void bond_add_vlans_on_slave(struct bonding *bond, struct net_device *slave_dev) { struct vlan_entry *vlan; + const struct net_device_ops *slave_ops = slave_dev->netdev_ops; write_lock_bh(&bond->lock); - if (list_empty(&bond->vlan_list)) { + if (list_empty(&bond->vlan_list)) goto out; - } if ((slave_dev->features & NETIF_F_HW_VLAN_RX) && - slave_dev->vlan_rx_register) { - slave_dev->vlan_rx_register(slave_dev, bond->vlgrp); - } + slave_ops->ndo_vlan_rx_register) + slave_ops->ndo_vlan_rx_register(slave_dev, bond->vlgrp); if (!(slave_dev->features & NETIF_F_HW_VLAN_FILTER) || - !(slave_dev->vlan_rx_add_vid)) { + !(slave_ops->ndo_vlan_rx_add_vid)) goto out; - } - list_for_each_entry(vlan, &bond->vlan_list, vlan_list) { - slave_dev->vlan_rx_add_vid(slave_dev, vlan->vlan_id); - } + list_for_each_entry(vlan, &bond->vlan_list, vlan_list) + slave_ops->ndo_vlan_rx_add_vid(slave_dev, vlan->vlan_id); out: write_unlock_bh(&bond->lock); @@ -550,34 +553,32 @@ out: static void bond_del_vlans_from_slave(struct bonding *bond, struct net_device *slave_dev) { + const struct net_device_ops *slave_ops = slave_dev->netdev_ops; struct vlan_entry *vlan; struct net_device *vlan_dev; write_lock_bh(&bond->lock); - if (list_empty(&bond->vlan_list)) { + if (list_empty(&bond->vlan_list)) goto out; - } if (!(slave_dev->features & NETIF_F_HW_VLAN_FILTER) || - !(slave_dev->vlan_rx_kill_vid)) { + !(slave_ops->ndo_vlan_rx_kill_vid)) goto unreg; - } list_for_each_entry(vlan, &bond->vlan_list, vlan_list) { /* Save and then restore vlan_dev in the grp array, * since the slave's driver might clear it. */ vlan_dev = vlan_group_get_device(bond->vlgrp, vlan->vlan_id); - slave_dev->vlan_rx_kill_vid(slave_dev, vlan->vlan_id); + slave_ops->ndo_vlan_rx_kill_vid(slave_dev, vlan->vlan_id); vlan_group_set_device(bond->vlgrp, vlan->vlan_id, vlan_dev); } unreg: if ((slave_dev->features & NETIF_F_HW_VLAN_RX) && - slave_dev->vlan_rx_register) { - slave_dev->vlan_rx_register(slave_dev, NULL); - } + slave_ops->ndo_vlan_rx_register) + slave_ops->ndo_vlan_rx_register(slave_dev, NULL); out: write_unlock_bh(&bond->lock); @@ -686,15 +687,15 @@ static int bond_update_speed_duplex(struct slave *slave) */ static int bond_check_dev_link(struct bonding *bond, struct net_device *slave_dev, int reporting) { + const struct net_device_ops *slave_ops = slave_dev->netdev_ops; static int (* ioctl)(struct net_device *, struct ifreq *, int); struct ifreq ifr; struct mii_ioctl_data *mii; - if (bond->params.use_carrier) { + if (bond->params.use_carrier) return netif_carrier_ok(slave_dev) ? BMSR_LSTATUS : 0; - } - ioctl = slave_dev->do_ioctl; + ioctl = slave_ops->ndo_do_ioctl; if (ioctl) { /* TODO: set pointer to correct ioctl on a per team member */ /* bases to make this more efficient. that is, once */ @@ -927,7 +928,7 @@ static int bond_mc_list_copy(struct dev_mc_list *mc_list, struct bonding *bond, */ static void bond_mc_list_flush(struct net_device *bond_dev, struct net_device *slave_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct dev_mc_list *dmi; for (dmi = bond_dev->mc_list; dmi; dmi = dmi->next) { @@ -1164,10 +1165,8 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active) bond_3ad_handle_link_change(new_active, BOND_LINK_UP); } - if ((bond->params.mode == BOND_MODE_TLB) || - (bond->params.mode == BOND_MODE_ALB)) { + if (bond_is_lb(bond)) bond_alb_handle_link_change(bond, new_active, BOND_LINK_UP); - } } else { if (USES_PRIMARY(bond->params.mode)) { printk(KERN_INFO DRV_NAME @@ -1182,8 +1181,7 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active) bond_mc_swap(bond, new_active, old_active); } - if ((bond->params.mode == BOND_MODE_TLB) || - (bond->params.mode == BOND_MODE_ALB)) { + if (bond_is_lb(bond)) { bond_alb_handle_active_change(bond, new_active); if (old_active) bond_set_slave_inactive_flags(old_active); @@ -1208,6 +1206,9 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active) bond->send_grat_arp = bond->params.num_grat_arp; bond_send_gratuitous_arp(bond); + bond->send_unsol_na = bond->params.num_unsol_na; + bond_send_unsolicited_na(bond); + write_unlock_bh(&bond->curr_slave_lock); read_unlock(&bond->lock); @@ -1315,9 +1316,9 @@ static void bond_detach_slave(struct bonding *bond, struct slave *slave) static int bond_sethwaddr(struct net_device *bond_dev, struct net_device *slave_dev) { - dprintk("bond_dev=%p\n", bond_dev); - dprintk("slave_dev=%p\n", slave_dev); - dprintk("slave_dev->addr_len=%d\n", slave_dev->addr_len); + pr_debug("bond_dev=%p\n", bond_dev); + pr_debug("slave_dev=%p\n", slave_dev); + pr_debug("slave_dev->addr_len=%d\n", slave_dev->addr_len); memcpy(bond_dev->dev_addr, slave_dev->dev_addr, slave_dev->addr_len); return 0; } @@ -1364,14 +1365,12 @@ done: return 0; } - static void bond_setup_by_slave(struct net_device *bond_dev, struct net_device *slave_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); - bond_dev->neigh_setup = slave_dev->neigh_setup; - bond_dev->header_ops = slave_dev->header_ops; + bond_dev->header_ops = slave_dev->header_ops; bond_dev->type = slave_dev->type; bond_dev->hard_header_len = slave_dev->hard_header_len; @@ -1385,7 +1384,8 @@ static void bond_setup_by_slave(struct net_device *bond_dev, /* enslave device <slave> to bond device <master> */ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); + const struct net_device_ops *slave_ops = slave_dev->netdev_ops; struct slave *new_slave = NULL; struct dev_mc_list *dmi; struct sockaddr addr; @@ -1394,7 +1394,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) int res = 0; if (!bond->params.use_carrier && slave_dev->ethtool_ops == NULL && - slave_dev->do_ioctl == NULL) { + slave_ops->ndo_do_ioctl == NULL) { printk(KERN_WARNING DRV_NAME ": %s: Warning: no link monitoring support for %s\n", bond_dev->name, slave_dev->name); @@ -1409,14 +1409,14 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) /* already enslaved */ if (slave_dev->flags & IFF_SLAVE) { - dprintk("Error, Device was already enslaved\n"); + pr_debug("Error, Device was already enslaved\n"); return -EBUSY; } /* vlan challenged mutual exclusion */ /* no need to lock since we're protected by rtnl_lock */ if (slave_dev->features & NETIF_F_VLAN_CHALLENGED) { - dprintk("%s: NETIF_F_VLAN_CHALLENGED\n", slave_dev->name); + pr_debug("%s: NETIF_F_VLAN_CHALLENGED\n", slave_dev->name); if (!list_empty(&bond->vlan_list)) { printk(KERN_ERR DRV_NAME ": %s: Error: cannot enslave VLAN " @@ -1434,7 +1434,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) bond_dev->features |= NETIF_F_VLAN_CHALLENGED; } } else { - dprintk("%s: ! NETIF_F_VLAN_CHALLENGED\n", slave_dev->name); + pr_debug("%s: ! NETIF_F_VLAN_CHALLENGED\n", slave_dev->name); if (bond->slave_cnt == 0) { /* First slave, and it is not VLAN challenged, * so remove the block of adding VLANs over the bond. @@ -1476,7 +1476,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) goto err_undo_flags; } - if (slave_dev->set_mac_address == NULL) { + if (slave_ops->ndo_set_mac_address == NULL) { if (bond->slave_cnt == 0) { printk(KERN_WARNING DRV_NAME ": %s: Warning: The first slave device " @@ -1522,28 +1522,27 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) addr.sa_family = slave_dev->type; res = dev_set_mac_address(slave_dev, &addr); if (res) { - dprintk("Error %d calling set_mac_address\n", res); + pr_debug("Error %d calling set_mac_address\n", res); goto err_free; } } res = netdev_set_master(slave_dev, bond_dev); if (res) { - dprintk("Error %d calling netdev_set_master\n", res); + pr_debug("Error %d calling netdev_set_master\n", res); goto err_restore_mac; } /* open the slave since the application closed it */ res = dev_open(slave_dev); if (res) { - dprintk("Openning slave %s failed\n", slave_dev->name); + pr_debug("Openning slave %s failed\n", slave_dev->name); goto err_unset_master; } new_slave->dev = slave_dev; slave_dev->priv_flags |= IFF_BONDING; - if ((bond->params.mode == BOND_MODE_TLB) || - (bond->params.mode == BOND_MODE_ALB)) { + if (bond_is_lb(bond)) { /* bond_alb_init_slave() must be called before all other stages since * it might fail and we do not want to have to undo everything */ @@ -1641,18 +1640,18 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) if (!bond->params.miimon || (bond_check_dev_link(bond, slave_dev, 0) == BMSR_LSTATUS)) { if (bond->params.updelay) { - dprintk("Initial state of slave_dev is " + pr_debug("Initial state of slave_dev is " "BOND_LINK_BACK\n"); new_slave->link = BOND_LINK_BACK; new_slave->delay = bond->params.updelay; } else { - dprintk("Initial state of slave_dev is " + pr_debug("Initial state of slave_dev is " "BOND_LINK_UP\n"); new_slave->link = BOND_LINK_UP; } new_slave->jiffies = jiffies; } else { - dprintk("Initial state of slave_dev is " + pr_debug("Initial state of slave_dev is " "BOND_LINK_DOWN\n"); new_slave->link = BOND_LINK_DOWN; } @@ -1713,7 +1712,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) bond_set_slave_inactive_flags(new_slave); break; default: - dprintk("This slave is always active in trunk mode\n"); + pr_debug("This slave is always active in trunk mode\n"); /* always active in trunk mode */ new_slave->state = BOND_STATE_ACTIVE; @@ -1787,11 +1786,10 @@ err_undo_flags: */ int bond_release(struct net_device *bond_dev, struct net_device *slave_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct slave *slave, *oldcurrent; struct sockaddr addr; int mac_addr_differ; - DECLARE_MAC_BUF(mac); /* slave is not a slave or master is not master of this slave */ if (!(slave_dev->flags & IFF_SLAVE) || @@ -1820,11 +1818,11 @@ int bond_release(struct net_device *bond_dev, struct net_device *slave_dev) if (!mac_addr_differ && (bond->slave_cnt > 1)) printk(KERN_WARNING DRV_NAME ": %s: Warning: the permanent HWaddr of %s - " - "%s - is still in use by %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, - print_mac(mac, slave->perm_hwaddr), + slave->perm_hwaddr, bond_dev->name, slave_dev->name); } @@ -1860,8 +1858,7 @@ int bond_release(struct net_device *bond_dev, struct net_device *slave_dev) bond_change_active_slave(bond, NULL); } - if ((bond->params.mode == BOND_MODE_TLB) || - (bond->params.mode == BOND_MODE_ALB)) { + if (bond_is_lb(bond)) { /* Must be called only after the slave has been * detached from the list and the curr_active_slave * has been cleared (if our_slave == old_current), @@ -1981,7 +1978,7 @@ void bond_destroy(struct bonding *bond) static void bond_destructor(struct net_device *bond_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); if (bond->wq) destroy_workqueue(bond->wq); @@ -1999,7 +1996,7 @@ static void bond_destructor(struct net_device *bond_dev) */ int bond_release_and_destroy(struct net_device *bond_dev, struct net_device *slave_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); int ret; ret = bond_release(bond_dev, slave_dev); @@ -2016,7 +2013,7 @@ int bond_release_and_destroy(struct net_device *bond_dev, struct net_device *sl */ static int bond_release_all(struct net_device *bond_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct slave *slave; struct net_device *slave_dev; struct sockaddr addr; @@ -2050,8 +2047,7 @@ static int bond_release_all(struct net_device *bond_dev) */ write_unlock_bh(&bond->lock); - if ((bond->params.mode == BOND_MODE_TLB) || - (bond->params.mode == BOND_MODE_ALB)) { + if (bond_is_lb(bond)) { /* must be called only after the slave * has been detached from the list */ @@ -2147,7 +2143,7 @@ out: */ static int bond_ioctl_change_active(struct net_device *bond_dev, struct net_device *slave_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct slave *old_active = NULL; struct slave *new_active = NULL; int res = 0; @@ -2196,7 +2192,7 @@ static int bond_ioctl_change_active(struct net_device *bond_dev, struct net_devi static int bond_info_query(struct net_device *bond_dev, struct ifbond *info) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); info->bond_mode = bond->params.mode; info->miimon = bond->params.miimon; @@ -2210,7 +2206,7 @@ static int bond_info_query(struct net_device *bond_dev, struct ifbond *info) static int bond_slave_info_query(struct net_device *bond_dev, struct ifslave *info) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct slave *slave; int i, found = 0; @@ -2378,8 +2374,7 @@ static void bond_miimon_commit(struct bonding *bond) if (bond->params.mode == BOND_MODE_8023AD) bond_3ad_handle_link_change(slave, BOND_LINK_UP); - if ((bond->params.mode == BOND_MODE_TLB) || - (bond->params.mode == BOND_MODE_ALB)) + if (bond_is_lb(bond)) bond_alb_handle_link_change(bond, slave, BOND_LINK_UP); @@ -2464,6 +2459,12 @@ void bond_mii_monitor(struct work_struct *work) read_unlock(&bond->curr_slave_lock); } + if (bond->send_unsol_na) { + read_lock(&bond->curr_slave_lock); + bond_send_unsolicited_na(bond); + read_unlock(&bond->curr_slave_lock); + } + if (bond_miimon_inspect(bond)) { read_unlock(&bond->lock); rtnl_lock(); @@ -2532,7 +2533,7 @@ static void bond_arp_send(struct net_device *slave_dev, int arp_op, __be32 dest_ { struct sk_buff *skb; - dprintk("arp %d on slave %s: dst %x src %x vid %d\n", arp_op, + pr_debug("arp %d on slave %s: dst %x src %x vid %d\n", arp_op, slave_dev->name, dest_ip, src_ip, vlan_id); skb = arp_create(arp_op, ETH_P_ARP, dest_ip, slave_dev, src_ip, @@ -2565,9 +2566,9 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave) for (i = 0; (i < BOND_MAX_ARP_TARGETS); i++) { if (!targets[i]) continue; - dprintk("basa: target %x\n", targets[i]); + pr_debug("basa: target %x\n", targets[i]); if (list_empty(&bond->vlan_list)) { - dprintk("basa: empty vlan: arp_send\n"); + pr_debug("basa: empty vlan: arp_send\n"); bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], bond->master_ip, 0); continue; @@ -2586,8 +2587,8 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave) if (rv) { if (net_ratelimit()) { printk(KERN_WARNING DRV_NAME - ": %s: no route to arp_ip_target %u.%u.%u.%u\n", - bond->dev->name, NIPQUAD(fl.fl4_dst)); + ": %s: no route to arp_ip_target %pI4\n", + bond->dev->name, &fl.fl4_dst); } continue; } @@ -2597,7 +2598,7 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave) */ if (rt->u.dst.dev == bond->dev) { ip_rt_put(rt); - dprintk("basa: rtdev == bond->dev: arp_send\n"); + pr_debug("basa: rtdev == bond->dev: arp_send\n"); bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], bond->master_ip, 0); continue; @@ -2608,7 +2609,7 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave) vlan_dev = vlan_group_get_device(bond->vlgrp, vlan->vlan_id); if (vlan_dev == rt->u.dst.dev) { vlan_id = vlan->vlan_id; - dprintk("basa: vlan match on %s %d\n", + pr_debug("basa: vlan match on %s %d\n", vlan_dev->name, vlan_id); break; } @@ -2623,8 +2624,8 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave) if (net_ratelimit()) { printk(KERN_WARNING DRV_NAME - ": %s: no path to arp_ip_target %u.%u.%u.%u via rt.dev %s\n", - bond->dev->name, NIPQUAD(fl.fl4_dst), + ": %s: no path to arp_ip_target %pI4 via rt.dev %s\n", + bond->dev->name, &fl.fl4_dst, rt->u.dst.dev ? rt->u.dst.dev->name : "NULL"); } ip_rt_put(rt); @@ -2643,7 +2644,7 @@ static void bond_send_gratuitous_arp(struct bonding *bond) struct vlan_entry *vlan; struct net_device *vlan_dev; - dprintk("bond_send_grat_arp: bond %s slave %s\n", bond->dev->name, + pr_debug("bond_send_grat_arp: bond %s slave %s\n", bond->dev->name, slave ? slave->dev->name : "NULL"); if (!slave || !bond->send_grat_arp || @@ -2673,10 +2674,8 @@ static void bond_validate_arp(struct bonding *bond, struct slave *slave, __be32 targets = bond->params.arp_targets; for (i = 0; (i < BOND_MAX_ARP_TARGETS) && targets[i]; i++) { - dprintk("bva: sip %u.%u.%u.%u tip %u.%u.%u.%u t[%d] " - "%u.%u.%u.%u bhti(tip) %d\n", - NIPQUAD(sip), NIPQUAD(tip), i, NIPQUAD(targets[i]), - bond_has_this_ip(bond, tip)); + pr_debug("bva: sip %pI4 tip %pI4 t[%d] %pI4 bhti(tip) %d\n", + &sip, &tip, i, &targets[i], bond_has_this_ip(bond, tip)); if (sip == targets[i]) { if (bond_has_this_ip(bond, tip)) slave->last_arp_rx = jiffies; @@ -2699,10 +2698,10 @@ static int bond_arp_rcv(struct sk_buff *skb, struct net_device *dev, struct pack if (!(dev->priv_flags & IFF_BONDING) || !(dev->flags & IFF_MASTER)) goto out; - bond = dev->priv; + bond = netdev_priv(dev); read_lock(&bond->lock); - dprintk("bond_arp_rcv: bond %s skb->dev %s orig_dev %s\n", + pr_debug("bond_arp_rcv: bond %s skb->dev %s orig_dev %s\n", bond->dev->name, skb->dev ? skb->dev->name : "NULL", orig_dev ? orig_dev->name : "NULL"); @@ -2728,10 +2727,10 @@ static int bond_arp_rcv(struct sk_buff *skb, struct net_device *dev, struct pack arp_ptr += 4 + dev->addr_len; memcpy(&tip, arp_ptr, 4); - dprintk("bond_arp_rcv: %s %s/%d av %d sv %d sip %u.%u.%u.%u" - " tip %u.%u.%u.%u\n", bond->dev->name, slave->dev->name, - slave->state, bond->params.arp_validate, - slave_do_arp_validate(bond, slave), NIPQUAD(sip), NIPQUAD(tip)); + pr_debug("bond_arp_rcv: %s %s/%d av %d sv %d sip %pI4 tip %pI4\n", + bond->dev->name, slave->dev->name, slave->state, + bond->params.arp_validate, slave_do_arp_validate(bond, slave), + &sip, &tip); /* * Backup slaves won't see the ARP reply, but do come through @@ -3161,6 +3160,12 @@ void bond_activebackup_arp_mon(struct work_struct *work) read_unlock(&bond->curr_slave_lock); } + if (bond->send_unsol_na) { + read_lock(&bond->curr_slave_lock); + bond_send_unsolicited_na(bond); + read_unlock(&bond->curr_slave_lock); + } + if (bond_ab_arp_inspect(bond, delta_in_ticks)) { read_unlock(&bond->lock); rtnl_lock(); @@ -3239,7 +3244,6 @@ static void bond_info_show_master(struct seq_file *seq) struct bonding *bond = seq->private; struct slave *curr; int i; - u32 target; read_lock(&bond->curr_slave_lock); curr = bond->curr_active_slave; @@ -3293,8 +3297,7 @@ static void bond_info_show_master(struct seq_file *seq) continue; if (printed) seq_printf(seq, ","); - target = ntohl(bond->params.arp_targets[i]); - seq_printf(seq, " %d.%d.%d.%d", HIPQUAD(target)); + seq_printf(seq, " %pI4", &bond->params.arp_targets[i]); printed = 1; } seq_printf(seq, "\n"); @@ -3302,11 +3305,12 @@ static void bond_info_show_master(struct seq_file *seq) if (bond->params.mode == BOND_MODE_8023AD) { struct ad_info ad_info; - DECLARE_MAC_BUF(mac); seq_puts(seq, "\n802.3ad info\n"); seq_printf(seq, "LACP rate: %s\n", (bond->params.lacp_fast) ? "fast" : "slow"); + seq_printf(seq, "Aggregator selection policy (ad_select): %s\n", + ad_select_tbl[bond->params.ad_select].modename); if (bond_3ad_get_active_agg_info(bond, &ad_info)) { seq_printf(seq, "bond %s has no active aggregator\n", @@ -3322,8 +3326,8 @@ static void bond_info_show_master(struct seq_file *seq) ad_info.actor_key); seq_printf(seq, "\tPartner Key: %d\n", ad_info.partner_key); - seq_printf(seq, "\tPartner Mac Address: %s\n", - print_mac(mac, ad_info.partner_system)); + seq_printf(seq, "\tPartner Mac Address: %pM\n", + ad_info.partner_system); } } } @@ -3331,7 +3335,6 @@ static void bond_info_show_master(struct seq_file *seq) static void bond_info_show_slave(struct seq_file *seq, const struct slave *slave) { struct bonding *bond = seq->private; - DECLARE_MAC_BUF(mac); seq_printf(seq, "\nSlave Interface: %s\n", slave->dev->name); seq_printf(seq, "MII Status: %s\n", @@ -3339,9 +3342,7 @@ static void bond_info_show_slave(struct seq_file *seq, const struct slave *slave seq_printf(seq, "Link Failure Count: %u\n", slave->link_failure_count); - seq_printf(seq, - "Permanent HW addr: %s\n", - print_mac(mac, slave->perm_hwaddr)); + seq_printf(seq, "Permanent HW addr: %pM\n", slave->perm_hwaddr); if (bond->params.mode == BOND_MODE_8023AD) { const struct aggregator *agg @@ -3506,7 +3507,7 @@ static int bond_event_changename(struct bonding *bond) static int bond_master_netdev_event(unsigned long event, struct net_device *bond_dev) { - struct bonding *event_bond = bond_dev->priv; + struct bonding *event_bond = netdev_priv(bond_dev); switch (event) { case NETDEV_CHANGENAME: @@ -3524,7 +3525,7 @@ static int bond_master_netdev_event(unsigned long event, struct net_device *bond static int bond_slave_netdev_event(unsigned long event, struct net_device *slave_dev) { struct net_device *bond_dev = slave_dev->master; - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); switch (event) { case NETDEV_UNREGISTER: @@ -3591,7 +3592,7 @@ static int bond_netdev_event(struct notifier_block *this, unsigned long event, v if (dev_net(event_dev) != &init_net) return NOTIFY_DONE; - dprintk("event_dev: %s, event: %lx\n", + pr_debug("event_dev: %s, event: %lx\n", (event_dev ? event_dev->name : "None"), event); @@ -3599,12 +3600,12 @@ static int bond_netdev_event(struct notifier_block *this, unsigned long event, v return NOTIFY_DONE; if (event_dev->flags & IFF_MASTER) { - dprintk("IFF_MASTER\n"); + pr_debug("IFF_MASTER\n"); return bond_master_netdev_event(event, event_dev); } if (event_dev->flags & IFF_SLAVE) { - dprintk("IFF_SLAVE\n"); + pr_debug("IFF_SLAVE\n"); return bond_slave_netdev_event(event, event_dev); } @@ -3775,12 +3776,11 @@ static int bond_xmit_hash_policy_l2(struct sk_buff *skb, static int bond_open(struct net_device *bond_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); bond->kill_timers = 0; - if ((bond->params.mode == BOND_MODE_TLB) || - (bond->params.mode == BOND_MODE_ALB)) { + if (bond_is_lb(bond)) { /* bond_alb_initialize must be called before the timer * is started. */ @@ -3816,6 +3816,7 @@ static int bond_open(struct net_device *bond_dev) queue_delayed_work(bond->wq, &bond->ad_work, 0); /* register to receive LACPDUs */ bond_register_lacpdu(bond); + bond_3ad_initiate_agg_selection(bond, 1); } return 0; @@ -3823,7 +3824,7 @@ static int bond_open(struct net_device *bond_dev) static int bond_close(struct net_device *bond_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); if (bond->params.mode == BOND_MODE_8023AD) { /* Unregister the receive of LACPDUs */ @@ -3836,6 +3837,7 @@ static int bond_close(struct net_device *bond_dev) write_lock_bh(&bond->lock); bond->send_grat_arp = 0; + bond->send_unsol_na = 0; /* signal timers not to re-arm */ bond->kill_timers = 1; @@ -3863,8 +3865,7 @@ static int bond_close(struct net_device *bond_dev) } - if ((bond->params.mode == BOND_MODE_TLB) || - (bond->params.mode == BOND_MODE_ALB)) { + if (bond_is_lb(bond)) { /* Must be called only after all * slaves have been released */ @@ -3876,8 +3877,8 @@ static int bond_close(struct net_device *bond_dev) static struct net_device_stats *bond_get_stats(struct net_device *bond_dev) { - struct bonding *bond = bond_dev->priv; - struct net_device_stats *stats = &(bond->stats), *sstats; + struct bonding *bond = netdev_priv(bond_dev); + struct net_device_stats *stats = &bond->stats; struct net_device_stats local_stats; struct slave *slave; int i; @@ -3887,7 +3888,8 @@ static struct net_device_stats *bond_get_stats(struct net_device *bond_dev) read_lock_bh(&bond->lock); bond_for_each_slave(bond, slave, i) { - sstats = slave->dev->get_stats(slave->dev); + const struct net_device_stats *sstats = dev_get_stats(slave->dev); + local_stats.rx_packets += sstats->rx_packets; local_stats.rx_bytes += sstats->rx_bytes; local_stats.rx_errors += sstats->rx_errors; @@ -3932,7 +3934,7 @@ static int bond_do_ioctl(struct net_device *bond_dev, struct ifreq *ifr, int cmd struct mii_ioctl_data *mii = NULL; int res = 0; - dprintk("bond_ioctl: master=%s, cmd=%d\n", + pr_debug("bond_ioctl: master=%s, cmd=%d\n", bond_dev->name, cmd); switch (cmd) { @@ -3954,7 +3956,7 @@ static int bond_do_ioctl(struct net_device *bond_dev, struct ifreq *ifr, int cmd } if (mii->reg_num == 1) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); mii->val_out = 0; read_lock(&bond->lock); read_lock(&bond->curr_slave_lock); @@ -4010,12 +4012,12 @@ static int bond_do_ioctl(struct net_device *bond_dev, struct ifreq *ifr, int cmd down_write(&(bonding_rwsem)); slave_dev = dev_get_by_name(&init_net, ifr->ifr_slave); - dprintk("slave_dev=%p: \n", slave_dev); + pr_debug("slave_dev=%p: \n", slave_dev); if (!slave_dev) { res = -ENODEV; } else { - dprintk("slave_dev->name=%s: \n", slave_dev->name); + pr_debug("slave_dev->name=%s: \n", slave_dev->name); switch (cmd) { case BOND_ENSLAVE_OLD: case SIOCBONDENSLAVE: @@ -4046,7 +4048,7 @@ static int bond_do_ioctl(struct net_device *bond_dev, struct ifreq *ifr, int cmd static void bond_set_multicast_list(struct net_device *bond_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct dev_mc_list *dmi; /* @@ -4102,17 +4104,31 @@ static void bond_set_multicast_list(struct net_device *bond_dev) read_unlock(&bond->lock); } +static int bond_neigh_setup(struct net_device *dev, struct neigh_parms *parms) +{ + struct bonding *bond = netdev_priv(dev); + struct slave *slave = bond->first_slave; + + if (slave) { + const struct net_device_ops *slave_ops + = slave->dev->netdev_ops; + if (slave_ops->ndo_neigh_setup) + return slave_ops->ndo_neigh_setup(dev, parms); + } + return 0; +} + /* * Change the MTU of all of a master's slaves to match the master */ static int bond_change_mtu(struct net_device *bond_dev, int new_mtu) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct slave *slave, *stop_at; int res = 0; int i; - dprintk("bond=%p, name=%s, new_mtu=%d\n", bond, + 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 @@ -4131,8 +4147,8 @@ static int bond_change_mtu(struct net_device *bond_dev, int new_mtu) */ bond_for_each_slave(bond, slave, i) { - dprintk("s %p s->p %p c_m %p\n", slave, - slave->prev, slave->dev->change_mtu); + pr_debug("s %p s->p %p c_m %p\n", slave, + slave->prev, slave->dev->netdev_ops->ndo_change_mtu); res = dev_set_mtu(slave->dev, new_mtu); @@ -4145,7 +4161,7 @@ static int bond_change_mtu(struct net_device *bond_dev, int new_mtu) * means changing their mtu from timer context, which * is probably not a good idea. */ - dprintk("err %d %s\n", res, slave->dev->name); + pr_debug("err %d %s\n", res, slave->dev->name); goto unwind; } } @@ -4162,7 +4178,7 @@ unwind: tmp_res = dev_set_mtu(slave->dev, bond_dev->mtu); if (tmp_res) { - dprintk("unwind err %d dev %s\n", tmp_res, + pr_debug("unwind err %d dev %s\n", tmp_res, slave->dev->name); } } @@ -4179,13 +4195,17 @@ unwind: */ static int bond_set_mac_address(struct net_device *bond_dev, void *addr) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct sockaddr *sa = addr, tmp_sa; struct slave *slave, *stop_at; int res = 0; int i; - dprintk("bond=%p, name=%s\n", bond, (bond_dev ? bond_dev->name : "None")); + if (bond->params.mode == BOND_MODE_ALB) + return bond_alb_set_mac_address(bond_dev, addr); + + + pr_debug("bond=%p, name=%s\n", bond, (bond_dev ? bond_dev->name : "None")); /* * If fail_over_mac is set to active, do nothing and return @@ -4214,11 +4234,12 @@ static int bond_set_mac_address(struct net_device *bond_dev, void *addr) */ bond_for_each_slave(bond, slave, i) { - dprintk("slave %p %s\n", slave, slave->dev->name); + const struct net_device_ops *slave_ops = slave->dev->netdev_ops; + pr_debug("slave %p %s\n", slave, slave->dev->name); - if (slave->dev->set_mac_address == NULL) { + if (slave_ops->ndo_set_mac_address == NULL) { res = -EOPNOTSUPP; - dprintk("EOPNOTSUPP %s\n", slave->dev->name); + pr_debug("EOPNOTSUPP %s\n", slave->dev->name); goto unwind; } @@ -4230,7 +4251,7 @@ static int bond_set_mac_address(struct net_device *bond_dev, void *addr) * breakage anyway until ARP finish * updating, so... */ - dprintk("err %d %s\n", res, slave->dev->name); + pr_debug("err %d %s\n", res, slave->dev->name); goto unwind; } } @@ -4250,7 +4271,7 @@ unwind: tmp_res = dev_set_mac_address(slave->dev, &tmp_sa); if (tmp_res) { - dprintk("unwind err %d dev %s\n", tmp_res, + pr_debug("unwind err %d dev %s\n", tmp_res, slave->dev->name); } } @@ -4260,7 +4281,7 @@ unwind: static int bond_xmit_roundrobin(struct sk_buff *skb, struct net_device *bond_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct slave *slave, *start_at; int i, slave_no, res = 1; @@ -4309,7 +4330,7 @@ out: */ static int bond_xmit_activebackup(struct sk_buff *skb, struct net_device *bond_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); int res = 1; read_lock(&bond->lock); @@ -4341,7 +4362,7 @@ out: */ static int bond_xmit_xor(struct sk_buff *skb, struct net_device *bond_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct slave *slave, *start_at; int slave_no; int i; @@ -4387,7 +4408,7 @@ out: */ static int bond_xmit_broadcast(struct sk_buff *skb, struct net_device *bond_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); struct slave *slave, *start_at; struct net_device *tx_dev = NULL; int i; @@ -4463,6 +4484,35 @@ static void bond_set_xmit_hash_policy(struct bonding *bond) } } +static int bond_start_xmit(struct sk_buff *skb, struct net_device *dev) +{ + const struct bonding *bond = netdev_priv(dev); + + switch (bond->params.mode) { + case BOND_MODE_ROUNDROBIN: + return bond_xmit_roundrobin(skb, dev); + case BOND_MODE_ACTIVEBACKUP: + return bond_xmit_activebackup(skb, dev); + case BOND_MODE_XOR: + return bond_xmit_xor(skb, dev); + case BOND_MODE_BROADCAST: + return bond_xmit_broadcast(skb, 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); + default: + /* Should never happen, mode already checked */ + printk(KERN_ERR DRV_NAME ": %s: Error: Unknown bonding mode %d\n", + dev->name, bond->params.mode); + WARN_ON_ONCE(1); + dev_kfree_skb(skb); + return NETDEV_TX_OK; + } +} + + /* * set bond mode specific net device operations */ @@ -4472,29 +4522,22 @@ void bond_set_mode_ops(struct bonding *bond, int mode) switch (mode) { case BOND_MODE_ROUNDROBIN: - bond_dev->hard_start_xmit = bond_xmit_roundrobin; break; case BOND_MODE_ACTIVEBACKUP: - bond_dev->hard_start_xmit = bond_xmit_activebackup; break; case BOND_MODE_XOR: - bond_dev->hard_start_xmit = bond_xmit_xor; bond_set_xmit_hash_policy(bond); break; case BOND_MODE_BROADCAST: - bond_dev->hard_start_xmit = bond_xmit_broadcast; break; case BOND_MODE_8023AD: bond_set_master_3ad_flags(bond); - bond_dev->hard_start_xmit = bond_3ad_xmit_xor; bond_set_xmit_hash_policy(bond); break; case BOND_MODE_ALB: bond_set_master_alb_flags(bond); /* FALLTHRU */ case BOND_MODE_TLB: - bond_dev->hard_start_xmit = bond_alb_xmit; - bond_dev->set_mac_address = bond_alb_set_mac_address; break; default: /* Should never happen, mode already checked */ @@ -4524,15 +4567,30 @@ static const struct ethtool_ops bond_ethtool_ops = { .get_flags = ethtool_op_get_flags, }; +static const struct net_device_ops bond_netdev_ops = { + .ndo_open = bond_open, + .ndo_stop = bond_close, + .ndo_start_xmit = bond_start_xmit, + .ndo_get_stats = bond_get_stats, + .ndo_do_ioctl = bond_do_ioctl, + .ndo_set_multicast_list = bond_set_multicast_list, + .ndo_change_mtu = bond_change_mtu, + .ndo_set_mac_address = bond_set_mac_address, + .ndo_neigh_setup = bond_neigh_setup, + .ndo_vlan_rx_register = bond_vlan_rx_register, + .ndo_vlan_rx_add_vid = bond_vlan_rx_add_vid, + .ndo_vlan_rx_kill_vid = bond_vlan_rx_kill_vid, +}; + /* * Does not allocate but creates a /proc entry. * Allowed to fail. */ static int bond_init(struct net_device *bond_dev, struct bond_params *params) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); - dprintk("Begin bond_init for %s\n", bond_dev->name); + pr_debug("Begin bond_init for %s\n", bond_dev->name); /* initialize rwlocks */ rwlock_init(&bond->lock); @@ -4551,20 +4609,13 @@ static int bond_init(struct net_device *bond_dev, struct bond_params *params) bond->primary_slave = NULL; bond->dev = bond_dev; bond->send_grat_arp = 0; + bond->send_unsol_na = 0; bond->setup_by_slave = 0; INIT_LIST_HEAD(&bond->vlan_list); /* Initialize the device entry points */ - bond_dev->open = bond_open; - bond_dev->stop = bond_close; - bond_dev->get_stats = bond_get_stats; - bond_dev->do_ioctl = bond_do_ioctl; + bond_dev->netdev_ops = &bond_netdev_ops; bond_dev->ethtool_ops = &bond_ethtool_ops; - bond_dev->set_multicast_list = bond_set_multicast_list; - bond_dev->change_mtu = bond_change_mtu; - bond_dev->set_mac_address = bond_set_mac_address; - bond_dev->validate_addr = NULL; - bond_set_mode_ops(bond, bond->params.mode); bond_dev->destructor = bond_destructor; @@ -4573,6 +4624,8 @@ static int bond_init(struct net_device *bond_dev, struct bond_params *params) bond_dev->tx_queue_len = 0; bond_dev->flags |= IFF_MASTER|IFF_MULTICAST; bond_dev->priv_flags |= IFF_BONDING; + if (bond->params.arp_interval) + bond_dev->priv_flags |= IFF_MASTER_ARPMON; /* At first, we block adding VLANs. That's the only way to * prevent problems that occur when adding VLANs over an @@ -4591,9 +4644,6 @@ static int bond_init(struct net_device *bond_dev, struct bond_params *params) * when there are slaves that are not hw accel * capable */ - bond_dev->vlan_rx_register = bond_vlan_rx_register; - bond_dev->vlan_rx_add_vid = bond_vlan_rx_add_vid; - bond_dev->vlan_rx_kill_vid = bond_vlan_rx_kill_vid; bond_dev->features |= (NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX | NETIF_F_HW_VLAN_FILTER); @@ -4632,7 +4682,7 @@ static void bond_work_cancel_all(struct bonding *bond) */ static void bond_deinit(struct net_device *bond_dev) { - struct bonding *bond = bond_dev->priv; + struct bonding *bond = netdev_priv(bond_dev); list_del(&bond->bond_list); @@ -4672,7 +4722,7 @@ static void bond_free_all(void) * 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, struct bond_parm_tbl *tbl) +int bond_parse_parm(const char *buf, const struct bond_parm_tbl *tbl) { int mode = -1, i, rv; char *p, modestr[BOND_MAX_MODENAME_LEN + 1] = { 0, }; @@ -4751,6 +4801,23 @@ static int bond_check_params(struct bond_params *params) } } + if (ad_select) { + params->ad_select = bond_parse_parm(ad_select, ad_select_tbl); + if (params->ad_select == -1) { + printk(KERN_ERR DRV_NAME + ": Error: Invalid ad_select \"%s\"\n", + ad_select == NULL ? "NULL" : ad_select); + return -EINVAL; + } + + if (bond_mode != BOND_MODE_8023AD) { + printk(KERN_WARNING DRV_NAME + ": ad_select param only affects 802.3ad mode\n"); + } + } else { + params->ad_select = BOND_AD_STABLE; + } + if (max_bonds < 0 || max_bonds > INT_MAX) { printk(KERN_WARNING DRV_NAME ": Warning: max_bonds (%d) not in range %d-%d, so it " @@ -4798,6 +4865,13 @@ static int bond_check_params(struct bond_params *params) num_grat_arp = 1; } + if (num_unsol_na < 0 || num_unsol_na > 255) { + printk(KERN_WARNING DRV_NAME + ": Warning: num_unsol_na (%d) not in range 0-255 so it " + "was reset to 1 \n", num_unsol_na); + num_unsol_na = 1; + } + /* reset values for 802.3ad */ if (bond_mode == BOND_MODE_8023AD) { if (!miimon) { @@ -4999,6 +5073,7 @@ static int bond_check_params(struct bond_params *params) params->xmit_policy = xmit_hashtype; params->miimon = miimon; params->num_grat_arp = num_grat_arp; + params->num_unsol_na = num_unsol_na; params->arp_interval = arp_interval; params->arp_validate = arp_validate_value; params->updelay = updelay; @@ -5099,7 +5174,7 @@ int bond_create(char *name, struct bond_params *params) up_write(&bonding_rwsem); rtnl_unlock(); /* allows sysfs registration of net device */ - res = bond_create_sysfs_entry(bond_dev->priv); + res = bond_create_sysfs_entry(netdev_priv(bond_dev)); if (res < 0) { rtnl_lock(); down_write(&bonding_rwsem); @@ -5151,6 +5226,7 @@ static int __init bonding_init(void) register_netdevice_notifier(&bond_netdev_notifier); register_inetaddr_notifier(&bond_inetaddr_notifier); + bond_register_ipv6_notifier(); goto out; err: @@ -5173,6 +5249,7 @@ static void __exit bonding_exit(void) { unregister_netdevice_notifier(&bond_netdev_notifier); unregister_inetaddr_notifier(&bond_inetaddr_notifier); + bond_unregister_ipv6_notifier(); bond_destroy_sysfs(); diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c index 3bdb4738252..18cf4787874 100644 --- a/drivers/net/bonding/bond_sysfs.c +++ b/drivers/net/bonding/bond_sysfs.c @@ -36,22 +36,13 @@ #include <linux/rtnetlink.h> #include <net/net_namespace.h> -/* #define BONDING_DEBUG 1 */ #include "bonding.h" + #define to_dev(obj) container_of(obj,struct device,kobj) -#define to_bond(cd) ((struct bonding *)(to_net_dev(cd)->priv)) +#define to_bond(cd) ((struct bonding *)(netdev_priv(to_net_dev(cd)))) /*---------------------------- Declarations -------------------------------*/ - -extern struct list_head bond_dev_list; -extern struct bond_params bonding_defaults; -extern struct bond_parm_tbl bond_mode_tbl[]; -extern struct bond_parm_tbl bond_lacp_tbl[]; -extern struct bond_parm_tbl xmit_hashtype_tbl[]; -extern struct bond_parm_tbl arp_validate_tbl[]; -extern struct bond_parm_tbl fail_over_mac_tbl[]; - static int expected_refcount = -1; /*--------------------------- Data Structures -----------------------------*/ @@ -316,18 +307,12 @@ static ssize_t bonding_store_slaves(struct device *d, /* Set the slave's MTU to match the bond */ original_mtu = dev->mtu; - if (dev->mtu != bond->dev->mtu) { - if (dev->change_mtu) { - res = dev->change_mtu(dev, - bond->dev->mtu); - if (res) { - ret = res; - goto out; - } - } else { - dev->mtu = bond->dev->mtu; - } + res = dev_set_mtu(dev, bond->dev->mtu); + if (res) { + ret = res; + goto out; } + res = bond_enslave(bond->dev, dev); bond_for_each_slave(bond, slave, i) if (strnicmp(slave->dev->name, ifname, IFNAMSIZ) == 0) @@ -356,11 +341,7 @@ static ssize_t bonding_store_slaves(struct device *d, goto out; } /* set the slave MTU to the default */ - if (dev->change_mtu) { - dev->change_mtu(dev, original_mtu); - } else { - dev->mtu = original_mtu; - } + dev_set_mtu(dev, original_mtu); } else { printk(KERN_ERR DRV_NAME ": unable to remove non-existent slave %s for bond %s.\n", @@ -620,6 +601,8 @@ static ssize_t bonding_store_arp_interval(struct device *d, ": %s: Setting ARP monitoring interval to %d.\n", bond->dev->name, new_value); bond->params.arp_interval = new_value; + if (bond->params.arp_interval) + bond->dev->priv_flags |= IFF_MASTER_ARPMON; if (bond->params.miimon) { printk(KERN_INFO DRV_NAME ": %s: ARP monitoring cannot be used with MII monitoring. " @@ -672,8 +655,8 @@ static ssize_t bonding_show_arp_targets(struct device *d, for (i = 0; i < BOND_MAX_ARP_TARGETS; i++) { if (bond->params.arp_targets[i]) - res += sprintf(buf + res, "%u.%u.%u.%u ", - NIPQUAD(bond->params.arp_targets[i])); + res += sprintf(buf + res, "%pI4 ", + &bond->params.arp_targets[i]); } if (res) buf[res-1] = '\n'; /* eat the leftover space */ @@ -695,8 +678,8 @@ static ssize_t bonding_store_arp_targets(struct device *d, if (buf[0] == '+') { if ((newtarget == 0) || (newtarget == htonl(INADDR_BROADCAST))) { printk(KERN_ERR DRV_NAME - ": %s: invalid ARP target %u.%u.%u.%u specified for addition\n", - bond->dev->name, NIPQUAD(newtarget)); + ": %s: invalid ARP target %pI4 specified for addition\n", + bond->dev->name, &newtarget); ret = -EINVAL; goto out; } @@ -704,8 +687,8 @@ static ssize_t bonding_store_arp_targets(struct device *d, for (i = 0; (i < BOND_MAX_ARP_TARGETS); i++) { if (targets[i] == newtarget) { /* duplicate */ printk(KERN_ERR DRV_NAME - ": %s: ARP target %u.%u.%u.%u is already present\n", - bond->dev->name, NIPQUAD(newtarget)); + ": %s: ARP target %pI4 is already present\n", + bond->dev->name, &newtarget); if (done) targets[i] = 0; ret = -EINVAL; @@ -713,8 +696,8 @@ static ssize_t bonding_store_arp_targets(struct device *d, } if (targets[i] == 0 && !done) { printk(KERN_INFO DRV_NAME - ": %s: adding ARP target %d.%d.%d.%d.\n", - bond->dev->name, NIPQUAD(newtarget)); + ": %s: adding ARP target %pI4.\n", + bond->dev->name, &newtarget); done = 1; targets[i] = newtarget; } @@ -731,8 +714,8 @@ static ssize_t bonding_store_arp_targets(struct device *d, else if (buf[0] == '-') { if ((newtarget == 0) || (newtarget == htonl(INADDR_BROADCAST))) { printk(KERN_ERR DRV_NAME - ": %s: invalid ARP target %d.%d.%d.%d specified for removal\n", - bond->dev->name, NIPQUAD(newtarget)); + ": %s: invalid ARP target %pI4 specified for removal\n", + bond->dev->name, &newtarget); ret = -EINVAL; goto out; } @@ -740,16 +723,16 @@ static ssize_t bonding_store_arp_targets(struct device *d, for (i = 0; (i < BOND_MAX_ARP_TARGETS); i++) { if (targets[i] == newtarget) { printk(KERN_INFO DRV_NAME - ": %s: removing ARP target %d.%d.%d.%d.\n", - bond->dev->name, NIPQUAD(newtarget)); + ": %s: removing ARP target %pI4.\n", + bond->dev->name, &newtarget); targets[i] = 0; done = 1; } } if (!done) { printk(KERN_INFO DRV_NAME - ": %s: unable to remove nonexistent ARP target %d.%d.%d.%d.\n", - bond->dev->name, NIPQUAD(newtarget)); + ": %s: unable to remove nonexistent ARP target %pI4.\n", + bond->dev->name, &newtarget); ret = -EINVAL; goto out; } @@ -942,6 +925,53 @@ out: } static DEVICE_ATTR(lacp_rate, S_IRUGO | S_IWUSR, bonding_show_lacp, bonding_store_lacp); +static ssize_t bonding_show_ad_select(struct device *d, + struct device_attribute *attr, + char *buf) +{ + struct bonding *bond = to_bond(d); + + 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 = count; + struct bonding *bond = to_bond(d); + + if (bond->dev->flags & IFF_UP) { + printk(KERN_ERR DRV_NAME + ": %s: Unable to update ad_select because interface " + "is up.\n", bond->dev->name); + ret = -EPERM; + goto out; + } + + new_value = bond_parse_parm(buf, ad_select_tbl); + + if (new_value != -1) { + bond->params.ad_select = new_value; + printk(KERN_INFO DRV_NAME + ": %s: Setting ad_select to %s (%d).\n", + bond->dev->name, ad_select_tbl[new_value].modename, + new_value); + } else { + printk(KERN_ERR DRV_NAME + ": %s: Ignoring invalid ad_select value %.*s.\n", + bond->dev->name, (int)strlen(buf) - 1, buf); + ret = -EINVAL; + } +out: + return ret; +} + +static DEVICE_ATTR(ad_select, S_IRUGO | S_IWUSR, bonding_show_ad_select, bonding_store_ad_select); + /* * Show and set the number of grat ARP to send after a failover event. */ @@ -981,6 +1011,47 @@ out: return ret; } static DEVICE_ATTR(num_grat_arp, S_IRUGO | S_IWUSR, bonding_show_n_grat_arp, bonding_store_n_grat_arp); + +/* + * Show and set the number of unsolicted NA's to send after a failover event. + */ +static ssize_t bonding_show_n_unsol_na(struct device *d, + struct device_attribute *attr, + char *buf) +{ + struct bonding *bond = to_bond(d); + + return sprintf(buf, "%d\n", bond->params.num_unsol_na); +} + +static ssize_t bonding_store_n_unsol_na(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) { + printk(KERN_ERR DRV_NAME + ": %s: no num_unsol_na value specified.\n", + bond->dev->name); + ret = -EINVAL; + goto out; + } + if (new_value < 0 || new_value > 255) { + printk(KERN_ERR DRV_NAME + ": %s: Invalid num_unsol_na value %d not in range 0-255; rejected.\n", + bond->dev->name, new_value); + ret = -EINVAL; + goto out; + } else { + bond->params.num_unsol_na = new_value; + } +out: + return ret; +} +static DEVICE_ATTR(num_unsol_na, S_IRUGO | S_IWUSR, bonding_show_n_unsol_na, bonding_store_n_unsol_na); + /* * Show and set the MII monitor interval. There are two tricky bits * here. First, if MII monitoring is activated, then we must disable @@ -1039,6 +1110,7 @@ static ssize_t bonding_store_miimon(struct device *d, "ARP monitoring. Disabling ARP monitoring...\n", bond->dev->name); bond->params.arp_interval = 0; + bond->dev->priv_flags &= ~IFF_MASTER_ARPMON; if (bond->params.arp_validate) { bond_unregister_arp(bond); bond->params.arp_validate = @@ -1391,13 +1463,11 @@ static ssize_t bonding_show_ad_partner_mac(struct device *d, { int count = 0; struct bonding *bond = to_bond(d); - DECLARE_MAC_BUF(mac); if (bond->params.mode == BOND_MODE_8023AD) { struct ad_info ad_info; if (!bond_3ad_get_active_agg_info(bond, &ad_info)) { - count = sprintf(buf,"%s\n", - print_mac(mac, ad_info.partner_system)); + count = sprintf(buf, "%pM\n", ad_info.partner_system); } } @@ -1417,8 +1487,10 @@ static struct attribute *per_bond_attrs[] = { &dev_attr_downdelay.attr, &dev_attr_updelay.attr, &dev_attr_lacp_rate.attr, + &dev_attr_ad_select.attr, &dev_attr_xmit_hash_policy.attr, &dev_attr_num_grat_arp.attr, + &dev_attr_num_unsol_na.attr, &dev_attr_miimon.attr, &dev_attr_primary.attr, &dev_attr_use_carrier.attr, diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h index ffb668dd6d3..ca849d2adf9 100644 --- a/drivers/net/bonding/bonding.h +++ b/drivers/net/bonding/bonding.h @@ -19,23 +19,18 @@ #include <linux/proc_fs.h> #include <linux/if_bonding.h> #include <linux/kobject.h> +#include <linux/in6.h> #include "bond_3ad.h" #include "bond_alb.h" -#define DRV_VERSION "3.3.0" -#define DRV_RELDATE "June 10, 2008" +#define DRV_VERSION "3.5.0" +#define DRV_RELDATE "November 4, 2008" #define DRV_NAME "bonding" #define DRV_DESCRIPTION "Ethernet Channel Bonding Driver" #define BOND_MAX_ARP_TARGETS 16 -#ifdef BONDING_DEBUG -#define dprintk(fmt, args...) \ - printk(KERN_DEBUG \ - DRV_NAME ": %s() %d: " fmt, __func__, __LINE__ , ## args ) -#else -#define dprintk(fmt, args...) -#endif /* BONDING_DEBUG */ +extern struct list_head bond_dev_list; #define IS_UP(dev) \ ((((dev)->flags & IFF_UP) == IFF_UP) && \ @@ -126,6 +121,7 @@ struct bond_params { int xmit_policy; int miimon; int num_grat_arp; + int num_unsol_na; int arp_interval; int arp_validate; int use_carrier; @@ -133,6 +129,7 @@ struct bond_params { int updelay; int downdelay; int lacp_fast; + int ad_select; char primary[IFNAMSIZ]; __be32 arp_targets[BOND_MAX_ARP_TARGETS]; }; @@ -148,6 +145,9 @@ struct vlan_entry { struct list_head vlan_list; __be32 vlan_ip; unsigned short vlan_id; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + struct in6_addr vlan_ipv6; +#endif }; struct slave { @@ -195,6 +195,7 @@ struct bonding { rwlock_t curr_slave_lock; s8 kill_timers; s8 send_grat_arp; + s8 send_unsol_na; s8 setup_by_slave; struct net_device_stats stats; #ifdef CONFIG_PROC_FS @@ -218,6 +219,9 @@ struct bonding { struct delayed_work arp_work; struct delayed_work alb_work; struct delayed_work ad_work; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + struct in6_addr master_ipv6; +#endif }; /** @@ -245,7 +249,13 @@ static inline struct bonding *bond_get_bond_by_slave(struct slave *slave) return NULL; } - return (struct bonding *)slave->dev->master->priv; + return (struct bonding *)netdev_priv(slave->dev->master); +} + +static inline bool bond_is_lb(const struct bonding *bond) +{ + return bond->params.mode == BOND_MODE_TLB + || bond->params.mode == BOND_MODE_ALB; } #define BOND_FOM_NONE 0 @@ -275,7 +285,7 @@ static inline unsigned long slave_last_rx(struct bonding *bond, static inline void bond_set_slave_inactive_flags(struct slave *slave) { - struct bonding *bond = slave->dev->master->priv; + struct bonding *bond = netdev_priv(slave->dev->master); if (bond->params.mode != BOND_MODE_TLB && bond->params.mode != BOND_MODE_ALB) slave->state = BOND_STATE_BACKUP; @@ -327,7 +337,7 @@ void bond_mii_monitor(struct work_struct *); void bond_loadbalance_arp_mon(struct work_struct *); void bond_activebackup_arp_mon(struct work_struct *); void bond_set_mode_ops(struct bonding *bond, int mode); -int bond_parse_parm(const char *mode_arg, struct bond_parm_tbl *tbl); +int bond_parse_parm(const char *mode_arg, const struct bond_parm_tbl *tbl); void bond_select_active_slave(struct bonding *bond); void bond_change_active_slave(struct bonding *bond, struct slave *new_active); void bond_register_arp(struct bonding *); @@ -335,11 +345,35 @@ void bond_unregister_arp(struct bonding *); /* exported from bond_main.c */ extern struct list_head bond_dev_list; -extern struct bond_parm_tbl bond_lacp_tbl[]; -extern struct bond_parm_tbl bond_mode_tbl[]; -extern struct bond_parm_tbl xmit_hashtype_tbl[]; -extern struct bond_parm_tbl arp_validate_tbl[]; -extern struct bond_parm_tbl fail_over_mac_tbl[]; +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 fail_over_mac_tbl[]; +extern struct bond_params bonding_defaults; +extern struct bond_parm_tbl ad_select_tbl[]; + +/* exported from bond_sysfs.c */ +extern struct rw_semaphore bonding_rwsem; + +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) +void bond_send_unsolicited_na(struct bonding *bond); +void bond_register_ipv6_notifier(void); +void bond_unregister_ipv6_notifier(void); +#else +static inline void bond_send_unsolicited_na(struct bonding *bond) +{ + return; +} +static inline void bond_register_ipv6_notifier(void) +{ + return; +} +static inline void bond_unregister_ipv6_notifier(void) +{ + return; +} +#endif #endif /* _LINUX_BONDING_H */ |