From: Andy Gospodarek <gospo@redhat.com>
Date: Thu, 23 Apr 2009 14:38:23 -0400
Subject: [net] bonding: update to upstream version 3.4.0
Message-id: 20090423183811.GA26634@gospo.rdu.redhat.com
O-Subject: [RHEL5.4 PATCH 1/2] bonding: update to upstream version 3.4.0
Bugzilla: 462632
RH-Acked-by: John W. Linville <linville@redhat.com>
RH-Acked-by: Neil Horman <nhorman@redhat.com>
RH-Acked-by: Prarit Bhargava <prarit@redhat.com>
One of the main reasons to move this far forward with the bonding driver
is to add this patch:
commit 305d552accae6afb859c493ebc7d98ca3371dae2
Author: Brian Haley <brian.haley@hp.com>
Date: Tue Nov 4 17:51:14 2008 -0800
bonding: send IPv6 neighbor advertisement on failover
There several critical fixes that I felt we should include, so instead
of simply cherry-picking and hacking the above commit, I did a full
backport and included all the changes since the 3.2.4 version of the
driver.
One of the major cleanups done in this patch is the inclusion of this
patch:
commit b63365a2d60268a3988285d6c3c6003d7066f93a
Author: Herbert Xu <herbert@gondor.apana.org.au>
Date: Thu Oct 23 01:11:29 2008 -0700
net: Fix disjunct computation of netdev features
It was needed to correct changes that we took previously so that
checksumming values for bridges and bonds are set correctly based on the
capabilities of all the devices in the bond/bridge. So far testing
looks good on this, but I need to do a bit more before I'm convinced.
I have been carrying most of this patch in my test kernels for quite a
while and received positive feedback. I have also gotten positive
feedback on this specific patch and it's ability to resolve the ipv6
issues.
This backport will resolve RHBZ 462632 and RHBZ 475523.
diff --git a/Documentation/networking/bonding.txt b/Documentation/networking/bonding.txt
index 2c9a7b2..772074b 100644
--- a/Documentation/networking/bonding.txt
+++ b/Documentation/networking/bonding.txt
@@ -338,7 +338,8 @@ max_bonds
Specifies the number of bonding devices to create for this
instance of the bonding driver. E.g., if max_bonds is 3, and
the bonding driver is not already loaded, then bond0, bond1
- and bond2 will be created. The default value is 1.
+ and bond2 will be created. The default value is 1. Specifying
+ a value of 0 will load bonding, but will not create any devices.
miimon
@@ -501,6 +502,16 @@ mode
swapped with the new curr_active_slave that was
chosen.
+num_unsol_na
+
+ Specifies the number of unsolicited IPv6 Neighbor Advertisements
+ to be issued after a failover event. One unsolicited NA is issued
+ immediately after the failover.
+
+ The valid range is 0 - 255; the default value is 1. This option
+ affects only the active-backup mode. This option was added for
+ bonding version 3.4.0.
+
primary
A string (eth0, eth2, etc) specifying which slave is the
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index a4e94ad..786902e 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -60,6 +60,7 @@ config DUMMY
config BONDING
tristate "Bonding driver support"
depends on INET
+ depends on IPV6 || IPV6=n
---help---
Say 'Y' or 'M' if you wish to be able to 'bond' multiple Ethernet
Channels together. This is called 'Etherchannel' by Cisco,
diff --git a/drivers/net/bonding/Makefile b/drivers/net/bonding/Makefile
index 5cdae2b..6f9c6fa 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 e437915..df81cfe 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -2105,6 +2105,7 @@ void bond_3ad_state_machine_handler(void *work_data)
aggregator = __get_first_agg(port);
ad_agg_selection_logic(aggregator);
}
+ bond_3ad_set_carrier(bond);
}
// for each port run the state machines
diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
index 99ef35b..61be99b 100644
--- a/drivers/net/bonding/bond_alb.c
+++ b/drivers/net/bonding/bond_alb.c
@@ -38,6 +38,7 @@
#include <linux/in.h>
#include <net/ipx.h>
#include <net/arp.h>
+#include <net/ipv6.h>
#include <asm/byteorder.h>
#include "bonding.h"
#include "bond_alb.h"
@@ -81,6 +82,7 @@
#define RLB_PROMISC_TIMEOUT 10*ALB_TIMER_TICKS_PER_SEC
static const u8 mac_bcast[ETH_ALEN] = {0xff,0xff,0xff,0xff,0xff,0xff};
+static const u8 mac_v6_allmcast[ETH_ALEN] = {0x33,0x33,0x00,0x00,0x00,0x01};
static const int alb_delta_in_ticks = HZ / ALB_TIMER_TICKS_PER_SEC;
#pragma pack(1)
@@ -712,7 +714,7 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond)
struct arp_pkt *arp = arp_pkt(skb);
struct slave *tx_slave = NULL;
- if (arp->op_code == __constant_htons(ARPOP_REPLY)) {
+ if (arp->op_code == htons(ARPOP_REPLY)) {
/* the arp must be sent on the selected
* rx channel
*/
@@ -721,7 +723,7 @@ static struct slave *rlb_arp_xmit(struct sk_buff *skb, struct bonding *bond)
memcpy(arp->mac_src,tx_slave->dev->dev_addr, ETH_ALEN);
}
dprintk("Server sent ARP Reply packet\n");
- } else if (arp->op_code == __constant_htons(ARPOP_REQUEST)) {
+ } else if (arp->op_code == htons(ARPOP_REQUEST)) {
/* Create an entry in the rx_hashtbl for this client as a
* place holder.
* When the arp reply is received the entry will be updated
@@ -1292,6 +1294,7 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
u32 hash_index = 0;
const u8 *hash_start = NULL;
int res = 1;
+ struct ipv6hdr *ip6hdr;
ip_bcast = htonl(0xffffffff);
skb_reset_mac_header(skb);
@@ -1322,11 +1325,32 @@ int bond_alb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
}
break;
case ETH_P_IPV6:
+ /* IPv6 doesn't really use broadcast mac address, but leave
+ * that here just in case.
+ */
if (memcmp(eth_data->h_dest, mac_bcast, ETH_ALEN) == 0) {
do_tx_balance = 0;
break;
}
+ /* IPv6 uses all-nodes multicast as an equivalent to
+ * broadcasts in IPv4.
+ */
+ if (memcmp(eth_data->h_dest, mac_v6_allmcast, ETH_ALEN) == 0) {
+ do_tx_balance = 0;
+ break;
+ }
+
+ /* Additianally, DAD probes should not be tx-balanced as that
+ * will lead to false positives for duplicate addresses and
+ * prevent address configuration from working.
+ */
+ ip6hdr = ipv6_hdr(skb);
+ if (ipv6_addr_any(&ip6hdr->saddr)) {
+ do_tx_balance = 0;
+ break;
+ }
+
hash_start = (char *)&(ipv6_hdr(skb)->daddr);
hash_size = sizeof(ipv6_hdr(skb)->daddr);
break;
diff --git a/drivers/net/bonding/bond_ipv6.c b/drivers/net/bonding/bond_ipv6.c
new file mode 100644
index 0000000..79e7016
--- /dev/null
+++ b/drivers/net/bonding/bond_ipv6.c
@@ -0,0 +1,221 @@
+/*
+ * 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.
+ *
+ */
+
+//#define BONDING_DEBUG 1
+
+#include <linux/types.h>
+#include <linux/if_vlan.h>
+#include <net/ipv6.h>
+#include <net/ndisc.h>
+#include <net/addrconf.h>
+#include "bonding.h"
+
+static inline struct net_device *vlan_group_get_device(struct vlan_group *vg,
+ int vlan_id)
+{
+ return vg->vlan_devices[vlan_id];
+}
+
+/*
+ * 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);
+
+ dprintk("ipv6 na on slave %s: dest %pI6, src %pI6\n",
+ slave->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;
+
+ dprintk("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;
+
+ 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 4426166..089e658 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -88,6 +88,8 @@
#define BOND_LINK_ARP_INTERV 0
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;
@@ -99,11 +101,15 @@ static char *xmit_hash_policy = NULL;
static int arp_interval = BOND_LINK_ARP_INTERV;
static char *arp_ip_target[BOND_MAX_ARP_TARGETS] = { NULL, };
static char *arp_validate = NULL;
-static int fail_over_mac = 0;
+static char *fail_over_mac = NULL;
struct bond_params bonding_defaults;
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);
@@ -133,8 +139,8 @@ module_param_array(arp_ip_target, charp, NULL, 0);
MODULE_PARM_DESC(arp_ip_target, "arp targets in n.n.n.n form");
module_param(arp_validate, charp, 0);
MODULE_PARM_DESC(arp_validate, "validate src/dst of ARP probes: none (default), active, backup or all");
-module_param(fail_over_mac, int, 0);
-MODULE_PARM_DESC(fail_over_mac, "For active-backup, do not set all slaves to the same MAC. 0 of off (default), 1 for on.");
+module_param(fail_over_mac, charp, 0);
+MODULE_PARM_DESC(fail_over_mac, "For active-backup, do not set all slaves to the same MAC. none (default), active or follow");
/*----------------------------- Global variables ----------------------------*/
@@ -187,6 +193,13 @@ struct bond_parm_tbl arp_validate_tbl[] = {
{ NULL, -1},
};
+struct bond_parm_tbl fail_over_mac_tbl[] = {
+{ "none", BOND_FOM_NONE},
+{ "active", BOND_FOM_ACTIVE},
+{ "follow", BOND_FOM_FOLLOW},
+{ NULL, -1},
+};
+
/*-------------------------- Forward declarations ---------------------------*/
static void bond_send_gratuitous_arp(struct bonding *bond);
@@ -232,14 +245,13 @@ static int bond_add_vlan(struct bonding *bond, unsigned short vlan_id)
dprintk("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);
@@ -261,14 +273,14 @@ static int bond_add_vlan(struct bonding *bond, unsigned short vlan_id)
*/
static int bond_del_vlan(struct bonding *bond, unsigned short vlan_id)
{
- struct vlan_entry *vlan, *next;
+ struct vlan_entry *vlan;
int res = -ENODEV;
dprintk("bond: %s, vlan id %d\n", bond->dev->name, vlan_id);
write_lock_bh(&bond->lock);
- list_for_each_entry_safe(vlan, next, &bond->vlan_list, vlan_list) {
+ list_for_each_entry(vlan, &bond->vlan_list, vlan_list) {
if (vlan->vlan_id == vlan_id) {
list_del(&vlan->vlan_list);
@@ -1002,6 +1014,82 @@ static void bond_mc_swap(struct bonding *bond, struct slave *new_active, struct
}
}
+/*
+ * bond_do_fail_over_mac
+ *
+ * Perform special MAC address swapping for fail_over_mac settings
+ *
+ * Called with RTNL, bond->lock for read, curr_slave_lock for write_bh.
+ */
+static void bond_do_fail_over_mac(struct bonding *bond,
+ struct slave *new_active,
+ struct slave *old_active)
+{
+ u8 tmp_mac[ETH_ALEN];
+ struct sockaddr saddr;
+ int rv;
+
+ switch (bond->params.fail_over_mac) {
+ case BOND_FOM_ACTIVE:
+ if (new_active)
+ memcpy(bond->dev->dev_addr, new_active->dev->dev_addr,
+ new_active->dev->addr_len);
+ break;
+ case BOND_FOM_FOLLOW:
+ /*
+ * if new_active && old_active, swap them
+ * if just old_active, do nothing (going to no active slave)
+ * if just new_active, set new_active to bond's MAC
+ */
+ if (!new_active)
+ return;
+
+ write_unlock_bh(&bond->curr_slave_lock);
+ read_unlock(&bond->lock);
+
+ if (old_active) {
+ memcpy(tmp_mac, new_active->dev->dev_addr, ETH_ALEN);
+ memcpy(saddr.sa_data, old_active->dev->dev_addr,
+ ETH_ALEN);
+ saddr.sa_family = new_active->dev->type;
+ } else {
+ memcpy(saddr.sa_data, bond->dev->dev_addr, ETH_ALEN);
+ saddr.sa_family = bond->dev->type;
+ }
+
+ rv = dev_set_mac_address(new_active->dev, &saddr);
+ if (rv) {
+ printk(KERN_ERR DRV_NAME
+ ": %s: Error %d setting MAC of slave %s\n",
+ bond->dev->name, -rv, new_active->dev->name);
+ goto out;
+ }
+
+ if (!old_active)
+ goto out;
+
+ memcpy(saddr.sa_data, tmp_mac, ETH_ALEN);
+ saddr.sa_family = old_active->dev->type;
+
+ rv = dev_set_mac_address(old_active->dev, &saddr);
+ if (rv)
+ printk(KERN_ERR DRV_NAME
+ ": %s: Error %d setting MAC of slave %s\n",
+ bond->dev->name, -rv, new_active->dev->name);
+out:
+ read_lock(&bond->lock);
+ write_lock_bh(&bond->curr_slave_lock);
+ break;
+ default:
+ printk(KERN_ERR DRV_NAME
+ ": %s: bond_do_fail_over_mac impossible: bad policy %d\n",
+ bond->dev->name, bond->params.fail_over_mac);
+ break;
+ }
+
+}
+
+
/**
* find_best_interface - select the best available slave to be the active one
* @bond: our bonding struct
@@ -1069,7 +1157,8 @@ static struct slave *bond_find_best_slave(struct bonding *bond)
* because it is apparently the best available slave we have, even though its
* updelay hasn't timed out yet.
*
- * Warning: Caller must hold curr_slave_lock for writing.
+ * If new_active is not NULL, caller must hold bond->lock for read and
+ * curr_slave_lock for write_bh.
*/
void bond_change_active_slave(struct bonding *bond, struct slave *new_active)
{
@@ -1080,6 +1169,8 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active)
}
if (new_active) {
+ new_active->jiffies = jiffies;
+
if (new_active->link == BOND_LINK_BACK) {
if (USES_PRIMARY(bond->params.mode)) {
printk(KERN_INFO DRV_NAME
@@ -1091,7 +1182,6 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active)
new_active->delay = 0;
new_active->link = BOND_LINK_UP;
- new_active->jiffies = jiffies;
if (bond->params.mode == BOND_MODE_8023AD) {
bond_3ad_handle_link_change(new_active, BOND_LINK_UP);
@@ -1133,22 +1223,17 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active)
if (new_active) {
bond_set_slave_active_flags(new_active);
- }
- /* when bonding does not set the slave MAC address, the bond MAC
- * address is the one of the active slave.
- */
- if (new_active && bond->params.fail_over_mac)
- memcpy(bond->dev->dev_addr, new_active->dev->dev_addr,
- new_active->dev->addr_len);
- if (bond->curr_active_slave &&
- test_bit(__LINK_STATE_LINKWATCH_PENDING,
- &bond->curr_active_slave->dev->state)) {
- dprintk("delaying gratuitous arp on %s\n",
- bond->curr_active_slave->dev->name);
- bond->send_grat_arp = 1;
- } else
+ if (bond->params.fail_over_mac)
+ bond_do_fail_over_mac(bond, new_active,
+ old_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);
+ }
}
}
@@ -1161,7 +1246,7 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active)
* - The primary_slave has got its link back.
* - A slave has got its link back and there's no old curr_active_slave.
*
- * Warning: Caller must hold curr_slave_lock for writing.
+ * Caller must hold bond->lock for read and curr_slave_lock for write_bh.
*/
void bond_select_active_slave(struct bonding *bond)
{
@@ -1273,18 +1358,24 @@ static int bond_compute_features(struct bonding *bond)
int i;
features &= ~(NETIF_F_ALL_CSUM | BOND_VLAN_FEATURES);
- features |= NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_HIGHDMA |
- NETIF_F_GSO_MASK | NETIF_F_NO_CSUM;
+ features |= NETIF_F_GSO_MASK | NETIF_F_NO_CSUM;
+
+ if (!bond->first_slave)
+ goto done;
+
+ features &= ~NETIF_F_ONE_FOR_ALL;
bond_for_each_slave(bond, slave, i) {
- features = netdev_compute_features(features,
- slave->dev->features);
+ features = netdev_increment_features(features,
+ slave->dev->features,
+ NETIF_F_ONE_FOR_ALL);
if (slave->dev->hard_header_len > max_hard_header_len)
max_hard_header_len = slave->dev->hard_header_len;
}
+done:
features |= (bond_dev->features & BOND_VLAN_FEATURES);
- bond_dev->features = features;
+ bond_dev->features = netdev_fix_features(features, NULL);
bond_dev->hard_header_len = max_hard_header_len;
return 0;
@@ -1369,14 +1460,14 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
printk(KERN_WARNING DRV_NAME
": %s: Warning: The first slave device "
"specified does not support setting the MAC "
- "address. Enabling the fail_over_mac option.",
+ "address. Setting fail_over_mac to active.",
bond_dev->name);
- bond->params.fail_over_mac = 1;
- } else if (!bond->params.fail_over_mac) {
+ bond->params.fail_over_mac = BOND_FOM_ACTIVE;
+ } else if (bond->params.fail_over_mac != BOND_FOM_ACTIVE) {
printk(KERN_ERR DRV_NAME
": %s: Error: The slave device specified "
"does not support setting the MAC address, "
- "but fail_over_mac is not enabled.\n"
+ "but fail_over_mac is not set to active.\n"
, bond_dev->name);
res = -EOPNOTSUPP;
goto err_undo_flags;
@@ -1425,13 +1516,13 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
res = netdev_set_master(slave_dev, bond_dev);
if (res) {
dprintk("Error %d calling netdev_set_master\n", res);
- goto err_close;
+ 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);
- goto err_restore_mac;
+ goto err_unset_master;
}
new_slave->dev = slave_dev;
@@ -1444,7 +1535,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
*/
res = bond_alb_init_slave(bond, new_slave);
if (res) {
- goto err_unset_master;
+ goto err_close;
}
}
@@ -1490,6 +1581,10 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
bond_compute_features(bond);
+ write_unlock_bh(&bond->lock);
+
+ read_lock(&bond->lock);
+
new_slave->last_arp_rx = jiffies;
if (bond->params.miimon && !bond->params.use_carrier) {
@@ -1566,6 +1661,8 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
}
}
+ write_lock_bh(&bond->curr_slave_lock);
+
switch (bond->params.mode) {
case BOND_MODE_ACTIVEBACKUP:
bond_set_slave_inactive_flags(new_slave);
@@ -1613,13 +1710,15 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
break;
} /* switch(bond_mode) */
+ write_unlock_bh(&bond->curr_slave_lock);
+
bond_set_carrier(bond);
- write_unlock_bh(&bond->lock);
+ read_unlock(&bond->lock);
res = bond_create_slave_symlinks(bond_dev, slave_dev);
if (res)
- goto err_unset_master;
+ goto err_close;
printk(KERN_INFO DRV_NAME
": %s: enslaving %s as a%s interface with a%s link.\n",
@@ -1631,14 +1730,18 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
return 0;
/* Undo stages on error */
-err_unset_master:
- netdev_set_master(slave_dev, NULL);
-
err_close:
dev_close(slave_dev);
+err_unset_master:
+ netdev_set_master(slave_dev, NULL);
+
err_restore_mac:
if (!bond->params.fail_over_mac) {
+ /* XXX TODO - fom follow mode needs to change master's
+ * MAC if this slave's MAC is in use by the bond, or at
+ * least print a warning.
+ */
memcpy(addr.sa_data, new_slave->perm_hwaddr, ETH_ALEN);
addr.sa_family = slave_dev->type;
dev_set_mac_address(slave_dev, &addr);
@@ -1692,26 +1795,24 @@ int bond_release(struct net_device *bond_dev, struct net_device *slave_dev)
return -EINVAL;
}
- mac_addr_differ = memcmp(bond_dev->dev_addr,
- slave->perm_hwaddr,
- ETH_ALEN);
- if (!mac_addr_differ && (bond->slave_cnt > 1)) {
- printk(KERN_WARNING DRV_NAME
- ": %s: Warning: the permanent HWaddr of %s "
- "- %02X:%02X:%02X:%02X:%02X:%02X - is "
- "still in use by %s. Set the HWaddr of "
- "%s to a different address to avoid "
- "conflicts.\n",
- bond_dev->name,
- slave_dev->name,
- slave->perm_hwaddr[0],
- slave->perm_hwaddr[1],
- slave->perm_hwaddr[2],
- slave->perm_hwaddr[3],
- slave->perm_hwaddr[4],
- slave->perm_hwaddr[5],
- bond_dev->name,
- slave_dev->name);
+ if (!bond->params.fail_over_mac) {
+ mac_addr_differ = memcmp(bond_dev->dev_addr, slave->perm_hwaddr,
+ ETH_ALEN);
+ if (!mac_addr_differ && (bond->slave_cnt > 1))
+ printk(KERN_WARNING DRV_NAME
+ ": %s: Warning: the permanent HWaddr of %s "
+ "- %02X:%02X:%02X:%02X:%02X:%02X - is "
+ "still in use by %s. Set the HWaddr of "
+ "%s to a different address to avoid "
+ "conflicts.\n",
+ bond_dev->name, slave_dev->name,
+ slave->perm_hwaddr[0],
+ slave->perm_hwaddr[1],
+ slave->perm_hwaddr[2],
+ slave->perm_hwaddr[3],
+ slave->perm_hwaddr[4],
+ slave->perm_hwaddr[5],
+ bond_dev->name, slave_dev->name);
}
/* Inform AD package of unbinding of slave. */
@@ -1838,7 +1939,7 @@ int bond_release(struct net_device *bond_dev, struct net_device *slave_dev)
/* close slave before restoring its mac address */
dev_close(slave_dev);
- if (!bond->params.fail_over_mac) {
+ if (bond->params.fail_over_mac != BOND_FOM_ACTIVE) {
/* restore original ("permanent") mac address */
memcpy(addr.sa_data, slave->perm_hwaddr, ETH_ALEN);
addr.sa_family = slave_dev->type;
@@ -2129,295 +2230,250 @@ static int bond_slave_info_query(struct net_device *bond_dev, struct ifslave *in
/*-------------------------------- Monitoring -------------------------------*/
-/*
- * if !have_locks, return nonzero if a failover is necessary. if
- * have_locks, do whatever failover activities are needed.
- *
- * This is to separate the inspection and failover steps for locking
- * purposes; failover requires rtnl, but acquiring it for every
- * inspection is undesirable, so a wrapper first does inspection, and
- * the acquires the necessary locks and calls again to perform
- * failover if needed. Since all locks are dropped, a complete
- * restart is needed between calls.
- */
-static int __bond_mii_monitor(struct bonding *bond, int have_locks)
-{
- struct slave *slave, *oldcurrent;
- int do_failover = 0;
- int i;
-
- if (bond->slave_cnt == 0)
- goto out;
-
- /* we will try to read the link status of each of our slaves, and
- * set their IFF_RUNNING flag appropriately. For each slave not
- * supporting MII status, we won't do anything so that a user-space
- * program could monitor the link itself if needed.
- */
- if (bond->send_grat_arp) {
- if (bond->curr_active_slave && test_bit(__LINK_STATE_LINKWATCH_PENDING,
- &bond->curr_active_slave->dev->state))
- dprintk("Needs to send gratuitous arp but not yet\n");
- else {
- dprintk("sending delayed gratuitous arp on on %s\n",
- bond->curr_active_slave->dev->name);
- bond_send_gratuitous_arp(bond);
- bond->send_grat_arp = 0;
- }
- }
- read_lock(&bond->curr_slave_lock);
- oldcurrent = bond->curr_active_slave;
- read_unlock(&bond->curr_slave_lock);
+static int bond_miimon_inspect(struct bonding *bond)
+{
+ struct slave *slave;
+ int i, link_state, commit = 0;
bond_for_each_slave(bond, slave, i) {
- struct net_device *slave_dev = slave->dev;
- int link_state;
- u16 old_speed = slave->speed;
- u8 old_duplex = slave->duplex;
+ slave->new_link = BOND_LINK_NOCHANGE;
- link_state = bond_check_dev_link(bond, slave_dev, 0);
+ link_state = bond_check_dev_link(bond, slave->dev, 0);
switch (slave->link) {
- case BOND_LINK_UP: /* the link was up */
- if (link_state == BMSR_LSTATUS) {
- if (!oldcurrent) {
- if (!have_locks)
- return 1;
- do_failover = 1;
- }
- break;
- } else { /* link going down */
- slave->link = BOND_LINK_FAIL;
- slave->delay = bond->params.downdelay;
-
- if (slave->link_failure_count < UINT_MAX) {
- slave->link_failure_count++;
- }
+ case BOND_LINK_UP:
+ if (link_state)
+ continue;
- if (bond->params.downdelay) {
- printk(KERN_INFO DRV_NAME
- ": %s: link status down for %s "
- "interface %s, disabling it in "
- "%d ms.\n",
- bond->dev->name,
- IS_UP(slave_dev)
- ? ((bond->params.mode == BOND_MODE_ACTIVEBACKUP)
- ? ((slave == oldcurrent)
- ? "active " : "backup ")
- : "")
- : "idle ",
- slave_dev->name,
- bond->params.downdelay * bond->params.miimon);
- }
+ slave->link = BOND_LINK_FAIL;
+ slave->delay = bond->params.downdelay;
+ if (slave->delay) {
+ printk(KERN_INFO DRV_NAME
+ ": %s: link status down for %s"
+ "interface %s, disabling it in %d ms.\n",
+ bond->dev->name,
+ (bond->params.mode ==
+ BOND_MODE_ACTIVEBACKUP) ?
+ ((slave->state == BOND_STATE_ACTIVE) ?
+ "active " : "backup ") : "",
+ slave->dev->name,
+ bond->params.downdelay * bond->params.miimon);
}
- /* no break ! fall through the BOND_LINK_FAIL test to
- ensure proper action to be taken
- */
- case BOND_LINK_FAIL: /* the link has just gone down */
- if (link_state != BMSR_LSTATUS) {
- /* link stays down */
- if (slave->delay <= 0) {
- if (!have_locks)
- return 1;
-
- /* link down for too long time */
- slave->link = BOND_LINK_DOWN;
-
- /* in active/backup mode, we must
- * completely disable this interface
- */
- if ((bond->params.mode == BOND_MODE_ACTIVEBACKUP) ||
- (bond->params.mode == BOND_MODE_8023AD)) {
- bond_set_slave_inactive_flags(slave);
- }
-
- printk(KERN_INFO DRV_NAME
- ": %s: link status definitely "
- "down for interface %s, "
- "disabling it\n",
- bond->dev->name,
- slave_dev->name);
-
- /* notify ad that the link status has changed */
- if (bond->params.mode == BOND_MODE_8023AD) {
- bond_3ad_handle_link_change(slave, BOND_LINK_DOWN);
- }
-
- if ((bond->params.mode == BOND_MODE_TLB) ||
- (bond->params.mode == BOND_MODE_ALB)) {
- bond_alb_handle_link_change(bond, slave, BOND_LINK_DOWN);
- }
-
- if (slave == oldcurrent) {
- do_failover = 1;
- }
- } else {
- slave->delay--;
- }
- } else {
- /* link up again */
- slave->link = BOND_LINK_UP;
+ /*FALLTHRU*/
+ case BOND_LINK_FAIL:
+ if (link_state) {
+ /*
+ * recovered before downdelay expired
+ */
+ slave->link = BOND_LINK_UP;
slave->jiffies = jiffies;
printk(KERN_INFO DRV_NAME
": %s: link status up again after %d "
"ms for interface %s.\n",
bond->dev->name,
- (bond->params.downdelay - slave->delay) * bond->params.miimon,
- slave_dev->name);
+ (bond->params.downdelay - slave->delay) *
+ bond->params.miimon,
+ slave->dev->name);
+ continue;
}
- break;
- case BOND_LINK_DOWN: /* the link was down */
- if (link_state != BMSR_LSTATUS) {
- /* the link stays down, nothing more to do */
- break;
- } else { /* link going up */
- slave->link = BOND_LINK_BACK;
- slave->delay = bond->params.updelay;
- if (bond->params.updelay) {
- /* if updelay == 0, no need to
- advertise about a 0 ms delay */
- printk(KERN_INFO DRV_NAME
- ": %s: link status up for "
- "interface %s, enabling it "
- "in %d ms.\n",
- bond->dev->name,
- slave_dev->name,
- bond->params.updelay * bond->params.miimon);
- }
+ if (slave->delay <= 0) {
+ slave->new_link = BOND_LINK_DOWN;
+ commit++;
+ continue;
}
- /* no break ! fall through the BOND_LINK_BACK state in
- case there's something to do.
- */
- case BOND_LINK_BACK: /* the link has just come back */
- if (link_state != BMSR_LSTATUS) {
- /* link down again */
- slave->link = BOND_LINK_DOWN;
+ slave->delay--;
+ break;
+
+ case BOND_LINK_DOWN:
+ if (!link_state)
+ continue;
+
+ slave->link = BOND_LINK_BACK;
+ slave->delay = bond->params.updelay;
+
+ if (slave->delay) {
+ printk(KERN_INFO DRV_NAME
+ ": %s: link status up for "
+ "interface %s, enabling it in %d ms.\n",
+ bond->dev->name, slave->dev->name,
+ bond->params.updelay *
+ bond->params.miimon);
+ }
+ /*FALLTHRU*/
+ case BOND_LINK_BACK:
+ if (!link_state) {
+ slave->link = BOND_LINK_DOWN;
printk(KERN_INFO DRV_NAME
": %s: link status down again after %d "
"ms for interface %s.\n",
bond->dev->name,
- (bond->params.updelay - slave->delay) * bond->params.miimon,
- slave_dev->name);
- } else {
- /* link stays up */
- if (slave->delay == 0) {
- if (!have_locks)
- return 1;
-
- /* now the link has been up for long time enough */
- slave->link = BOND_LINK_UP;
- slave->jiffies = jiffies;
-
- if (bond->params.mode == BOND_MODE_8023AD) {
- /* prevent it from being the active one */
- slave->state = BOND_STATE_BACKUP;
- } else if (bond->params.mode != BOND_MODE_ACTIVEBACKUP) {
- /* make it immediately active */
- slave->state = BOND_STATE_ACTIVE;
- } else if (slave != bond->primary_slave) {
- /* prevent it from being the active one */
- slave->state = BOND_STATE_BACKUP;
- }
+ (bond->params.updelay - slave->delay) *
+ bond->params.miimon,
+ slave->dev->name);
- printk(KERN_INFO DRV_NAME
- ": %s: link status definitely "
- "up for interface %s.\n",
- bond->dev->name,
- slave_dev->name);
-
- /* notify ad that the link status has changed */
- 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)) {
- bond_alb_handle_link_change(bond, slave, BOND_LINK_UP);
- }
-
- if ((!oldcurrent) ||
- (slave == bond->primary_slave)) {
- do_failover = 1;
- }
- } else {
- slave->delay--;
- }
+ continue;
}
+
+ if (slave->delay <= 0) {
+ slave->new_link = BOND_LINK_UP;
+ commit++;
+ continue;
+ }
+
+ slave->delay--;
break;
- default:
- /* Should not happen */
- printk(KERN_ERR DRV_NAME
- ": %s: Error: %s Illegal value (link=%d)\n",
- bond->dev->name,
- slave->dev->name,
- slave->link);
- goto out;
- } /* end of switch (slave->link) */
+ }
+ }
- bond_update_speed_duplex(slave);
+ return commit;
+}
- if (bond->params.mode == BOND_MODE_8023AD) {
- if (old_speed != slave->speed) {
- bond_3ad_adapter_speed_changed(slave);
- }
+static void bond_miimon_commit(struct bonding *bond)
+{
+ struct slave *slave;
+ int i;
- if (old_duplex != slave->duplex) {
- bond_3ad_adapter_duplex_changed(slave);
+ bond_for_each_slave(bond, slave, i) {
+ switch (slave->new_link) {
+ case BOND_LINK_NOCHANGE:
+ continue;
+
+ case BOND_LINK_UP:
+ slave->link = BOND_LINK_UP;
+ slave->jiffies = jiffies;
+
+ if (bond->params.mode == BOND_MODE_8023AD) {
+ /* prevent it from being the active one */
+ slave->state = BOND_STATE_BACKUP;
+ } else if (bond->params.mode != BOND_MODE_ACTIVEBACKUP) {
+ /* make it immediately active */
+ slave->state = BOND_STATE_ACTIVE;
+ } else if (slave != bond->primary_slave) {
+ /* prevent it from being the active one */
+ slave->state = BOND_STATE_BACKUP;
}
- }
- } /* end of for */
+ printk(KERN_INFO DRV_NAME
+ ": %s: link status definitely "
+ "up for interface %s.\n",
+ bond->dev->name, slave->dev->name);
- if (do_failover) {
- ASSERT_RTNL();
+ /* notify ad that the link status has changed */
+ if (bond->params.mode == BOND_MODE_8023AD)
+ bond_3ad_handle_link_change(slave, BOND_LINK_UP);
- write_lock_bh(&bond->curr_slave_lock);
+ if ((bond->params.mode == BOND_MODE_TLB) ||
+ (bond->params.mode == BOND_MODE_ALB))
+ bond_alb_handle_link_change(bond, slave,
+ BOND_LINK_UP);
- bond_select_active_slave(bond);
+ if (!bond->curr_active_slave ||
+ (slave == bond->primary_slave))
+ goto do_failover;
- write_unlock_bh(&bond->curr_slave_lock);
+ continue;
- } else
- bond_set_carrier(bond);
+ case BOND_LINK_DOWN:
+ if (slave->link_failure_count < UINT_MAX)
+ slave->link_failure_count++;
-out:
- return 0;
+ slave->link = BOND_LINK_DOWN;
+
+ if (bond->params.mode == BOND_MODE_ACTIVEBACKUP ||
+ bond->params.mode == BOND_MODE_8023AD)
+ bond_set_slave_inactive_flags(slave);
+
+ printk(KERN_INFO DRV_NAME
+ ": %s: link status definitely down for "
+ "interface %s, disabling it\n",
+ bond->dev->name, slave->dev->name);
+
+ if (bond->params.mode == BOND_MODE_8023AD)
+ bond_3ad_handle_link_change(slave,
+ BOND_LINK_DOWN);
+
+ if (bond->params.mode == BOND_MODE_TLB ||
+ bond->params.mode == BOND_MODE_ALB)
+ bond_alb_handle_link_change(bond, slave,
+ BOND_LINK_DOWN);
+
+ if (slave == bond->curr_active_slave)
+ goto do_failover;
+
+ continue;
+
+ default:
+ printk(KERN_ERR DRV_NAME
+ ": %s: invalid new link %d on slave %s\n",
+ bond->dev->name, slave->new_link,
+ slave->dev->name);
+ slave->new_link = BOND_LINK_NOCHANGE;
+
+ continue;
+ }
+
+do_failover:
+ ASSERT_RTNL();
+ write_lock_bh(&bond->curr_slave_lock);
+ bond_select_active_slave(bond);
+ write_unlock_bh(&bond->curr_slave_lock);
+ }
+
+ bond_set_carrier(bond);
}
/*
* bond_mii_monitor
*
* Really a wrapper that splits the mii monitor into two phases: an
- * inspection, then (if inspection indicates something needs to be
- * done) an acquisition of appropriate locks followed by another pass
- * to implement whatever link state changes are indicated.
+ * inspection, then (if inspection indicates something needs to be done)
+ * an acquisition of appropriate locks followed by a commit phase to
+ * implement whatever link state changes are indicated.
*/
void bond_mii_monitor(void *work_data)
{
struct bonding *bond = work_data;
- unsigned long delay;
read_lock(&bond->lock);
- if (bond->kill_timers) {
- read_unlock(&bond->lock);
- return;
+ if (bond->kill_timers)
+ goto out;
+
+ if (bond->slave_cnt == 0)
+ goto re_arm;
+
+ if (bond->send_grat_arp) {
+ read_lock(&bond->curr_slave_lock);
+ bond_send_gratuitous_arp(bond);
+ read_unlock(&bond->curr_slave_lock);
}
- if (__bond_mii_monitor(bond, 0)) {
+
+ 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();
read_lock(&bond->lock);
- __bond_mii_monitor(bond, 1);
+
+ bond_miimon_commit(bond);
+
read_unlock(&bond->lock);
rtnl_unlock(); /* might sleep, hold no other locks */
read_lock(&bond->lock);
}
- delay = ((bond->params.miimon * HZ) / 1000) ? : 1;
+re_arm:
+ if (bond->params.miimon)
+ queue_delayed_work(bond->wq, &bond->mii_work,
+ msecs_to_jiffies(bond->params.miimon));
+out:
read_unlock(&bond->lock);
- queue_delayed_work(bond->wq, &bond->mii_work, delay);
}
static __be32 bond_glean_dev_ip(struct net_device *dev)
@@ -2444,37 +2500,14 @@ out:
return addr;
}
-static int bond_has_ip(struct bonding *bond)
-{
- struct vlan_entry *vlan, *vlan_next;
-
- if (bond->master_ip)
- return 1;
-
- if (list_empty(&bond->vlan_list))
- return 0;
-
- list_for_each_entry_safe(vlan, vlan_next, &bond->vlan_list,
- vlan_list) {
- if (vlan->vlan_ip)
- return 1;
- }
-
- return 0;
-}
-
static int bond_has_this_ip(struct bonding *bond, __be32 ip)
{
- struct vlan_entry *vlan, *vlan_next;
+ struct vlan_entry *vlan;
if (ip == bond->master_ip)
return 1;
- if (list_empty(&bond->vlan_list))
- return 0;
-
- list_for_each_entry_safe(vlan, vlan_next, &bond->vlan_list,
- vlan_list) {
+ list_for_each_entry(vlan, &bond->vlan_list, vlan_list) {
if (ip == vlan->vlan_ip)
return 1;
}
@@ -2516,7 +2549,7 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
{
int i, vlan_id, rv;
__be32 *targets = bond->params.arp_targets;
- struct vlan_entry *vlan, *vlan_next;
+ struct vlan_entry *vlan;
struct net_device *vlan_dev;
struct flowi fl;
struct rtable *rt;
@@ -2563,8 +2596,7 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
}
vlan_id = 0;
- list_for_each_entry_safe(vlan, vlan_next, &bond->vlan_list,
- vlan_list) {
+ list_for_each_entry(vlan, &bond->vlan_list, vlan_list) {
vlan_dev = bond->vlgrp->vlan_devices[vlan->vlan_id];
if (vlan_dev == rt->u.dst.dev) {
vlan_id = vlan->vlan_id;
@@ -2594,6 +2626,8 @@ static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
/*
* Kick out a gratuitous ARP for an IP on the bonding master plus one
* for each VLAN above us.
+ *
+ * Caller must hold curr_slave_lock for read or better
*/
static void bond_send_gratuitous_arp(struct bonding *bond)
{
@@ -2603,9 +2637,13 @@ static void bond_send_gratuitous_arp(struct bonding *bond)
dprintk("bond_send_grat_arp: bond %s slave %s\n", bond->dev->name,
slave ? slave->dev->name : "NULL");
- if (!slave)
+
+ if (!slave || !bond->send_grat_arp ||
+ test_bit(__LINK_STATE_LINKWATCH_PENDING, &slave->dev->state))
return;
+ bond->send_grat_arp--;
+
if (bond->master_ip) {
bond_arp_send(slave->dev, ARPOP_REPLY, bond->master_ip,
bond->master_ip, 0);
@@ -2724,7 +2762,7 @@ void bond_loadbalance_arp_mon(void *work_data)
read_lock(&bond->lock);
- delta_in_ticks = (bond->params.arp_interval * HZ) / 1000;
+ delta_in_ticks = msecs_to_jiffies(bond->params.arp_interval);
if (bond->kill_timers) {
goto out;
@@ -2781,8 +2819,7 @@ void bond_loadbalance_arp_mon(void *work_data)
* if we don't know our ip yet
*/
if (time_after_eq(jiffies, slave->dev->trans_start + 2*delta_in_ticks) ||
- (time_after_eq(jiffies, slave->dev->last_rx + 2*delta_in_ticks) &&
- bond_has_ip(bond))) {
+ (time_after_eq(jiffies, slave->dev->last_rx + 2*delta_in_ticks))) {
slave->link = BOND_LINK_DOWN;
slave->state = BOND_STATE_BACKUP;
@@ -2830,245 +2867,310 @@ out:
}
/*
- * When using arp monitoring in active-backup mode, this function is
- * called to determine if any backup slaves have went down or a new
- * current slave needs to be found.
- * The backup slaves never generate traffic, they are considered up by merely
- * receiving traffic. If the current slave goes down, each backup slave will
- * be given the opportunity to tx/rx an arp before being taken down - this
- * prevents all slaves from being taken down due to the current slave not
- * sending any traffic for the backups to receive. The arps are not necessarily
- * necessary, any tx and rx traffic will keep the current slave up. While any
- * rx traffic will keep the backup slaves up, the current slave is responsible
- * for generating traffic to keep them up regardless of any other traffic they
- * may have received.
- * see loadbalance_arp_monitor for arp monitoring in load balancing mode
+ * Called to inspect slaves for active-backup mode ARP monitor link state
+ * changes. Sets new_link in slaves to specify what action should take
+ * place for the slave. Returns 0 if no changes are found, >0 if changes
+ * to link states must be committed.
+ *
+ * Called with bond->lock held for read.
*/
-void bond_activebackup_arp_mon(void *work_data)
+static int bond_ab_arp_inspect(struct bonding *bond, int delta_in_ticks)
{
- struct bonding *bond = work_data;
struct slave *slave;
- int delta_in_ticks;
- int i;
+ int i, commit = 0;
- read_lock(&bond->lock);
+ bond_for_each_slave(bond, slave, i) {
+ slave->new_link = BOND_LINK_NOCHANGE;
- delta_in_ticks = (bond->params.arp_interval * HZ) / 1000;
+ if (slave->link != BOND_LINK_UP) {
+ if (time_before_eq(jiffies, slave_last_rx(bond, slave) +
+ delta_in_ticks)) {
+ slave->new_link = BOND_LINK_UP;
+ commit++;
+ }
- if (bond->kill_timers) {
- goto out;
- }
+ continue;
+ }
- if (bond->slave_cnt == 0) {
- goto re_arm;
+ /*
+ * Give slaves 2*delta after being enslaved or made
+ * active. This avoids bouncing, as the last receive
+ * times need a full ARP monitor cycle to be updated.
+ */
+ if (!time_after_eq(jiffies, slave->jiffies +
+ 2 * delta_in_ticks))
+ continue;
+
+ /*
+ * Backup slave is down if:
+ * - No current_arp_slave AND
+ * - more than 3*delta since last receive AND
+ * - the bond has an IP address
+ *
+ * Note: a non-null current_arp_slave indicates
+ * the curr_active_slave went down and we are
+ * searching for a new one; under this condition
+ * we only take the curr_active_slave down - this
+ * gives each slave a chance to tx/rx traffic
+ * before being taken out
+ */
+ if (slave->state == BOND_STATE_BACKUP &&
+ !bond->current_arp_slave &&
+ time_after(jiffies, slave_last_rx(bond, slave) +
+ 3 * delta_in_ticks)) {
+ slave->new_link = BOND_LINK_DOWN;
+ commit++;
+ }
+
+ /*
+ * Active slave is down if:
+ * - more than 2*delta since transmitting OR
+ * - (more than 2*delta since receive AND
+ * the bond has an IP address)
+ */
+ if ((slave->state == BOND_STATE_ACTIVE) &&
+ (time_after_eq(jiffies, slave->dev->trans_start +
+ 2 * delta_in_ticks) ||
+ (time_after_eq(jiffies, slave_last_rx(bond, slave)
+ + 2 * delta_in_ticks)))) {
+ slave->new_link = BOND_LINK_DOWN;
+ commit++;
+ }
}
- /* determine if any slave has come up or any backup slave has
- * gone down
- * TODO: what about up/down delay in arp mode? it wasn't here before
- * so it can wait
+ read_lock(&bond->curr_slave_lock);
+
+ /*
+ * Trigger a commit if the primary option setting has changed.
*/
- bond_for_each_slave(bond, slave, i) {
- if (slave->link != BOND_LINK_UP) {
- if (time_before_eq(jiffies,
- slave_last_rx(bond, slave) + delta_in_ticks)) {
+ if (bond->primary_slave &&
+ (bond->primary_slave != bond->curr_active_slave) &&
+ (bond->primary_slave->link == BOND_LINK_UP))
+ commit++;
- slave->link = BOND_LINK_UP;
+ read_unlock(&bond->curr_slave_lock);
- write_lock_bh(&bond->curr_slave_lock);
+ return commit;
+}
- if ((!bond->curr_active_slave) &&
- time_before_eq(jiffies, slave->dev->trans_start + delta_in_ticks)) {
- bond_change_active_slave(bond, slave);
- bond->current_arp_slave = NULL;
- } else if (bond->curr_active_slave != slave) {
- /* this slave has just come up but we
- * already have a current slave; this
- * can also happen if bond_enslave adds
- * a new slave that is up while we are
- * searching for a new slave
- */
- bond_set_slave_inactive_flags(slave);
- bond->current_arp_slave = NULL;
- }
+/*
+ * Called to commit link state changes noted by inspection step of
+ * active-backup mode ARP monitor.
+ *
+ * Called with RTNL and bond->lock for read.
+ */
+static void bond_ab_arp_commit(struct bonding *bond, int delta_in_ticks)
+{
+ struct slave *slave;
+ int i;
- bond_set_carrier(bond);
+ bond_for_each_slave(bond, slave, i) {
+ switch (slave->new_link) {
+ case BOND_LINK_NOCHANGE:
+ continue;
- if (slave == bond->curr_active_slave) {
- printk(KERN_INFO DRV_NAME
- ": %s: %s is up and now the "
- "active interface\n",
- bond->dev->name,
- slave->dev->name);
- netif_carrier_on(bond->dev);
- } else {
- printk(KERN_INFO DRV_NAME
- ": %s: backup interface %s is "
- "now up\n",
- bond->dev->name,
- slave->dev->name);
- }
+ case BOND_LINK_UP:
+ write_lock_bh(&bond->curr_slave_lock);
- write_unlock_bh(&bond->curr_slave_lock);
- }
- } else {
- read_lock(&bond->curr_slave_lock);
+ if (!bond->curr_active_slave &&
+ time_before_eq(jiffies, slave->dev->trans_start +
+ delta_in_ticks)) {
+ slave->link = BOND_LINK_UP;
+ bond_change_active_slave(bond, slave);
+ bond->current_arp_slave = NULL;
- if ((slave != bond->curr_active_slave) &&
- (!bond->current_arp_slave) &&
- (time_after_eq(jiffies, slave_last_rx(bond, slave) + 3*delta_in_ticks) &&
- bond_has_ip(bond))) {
- /* a backup slave has gone down; three times
- * the delta allows the current slave to be
- * taken out before the backup slave.
- * note: a non-null current_arp_slave indicates
- * the curr_active_slave went down and we are
- * searching for a new one; under this
- * condition we only take the curr_active_slave
- * down - this gives each slave a chance to
- * tx/rx traffic before being taken out
+ printk(KERN_INFO DRV_NAME
+ ": %s: %s is up and now the "
+ "active interface\n",
+ bond->dev->name, slave->dev->name);
+
+ } else if (bond->curr_active_slave != slave) {
+ /* this slave has just come up but we
+ * already have a current slave; this can
+ * also happen if bond_enslave adds a new
+ * slave that is up while we are searching
+ * for a new slave
*/
+ slave->link = BOND_LINK_UP;
+ bond_set_slave_inactive_flags(slave);
+ bond->current_arp_slave = NULL;
- read_unlock(&bond->curr_slave_lock);
+ printk(KERN_INFO DRV_NAME
+ ": %s: backup interface %s is now up\n",
+ bond->dev->name, slave->dev->name);
+ }
- slave->link = BOND_LINK_DOWN;
+ write_unlock_bh(&bond->curr_slave_lock);
- if (slave->link_failure_count < UINT_MAX) {
- slave->link_failure_count++;
- }
+ break;
+
+ case BOND_LINK_DOWN:
+ if (slave->link_failure_count < UINT_MAX)
+ slave->link_failure_count++;
+
+ slave->link = BOND_LINK_DOWN;
+
+ if (slave == bond->curr_active_slave) {
+ printk(KERN_INFO DRV_NAME
+ ": %s: link status down for active "
+ "interface %s, disabling it\n",
+ bond->dev->name, slave->dev->name);
bond_set_slave_inactive_flags(slave);
+ write_lock_bh(&bond->curr_slave_lock);
+
+ bond_select_active_slave(bond);
+ if (bond->curr_active_slave)
+ bond->curr_active_slave->jiffies =
+ jiffies;
+
+ write_unlock_bh(&bond->curr_slave_lock);
+
+ bond->current_arp_slave = NULL;
+
+ } else if (slave->state == BOND_STATE_BACKUP) {
printk(KERN_INFO DRV_NAME
": %s: backup interface %s is now down\n",
- bond->dev->name,
- slave->dev->name);
- } else {
- read_unlock(&bond->curr_slave_lock);
+ bond->dev->name, slave->dev->name);
+
+ bond_set_slave_inactive_flags(slave);
}
+ break;
+
+ default:
+ printk(KERN_ERR DRV_NAME
+ ": %s: impossible: new_link %d on slave %s\n",
+ bond->dev->name, slave->new_link,
+ slave->dev->name);
}
}
- read_lock(&bond->curr_slave_lock);
- slave = bond->curr_active_slave;
- read_unlock(&bond->curr_slave_lock);
-
- if (slave) {
- /* if we have sent traffic in the past 2*arp_intervals but
- * haven't xmit and rx traffic in that time interval, select
- * a different slave. slave->jiffies is only updated when
- * a slave first becomes the curr_active_slave - not necessarily
- * after every arp; this ensures the slave has a full 2*delta
- * before being taken out. if a primary is being used, check
- * if it is up and needs to take over as the curr_active_slave
- */
- if ((time_after_eq(jiffies, slave->dev->trans_start + 2*delta_in_ticks) ||
- (time_after_eq(jiffies, slave_last_rx(bond, slave) + 2*delta_in_ticks) &&
- bond_has_ip(bond))) &&
- time_after_eq(jiffies, slave->jiffies + 2*delta_in_ticks)) {
+ /*
+ * No race with changes to primary via sysfs, as we hold rtnl.
+ */
+ if (bond->primary_slave &&
+ (bond->primary_slave != bond->curr_active_slave) &&
+ (bond->primary_slave->link == BOND_LINK_UP)) {
+ write_lock_bh(&bond->curr_slave_lock);
+ bond_change_active_slave(bond, bond->primary_slave);
+ write_unlock_bh(&bond->curr_slave_lock);
+ }
- slave->link = BOND_LINK_DOWN;
+ bond_set_carrier(bond);
+}
- if (slave->link_failure_count < UINT_MAX) {
- slave->link_failure_count++;
- }
+/*
+ * Send ARP probes for active-backup mode ARP monitor.
+ *
+ * Called with bond->lock held for read.
+ */
+static void bond_ab_arp_probe(struct bonding *bond)
+{
+ struct slave *slave;
+ int i;
- printk(KERN_INFO DRV_NAME
- ": %s: link status down for active interface "
- "%s, disabling it\n",
- bond->dev->name,
- slave->dev->name);
+ read_lock(&bond->curr_slave_lock);
- write_lock_bh(&bond->curr_slave_lock);
+ if (bond->current_arp_slave && bond->curr_active_slave)
+ printk("PROBE: c_arp %s && cas %s BAD\n",
+ bond->current_arp_slave->dev->name,
+ bond->curr_active_slave->dev->name);
- bond_select_active_slave(bond);
- slave = bond->curr_active_slave;
+ if (bond->curr_active_slave) {
+ bond_arp_send_all(bond, bond->curr_active_slave);
+ read_unlock(&bond->curr_slave_lock);
+ return;
+ }
- write_unlock_bh(&bond->curr_slave_lock);
+ read_unlock(&bond->curr_slave_lock);
- bond->current_arp_slave = slave;
+ /* if we don't have a curr_active_slave, search for the next available
+ * backup slave from the current_arp_slave and make it the candidate
+ * for becoming the curr_active_slave
+ */
- if (slave) {
- slave->jiffies = jiffies;
- }
- } else if ((bond->primary_slave) &&
- (bond->primary_slave != slave) &&
- (bond->primary_slave->link == BOND_LINK_UP)) {
- /* at this point, slave is the curr_active_slave */
- printk(KERN_INFO DRV_NAME
- ": %s: changing from interface %s to primary "
- "interface %s\n",
- bond->dev->name,
- slave->dev->name,
- bond->primary_slave->dev->name);
+ if (!bond->current_arp_slave) {
+ bond->current_arp_slave = bond->first_slave;
+ if (!bond->current_arp_slave)
+ return;
+ }
- /* primary is up so switch to it */
- write_lock_bh(&bond->curr_slave_lock);
- bond_change_active_slave(bond, bond->primary_slave);
- write_unlock_bh(&bond->curr_slave_lock);
+ bond_set_slave_inactive_flags(bond->current_arp_slave);
- slave = bond->primary_slave;
+ /* search for next candidate */
+ bond_for_each_slave_from(bond, slave, i, bond->current_arp_slave->next) {
+ if (IS_UP(slave->dev)) {
+ slave->link = BOND_LINK_BACK;
+ bond_set_slave_active_flags(slave);
+ bond_arp_send_all(bond, slave);
slave->jiffies = jiffies;
- } else {
- bond->current_arp_slave = NULL;
+ bond->current_arp_slave = slave;
+ break;
}
- /* the current slave must tx an arp to ensure backup slaves
- * rx traffic
+ /* if the link state is up at this point, we
+ * mark it down - this can happen if we have
+ * simultaneous link failures and
+ * reselect_active_interface doesn't make this
+ * one the current slave so it is still marked
+ * up when it is actually down
*/
- if (slave && bond_has_ip(bond)) {
- bond_arp_send_all(bond, slave);
+ if (slave->link == BOND_LINK_UP) {
+ slave->link = BOND_LINK_DOWN;
+ if (slave->link_failure_count < UINT_MAX)
+ slave->link_failure_count++;
+
+ bond_set_slave_inactive_flags(slave);
+
+ printk(KERN_INFO DRV_NAME
+ ": %s: backup interface %s is now down.\n",
+ bond->dev->name, slave->dev->name);
}
}
+}
- /* if we don't have a curr_active_slave, search for the next available
- * backup slave from the current_arp_slave and make it the candidate
- * for becoming the curr_active_slave
- */
- if (!slave) {
- if (!bond->current_arp_slave) {
- bond->current_arp_slave = bond->first_slave;
- }
-
- if (bond->current_arp_slave) {
- bond_set_slave_inactive_flags(bond->current_arp_slave);
-
- /* search for next candidate */
- bond_for_each_slave_from(bond, slave, i, bond->current_arp_slave->next) {
- if (IS_UP(slave->dev)) {
- slave->link = BOND_LINK_BACK;
- bond_set_slave_active_flags(slave);
- bond_arp_send_all(bond, slave);
- slave->jiffies = jiffies;
- bond->current_arp_slave = slave;
- break;
- }
+void bond_activebackup_arp_mon(void *work_data)
+{
+ struct bonding *bond = work_data;
+ int delta_in_ticks;
- /* if the link state is up at this point, we
- * mark it down - this can happen if we have
- * simultaneous link failures and
- * reselect_active_interface doesn't make this
- * one the current slave so it is still marked
- * up when it is actually down
- */
- if (slave->link == BOND_LINK_UP) {
- slave->link = BOND_LINK_DOWN;
- if (slave->link_failure_count < UINT_MAX) {
- slave->link_failure_count++;
- }
+ read_lock(&bond->lock);
- bond_set_slave_inactive_flags(slave);
+ if (bond->kill_timers)
+ goto out;
- printk(KERN_INFO DRV_NAME
- ": %s: backup interface %s is "
- "now down.\n",
- bond->dev->name,
- slave->dev->name);
- }
- }
- }
+ delta_in_ticks = msecs_to_jiffies(bond->params.arp_interval);
+
+ if (bond->slave_cnt == 0)
+ goto re_arm;
+
+ if (bond->send_grat_arp) {
+ read_lock(&bond->curr_slave_lock);
+ bond_send_gratuitous_arp(bond);
+ 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();
+ read_lock(&bond->lock);
+
+ bond_ab_arp_commit(bond, delta_in_ticks);
+
+ read_unlock(&bond->lock);
+ rtnl_unlock();
+ read_lock(&bond->lock);
+ }
+
+ bond_ab_arp_probe(bond);
+
re_arm:
if (bond->params.arp_interval) {
queue_delayed_work(bond->wq, &bond->arp_work, delta_in_ticks);
@@ -3146,7 +3248,8 @@ static void bond_info_show_master(struct seq_file *seq)
if (bond->params.mode == BOND_MODE_ACTIVEBACKUP &&
bond->params.fail_over_mac)
- seq_printf(seq, " (fail_over_mac)");
+ seq_printf(seq, " (fail_over_mac %s)",
+ fail_over_mac_tbl[bond->params.fail_over_mac].modename);
seq_printf(seq, "\n");
@@ -3525,10 +3628,10 @@ static int bond_inetaddr_event(struct notifier_block *this, unsigned long event,
{
struct in_ifaddr *ifa = ptr;
struct net_device *vlan_dev, *event_dev = ifa->ifa_dev->dev;
- struct bonding *bond, *bond_next;
- struct vlan_entry *vlan, *vlan_next;
+ struct bonding *bond;
+ struct vlan_entry *vlan;
- list_for_each_entry_safe(bond, bond_next, &bond_dev_list, bond_list) {
+ list_for_each_entry(bond, &bond_dev_list, bond_list) {
if (bond->dev == event_dev) {
switch (event) {
case NETDEV_UP:
@@ -3542,11 +3645,7 @@ static int bond_inetaddr_event(struct notifier_block *this, unsigned long event,
}
}
- if (list_empty(&bond->vlan_list))
- continue;
-
- list_for_each_entry_safe(vlan, vlan_next, &bond->vlan_list,
- vlan_list) {
+ list_for_each_entry(vlan, &bond->vlan_list, vlan_list) {
vlan_dev = bond->vlgrp->vlan_devices[vlan->vlan_id];
if (vlan_dev == event_dev) {
switch (event) {
@@ -3628,7 +3727,7 @@ static int bond_xmit_hash_policy_l23(struct sk_buff *skb,
struct ethhdr *data = (struct ethhdr *)skb->data;
struct iphdr *iph = ip_hdr(skb);
- if (skb->protocol == __constant_htons(ETH_P_IP)) {
+ if (skb->protocol == htons(ETH_P_IP)) {
return ((ntohl(iph->saddr ^ iph->daddr) & 0xffff) ^
(data->h_dest[5] ^ bond_dev->dev_addr[5])) % count;
}
@@ -3649,8 +3748,8 @@ static int bond_xmit_hash_policy_l34(struct sk_buff *skb,
__be16 *layer4hdr = (__be16 *)((u32 *)iph + iph->ihl);
int layer4_xor = 0;
- if (skb->protocol == __constant_htons(ETH_P_IP)) {
- if (!(iph->frag_off & __constant_htons(IP_MF|IP_OFFSET)) &&
+ if (skb->protocol == htons(ETH_P_IP)) {
+ if (!(iph->frag_off & htons(IP_MF|IP_OFFSET)) &&
(iph->protocol == IPPROTO_TCP ||
iph->protocol == IPPROTO_UDP)) {
layer4_xor = ntohs((*layer4hdr ^ *(layer4hdr + 1)));
@@ -3740,6 +3839,8 @@ 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;
@@ -4087,10 +4188,10 @@ static int bond_set_mac_address(struct net_device *bond_dev, void *addr)
dprintk("bond=%p, name=%s\n", bond, (bond_dev ? bond_dev->name : "None"));
/*
- * If fail_over_mac is enabled, do nothing and return success.
- * Returning an error causes ifenslave to fail.
+ * If fail_over_mac is set to active, do nothing and return
+ * success. Returning an error causes ifenslave to fail.
*/
- if (bond->params.fail_over_mac)
+ if (bond->params.fail_over_mac == BOND_FOM_ACTIVE)
return 0;
if (!is_valid_ether_addr(sa->sa_data)) {
@@ -4414,11 +4515,12 @@ static void bond_ethtool_get_drvinfo(struct net_device *bond_dev,
}
static struct ethtool_ops bond_ethtool_ops = {
+ .get_drvinfo = bond_ethtool_get_drvinfo,
+ .get_link = ethtool_op_get_link,
.get_tx_csum = ethtool_op_get_tx_csum,
+ .get_sg = ethtool_op_get_sg,
.get_tso = ethtool_op_get_tso,
.get_ufo = ethtool_op_get_ufo,
- .get_sg = ethtool_op_get_sg,
- .get_drvinfo = bond_ethtool_get_drvinfo,
};
/*
@@ -4448,6 +4550,7 @@ 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);
@@ -4469,6 +4572,8 @@ 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
@@ -4552,8 +4657,7 @@ static void bond_free_all(void)
bond_work_cancel_all(bond);
/* Release the bonded slaves */
bond_release_all(bond_dev);
- bond_deinit(bond_dev);
- unregister_netdevice(bond_dev);
+ bond_destroy(bond);
}
#ifdef CONFIG_PROC_FS
@@ -4598,7 +4702,7 @@ int bond_parse_parm(const char *buf, struct bond_parm_tbl *tbl)
static int bond_check_params(struct bond_params *params)
{
- int arp_validate_value;
+ int arp_validate_value, fail_over_mac_value;
/*
* Convert string parameters.
@@ -4648,11 +4752,11 @@ static int bond_check_params(struct bond_params *params)
}
}
- if (max_bonds < 1 || max_bonds > INT_MAX) {
+ if (max_bonds < 0 || max_bonds > INT_MAX) {
printk(KERN_WARNING DRV_NAME
": Warning: max_bonds (%d) not in range %d-%d, so it "
"was reset to BOND_DEFAULT_MAX_BONDS (%d)\n",
- max_bonds, 1, INT_MAX, BOND_DEFAULT_MAX_BONDS);
+ max_bonds, 0, INT_MAX, BOND_DEFAULT_MAX_BONDS);
max_bonds = BOND_DEFAULT_MAX_BONDS;
}
@@ -4688,6 +4792,20 @@ static int bond_check_params(struct bond_params *params)
use_carrier = 1;
}
+ if (num_grat_arp < 0 || num_grat_arp > 255) {
+ printk(KERN_WARNING DRV_NAME
+ ": Warning: num_grat_arp (%d) not in range 0-255 so it "
+ "was reset to 1 \n", num_grat_arp);
+ 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) {
@@ -4844,7 +4962,7 @@ static int bond_check_params(struct bond_params *params)
printk("\n");
- } else {
+ } else if (max_bonds) {
/* miimon and arp_interval not set, we need one so things
* work as expected, see bonding.txt for details
*/
@@ -4866,15 +4984,30 @@ static int bond_check_params(struct bond_params *params)
primary = NULL;
}
- if (fail_over_mac && (bond_mode != BOND_MODE_ACTIVEBACKUP))
- printk(KERN_WARNING DRV_NAME
- ": Warning: fail_over_mac only affects "
- "active-backup mode.\n");
+ if (fail_over_mac) {
+ fail_over_mac_value = bond_parse_parm(fail_over_mac,
+ fail_over_mac_tbl);
+ if (fail_over_mac_value == -1) {
+ printk(KERN_ERR DRV_NAME
+ ": Error: invalid fail_over_mac \"%s\"\n",
+ arp_validate == NULL ? "NULL" : arp_validate);
+ return -EINVAL;
+ }
+
+ if (bond_mode != BOND_MODE_ACTIVEBACKUP)
+ printk(KERN_WARNING DRV_NAME
+ ": Warning: fail_over_mac only affects "
+ "active-backup mode.\n");
+ } else {
+ fail_over_mac_value = BOND_FOM_NONE;
+ }
/* fill params struct with the proper values */
params->mode = bond_mode;
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;
@@ -4882,7 +5015,7 @@ static int bond_check_params(struct bond_params *params)
params->use_carrier = use_carrier;
params->lacp_fast = lacp_fast;
params->primary[0] = 0;
- params->fail_over_mac = fail_over_mac;
+ params->fail_over_mac = fail_over_mac_value;
if (primary) {
strncpy(params->primary, primary, IFNAMSIZ);
@@ -4901,10 +5034,10 @@ static struct lock_class_key bonding_netdev_xmit_lock_key;
* Caller must NOT hold rtnl_lock; we need to release it here before we
* set up our sysfs entries.
*/
-int bond_create(char *name, struct bond_params *params, struct bonding **newbond)
+int bond_create(char *name, struct bond_params *params)
{
struct net_device *bond_dev;
- struct bonding *bond, *nxt;
+ struct bonding *bond;
int res;
rtnl_lock();
@@ -4912,7 +5045,7 @@ int bond_create(char *name, struct bond_params *params, struct bonding **newbond
/* Check to see if the bond already exists. */
if (name) {
- list_for_each_entry_safe(bond, nxt, &bond_dev_list, bond_list)
+ list_for_each_entry(bond, &bond_dev_list, bond_list)
if (strnicmp(bond->dev->name, name, IFNAMSIZ) == 0) {
printk(KERN_ERR DRV_NAME
": cannot add bond %s; it already exists\n",
@@ -4957,9 +5090,6 @@ int bond_create(char *name, struct bond_params *params, struct bonding **newbond
lockdep_set_class(&bond_dev->_xmit_lock, &bonding_netdev_xmit_lock_key);
- if (newbond)
- *newbond = bond_dev->priv;
-
netif_carrier_off(bond_dev);
up_write(&bonding_rwsem);
@@ -4968,7 +5098,9 @@ int bond_create(char *name, struct bond_params *params, struct bonding **newbond
if (res < 0) {
rtnl_lock();
down_write(&bonding_rwsem);
- goto out_bond;
+ bond_deinit(bond_dev);
+ unregister_netdevice(bond_dev);
+ goto out_rtnl;
}
return 0;
@@ -4987,7 +5119,7 @@ static int __init bonding_init(void)
{
int i;
int res;
- struct bonding *bond, *nxt;
+ struct bonding *bond;
printk(KERN_INFO "%s", version);
@@ -5003,7 +5135,7 @@ static int __init bonding_init(void)
init_rwsem(&bonding_rwsem);
for (i = 0; i < max_bonds; i++) {
- res = bond_create(NULL, &bonding_defaults, NULL);
+ res = bond_create(NULL, &bonding_defaults);
if (res)
goto err;
}
@@ -5014,17 +5146,19 @@ 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:
- list_for_each_entry_safe(bond, nxt, &bond_dev_list, bond_list) {
+ list_for_each_entry(bond, &bond_dev_list, bond_list) {
bond_work_cancel_all(bond);
destroy_workqueue(bond->wq);
}
+ bond_destroy_sysfs();
+
rtnl_lock();
bond_free_all();
- bond_destroy_sysfs();
rtnl_unlock();
out:
return res;
@@ -5035,10 +5169,12 @@ 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();
rtnl_lock();
bond_free_all();
- bond_destroy_sysfs();
rtnl_unlock();
}
diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c
index b1f861f..4f4ce7b 100644
--- a/drivers/net/bonding/bond_sysfs.c
+++ b/drivers/net/bonding/bond_sysfs.c
@@ -51,9 +51,9 @@ 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;
-static struct class *netdev_class;
/*--------------------------- Data Structures -----------------------------*/
/* Bonding sysfs lock. Why can't we just use the subsystem lock?
@@ -112,7 +112,6 @@ static ssize_t bonding_store_bonds(struct class *cls, const char *buffer, size_t
char *ifname;
int rv, res = count;
struct bonding *bond;
- struct bonding *nxt;
sscanf(buffer, "%16s", command); /* IFNAMSIZ*/
ifname = command + 1;
@@ -123,7 +122,7 @@ static ssize_t bonding_store_bonds(struct class *cls, const char *buffer, size_t
if (command[0] == '+') {
printk(KERN_INFO DRV_NAME
": %s is being created...\n", ifname);
- rv = bond_create(ifname, &bonding_defaults, &bond);
+ rv = bond_create(ifname, &bonding_defaults);
if (rv) {
printk(KERN_INFO DRV_NAME ": Bond creation failed.\n");
res = rv;
@@ -135,7 +134,7 @@ static ssize_t bonding_store_bonds(struct class *cls, const char *buffer, size_t
rtnl_lock();
down_write(&bonding_rwsem);
- list_for_each_entry_safe(bond, nxt, &bond_dev_list, bond_list)
+ list_for_each_entry(bond, &bond_dev_list, bond_list)
if (strnicmp(bond->dev->name, ifname, IFNAMSIZ) == 0) {
/* check the ref count on the bond's kobject.
* If it's > expected, then there's a file open,
@@ -344,9 +343,6 @@ static ssize_t bonding_store_slaves(struct class_device *cd, const char *buffer,
if (dev) {
printk(KERN_INFO DRV_NAME ": %s: Removing slave %s\n",
bond->dev->name, dev->name);
- if (bond->setup_by_slave)
- res = bond_release_and_destroy(bond->dev, dev);
- else
res = bond_release(bond->dev, dev);
if (res) {
ret = res;
@@ -530,42 +526,37 @@ static ssize_t bonding_show_fail_over_mac(struct class_device *cd, char *buf)
{
struct bonding *bond = to_bond(cd);
- return sprintf(buf, "%d\n", bond->params.fail_over_mac) + 1;
+ return sprintf(buf, "%s %d\n",
+ fail_over_mac_tbl[bond->params.fail_over_mac].modename,
+ bond->params.fail_over_mac);
}
static ssize_t bonding_store_fail_over_mac(struct class_device *cd, const char *buf, size_t count)
{
int new_value;
- int ret = count;
struct bonding *bond = to_bond(cd);
if (bond->slave_cnt != 0) {
printk(KERN_ERR DRV_NAME
": %s: Can't alter fail_over_mac with slaves in bond.\n",
bond->dev->name);
- ret = -EPERM;
- goto out;
+ return -EPERM;
}
- if (sscanf(buf, "%d", &new_value) != 1) {
+ new_value = bond_parse_parm(buf, fail_over_mac_tbl);
+ if (new_value < 0) {
printk(KERN_ERR DRV_NAME
- ": %s: no fail_over_mac value specified.\n",
- bond->dev->name);
- ret = -EINVAL;
- goto out;
+ ": %s: Ignoring invalid fail_over_mac value %s.\n",
+ bond->dev->name, buf);
+ return -EINVAL;
}
- if ((new_value == 0) || (new_value == 1)) {
- bond->params.fail_over_mac = new_value;
- printk(KERN_INFO DRV_NAME ": %s: Setting fail_over_mac to %d.\n",
- bond->dev->name, new_value);
- } else {
- printk(KERN_INFO DRV_NAME
- ": %s: Ignoring invalid fail_over_mac value %d.\n",
- bond->dev->name, new_value);
- }
-out:
- return ret;
+ bond->params.fail_over_mac = new_value;
+ printk(KERN_INFO DRV_NAME ": %s: Setting fail_over_mac to %s (%d).\n",
+ bond->dev->name, fail_over_mac_tbl[new_value].modename,
+ new_value);
+
+ return count;
}
static CLASS_DEVICE_ATTR(fail_over_mac, S_IRUGO | S_IWUSR, bonding_show_fail_over_mac, bonding_store_fail_over_mac);
@@ -607,6 +598,8 @@ static ssize_t bonding_store_arp_interval(struct class_device *cd, const char *b
": %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. "
@@ -916,6 +909,80 @@ out:
static CLASS_DEVICE_ATTR(lacp_rate, S_IRUGO | S_IWUSR, bonding_show_lacp, bonding_store_lacp);
/*
+ * Show and set the number of grat ARP to send after a failover event.
+ */
+static ssize_t bonding_show_n_grat_arp(struct class_device *cd, char *buf)
+{
+ struct bonding *bond = to_bond(cd);
+
+ return sprintf(buf, "%d\n", bond->params.num_grat_arp);
+}
+
+static ssize_t bonding_store_n_grat_arp(struct class_device *cd, const char *buf, size_t count)
+{
+ int new_value, ret = count;
+ struct bonding *bond = to_bond(cd);
+
+ if (sscanf(buf, "%d", &new_value) != 1) {
+ printk(KERN_ERR DRV_NAME
+ ": %s: no num_grat_arp 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_grat_arp value %d not in range 0-255; rejected.\n",
+ bond->dev->name, new_value);
+ ret = -EINVAL;
+ goto out;
+ } else {
+ bond->params.num_grat_arp = new_value;
+ }
+out:
+ return ret;
+}
+
+static CLASS_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 class_device *cd,
+ char *buf)
+{
+ struct bonding *bond = to_bond(cd);
+
+ return sprintf(buf, "%d\n", bond->params.num_unsol_na);
+}
+
+static ssize_t bonding_store_n_unsol_na(struct class_device *cd, const char *buf, size_t count)
+{
+ int new_value, ret = count;
+ struct bonding *bond = to_bond(cd);
+
+ 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 CLASS_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
* ARP monitoring. Second, if the timer isn't running, we must
@@ -969,6 +1036,7 @@ static ssize_t bonding_store_miimon(struct class_device *cd, const char *buf, si
"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 =
@@ -1322,6 +1390,8 @@ static struct attribute *per_bond_attrs[] = {
&class_device_attr_slaves.attr,
&class_device_attr_mode.attr,
&class_device_attr_fail_over_mac.attr,
+ &class_device_attr_num_unsol_na.attr,
+ &class_device_attr_num_grat_arp.attr,
&class_device_attr_arp_validate.attr,
&class_device_attr_arp_interval.attr,
&class_device_attr_arp_ip_target.attr,
@@ -1353,19 +1423,9 @@ static struct attribute_group bonding_group = {
*/
int bond_create_sysfs(void)
{
- int ret = 0;
- struct bonding *firstbond;
-
- /* get the netdev class pointer */
- firstbond = container_of(bond_dev_list.next, struct bonding, bond_list);
- if (!firstbond)
- return -ENODEV;
-
- netdev_class = firstbond->dev->class_dev.class;
- if (!netdev_class)
- return -ENODEV;
+ int ret;
- ret = class_create_file(netdev_class, &class_attr_bonding_masters);
+ ret = netdev_class_create_file(&class_attr_bonding_masters);
/*
* Permit multiple loads of the module by ignoring failures to
* create the bonding_masters sysfs file. Bonding devices
@@ -1378,8 +1438,11 @@ int bond_create_sysfs(void)
* configure multiple bonding devices.
*/
if (ret == -EEXIST) {
- netdev_class = NULL;
- return 0;
+ /* Is someone being kinky and naming a device bonding_master? */
+ if (dev_get_by_name(class_attr_bonding_masters.attr.name))
+ printk(KERN_ERR
+ "network device named %s already exists in sysfs",
+ class_attr_bonding_masters.attr.name);
}
return ret;
@@ -1391,8 +1454,7 @@ int bond_create_sysfs(void)
*/
void bond_destroy_sysfs(void)
{
- if (netdev_class)
- class_remove_file(netdev_class, &class_attr_bonding_masters);
+ netdev_class_remove_file(&class_attr_bonding_masters);
}
/*
diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index 698c170..8a9c376 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -19,16 +19,19 @@
#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 "3.2.4"
-#define DRV_RELDATE "January 28, 2008"
+#define DRV_VERSION "3.4.0"
+#define DRV_RELDATE "October 7, 2008"
#define DRV_NAME "bonding"
#define DRV_DESCRIPTION "Ethernet Channel Bonding Driver"
#define BOND_MAX_ARP_TARGETS 16
+extern struct list_head bond_dev_list;
+
#ifdef BONDING_DEBUG
#define dprintk(fmt, args...) \
printk(KERN_DEBUG \
@@ -125,6 +128,8 @@ struct bond_params {
int mode;
int xmit_policy;
int miimon;
+ int num_grat_arp;
+ int num_unsol_na;
int arp_interval;
int arp_validate;
int use_carrier;
@@ -147,6 +152,9 @@ 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 {
@@ -157,6 +165,7 @@ struct slave {
unsigned long jiffies;
unsigned long last_arp_rx;
s8 link; /* one of BOND_LINK_XXXX */
+ s8 new_link;
s8 state; /* one of BOND_STATE_XXXX */
u32 original_flags;
u32 link_failure_count;
@@ -168,6 +177,16 @@ struct slave {
};
/*
+ * Link pseudo-state only used internally by monitors
+ */
+#define BOND_LINK_NOCHANGE -1
+
+/*
+ * Link pseudo-state only used internally by monitors
+ */
+#define BOND_LINK_NOCHANGE -1
+
+/*
* Here are the locking policies for the two bonding locks:
*
* 1) Get bond->lock when reading/writing slave list.
@@ -187,6 +206,7 @@ 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
@@ -210,6 +230,9 @@ struct bonding {
struct work_struct arp_work;
struct work_struct alb_work;
struct work_struct ad_work;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+ struct in6_addr master_ipv6;
+#endif
};
/**
@@ -240,6 +263,10 @@ static inline struct bonding *bond_get_bond_by_slave(struct slave *slave)
return (struct bonding *)slave->dev->master->priv;
}
+#define BOND_FOM_NONE 0
+#define BOND_FOM_ACTIVE 1
+#define BOND_FOM_FOLLOW 2
+
#define BOND_ARP_VALIDATE_NONE 0
#define BOND_ARP_VALIDATE_ACTIVE (1 << BOND_STATE_ACTIVE)
#define BOND_ARP_VALIDATE_BACKUP (1 << BOND_STATE_BACKUP)
@@ -300,7 +327,7 @@ static inline void bond_unset_master_alb_flags(struct bonding *bond)
struct vlan_entry *bond_next_vlan(struct bonding *bond, struct vlan_entry *curr);
int bond_dev_queue_xmit(struct bonding *bond, struct sk_buff *skb, struct net_device *slave_dev);
-int bond_create(char *name, struct bond_params *params, struct bonding **newbond);
+int bond_create(char *name, struct bond_params *params);
void bond_destroy(struct bonding *bond);
int bond_release_and_destroy(struct net_device *bond_dev, struct net_device *slave_dev);
int bond_create_sysfs(void);
@@ -321,5 +348,31 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active);
void bond_register_arp(struct bonding *);
void bond_unregister_arp(struct bonding *);
-#endif /* _LINUX_BONDING_H */
+/* 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[];
+
+#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 */
diff --git a/include/linux/if.h b/include/linux/if.h
index 554e1a8..2b1704d 100644
--- a/include/linux/if.h
+++ b/include/linux/if.h
@@ -62,6 +62,7 @@
#define IFF_BONDING 0x20 /* bonding master or slave */
#define IFF_SLAVE_NEEDARP 0x40 /* need ARPs for validation */
#define IFF_EXTENDED 0x80 /* Extended data available */
+#define IFF_MASTER_ARPMON 0x100 /* bonding master, ARP mon in use */
#define IF_GET_IFACE 0x0001 /* for querying only */
#define IF_GET_PROTO 0x0002
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c5ce762..41a0c78 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -328,6 +328,14 @@ struct net_device
#define NETIF_F_GEN_CSUM (NETIF_F_NO_CSUM | NETIF_F_HW_CSUM)
#define NETIF_F_ALL_CSUM (NETIF_F_IP_CSUM | NETIF_F_GEN_CSUM)
+ /*
+ * If one device supports one of these features, then enable them
+ * for all in netdev_increment_features.
+ */
+#define NETIF_F_ONE_FOR_ALL (NETIF_F_GSO_SOFTWARE | NETIF_F_GSO_ROBUST | \
+ NETIF_F_SG | NETIF_F_HIGHDMA | \
+ NETIF_F_FRAGLIST)
+
struct net_device *next_sched;
/* Interface index. Unique device identifier */
@@ -1030,9 +1038,14 @@ extern void *dev_seq_next(struct seq_file *seq, void *v, loff_t *pos);
extern void dev_seq_stop(struct seq_file *seq, void *v);
#endif
+extern int netdev_class_create_file(struct class_attribute *class_attr);
+extern void netdev_class_remove_file(struct class_attribute *class_attr);
+
extern void linkwatch_run_queue(void);
-extern int netdev_compute_features(unsigned long all, unsigned long one);
+unsigned long netdev_increment_features(unsigned long all, unsigned long one,
+ unsigned long mask);
+unsigned long netdev_fix_features(unsigned long features, const char *name);
static inline int net_gso_ok(int features, int gso_type)
{
@@ -1061,22 +1074,26 @@ static inline int skb_bond_should_drop(struct sk_buff *skb)
struct net_device *dev = skb->dev;
struct net_device *master = dev->master;
- if (master &&
- (dev->priv_flags & IFF_SLAVE_INACTIVE)) {
- if ((dev->priv_flags & IFF_SLAVE_NEEDARP) &&
- skb->protocol == __constant_htons(ETH_P_ARP))
- return 0;
+ if (master) {
+ if (master->priv_flags & IFF_MASTER_ARPMON)
+ dev->last_rx = jiffies;
- if (master->priv_flags & IFF_MASTER_ALB) {
- if (skb->pkt_type != PACKET_BROADCAST &&
- skb->pkt_type != PACKET_MULTICAST)
+ if (dev->priv_flags & IFF_SLAVE_INACTIVE) {
+ if ((dev->priv_flags & IFF_SLAVE_NEEDARP) &&
+ skb->protocol == __constant_htons(ETH_P_ARP))
return 0;
- }
- if (master->priv_flags & IFF_MASTER_8023AD &&
- skb->protocol == __constant_htons(ETH_P_SLOW))
- return 0;
- return 1;
+ if (master->priv_flags & IFF_MASTER_ALB) {
+ if (skb->pkt_type != PACKET_BROADCAST &&
+ skb->pkt_type != PACKET_MULTICAST)
+ return 0;
+ }
+ if (master->priv_flags & IFF_MASTER_8023AD &&
+ skb->protocol == __constant_htons(ETH_P_SLOW))
+ return 0;
+
+ return 1;
+ }
}
return 0;
}
diff --git a/include/net/ndisc.h b/include/net/ndisc.h
index d3915da..99afe7a 100644
--- a/include/net/ndisc.h
+++ b/include/net/ndisc.h
@@ -101,6 +101,20 @@ extern void ndisc_send_redirect(struct sk_buff *skb,
extern int ndisc_mc_map(struct in6_addr *addr, char *buf, struct net_device *dev, int dir);
+extern struct sk_buff *ndisc_build_skb(struct net_device *dev,
+ struct in6_addr *daddr,
+ struct in6_addr *saddr,
+ struct icmp6hdr *icmp6h,
+ const struct in6_addr *target,
+ int llinfo);
+
+extern void ndisc_send_skb(struct sk_buff *skb,
+ struct net_device *dev,
+ struct neighbour *neigh,
+ struct in6_addr *daddr,
+ struct in6_addr *saddr,
+ struct icmp6hdr *icmp6h);
+
struct rt6_info * dflt_rt_lookup(void);
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index efabbb5..cef123b 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -185,5 +185,5 @@ void br_dev_setup(struct net_device *dev)
dev->priv_flags = IFF_EBRIDGE;
dev->features = NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_HIGHDMA |
- NETIF_F_GSO_MASK | NETIF_F_NO_CSUM | NETIF_F_LLTX;
+ NETIF_F_GSO_MASK | NETIF_F_NO_CSUM | NETIF_F_LLTX | NETIF_F_GSO;
}
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index 95ea4d6..2fba454 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -358,15 +358,21 @@ int br_min_mtu(const struct net_bridge *br)
void br_features_recompute(struct net_bridge *br)
{
struct net_bridge_port *p;
- unsigned long features;
+ unsigned long features, mask;
- features = br->feature_mask;
+ features = mask = br->feature_mask;
+ if (list_empty(&br->port_list))
+ goto done;
+
+ features &= ~NETIF_F_ONE_FOR_ALL;
list_for_each_entry(p, &br->port_list, list) {
- features = netdev_compute_features(features, p->dev->features);
+ features = netdev_increment_features(features,
+ p->dev->features, mask);
}
- br->dev->features = features;
+done:
+ br->dev->features = netdev_fix_features(features, NULL);
}
/* called with RTNL */
diff --git a/net/core/dev.c b/net/core/dev.c
index 4bda12b..16ac4df 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2932,6 +2932,46 @@ static inline void net_set_todo(struct net_device *dev)
spin_unlock(&net_todo_list_lock);
}
+unsigned long netdev_fix_features(unsigned long features, const char *name)
+{
+ /* Fix illegal SG+CSUM combinations. */
+ if ((features & NETIF_F_SG) &&
+ !(features & NETIF_F_ALL_CSUM)) {
+ if (name)
+ printk(KERN_NOTICE "%s: Dropping NETIF_F_SG since no "
+ "checksum feature.\n", name);
+ features &= ~NETIF_F_SG;
+ }
+
+ /* TSO requires that SG is present as well. */
+ if ((features & NETIF_F_TSO) && !(features & NETIF_F_SG)) {
+ if (name)
+ printk(KERN_NOTICE "%s: Dropping NETIF_F_TSO since no "
+ "SG feature.\n", name);
+ features &= ~NETIF_F_TSO;
+ }
+
+ if (features & NETIF_F_UFO) {
+ if (!(features & NETIF_F_GEN_CSUM)) {
+ if (name)
+ printk(KERN_ERR "%s: Dropping NETIF_F_UFO "
+ "since no NETIF_F_HW_CSUM feature.\n",
+ name);
+ features &= ~NETIF_F_UFO;
+ }
+
+ if (!(features & NETIF_F_SG)) {
+ if (name)
+ printk(KERN_ERR "%s: Dropping NETIF_F_UFO "
+ "since no NETIF_F_SG feature.\n", name);
+ features &= ~NETIF_F_UFO;
+ }
+ }
+
+ return features;
+}
+EXPORT_SYMBOL(netdev_fix_features);
+
/**
* register_netdevice - register a network device
* @dev: device to register
@@ -3004,37 +3044,9 @@ int register_netdevice(struct net_device *dev)
ret = -EEXIST;
goto out_err;
}
- }
-
- /* Fix illegal SG+CSUM combinations. */
- if ((dev->features & NETIF_F_SG) &&
- !(dev->features & NETIF_F_ALL_CSUM)) {
- printk(KERN_NOTICE "%s: Dropping NETIF_F_SG since no checksum feature.\n",
- dev->name);
- dev->features &= ~NETIF_F_SG;
}
- /* TSO requires that SG is present as well. */
- if ((dev->features & NETIF_F_TSO) &&
- !(dev->features & NETIF_F_SG)) {
- printk(KERN_NOTICE "%s: Dropping NETIF_F_TSO since no SG feature.\n",
- dev->name);
- dev->features &= ~NETIF_F_TSO;
- }
- if (dev->features & NETIF_F_UFO) {
- if (!(dev->features & NETIF_F_HW_CSUM)) {
- printk(KERN_ERR "%s: Dropping NETIF_F_UFO since no "
- "NETIF_F_HW_CSUM feature.\n",
- dev->name);
- dev->features &= ~NETIF_F_UFO;
- }
- if (!(dev->features & NETIF_F_SG)) {
- printk(KERN_ERR "%s: Dropping NETIF_F_UFO since no "
- "NETIF_F_SG feature.\n",
- dev->name);
- dev->features &= ~NETIF_F_UFO;
- }
- }
+ dev->features = netdev_fix_features(dev->features, dev->name);
/*
* nil rebuild_header routine,
@@ -3598,48 +3610,45 @@ static int __init netdev_dma_register(void) { return -ENODEV; }
*/
/**
- * netdev_compute_feature - compute conjunction of two feature sets
- * @all: first feature set
- * @one: second feature set
+ * netdev_increment_features - increment feature set by one
+ * @all: current feature set
+ * @one: new feature set
+ * @mask: mask feature set
*
* Computes a new feature set after adding a device with feature set
- * @one to the master device with current feature set @all. Returns
- * the new feature set.
- */
-int netdev_compute_features(unsigned long all, unsigned long one)
-{
- /* if device needs checksumming, downgrade to hw checksumming */
- if (all & NETIF_F_NO_CSUM && !(one & NETIF_F_NO_CSUM))
- all ^= NETIF_F_NO_CSUM | NETIF_F_HW_CSUM;
-
- /* if device can't do all checksum, downgrade to ipv4/ipv6 */
- if (all & NETIF_F_HW_CSUM && !(one & NETIF_F_HW_CSUM))
- all ^= NETIF_F_HW_CSUM | NETIF_F_IP_CSUM;
-
- if (one & NETIF_F_GSO)
- one |= NETIF_F_GSO_SOFTWARE;
- one |= NETIF_F_GSO;
-
- /*
- * If even one device supports a GSO protocol with software fallback,
- * enable it for all.
- */
- all |= one & NETIF_F_GSO_SOFTWARE;
+ * @one to the master device with current feature set @all. Will not
+ * enable anything that is off in @mask. Returns the new feature set.
+ */
+unsigned long netdev_increment_features(unsigned long all, unsigned long one,
+ unsigned long mask)
+{
+ /* If device needs checksumming, downgrade to it. */
+ if (all & NETIF_F_NO_CSUM && !(one & NETIF_F_NO_CSUM))
+ all ^= NETIF_F_NO_CSUM | (one & NETIF_F_ALL_CSUM);
+ else if (mask & NETIF_F_ALL_CSUM) {
+ /* If one device supports v4/v6 checksumming, set for all. */
+ if (one & (NETIF_F_IP_CSUM) &&
+ !(all & NETIF_F_GEN_CSUM)) {
+ all &= ~NETIF_F_ALL_CSUM;
+ all |= one & (NETIF_F_IP_CSUM);
+ }
- /* If even one device supports robust GSO, enable it for all. */
- if (one & NETIF_F_GSO_ROBUST)
- all |= NETIF_F_GSO_ROBUST;
+ /* If one device supports hw checksumming, set for all. */
+ if (one & NETIF_F_GEN_CSUM && !(all & NETIF_F_GEN_CSUM)) {
+ all &= ~NETIF_F_ALL_CSUM;
+ all |= NETIF_F_HW_CSUM;
+ }
+ }
- all &= one | NETIF_F_LLTX;
+ one |= NETIF_F_ALL_CSUM;
- if (!(all & NETIF_F_ALL_CSUM))
- all &= ~NETIF_F_SG;
- if (!(all & NETIF_F_SG))
- all &= ~NETIF_F_GSO_MASK;
+ one |= all & NETIF_F_ONE_FOR_ALL;
+ all &= one | NETIF_F_LLTX | NETIF_F_GSO;
+ all |= one & mask & NETIF_F_ONE_FOR_ALL;
return all;
}
-EXPORT_SYMBOL(netdev_compute_features);
+EXPORT_SYMBOL(netdev_increment_features);
/*
* This is called single threaded during boot, so no need
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index 3ab3057..7d65841 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -446,6 +446,19 @@ static struct class net_class = {
* Delete sysfs entries but hold kobject reference until after all
* netdev references are gone.
*/
+int netdev_class_create_file(struct class_attribute *class_attr)
+{
+ return class_create_file(&net_class, class_attr);
+}
+
+void netdev_class_remove_file(struct class_attribute *class_attr)
+{
+ class_remove_file(&net_class, class_attr);
+}
+
+EXPORT_SYMBOL(netdev_class_create_file);
+EXPORT_SYMBOL(netdev_class_remove_file);
+
void netdev_unregister_sysfs(struct net_device * net)
{
struct class_device *dev = &(net->class_dev);
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index c16e097..1bfc483 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -425,6 +425,134 @@ static inline void ndisc_flow_init(struct flowi *fl, u8 type,
security_sk_classify_flow(ndisc_socket->sk, fl);
}
+struct sk_buff *ndisc_build_skb(struct net_device *dev,
+ struct in6_addr *daddr,
+ struct in6_addr *saddr,
+ struct icmp6hdr *icmp6h,
+ const struct in6_addr *target,
+ int llinfo)
+{
+ struct sock *sk = ndisc_socket->sk;
+ struct sk_buff *skb;
+ struct icmp6hdr *hdr;
+ int len;
+ int err;
+ u8 *opt;
+
+ if (!dev->addr_len)
+ llinfo = 0;
+
+ len = sizeof(struct icmp6hdr) + (target ? sizeof(*target) : 0);
+ if (llinfo)
+ len += ndisc_opt_addr_space(dev);
+
+ skb = sock_alloc_send_skb(sk,
+ (MAX_HEADER + sizeof(struct ipv6hdr) +
+ len + LL_RESERVED_SPACE(dev)),
+ 1, &err);
+ if (!skb) {
+ ND_PRINTK0(KERN_ERR
+ "ICMPv6 ND: %s() failed to allocate an skb.\n",
+ __FUNCTION__);
+ return NULL;
+ }
+
+ skb_reserve(skb, LL_RESERVED_SPACE(dev));
+ ip6_nd_hdr(sk, skb, dev, saddr, daddr, IPPROTO_ICMPV6, len);
+
+ skb->h.raw = skb->tail;
+ skb_put(skb, len);
+
+ hdr = (struct icmp6hdr *)skb_transport_header(skb);
+ memcpy(hdr, icmp6h, sizeof(*hdr));
+
+ opt = skb_transport_header(skb) + sizeof(struct icmp6hdr);
+ if (target) {
+ ipv6_addr_copy((struct in6_addr *)opt, target);
+ opt += sizeof(*target);
+ }
+
+ if (llinfo)
+ ndisc_fill_addr_option(opt, llinfo, dev->dev_addr,
+ dev->addr_len, dev->type);
+
+ hdr->icmp6_cksum = csum_ipv6_magic(saddr, daddr, len,
+ IPPROTO_ICMPV6,
+ csum_partial((__u8 *) hdr,
+ len, 0));
+
+ return skb;
+}
+
+EXPORT_SYMBOL(ndisc_build_skb);
+
+void ndisc_send_skb(struct sk_buff *skb,
+ struct net_device *dev,
+ struct neighbour *neigh,
+ struct in6_addr *daddr,
+ struct in6_addr *saddr,
+ struct icmp6hdr *icmp6h)
+{
+ struct flowi fl;
+ struct dst_entry *dst;
+ struct inet6_dev *idev;
+ int err;
+ u8 type;
+
+ type = icmp6h->icmp6_type;
+
+ ndisc_flow_init(&fl, icmp6h->icmp6_type, saddr, daddr, dev->ifindex);
+
+ dst = ndisc_dst_alloc(dev, neigh, daddr, ip6_output);
+ if (!dst) {
+ kfree_skb(skb);
+ return;
+ }
+
+ err = xfrm_lookup(&dst, &fl, NULL, 0);
+ if (err < 0) {
+ kfree_skb(skb);
+ return;
+ }
+
+ skb->dst = dst;
+
+ idev = in6_dev_get(dst->dev);
+ IP6_INC_STATS(idev, IPSTATS_MIB_OUTREQUESTS);
+
+ err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, dst->dev, dst_output);
+ if (!err) {
+ ICMP6_INC_STATS(idev, type);
+ ICMP6_INC_STATS(idev, ICMP6_MIB_OUTMSGS);
+ }
+
+ if (likely(idev != NULL))
+ in6_dev_put(idev);
+}
+
+
+
+EXPORT_SYMBOL(ndisc_send_skb);
+
+/*
+ * Send a Neighbour Discover packet
+ */
+static void __ndisc_send(struct net_device *dev,
+ struct neighbour *neigh,
+ struct in6_addr *daddr,
+ struct in6_addr *saddr,
+ struct icmp6hdr *icmp6h, const struct in6_addr *target,
+ int llinfo)
+{
+ struct sk_buff *skb;
+
+ skb = ndisc_build_skb(dev, daddr, saddr, icmp6h, target, llinfo);
+ if (!skb)
+ return;
+
+ ndisc_send_skb(skb, dev, neigh, daddr, saddr, icmp6h);
+}
+
static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh,
struct in6_addr *daddr, struct in6_addr *solicited_addr,
int router, int solicited, int override, int inc_opt)
distrib
>
Scientific%20Linux
>
5x
>
x86_64
>
by-pkgid
>
27922b4260f65d317aabda37e42bbbff
>
files
>
2501
