linux_dsm_epyc7002/net/bridge/br_if.c
Eric Dumazet ec1e5610c0 bridge: add RCU annotations to bridge port lookup
br_port_get() renamed to br_port_get_rtnl() to make clear RTNL is held.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2010-11-15 11:13:18 -08:00

506 lines
10 KiB
C

/*
* Userspace interface
* Linux ethernet bridge
*
* Authors:
* Lennert Buytenhek <buytenh@gnu.org>
*
* 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.
*/
#include <linux/kernel.h>
#include <linux/netdevice.h>
#include <linux/netpoll.h>
#include <linux/ethtool.h>
#include <linux/if_arp.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/rtnetlink.h>
#include <linux/if_ether.h>
#include <linux/slab.h>
#include <net/sock.h>
#include "br_private.h"
/*
* Determine initial path cost based on speed.
* using recommendations from 802.1d standard
*
* Since driver might sleep need to not be holding any locks.
*/
static int port_cost(struct net_device *dev)
{
if (dev->ethtool_ops && dev->ethtool_ops->get_settings) {
struct ethtool_cmd ecmd = { .cmd = ETHTOOL_GSET, };
if (!dev->ethtool_ops->get_settings(dev, &ecmd)) {
switch(ecmd.speed) {
case SPEED_10000:
return 2;
case SPEED_1000:
return 4;
case SPEED_100:
return 19;
case SPEED_10:
return 100;
}
}
}
/* Old silly heuristics based on name */
if (!strncmp(dev->name, "lec", 3))
return 7;
if (!strncmp(dev->name, "plip", 4))
return 2500;
return 100; /* assume old 10Mbps */
}
/* Check for port carrier transistions. */
void br_port_carrier_check(struct net_bridge_port *p)
{
struct net_device *dev = p->dev;
struct net_bridge *br = p->br;
if (netif_running(dev) && netif_carrier_ok(dev))
p->path_cost = port_cost(dev);
if (!netif_running(br->dev))
return;
spin_lock_bh(&br->lock);
if (netif_running(dev) && netif_carrier_ok(dev)) {
if (p->state == BR_STATE_DISABLED)
br_stp_enable_port(p);
} else {
if (p->state != BR_STATE_DISABLED)
br_stp_disable_port(p);
}
spin_unlock_bh(&br->lock);
}
static void release_nbp(struct kobject *kobj)
{
struct net_bridge_port *p
= container_of(kobj, struct net_bridge_port, kobj);
kfree(p);
}
static struct kobj_type brport_ktype = {
#ifdef CONFIG_SYSFS
.sysfs_ops = &brport_sysfs_ops,
#endif
.release = release_nbp,
};
static void destroy_nbp(struct net_bridge_port *p)
{
struct net_device *dev = p->dev;
p->br = NULL;
p->dev = NULL;
dev_put(dev);
kobject_put(&p->kobj);
}
static void destroy_nbp_rcu(struct rcu_head *head)
{
struct net_bridge_port *p =
container_of(head, struct net_bridge_port, rcu);
destroy_nbp(p);
}
/* Delete port(interface) from bridge is done in two steps.
* via RCU. First step, marks device as down. That deletes
* all the timers and stops new packets from flowing through.
*
* Final cleanup doesn't occur until after all CPU's finished
* processing packets.
*
* Protected from multiple admin operations by RTNL mutex
*/
static void del_nbp(struct net_bridge_port *p)
{
struct net_bridge *br = p->br;
struct net_device *dev = p->dev;
sysfs_remove_link(br->ifobj, p->dev->name);
dev_set_promiscuity(dev, -1);
spin_lock_bh(&br->lock);
br_stp_disable_port(p);
spin_unlock_bh(&br->lock);
br_ifinfo_notify(RTM_DELLINK, p);
br_fdb_delete_by_port(br, p, 1);
list_del_rcu(&p->list);
dev->priv_flags &= ~IFF_BRIDGE_PORT;
netdev_rx_handler_unregister(dev);
br_multicast_del_port(p);
kobject_uevent(&p->kobj, KOBJ_REMOVE);
kobject_del(&p->kobj);
br_netpoll_disable(p);
call_rcu(&p->rcu, destroy_nbp_rcu);
}
/* called with RTNL */
static void del_br(struct net_bridge *br, struct list_head *head)
{
struct net_bridge_port *p, *n;
list_for_each_entry_safe(p, n, &br->port_list, list) {
del_nbp(p);
}
del_timer_sync(&br->gc_timer);
br_sysfs_delbr(br->dev);
unregister_netdevice_queue(br->dev, head);
}
static struct net_device *new_bridge_dev(struct net *net, const char *name)
{
struct net_bridge *br;
struct net_device *dev;
dev = alloc_netdev(sizeof(struct net_bridge), name,
br_dev_setup);
if (!dev)
return NULL;
dev_net_set(dev, net);
br = netdev_priv(dev);
br->dev = dev;
br->stats = alloc_percpu(struct br_cpu_netstats);
if (!br->stats) {
free_netdev(dev);
return NULL;
}
spin_lock_init(&br->lock);
INIT_LIST_HEAD(&br->port_list);
spin_lock_init(&br->hash_lock);
br->bridge_id.prio[0] = 0x80;
br->bridge_id.prio[1] = 0x00;
memcpy(br->group_addr, br_group_address, ETH_ALEN);
br->feature_mask = dev->features;
br->stp_enabled = BR_NO_STP;
br->designated_root = br->bridge_id;
br->root_path_cost = 0;
br->root_port = 0;
br->bridge_max_age = br->max_age = 20 * HZ;
br->bridge_hello_time = br->hello_time = 2 * HZ;
br->bridge_forward_delay = br->forward_delay = 15 * HZ;
br->topology_change = 0;
br->topology_change_detected = 0;
br->ageing_time = 300 * HZ;
br_netfilter_rtable_init(br);
br_stp_timer_init(br);
br_multicast_init(br);
return dev;
}
/* find an available port number */
static int find_portno(struct net_bridge *br)
{
int index;
struct net_bridge_port *p;
unsigned long *inuse;
inuse = kcalloc(BITS_TO_LONGS(BR_MAX_PORTS), sizeof(unsigned long),
GFP_KERNEL);
if (!inuse)
return -ENOMEM;
set_bit(0, inuse); /* zero is reserved */
list_for_each_entry(p, &br->port_list, list) {
set_bit(p->port_no, inuse);
}
index = find_first_zero_bit(inuse, BR_MAX_PORTS);
kfree(inuse);
return (index >= BR_MAX_PORTS) ? -EXFULL : index;
}
/* called with RTNL but without bridge lock */
static struct net_bridge_port *new_nbp(struct net_bridge *br,
struct net_device *dev)
{
int index;
struct net_bridge_port *p;
index = find_portno(br);
if (index < 0)
return ERR_PTR(index);
p = kzalloc(sizeof(*p), GFP_KERNEL);
if (p == NULL)
return ERR_PTR(-ENOMEM);
p->br = br;
dev_hold(dev);
p->dev = dev;
p->path_cost = port_cost(dev);
p->priority = 0x8000 >> BR_PORT_BITS;
p->port_no = index;
p->flags = 0;
br_init_port(p);
p->state = BR_STATE_DISABLED;
br_stp_port_timer_init(p);
br_multicast_add_port(p);
return p;
}
static struct device_type br_type = {
.name = "bridge",
};
int br_add_bridge(struct net *net, const char *name)
{
struct net_device *dev;
int ret;
dev = new_bridge_dev(net, name);
if (!dev)
return -ENOMEM;
rtnl_lock();
if (strchr(dev->name, '%')) {
ret = dev_alloc_name(dev, dev->name);
if (ret < 0)
goto out_free;
}
SET_NETDEV_DEVTYPE(dev, &br_type);
ret = register_netdevice(dev);
if (ret)
goto out_free;
ret = br_sysfs_addbr(dev);
if (ret)
unregister_netdevice(dev);
out:
rtnl_unlock();
return ret;
out_free:
free_netdev(dev);
goto out;
}
int br_del_bridge(struct net *net, const char *name)
{
struct net_device *dev;
int ret = 0;
rtnl_lock();
dev = __dev_get_by_name(net, name);
if (dev == NULL)
ret = -ENXIO; /* Could not find device */
else if (!(dev->priv_flags & IFF_EBRIDGE)) {
/* Attempt to delete non bridge device! */
ret = -EPERM;
}
else if (dev->flags & IFF_UP) {
/* Not shutdown yet. */
ret = -EBUSY;
}
else
del_br(netdev_priv(dev), NULL);
rtnl_unlock();
return ret;
}
/* MTU of the bridge pseudo-device: ETH_DATA_LEN or the minimum of the ports */
int br_min_mtu(const struct net_bridge *br)
{
const struct net_bridge_port *p;
int mtu = 0;
ASSERT_RTNL();
if (list_empty(&br->port_list))
mtu = ETH_DATA_LEN;
else {
list_for_each_entry(p, &br->port_list, list) {
if (!mtu || p->dev->mtu < mtu)
mtu = p->dev->mtu;
}
}
return mtu;
}
/*
* Recomputes features using slave's features
*/
void br_features_recompute(struct net_bridge *br)
{
struct net_bridge_port *p;
unsigned long features, 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_increment_features(features,
p->dev->features, mask);
}
done:
br->dev->features = netdev_fix_features(features, NULL);
}
/* called with RTNL */
int br_add_if(struct net_bridge *br, struct net_device *dev)
{
struct net_bridge_port *p;
int err = 0;
/* Don't allow bridging non-ethernet like devices */
if ((dev->flags & IFF_LOOPBACK) ||
dev->type != ARPHRD_ETHER || dev->addr_len != ETH_ALEN)
return -EINVAL;
/* No bridging of bridges */
if (dev->netdev_ops->ndo_start_xmit == br_dev_xmit)
return -ELOOP;
/* Device is already being bridged */
if (br_port_exists(dev))
return -EBUSY;
/* No bridging devices that dislike that (e.g. wireless) */
if (dev->priv_flags & IFF_DONT_BRIDGE)
return -EOPNOTSUPP;
p = new_nbp(br, dev);
if (IS_ERR(p))
return PTR_ERR(p);
err = dev_set_promiscuity(dev, 1);
if (err)
goto put_back;
err = kobject_init_and_add(&p->kobj, &brport_ktype, &(dev->dev.kobj),
SYSFS_BRIDGE_PORT_ATTR);
if (err)
goto err0;
err = br_fdb_insert(br, p, dev->dev_addr);
if (err)
goto err1;
err = br_sysfs_addif(p);
if (err)
goto err2;
if (br_netpoll_info(br) && ((err = br_netpoll_enable(p))))
goto err3;
err = netdev_rx_handler_register(dev, br_handle_frame, p);
if (err)
goto err3;
dev->priv_flags |= IFF_BRIDGE_PORT;
dev_disable_lro(dev);
list_add_rcu(&p->list, &br->port_list);
spin_lock_bh(&br->lock);
br_stp_recalculate_bridge_id(br);
br_features_recompute(br);
if ((dev->flags & IFF_UP) && netif_carrier_ok(dev) &&
(br->dev->flags & IFF_UP))
br_stp_enable_port(p);
spin_unlock_bh(&br->lock);
br_ifinfo_notify(RTM_NEWLINK, p);
dev_set_mtu(br->dev, br_min_mtu(br));
kobject_uevent(&p->kobj, KOBJ_ADD);
return 0;
err3:
sysfs_remove_link(br->ifobj, p->dev->name);
err2:
br_fdb_delete_by_port(br, p, 1);
err1:
kobject_put(&p->kobj);
p = NULL; /* kobject_put frees */
err0:
dev_set_promiscuity(dev, -1);
put_back:
dev_put(dev);
kfree(p);
return err;
}
/* called with RTNL */
int br_del_if(struct net_bridge *br, struct net_device *dev)
{
struct net_bridge_port *p;
p = br_port_get_rtnl(dev);
if (!p || p->br != br)
return -EINVAL;
del_nbp(p);
spin_lock_bh(&br->lock);
br_stp_recalculate_bridge_id(br);
br_features_recompute(br);
spin_unlock_bh(&br->lock);
return 0;
}
void __net_exit br_net_exit(struct net *net)
{
struct net_device *dev;
LIST_HEAD(list);
rtnl_lock();
for_each_netdev(net, dev)
if (dev->priv_flags & IFF_EBRIDGE)
del_br(netdev_priv(dev), &list);
unregister_netdevice_many(&list);
rtnl_unlock();
}