[NET]: Make the device list and device lookups per namespace.
This patch makes most of the generic device layer network
namespace safe. This patch makes dev_base_head a
network namespace variable, and then it picks up
a few associated variables. The functions:
dev_getbyhwaddr
dev_getfirsthwbytype
dev_get_by_flags
dev_get_by_name
__dev_get_by_name
dev_get_by_index
__dev_get_by_index
dev_ioctl
dev_ethtool
dev_load
wireless_process_ioctl
were modified to take a network namespace argument, and
deal with it.
vlan_ioctl_set and brioctl_set were modified so their
hooks will receive a network namespace argument.
So basically anthing in the core of the network stack that was
affected to by the change of dev_base was modified to handle
multiple network namespaces. The rest of the network stack was
simply modified to explicitly use &init_net the initial network
namespace. This can be fixed when those components of the network
stack are modified to handle multiple network namespaces.
For now the ifindex generator is left global.
Fundametally ifindex numbers are per namespace, or else
we will have corner case problems with migration when
we get that far.
At the same time there are assumptions in the network stack
that the ifindex of a network device won't change. Making
the ifindex number global seems a good compromise until
the network stack can cope with ifindex changes when
you change namespaces, and the like.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index a11e7a5..3a68300 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -981,7 +981,7 @@
if (mask && mask != htonl(0xFFFFFFFF))
return -EINVAL;
if (!dev && (r->arp_flags & ATF_COM)) {
- dev = dev_getbyhwaddr(r->arp_ha.sa_family, r->arp_ha.sa_data);
+ dev = dev_getbyhwaddr(&init_net, r->arp_ha.sa_family, r->arp_ha.sa_data);
if (!dev)
return -ENODEV;
}
@@ -1169,7 +1169,7 @@
rtnl_lock();
if (r.arp_dev[0]) {
err = -ENODEV;
- if ((dev = __dev_get_by_name(r.arp_dev)) == NULL)
+ if ((dev = __dev_get_by_name(&init_net, r.arp_dev)) == NULL)
goto out;
/* Mmmm... It is wrong... ARPHRD_NETROM==0 */
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index c5eb1a29..721b89b 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -420,7 +420,7 @@
struct net_device *dev;
struct in_device *in_dev = NULL;
read_lock(&dev_base_lock);
- dev = __dev_get_by_index(ifindex);
+ dev = __dev_get_by_index(&init_net, ifindex);
if (dev)
in_dev = in_dev_get(dev);
read_unlock(&dev_base_lock);
@@ -506,7 +506,7 @@
goto errout;
}
- dev = __dev_get_by_index(ifm->ifa_index);
+ dev = __dev_get_by_index(&init_net, ifm->ifa_index);
if (dev == NULL) {
err = -ENODEV;
goto errout;
@@ -628,7 +628,7 @@
*colon = 0;
#ifdef CONFIG_KMOD
- dev_load(ifr.ifr_name);
+ dev_load(&init_net, ifr.ifr_name);
#endif
switch (cmd) {
@@ -669,7 +669,7 @@
rtnl_lock();
ret = -ENODEV;
- if ((dev = __dev_get_by_name(ifr.ifr_name)) == NULL)
+ if ((dev = __dev_get_by_name(&init_net, ifr.ifr_name)) == NULL)
goto done;
if (colon)
@@ -909,7 +909,7 @@
*/
read_lock(&dev_base_lock);
rcu_read_lock();
- for_each_netdev(dev) {
+ for_each_netdev(&init_net, dev) {
if ((in_dev = __in_dev_get_rcu(dev)) == NULL)
continue;
@@ -988,7 +988,7 @@
read_lock(&dev_base_lock);
rcu_read_lock();
- for_each_netdev(dev) {
+ for_each_netdev(&init_net, dev) {
if ((in_dev = __in_dev_get_rcu(dev))) {
addr = confirm_addr_indev(in_dev, dst, local, scope);
if (addr)
@@ -1185,7 +1185,7 @@
s_ip_idx = ip_idx = cb->args[1];
idx = 0;
- for_each_netdev(dev) {
+ for_each_netdev(&init_net, dev) {
if (idx < s_idx)
goto cont;
if (idx > s_idx)
@@ -1244,7 +1244,7 @@
struct net_device *dev;
read_lock(&dev_base_lock);
- for_each_netdev(dev) {
+ for_each_netdev(&init_net, dev) {
struct in_device *in_dev;
rcu_read_lock();
in_dev = __in_dev_get_rcu(dev);
@@ -1333,7 +1333,7 @@
IPV4_DEVCONF_DFLT(FORWARDING) = on;
read_lock(&dev_base_lock);
- for_each_netdev(dev) {
+ for_each_netdev(&init_net, dev) {
struct in_device *in_dev;
rcu_read_lock();
in_dev = __in_dev_get_rcu(dev);
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 140bf7a..df17aab 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -334,7 +334,7 @@
colon = strchr(devname, ':');
if (colon)
*colon = 0;
- dev = __dev_get_by_name(devname);
+ dev = __dev_get_by_name(&init_net, devname);
if (!dev)
return -ENODEV;
cfg->fc_oif = dev->ifindex;
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index c434119..d30fb68 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -533,7 +533,7 @@
return -EINVAL;
if (inet_addr_type(nh->nh_gw) != RTN_UNICAST)
return -EINVAL;
- if ((dev = __dev_get_by_index(nh->nh_oif)) == NULL)
+ if ((dev = __dev_get_by_index(&init_net, nh->nh_oif)) == NULL)
return -ENODEV;
if (!(dev->flags&IFF_UP))
return -ENETDOWN;
@@ -799,7 +799,7 @@
if (nhs != 1 || nh->nh_gw)
goto err_inval;
nh->nh_scope = RT_SCOPE_NOWHERE;
- nh->nh_dev = dev_get_by_index(fi->fib_nh->nh_oif);
+ nh->nh_dev = dev_get_by_index(&init_net, fi->fib_nh->nh_oif);
err = -ENODEV;
if (nh->nh_dev == NULL)
goto failure;
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 02a899b..68a2267 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -517,7 +517,7 @@
struct net_device *dev = NULL;
if (rt->fl.iif && sysctl_icmp_errors_use_inbound_ifaddr)
- dev = dev_get_by_index(rt->fl.iif);
+ dev = dev_get_by_index(&init_net, rt->fl.iif);
if (dev) {
saddr = inet_select_addr(dev, 0, RT_SCOPE_LINK);
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index d78599a..ad500a4 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -2292,7 +2292,7 @@
struct igmp_mc_iter_state *state = igmp_mc_seq_private(seq);
state->in_dev = NULL;
- for_each_netdev(state->dev) {
+ for_each_netdev(&init_net, state->dev) {
struct in_device *in_dev;
in_dev = in_dev_get(state->dev);
if (!in_dev)
@@ -2454,7 +2454,7 @@
state->idev = NULL;
state->im = NULL;
- for_each_netdev(state->dev) {
+ for_each_netdev(&init_net, state->dev) {
struct in_device *idev;
idev = in_dev_get(state->dev);
if (unlikely(idev == NULL))
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 0231bdc..fabb86d 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -292,7 +292,7 @@
if ((qp->last_in&FIRST_IN) && qp->fragments != NULL) {
struct sk_buff *head = qp->fragments;
/* Send an ICMP "Fragment Reassembly Timeout" message. */
- if ((head->dev = dev_get_by_index(qp->iif)) != NULL) {
+ if ((head->dev = dev_get_by_index(&init_net, qp->iif)) != NULL) {
icmp_send(head, ICMP_TIME_EXCEEDED, ICMP_EXC_FRAGTIME, 0);
dev_put(head->dev);
}
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 5c14ed6..3106225 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -262,7 +262,7 @@
int i;
for (i=1; i<100; i++) {
sprintf(name, "gre%d", i);
- if (__dev_get_by_name(name) == NULL)
+ if (__dev_get_by_name(&init_net, name) == NULL)
break;
}
if (i==100)
@@ -1196,7 +1196,7 @@
}
if (!tdev && tunnel->parms.link)
- tdev = __dev_get_by_index(tunnel->parms.link);
+ tdev = __dev_get_by_index(&init_net, tunnel->parms.link);
if (tdev) {
hlen = tdev->hard_header_len;
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index 6b420ae..b2b3053 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -602,7 +602,7 @@
dev_put(dev);
}
} else
- dev = __dev_get_by_index(mreq.imr_ifindex);
+ dev = __dev_get_by_index(&init_net, mreq.imr_ifindex);
err = -EADDRNOTAVAIL;
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index 08ff623..4303851 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -193,7 +193,7 @@
if (dev_change_flags(&loopback_dev, loopback_dev.flags | IFF_UP) < 0)
printk(KERN_ERR "IP-Config: Failed to open %s\n", loopback_dev.name);
- for_each_netdev(dev) {
+ for_each_netdev(&init_net, dev) {
if (dev == &loopback_dev)
continue;
if (user_dev_name[0] ? !strcmp(dev->name, user_dev_name) :
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index 3964372..652bd86 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -225,7 +225,7 @@
int i;
for (i=1; i<100; i++) {
sprintf(name, "tunl%d", i);
- if (__dev_get_by_name(name) == NULL)
+ if (__dev_get_by_name(&init_net, name) == NULL)
break;
}
if (i==100)
@@ -822,7 +822,7 @@
}
if (!tdev && tunnel->parms.link)
- tdev = __dev_get_by_index(tunnel->parms.link);
+ tdev = __dev_get_by_index(&init_net, tunnel->parms.link);
if (tdev) {
dev->hard_header_len = tdev->hard_header_len + sizeof(struct iphdr);
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 0365988..b8b4b49 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -125,7 +125,7 @@
{
struct net_device *dev;
- dev = __dev_get_by_name("tunl0");
+ dev = __dev_get_by_name(&init_net, "tunl0");
if (dev) {
int err;
@@ -149,7 +149,7 @@
dev = NULL;
- if (err == 0 && (dev = __dev_get_by_name(p.name)) != NULL) {
+ if (err == 0 && (dev = __dev_get_by_name(&init_net, p.name)) != NULL) {
dev->flags |= IFF_MULTICAST;
in_dev = __in_dev_get_rtnl(dev);
diff --git a/net/ipv4/ipvs/ip_vs_sync.c b/net/ipv4/ipvs/ip_vs_sync.c
index 356f067..1960747 100644
--- a/net/ipv4/ipvs/ip_vs_sync.c
+++ b/net/ipv4/ipvs/ip_vs_sync.c
@@ -387,7 +387,7 @@
struct net_device *dev;
struct inet_sock *inet = inet_sk(sk);
- if ((dev = __dev_get_by_name(ifname)) == NULL)
+ if ((dev = __dev_get_by_name(&init_net, ifname)) == NULL)
return -ENODEV;
if (sk->sk_bound_dev_if && dev->ifindex != sk->sk_bound_dev_if)
@@ -412,7 +412,7 @@
int num;
if (sync_state == IP_VS_STATE_MASTER) {
- if ((dev = __dev_get_by_name(ip_vs_master_mcast_ifn)) == NULL)
+ if ((dev = __dev_get_by_name(&init_net, ip_vs_master_mcast_ifn)) == NULL)
return -ENODEV;
num = (dev->mtu - sizeof(struct iphdr) -
@@ -423,7 +423,7 @@
IP_VS_DBG(7, "setting the maximum length of sync sending "
"message %d.\n", sync_send_mesg_maxlen);
} else if (sync_state == IP_VS_STATE_BACKUP) {
- if ((dev = __dev_get_by_name(ip_vs_backup_mcast_ifn)) == NULL)
+ if ((dev = __dev_get_by_name(&init_net, ip_vs_backup_mcast_ifn)) == NULL)
return -ENODEV;
sync_recv_mesg_maxlen = dev->mtu -
@@ -451,7 +451,7 @@
memset(&mreq, 0, sizeof(mreq));
memcpy(&mreq.imr_multiaddr, addr, sizeof(struct in_addr));
- if ((dev = __dev_get_by_name(ifname)) == NULL)
+ if ((dev = __dev_get_by_name(&init_net, ifname)) == NULL)
return -ENODEV;
if (sk->sk_bound_dev_if && dev->ifindex != sk->sk_bound_dev_if)
return -EINVAL;
@@ -472,7 +472,7 @@
__be32 addr;
struct sockaddr_in sin;
- if ((dev = __dev_get_by_name(ifname)) == NULL)
+ if ((dev = __dev_get_by_name(&init_net, ifname)) == NULL)
return -ENODEV;
addr = inet_select_addr(dev, 0, RT_SCOPE_UNIVERSE);
diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c
index 50fc9e0..27f14e1 100644
--- a/net/ipv4/netfilter/ipt_CLUSTERIP.c
+++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c
@@ -401,7 +401,7 @@
return false;
}
- dev = dev_get_by_name(e->ip.iniface);
+ dev = dev_get_by_name(&init_net, e->ip.iniface);
if (!dev) {
printk(KERN_WARNING "CLUSTERIP: no such interface %s\n", e->ip.iniface);
return false;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index efd2a92..396c631 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -2213,7 +2213,7 @@
if (oldflp->oif) {
- dev_out = dev_get_by_index(oldflp->oif);
+ dev_out = dev_get_by_index(&init_net, oldflp->oif);
err = -ENODEV;
if (dev_out == NULL)
goto out;
@@ -2592,7 +2592,7 @@
if (iif) {
struct net_device *dev;
- dev = __dev_get_by_index(iif);
+ dev = __dev_get_by_index(&init_net, iif);
if (dev == NULL) {
err = -ENODEV;
goto errout_free;