This patch makes loopback_dev per network namespace. Adding
code to create a different loopback device for each network
namespace and adding the code to free a loopback device
when a network namespace exits.
This patch modifies all users the loopback_dev so they
access it as init_net.loopback_dev, keeping all of the
code compiling and working. A later pass will be needed to
update the users to use something other than the initial network
namespace.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
drivers/net/loopback.c | 26 +++++++++++++++++++++-----
include/linux/netdevice.h | 1 -
include/net/net_namespace.h | 3 +++
net/core/dst.c | 5 +++--
net/decnet/dn_dev.c | 4 ++--
net/decnet/dn_route.c | 14 +++++++-------
net/ipv4/route.c | 18 +++++++++---------
net/ipv4/xfrm4_policy.c | 2 +-
net/ipv6/addrconf.c | 18 +++++++++---------
net/ipv6/netfilter/ip6t_REJECT.c | 2 +-
net/ipv6/route.c | 12 ++++++------
net/ipv6/xfrm6_policy.c | 2 +-
net/xfrm/xfrm_policy.c | 2 +-
13 files changed, 64 insertions(+), 45 deletions(-)
diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c
index f3018bb..0f9d8c6 100644
--- a/drivers/net/loopback.c
+++ b/drivers/net/loopback.c
@@ -57,6 +57,7 @@
#include <linux/ip.h>
#include <linux/tcp.h>
#include <linux/percpu.h>
+#include <net/net_namespace.h>
struct pcpu_lstats {
unsigned long packets;
@@ -252,7 +253,7 @@ static void loopback_setup(struct net_device *dev)
}
/* Setup and register the loopback device. */
-static int __init loopback_init(void)
+static int loopback_net_init(struct net *net)
{
struct net_device *dev;
int err;
@@ -262,12 +263,13 @@ static int __init loopback_init(void)
if (!dev)
goto out;
+ dev->nd_net = net;
err = register_netdev(dev);
if (err)
goto out_free_netdev;
err = 0;
- loopback_dev = dev;
+ net->loopback_dev = dev;
out:
if (err)
@@ -279,7 +281,21 @@ out_free_netdev:
goto out;
}
-fs_initcall(loopback_init);
+static void loopback_net_exit(struct net *net)
+{
+ struct net_device *dev = net->loopback_dev;
+
+ unregister_netdev(dev);
+}
+
+static struct pernet_operations loopback_net_ops = {
+ .init = loopback_net_init,
+ .exit = loopback_net_exit,
+};
+
+static int __init loopback_init(void)
+{
+ return register_pernet_device(&loopback_net_ops);
+}
-struct net_device *loopback_dev;
-EXPORT_SYMBOL(loopback_dev);
+fs_initcall(loopback_init);
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 2088097..71cf409 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -742,7 +742,6 @@ struct packet_type {
#include <linux/interrupt.h>
#include <linux/notifier.h>
-extern struct net_device *loopback_dev; /* The loopback */
extern rwlock_t dev_base_lock; /* Device list lock */
diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index 3ea4194..13b0e3b 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -9,6 +9,7 @@
#include <linux/list.h>
struct proc_dir_entry;
+struct net_device;
struct net {
atomic_t count; /* To decided when the network
* namespace should be freed.
@@ -23,6 +24,8 @@ struct net {
struct proc_dir_entry *proc_net_stat;
struct proc_dir_entry *proc_net_root;
+ struct net_device *loopback_dev; /* The loopback */
+
struct list_head dev_base_head;
struct hlist_head *dev_name_head;
struct hlist_head *dev_index_head;
diff --git a/net/core/dst.c b/net/core/dst.c
index ad5ffa1..16958e6 100644
--- a/net/core/dst.c
+++ b/net/core/dst.c
@@ -18,6 +18,7 @@
#include <linux/types.h>
#include <net/net_namespace.h>
+#include <net/net_namespace.h>
#include <net/dst.h>
/*
@@ -278,11 +279,11 @@ static inline void dst_ifdown(struct dst_entry *dst, struct net_device *dev,
if (!unregister) {
dst->input = dst->output = dst_discard;
} else {
- dst->dev = loopback_dev;
+ dst->dev = init_net.loopback_dev;
dev_hold(dst->dev);
dev_put(dev);
if (dst->neighbour && dst->neighbour->dev == dev) {
- dst->neighbour->dev = loopback_dev;
+ dst->neighbour->dev = init_net.loopback_dev;
dev_put(dev);
dev_hold(dst->neighbour->dev);
}
diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c
index bcaf4c5..26130af 100644
--- a/net/decnet/dn_dev.c
+++ b/net/decnet/dn_dev.c
@@ -869,10 +869,10 @@ last_chance:
rv = dn_dev_get_first(dev, addr);
read_unlock(&dev_base_lock);
dev_put(dev);
- if (rv == 0 || dev == loopback_dev)
+ if (rv == 0 || dev == init_net.loopback_dev)
return rv;
}
- dev = loopback_dev;
+ dev = init_net.loopback_dev;
dev_hold(dev);
goto last_chance;
}
diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c
index 96fe0aa..b7ebf99 100644
--- a/net/decnet/dn_route.c
+++ b/net/decnet/dn_route.c
@@ -887,7 +887,7 @@ static int dn_route_output_slow(struct dst_entry **pprt, const struct flowi *old
.scope = RT_SCOPE_UNIVERSE,
} },
.mark = oldflp->mark,
- .iif = loopback_dev->ifindex,
+ .iif = init_net.loopback_dev->ifindex,
.oif = oldflp->oif };
struct dn_route *rt = NULL;
struct net_device *dev_out = NULL, *dev;
@@ -904,7 +904,7 @@ static int dn_route_output_slow(struct dst_entry **pprt, const struct flowi *old
"dn_route_output_slow: dst=%04x src=%04x mark=%d"
" iif=%d oif=%d\n", dn_ntohs(oldflp->fld_dst),
dn_ntohs(oldflp->fld_src),
- oldflp->mark, loopback_dev->ifindex, oldflp->oif);
+ oldflp->mark, init_net.loopback_dev->ifindex, oldflp->oif);
/* If we have an output interface, verify its a DECnet device */
if (oldflp->oif) {
@@ -957,7 +957,7 @@ source_ok:
err = -EADDRNOTAVAIL;
if (dev_out)
dev_put(dev_out);
- dev_out = loopback_dev;
+ dev_out = init_net.loopback_dev;
dev_hold(dev_out);
if (!fl.fld_dst) {
fl.fld_dst =
@@ -966,7 +966,7 @@ source_ok:
if (!fl.fld_dst)
goto out;
}
- fl.oif = loopback_dev->ifindex;
+ fl.oif = init_net.loopback_dev->ifindex;
res.type = RTN_LOCAL;
goto make_route;
}
@@ -1012,7 +1012,7 @@ source_ok:
if (dev_out)
dev_put(dev_out);
if (dn_dev_islocal(neigh->dev, fl.fld_dst)) {
- dev_out = loopback_dev;
+ dev_out = init_net.loopback_dev;
res.type = RTN_LOCAL;
} else {
dev_out = neigh->dev;
@@ -1033,7 +1033,7 @@ source_ok:
/* Possible improvement - check all devices for local addr */
if (dn_dev_islocal(dev_out, fl.fld_dst)) {
dev_put(dev_out);
- dev_out = loopback_dev;
+ dev_out = init_net.loopback_dev;
dev_hold(dev_out);
res.type = RTN_LOCAL;
goto select_source;
@@ -1069,7 +1069,7 @@ select_source:
fl.fld_src = fl.fld_dst;
if (dev_out)
dev_put(dev_out);
- dev_out = loopback_dev;
+ dev_out = init_net.loopback_dev;
dev_hold(dev_out);
fl.oif = dev_out->ifindex;
if (res.fi)
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index ca2878d..2a9b363 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1402,8 +1402,8 @@ static void ipv4_dst_ifdown(struct dst_entry *dst, struct net_device *dev,
{
struct rtable *rt = (struct rtable *) dst;
struct in_device *idev = rt->idev;
- if (dev != loopback_dev && idev && idev->dev == dev) {
- struct in_device *loopback_idev = in_dev_get(loopback_dev);
+ if (dev != init_net.loopback_dev && idev && idev->dev == dev) {
+ struct in_device *loopback_idev = in_dev_get(init_net.loopback_dev);
if (loopback_idev) {
rt->idev = loopback_idev;
in_dev_put(idev);
@@ -1555,7 +1555,7 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr,
#endif
rth->rt_iif =
rth->fl.iif = dev->ifindex;
- rth->u.dst.dev = loopback_dev;
+ rth->u.dst.dev = init_net.loopback_dev;
dev_hold(rth->u.dst.dev);
rth->idev = in_dev_get(rth->u.dst.dev);
rth->fl.oif = 0;
@@ -1812,7 +1812,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr,
if (res.type == RTN_LOCAL) {
int result;
result = fib_validate_source(saddr, daddr, tos,
- loopback_dev->ifindex,
+ init_net.loopback_dev->ifindex,
dev, &spec_dst, &itag);
if (result < 0)
goto martian_source;
@@ -1879,7 +1879,7 @@ local_input:
#endif
rth->rt_iif =
rth->fl.iif = dev->ifindex;
- rth->u.dst.dev = loopback_dev;
+ rth->u.dst.dev = init_net.loopback_dev;
dev_hold(rth->u.dst.dev);
rth->idev = in_dev_get(rth->u.dst.dev);
rth->rt_gateway = daddr;
@@ -2149,7 +2149,7 @@ static int ip_route_output_slow(struct rtable **rp, const struct flowi *oldflp)
RT_SCOPE_UNIVERSE),
} },
.mark = oldflp->mark,
- .iif = loopback_dev->ifindex,
+ .iif = init_net.loopback_dev->ifindex,
.oif = oldflp->oif };
struct fib_result res;
unsigned flags = 0;
@@ -2243,9 +2243,9 @@ static int ip_route_output_slow(struct rtable **rp, const struct flowi *oldflp)
fl.fl4_dst = fl.fl4_src = htonl(INADDR_LOOPBACK);
if (dev_out)
dev_put(dev_out);
- dev_out = loopback_dev;
+ dev_out = init_net.loopback_dev;
dev_hold(dev_out);
- fl.oif = loopback_dev->ifindex;
+ fl.oif = init_net.loopback_dev->ifindex;
res.type = RTN_LOCAL;
flags |= RTCF_LOCAL;
goto make_route;
@@ -2290,7 +2290,7 @@ static int ip_route_output_slow(struct rtable **rp, const struct flowi *oldflp)
fl.fl4_src = fl.fl4_dst;
if (dev_out)
dev_put(dev_out);
- dev_out = loopback_dev;
+ dev_out = init_net.loopback_dev;
dev_hold(dev_out);
fl.oif = dev_o
...