#include <linux/route.h>
#include <linux/netdevice.h>
#include <linux/in6.h>
+#include <linux/mroute6.h>
#include <linux/init.h>
#include <linux/if_arp.h>
#include <linux/proc_fs.h>
struct in6_addr *gwaddr, int ifindex);
#endif
-static struct dst_ops ip6_dst_ops = {
+static struct dst_ops ip6_dst_ops_template = {
.family = AF_INET6,
.protocol = __constant_htons(ETH_P_IPV6),
.gc = ip6_dst_gc,
.metrics = { [RTAX_HOPLIMIT - 1] = 255, },
.input = ip6_pkt_discard,
.output = ip6_pkt_discard_out,
- .ops = &ip6_dst_ops,
}
},
.rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP),
.metrics = { [RTAX_HOPLIMIT - 1] = 255, },
.input = ip6_pkt_prohibit,
.output = ip6_pkt_prohibit_out,
- .ops = &ip6_dst_ops,
}
},
.rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP),
.metrics = { [RTAX_HOPLIMIT - 1] = 255, },
.input = dst_discard,
.output = dst_discard,
- .ops = &ip6_dst_ops,
}
},
.rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP),
#endif
/* allocate dst with ip6_dst_ops */
-static __inline__ struct rt6_info *ip6_dst_alloc(void)
+static inline struct rt6_info *ip6_dst_alloc(struct dst_ops *ops)
{
- return (struct rt6_info *)dst_alloc(&ip6_dst_ops);
+ return (struct rt6_info *)dst_alloc(ops);
}
static void ip6_dst_destroy(struct dst_entry *dst)
struct rt6_info *rt = (struct rt6_info *)dst;
struct inet6_dev *idev = rt->rt6i_idev;
struct net_device *loopback_dev =
- dev->nd_net->loopback_dev;
+ dev_net(dev)->loopback_dev;
if (dev != loopback_dev && idev != NULL && idev->dev == dev) {
struct inet6_dev *loopback_idev =
struct net *net;
RT6_TRACE("%s(fn->leaf=%p, oif=%d)\n",
- __FUNCTION__, fn->leaf, oif);
+ __func__, fn->leaf, oif);
rt0 = fn->rr_ptr;
if (!rt0)
}
RT6_TRACE("%s() => %p\n",
- __FUNCTION__, match);
+ __func__, match);
- net = rt0->rt6i_dev->nd_net;
+ net = dev_net(rt0->rt6i_dev);
return (match ? match : net->ipv6.ip6_null_entry);
}
int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
struct in6_addr *gwaddr)
{
- struct net *net = dev->nd_net;
+ struct net *net = dev_net(dev);
struct route_info *rinfo = (struct route_info *) opt;
struct in6_addr prefix_buf, *prefix;
unsigned int pref;
}
-struct rt6_info *rt6_lookup(struct net *net, struct in6_addr *daddr,
- struct in6_addr *saddr, int oif, int strict)
+struct rt6_info *rt6_lookup(struct net *net, const struct in6_addr *daddr,
+ const struct in6_addr *saddr, int oif, int strict)
{
struct flowi fl = {
.oif = oif,
int ip6_ins_rt(struct rt6_info *rt)
{
struct nl_info info = {
- .nl_net = rt->rt6i_dev->nd_net,
+ .nl_net = dev_net(rt->rt6i_dev),
};
return __ip6_ins_rt(rt, &info);
}
void ip6_route_input(struct sk_buff *skb)
{
struct ipv6hdr *iph = ipv6_hdr(skb);
- struct net *net = skb->dev->nd_net;
+ struct net *net = dev_net(skb->dev);
int flags = RT6_LOOKUP_F_HAS_SADDR;
struct flowi fl = {
.iif = skb->dev->ifindex,
return ip6_pol_route(net, table, fl->oif, fl, flags);
}
-struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl)
+struct dst_entry * ip6_route_output(struct net *net, struct sock *sk,
+ struct flowi *fl)
{
int flags = 0;
if (!ipv6_addr_any(&fl->fl6_src))
flags |= RT6_LOOKUP_F_HAS_SADDR;
+ else if (sk) {
+ unsigned int prefs = inet6_sk(sk)->srcprefs;
+ if (prefs & IPV6_PREFER_SRC_TMP)
+ flags |= RT6_LOOKUP_F_SRCPREF_TMP;
+ if (prefs & IPV6_PREFER_SRC_PUBLIC)
+ flags |= RT6_LOOKUP_F_SRCPREF_PUBLIC;
+ if (prefs & IPV6_PREFER_SRC_COA)
+ flags |= RT6_LOOKUP_F_SRCPREF_COA;
+ }
- return fib6_rule_lookup(&init_net, fl, flags, ip6_pol_route_output);
+ return fib6_rule_lookup(net, fl, flags, ip6_pol_route_output);
}
EXPORT_SYMBOL(ip6_route_output);
struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
struct neighbour *neigh,
- struct in6_addr *addr)
+ const struct in6_addr *addr)
{
struct rt6_info *rt;
struct inet6_dev *idev = in6_dev_get(dev);
- struct net *net = dev->nd_net;
+ struct net *net = dev_net(dev);
if (unlikely(idev == NULL))
return NULL;
- rt = ip6_dst_alloc();
+ rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops);
if (unlikely(rt == NULL)) {
in6_dev_put(idev);
goto out;
static int ip6_dst_gc(struct dst_ops *ops)
{
- static unsigned expire = 30*HZ;
- static unsigned long last_gc;
unsigned long now = jiffies;
-
- if (time_after(last_gc + init_net.ipv6.sysctl.ip6_rt_gc_min_interval, now) &&
- atomic_read(&ip6_dst_ops.entries) <= init_net.ipv6.sysctl.ip6_rt_max_size)
+ struct net *net = ops->dst_net;
+ int rt_min_interval = net->ipv6.sysctl.ip6_rt_gc_min_interval;
+ int rt_max_size = net->ipv6.sysctl.ip6_rt_max_size;
+ int rt_elasticity = net->ipv6.sysctl.ip6_rt_gc_elasticity;
+ int rt_gc_timeout = net->ipv6.sysctl.ip6_rt_gc_timeout;
+ unsigned long rt_last_gc = net->ipv6.ip6_rt_last_gc;
+
+ if (time_after(rt_last_gc + rt_min_interval, now) &&
+ atomic_read(&ops->entries) <= rt_max_size)
goto out;
- expire++;
- fib6_run_gc(expire, &init_net);
- last_gc = now;
- if (atomic_read(&ip6_dst_ops.entries) < ip6_dst_ops.gc_thresh)
- expire = init_net.ipv6.sysctl.ip6_rt_gc_timeout>>1;
-
+ net->ipv6.ip6_rt_gc_expire++;
+ fib6_run_gc(net->ipv6.ip6_rt_gc_expire, net);
+ net->ipv6.ip6_rt_last_gc = now;
+ if (atomic_read(&ops->entries) < ops->gc_thresh)
+ net->ipv6.ip6_rt_gc_expire = rt_gc_timeout>>1;
out:
- expire -= expire>>init_net.ipv6.sysctl.ip6_rt_gc_elasticity;
- return (atomic_read(&ip6_dst_ops.entries) > init_net.ipv6.sysctl.ip6_rt_max_size);
+ net->ipv6.ip6_rt_gc_expire -= net->ipv6.ip6_rt_gc_expire>>rt_elasticity;
+ return (atomic_read(&ops->entries) > rt_max_size);
}
/* Clean host part of a prefix. Not necessary in radix tree,
return mtu;
}
-int ipv6_get_hoplimit(struct net_device *dev)
+int ip6_dst_hoplimit(struct dst_entry *dst)
{
- int hoplimit = ipv6_devconf.hop_limit;
- struct inet6_dev *idev;
-
- idev = in6_dev_get(dev);
- if (idev) {
- hoplimit = idev->cnf.hop_limit;
- in6_dev_put(idev);
+ int hoplimit = dst_metric(dst, RTAX_HOPLIMIT);
+ if (hoplimit < 0) {
+ struct net_device *dev = dst->dev;
+ struct inet6_dev *idev = in6_dev_get(dev);
+ if (idev) {
+ hoplimit = idev->cnf.hop_limit;
+ in6_dev_put(idev);
+ } else
+ hoplimit = ipv6_devconf.hop_limit;
}
return hoplimit;
}
goto out;
}
- rt = ip6_dst_alloc();
+ rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops);
if (rt == NULL) {
err = -ENOMEM;
rt->rt6i_idev = idev;
rt->rt6i_table = table;
- cfg->fc_nlinfo.nl_net = dev->nd_net;
+ cfg->fc_nlinfo.nl_net = dev_net(dev);
return __ip6_ins_rt(rt, &cfg->fc_nlinfo);
{
int err;
struct fib6_table *table;
- struct net *net = rt->rt6i_dev->nd_net;
+ struct net *net = dev_net(rt->rt6i_dev);
if (rt == net->ipv6.ip6_null_entry)
return -ENOENT;
int ip6_del_rt(struct rt6_info *rt)
{
struct nl_info info = {
- .nl_net = rt->rt6i_dev->nd_net,
+ .nl_net = dev_net(rt->rt6i_dev),
};
return __ip6_del_rt(rt, &info);
}
struct net_device *dev)
{
int flags = RT6_LOOKUP_F_HAS_SADDR;
- struct net *net = dev->nd_net;
+ struct net *net = dev_net(dev);
struct ip6rd_flowi rdfl = {
.fl = {
.oif = dev->ifindex,
{
struct rt6_info *rt, *nrt = NULL;
struct netevent_redirect netevent;
- struct net *net = neigh->dev->nd_net;
+ struct net *net = dev_net(neigh->dev);
rt = ip6_route_redirect(dest, src, saddr, neigh->dev);
nrt->rt6i_nexthop = neigh_clone(neigh);
/* Reset pmtu, it may be better */
nrt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(neigh->dev);
- nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(neigh->dev->nd_net,
+ nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dev_net(neigh->dev),
dst_mtu(&nrt->u.dst));
if (ip6_ins_rt(nrt))
struct net_device *dev, u32 pmtu)
{
struct rt6_info *rt, *nrt;
- struct net *net = dev->nd_net;
+ struct net *net = dev_net(dev);
int allfrag = 0;
rt = rt6_lookup(net, daddr, saddr, dev->ifindex, 0);
static struct rt6_info * ip6_rt_copy(struct rt6_info *ort)
{
- struct rt6_info *rt = ip6_dst_alloc();
+ struct net *net = dev_net(ort->rt6i_dev);
+ struct rt6_info *rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops);
if (rt) {
rt->u.dst.input = ort->u.dst.input;
struct rt6_info *rt;
struct fib6_table *table;
- table = fib6_get_table(dev->nd_net, RT6_TABLE_DFLT);
+ table = fib6_get_table(dev_net(dev), RT6_TABLE_DFLT);
if (table == NULL)
return NULL;
return rt;
}
-EXPORT_SYMBOL(rt6_get_dflt_router);
-
struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr,
struct net_device *dev,
unsigned int pref)
RTF_UP | RTF_EXPIRES | RTF_PREF(pref),
.fc_nlinfo.pid = 0,
.fc_nlinfo.nlh = NULL,
- .fc_nlinfo.nl_net = dev->nd_net,
+ .fc_nlinfo.nl_net = dev_net(dev),
};
ipv6_addr_copy(&cfg.fc_gateway, gwaddr);
const struct in6_addr *addr,
int anycast)
{
- struct net *net = idev->dev->nd_net;
- struct rt6_info *rt = ip6_dst_alloc();
+ struct net *net = dev_net(idev->dev);
+ struct rt6_info *rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops);
if (rt == NULL)
return ERR_PTR(-ENOMEM);
{
struct rt6_mtu_change_arg *arg = (struct rt6_mtu_change_arg *) p_arg;
struct inet6_dev *idev;
- struct net *net = arg->dev->nd_net;
+ struct net *net = dev_net(arg->dev);
/* In IPv6 pmtu discovery is not optional,
so that RTAX_MTU lock cannot disable it.
.mtu = mtu,
};
- fib6_clean_all(dev->nd_net, rt6_mtu_change_route, 0, &arg);
+ fib6_clean_all(dev_net(dev), rt6_mtu_change_route, 0, &arg);
}
static const struct nla_policy rtm_ipv6_policy[RTA_MAX+1] = {
cfg->fc_nlinfo.pid = NETLINK_CB(skb).pid;
cfg->fc_nlinfo.nlh = nlh;
- cfg->fc_nlinfo.nl_net = skb->sk->sk_net;
+ cfg->fc_nlinfo.nl_net = sock_net(skb->sk);
if (tb[RTA_GATEWAY]) {
nla_memcpy(&cfg->fc_gateway, tb[RTA_GATEWAY], 16);
static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
struct in6_addr *dst, struct in6_addr *src,
int iif, int type, u32 pid, u32 seq,
- int prefix, unsigned int flags)
+ int prefix, int nowait, unsigned int flags)
{
struct rtmsg *rtm;
struct nlmsghdr *nlh;
} else if (rtm->rtm_src_len)
NLA_PUT(skb, RTA_SRC, 16, &rt->rt6i_src.addr);
#endif
- if (iif)
- NLA_PUT_U32(skb, RTA_IIF, iif);
- else if (dst) {
+ if (iif) {
+#ifdef CONFIG_IPV6_MROUTE
+ if (ipv6_addr_is_multicast(&rt->rt6i_dst.addr)) {
+ int err = ip6mr_get_route(skb, rtm, nowait);
+ if (err <= 0) {
+ if (!nowait) {
+ if (err == 0)
+ return 0;
+ goto nla_put_failure;
+ } else {
+ if (err == -EMSGSIZE)
+ goto nla_put_failure;
+ }
+ }
+ } else
+#endif
+ NLA_PUT_U32(skb, RTA_IIF, iif);
+ } else if (dst) {
struct in6_addr saddr_buf;
if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev,
- dst, &saddr_buf) == 0)
+ dst, 0, &saddr_buf) == 0)
NLA_PUT(skb, RTA_PREFSRC, 16, &saddr_buf);
}
return rt6_fill_node(arg->skb, rt, NULL, NULL, 0, RTM_NEWROUTE,
NETLINK_CB(arg->cb->skb).pid, arg->cb->nlh->nlmsg_seq,
- prefix, NLM_F_MULTI);
+ prefix, 0, NLM_F_MULTI);
}
static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg)
{
- struct net *net = in_skb->sk->sk_net;
+ struct net *net = sock_net(in_skb->sk);
struct nlattr *tb[RTA_MAX+1];
struct rt6_info *rt;
struct sk_buff *skb;
skb_reset_mac_header(skb);
skb_reserve(skb, MAX_HEADER + sizeof(struct ipv6hdr));
- rt = (struct rt6_info*) ip6_route_output(NULL, &fl);
+ rt = (struct rt6_info*) ip6_route_output(net, NULL, &fl);
skb->dst = &rt->u.dst;
err = rt6_fill_node(skb, rt, &fl.fl6_dst, &fl.fl6_src, iif,
RTM_NEWROUTE, NETLINK_CB(in_skb).pid,
- nlh->nlmsg_seq, 0, 0);
+ nlh->nlmsg_seq, 0, 0, 0);
if (err < 0) {
kfree_skb(skb);
goto errout;
goto errout;
err = rt6_fill_node(skb, rt, NULL, NULL, 0,
- event, info->pid, seq, 0, 0);
+ event, info->pid, seq, 0, 0, 0);
if (err < 0) {
/* -EMSGSIZE implies BUG in rt6_nlmsg_size() */
WARN_ON(err == -EMSGSIZE);
unsigned long event, void *data)
{
struct net_device *dev = (struct net_device *)data;
- struct net *net = dev->nd_net;
+ struct net *net = dev_net(dev);
if (event == NETDEV_REGISTER && (dev->flags & IFF_LOOPBACK)) {
net->ipv6.ip6_null_entry->u.dst.dev = dev;
static int ipv6_route_open(struct inode *inode, struct file *file)
{
+ int err;
struct net *net = get_proc_net(inode);
if (!net)
return -ENXIO;
- return single_open(file, ipv6_route_show, net);
+
+ err = single_open(file, ipv6_route_show, net);
+ if (err < 0) {
+ put_net(net);
+ return err;
+ }
+
+ return 0;
}
static int ipv6_route_release(struct inode *inode, struct file *file)
net->ipv6.rt6_stats->fib_rt_alloc,
net->ipv6.rt6_stats->fib_rt_entries,
net->ipv6.rt6_stats->fib_rt_cache,
- atomic_read(&ip6_dst_ops.entries),
+ atomic_read(&net->ipv6.ip6_dst_ops->entries),
net->ipv6.rt6_stats->fib_discarded_routes);
return 0;
static int rt6_stats_seq_open(struct inode *inode, struct file *file)
{
+ int err;
struct net *net = get_proc_net(inode);
- return single_open(file, rt6_stats_seq_show, net);
+ if (!net)
+ return -ENXIO;
+
+ err = single_open(file, rt6_stats_seq_show, net);
+ if (err < 0) {
+ put_net(net);
+ return err;
+ }
+
+ return 0;
}
static int rt6_stats_seq_release(struct inode *inode, struct file *file)
{
.ctl_name = NET_IPV6_ROUTE_GC_THRESH,
.procname = "gc_thresh",
- .data = &ip6_dst_ops.gc_thresh,
+ .data = &ip6_dst_ops_template.gc_thresh,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = &proc_dointvec,
if (table) {
table[0].data = &net->ipv6.sysctl.flush_delay;
- /* table[1].data will be handled when we have
- routes per namespace */
+ table[1].data = &net->ipv6.ip6_dst_ops->gc_thresh;
table[2].data = &net->ipv6.sysctl.ip6_rt_max_size;
table[3].data = &net->ipv6.sysctl.ip6_rt_gc_min_interval;
table[4].data = &net->ipv6.sysctl.ip6_rt_gc_timeout;
int ret = 0;
ret = -ENOMEM;
+ net->ipv6.ip6_dst_ops = kmemdup(&ip6_dst_ops_template,
+ sizeof(*net->ipv6.ip6_dst_ops),
+ GFP_KERNEL);
+ if (!net->ipv6.ip6_dst_ops)
+ goto out;
+ net->ipv6.ip6_dst_ops->dst_net = net;
+
net->ipv6.ip6_null_entry = kmemdup(&ip6_null_entry_template,
sizeof(*net->ipv6.ip6_null_entry),
GFP_KERNEL);
if (!net->ipv6.ip6_null_entry)
- goto out;
+ goto out_ip6_dst_ops;
net->ipv6.ip6_null_entry->u.dst.path =
(struct dst_entry *)net->ipv6.ip6_null_entry;
+ net->ipv6.ip6_null_entry->u.dst.ops = net->ipv6.ip6_dst_ops;
#ifdef CONFIG_IPV6_MULTIPLE_TABLES
net->ipv6.ip6_prohibit_entry = kmemdup(&ip6_prohibit_entry_template,
}
net->ipv6.ip6_prohibit_entry->u.dst.path =
(struct dst_entry *)net->ipv6.ip6_prohibit_entry;
+ net->ipv6.ip6_prohibit_entry->u.dst.ops = net->ipv6.ip6_dst_ops;
net->ipv6.ip6_blk_hole_entry = kmemdup(&ip6_blk_hole_entry_template,
sizeof(*net->ipv6.ip6_blk_hole_entry),
}
net->ipv6.ip6_blk_hole_entry->u.dst.path =
(struct dst_entry *)net->ipv6.ip6_blk_hole_entry;
+ net->ipv6.ip6_blk_hole_entry->u.dst.ops = net->ipv6.ip6_dst_ops;
#endif
#ifdef CONFIG_PROC_FS
proc_net_fops_create(net, "ipv6_route", 0, &ipv6_route_proc_fops);
proc_net_fops_create(net, "rt6_stats", S_IRUGO, &rt6_stats_seq_fops);
#endif
+ net->ipv6.ip6_rt_gc_expire = 30*HZ;
+
ret = 0;
out:
return ret;
+
+out_ip6_dst_ops:
+ kfree(net->ipv6.ip6_dst_ops);
+ goto out;
}
static void ip6_route_net_exit(struct net *net)
kfree(net->ipv6.ip6_prohibit_entry);
kfree(net->ipv6.ip6_blk_hole_entry);
#endif
+ kfree(net->ipv6.ip6_dst_ops);
}
static struct pernet_operations ip6_route_net_ops = {
{
int ret;
- ip6_dst_ops.kmem_cachep =
+ ret = -ENOMEM;
+ ip6_dst_ops_template.kmem_cachep =
kmem_cache_create("ip6_dst_cache", sizeof(struct rt6_info), 0,
SLAB_HWCACHE_ALIGN, NULL);
- if (!ip6_dst_ops.kmem_cachep)
- return -ENOMEM;
-
- ip6_dst_blackhole_ops.kmem_cachep = ip6_dst_ops.kmem_cachep;
+ if (!ip6_dst_ops_template.kmem_cachep)
+ goto out;;
ret = register_pernet_subsys(&ip6_route_net_ops);
if (ret)
out_register_subsys:
unregister_pernet_subsys(&ip6_route_net_ops);
out_kmem_cache:
- kmem_cache_destroy(ip6_dst_ops.kmem_cachep);
+ kmem_cache_destroy(ip6_dst_ops_template.kmem_cachep);
goto out;
}
xfrm6_fini();
fib6_gc_cleanup();
unregister_pernet_subsys(&ip6_route_net_ops);
- kmem_cache_destroy(ip6_dst_ops.kmem_cachep);
+ kmem_cache_destroy(ip6_dst_ops_template.kmem_cachep);
}
projects / linux-2.6 / blobdiff