#include <net/tcp.h>
#include <net/icmp.h>
#include <net/xfrm.h>
-#include <net/ip_mp_alg.h>
#include <net/netevent.h>
#include <net/rtnetlink.h>
#ifdef CONFIG_SYSCTL
#define ECN_OR_COST(class) TC_PRIO_##class
-__u8 ip_tos2prio[16] = {
+const __u8 ip_tos2prio[16] = {
TC_PRIO_BESTEFFORT,
ECN_OR_COST(FILLER),
TC_PRIO_BESTEFFORT,
static __inline__ void rt_free(struct rtable *rt)
{
- multipath_remove(rt);
call_rcu_bh(&rt->u.dst.rcu_head, dst_rcu_free);
}
static __inline__ void rt_drop(struct rtable *rt)
{
- multipath_remove(rt);
ip_rt_put(rt);
call_rcu_bh(&rt->u.dst.rcu_head, dst_rcu_free);
}
(fl1->iif ^ fl2->iif)) == 0;
}
-#ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
-static struct rtable **rt_remove_balanced_route(struct rtable **chain_head,
- struct rtable *expentry,
- int *removed_count)
-{
- int passedexpired = 0;
- struct rtable **nextstep = NULL;
- struct rtable **rthp = chain_head;
- struct rtable *rth;
-
- if (removed_count)
- *removed_count = 0;
-
- while ((rth = *rthp) != NULL) {
- if (rth == expentry)
- passedexpired = 1;
-
- if (((*rthp)->u.dst.flags & DST_BALANCED) != 0 &&
- compare_keys(&(*rthp)->fl, &expentry->fl)) {
- if (*rthp == expentry) {
- *rthp = rth->u.dst.rt_next;
- continue;
- } else {
- *rthp = rth->u.dst.rt_next;
- rt_free(rth);
- if (removed_count)
- ++(*removed_count);
- }
- } else {
- if (!((*rthp)->u.dst.flags & DST_BALANCED) &&
- passedexpired && !nextstep)
- nextstep = &rth->u.dst.rt_next;
-
- rthp = &rth->u.dst.rt_next;
- }
- }
-
- rt_free(expentry);
- if (removed_count)
- ++(*removed_count);
-
- return nextstep;
-}
-#endif /* CONFIG_IP_ROUTE_MULTIPATH_CACHED */
-
-
/* This runs via a timer and thus is always in BH context. */
static void rt_check_expire(unsigned long dummy)
{
}
/* Cleanup aged off entries. */
-#ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
- /* remove all related balanced entries if necessary */
- if (rth->u.dst.flags & DST_BALANCED) {
- rthp = rt_remove_balanced_route(
- &rt_hash_table[i].chain,
- rth, NULL);
- if (!rthp)
- break;
- } else {
- *rthp = rth->u.dst.rt_next;
- rt_free(rth);
- }
-#else /* CONFIG_IP_ROUTE_MULTIPATH_CACHED */
*rthp = rth->u.dst.rt_next;
rt_free(rth);
-#endif /* CONFIG_IP_ROUTE_MULTIPATH_CACHED */
}
spin_unlock(rt_hash_lock_addr(i));
if (delay < 0)
delay = ip_rt_min_delay;
- /* flush existing multipath state*/
- multipath_flush();
-
spin_lock_bh(&rt_flush_lock);
if (del_timer(&rt_flush_timer) && delay > 0 && rt_deadline) {
rthp = &rth->u.dst.rt_next;
continue;
}
-#ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
- /* remove all related balanced entries
- * if necessary
- */
- if (rth->u.dst.flags & DST_BALANCED) {
- int r;
-
- rthp = rt_remove_balanced_route(
- &rt_hash_table[k].chain,
- rth,
- &r);
- goal -= r;
- if (!rthp)
- break;
- } else {
- *rthp = rth->u.dst.rt_next;
- rt_free(rth);
- goal--;
- }
-#else /* CONFIG_IP_ROUTE_MULTIPATH_CACHED */
*rthp = rth->u.dst.rt_next;
rt_free(rth);
goal--;
-#endif /* CONFIG_IP_ROUTE_MULTIPATH_CACHED */
}
spin_unlock_bh(rt_hash_lock_addr(k));
if (goal <= 0)
spin_lock_bh(rt_hash_lock_addr(hash));
while ((rth = *rthp) != NULL) {
-#ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
- if (!(rth->u.dst.flags & DST_BALANCED) &&
- compare_keys(&rth->fl, &rt->fl)) {
-#else
if (compare_keys(&rth->fl, &rt->fl)) {
-#endif
/* Put it first */
*rthp = rth->u.dst.rt_next;
/*
atomic_set(&rth->u.dst.__refcnt, 1);
rth->u.dst.flags= DST_HOST;
- if (in_dev->cnf.no_policy)
+ if (IN_DEV_CONF_GET(in_dev, NOPOLICY))
rth->u.dst.flags |= DST_NOPOLICY;
rth->fl.fl4_dst = daddr;
rth->rt_dst = daddr;
atomic_set(&rth->u.dst.__refcnt, 1);
rth->u.dst.flags= DST_HOST;
-#ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
- if (res->fi->fib_nhs > 1)
- rth->u.dst.flags |= DST_BALANCED;
-#endif
- if (in_dev->cnf.no_policy)
+ if (IN_DEV_CONF_GET(in_dev, NOPOLICY))
rth->u.dst.flags |= DST_NOPOLICY;
- if (out_dev->cnf.no_xfrm)
+ if (IN_DEV_CONF_GET(out_dev, NOXFRM))
rth->u.dst.flags |= DST_NOXFRM;
rth->fl.fl4_dst = daddr;
rth->rt_dst = daddr;
return err;
}
-static inline int ip_mkroute_input_def(struct sk_buff *skb,
- struct fib_result* res,
- const struct flowi *fl,
- struct in_device *in_dev,
- __be32 daddr, __be32 saddr, u32 tos)
+static inline int ip_mkroute_input(struct sk_buff *skb,
+ struct fib_result* res,
+ const struct flowi *fl,
+ struct in_device *in_dev,
+ __be32 daddr, __be32 saddr, u32 tos)
{
struct rtable* rth = NULL;
int err;
return rt_intern_hash(hash, rth, (struct rtable**)&skb->dst);
}
-static inline int ip_mkroute_input(struct sk_buff *skb,
- struct fib_result* res,
- const struct flowi *fl,
- struct in_device *in_dev,
- __be32 daddr, __be32 saddr, u32 tos)
-{
-#ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
- struct rtable* rth = NULL, *rtres;
- unsigned char hop, hopcount;
- int err = -EINVAL;
- unsigned int hash;
-
- if (res->fi)
- hopcount = res->fi->fib_nhs;
- else
- hopcount = 1;
-
- /* distinguish between multipath and singlepath */
- if (hopcount < 2)
- return ip_mkroute_input_def(skb, res, fl, in_dev, daddr,
- saddr, tos);
-
- /* add all alternatives to the routing cache */
- for (hop = 0; hop < hopcount; hop++) {
- res->nh_sel = hop;
-
- /* put reference to previous result */
- if (hop)
- ip_rt_put(rtres);
-
- /* create a routing cache entry */
- err = __mkroute_input(skb, res, in_dev, daddr, saddr, tos,
- &rth);
- if (err)
- return err;
-
- /* put it into the cache */
- hash = rt_hash(daddr, saddr, fl->iif);
- err = rt_intern_hash(hash, rth, &rtres);
- if (err)
- return err;
-
- /* forward hop information to multipath impl. */
- multipath_set_nhinfo(rth,
- FIB_RES_NETWORK(*res),
- FIB_RES_NETMASK(*res),
- res->prefixlen,
- &FIB_RES_NH(*res));
- }
- skb->dst = &rtres->u.dst;
- return err;
-#else /* CONFIG_IP_ROUTE_MULTIPATH_CACHED */
- return ip_mkroute_input_def(skb, res, fl, in_dev, daddr, saddr, tos);
-#endif /* CONFIG_IP_ROUTE_MULTIPATH_CACHED */
-}
-
-
/*
* NOTE. We drop all the packets that has local source
* addresses, because every properly looped back packet
atomic_set(&rth->u.dst.__refcnt, 1);
rth->u.dst.flags= DST_HOST;
- if (in_dev->cnf.no_policy)
+ if (IN_DEV_CONF_GET(in_dev, NOPOLICY))
rth->u.dst.flags |= DST_NOPOLICY;
rth->fl.fl4_dst = daddr;
rth->rt_dst = daddr;
atomic_set(&rth->u.dst.__refcnt, 1);
rth->u.dst.flags= DST_HOST;
-#ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
- if (res->fi) {
- rth->rt_multipath_alg = res->fi->fib_mp_alg;
- if (res->fi->fib_nhs > 1)
- rth->u.dst.flags |= DST_BALANCED;
- }
-#endif
- if (in_dev->cnf.no_xfrm)
+ if (IN_DEV_CONF_GET(in_dev, NOXFRM))
rth->u.dst.flags |= DST_NOXFRM;
- if (in_dev->cnf.no_policy)
+ if (IN_DEV_CONF_GET(in_dev, NOPOLICY))
rth->u.dst.flags |= DST_NOPOLICY;
rth->fl.fl4_dst = oldflp->fl4_dst;
return err;
}
-static inline int ip_mkroute_output_def(struct rtable **rp,
- struct fib_result* res,
- const struct flowi *fl,
- const struct flowi *oldflp,
- struct net_device *dev_out,
- unsigned flags)
+static inline int ip_mkroute_output(struct rtable **rp,
+ struct fib_result* res,
+ const struct flowi *fl,
+ const struct flowi *oldflp,
+ struct net_device *dev_out,
+ unsigned flags)
{
struct rtable *rth = NULL;
int err = __mkroute_output(&rth, res, fl, oldflp, dev_out, flags);
return err;
}
-static inline int ip_mkroute_output(struct rtable** rp,
- struct fib_result* res,
- const struct flowi *fl,
- const struct flowi *oldflp,
- struct net_device *dev_out,
- unsigned flags)
-{
-#ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
- unsigned char hop;
- unsigned hash;
- int err = -EINVAL;
- struct rtable *rth = NULL;
-
- if (res->fi && res->fi->fib_nhs > 1) {
- unsigned char hopcount = res->fi->fib_nhs;
-
- for (hop = 0; hop < hopcount; hop++) {
- struct net_device *dev2nexthop;
-
- res->nh_sel = hop;
-
- /* hold a work reference to the output device */
- dev2nexthop = FIB_RES_DEV(*res);
- dev_hold(dev2nexthop);
-
- /* put reference to previous result */
- if (hop)
- ip_rt_put(*rp);
-
- err = __mkroute_output(&rth, res, fl, oldflp,
- dev2nexthop, flags);
-
- if (err != 0)
- goto cleanup;
-
- hash = rt_hash(oldflp->fl4_dst, oldflp->fl4_src,
- oldflp->oif);
- err = rt_intern_hash(hash, rth, rp);
-
- /* forward hop information to multipath impl. */
- multipath_set_nhinfo(rth,
- FIB_RES_NETWORK(*res),
- FIB_RES_NETMASK(*res),
- res->prefixlen,
- &FIB_RES_NH(*res));
- cleanup:
- /* release work reference to output device */
- dev_put(dev2nexthop);
-
- if (err != 0)
- return err;
- }
- return err;
- } else {
- return ip_mkroute_output_def(rp, res, fl, oldflp, dev_out,
- flags);
- }
-#else /* CONFIG_IP_ROUTE_MULTIPATH_CACHED */
- return ip_mkroute_output_def(rp, res, fl, oldflp, dev_out, flags);
-#endif
-}
-
/*
* Major route resolver routine.
*/
/* It is equivalent to inet_addr_type(saddr) == RTN_LOCAL */
dev_out = ip_dev_find(oldflp->fl4_src);
- if ((dev_out == NULL) && !(sysctl_ip_nonlocal_bind))
+ if (dev_out == NULL)
goto out;
/* I removed check for oif == dev_out->oif here.
of another iface. --ANK
*/
- if (dev_out && oldflp->oif == 0
+ if (oldflp->oif == 0
&& (MULTICAST(oldflp->fl4_dst) || oldflp->fl4_dst == htonl(0xFFFFFFFF))) {
/* Special hack: user can direct multicasts
and limited broadcast via necessary interface
rth->fl.mark == flp->mark &&
!((rth->fl.fl4_tos ^ flp->fl4_tos) &
(IPTOS_RT_MASK | RTO_ONLINK))) {
-
- /* check for multipath routes and choose one if
- * necessary
- */
- if (multipath_select_route(flp, rth, rp)) {
- dst_hold(&(*rp)->u.dst);
- RT_CACHE_STAT_INC(out_hit);
- rcu_read_unlock_bh();
- return 0;
- }
-
rth->u.dst.lastuse = jiffies;
dst_hold(&rth->u.dst);
rth->u.dst.__use++;
EXPORT_SYMBOL_GPL(__ip_route_output_key);
+static void ipv4_rt_blackhole_update_pmtu(struct dst_entry *dst, u32 mtu)
+{
+}
+
+static struct dst_ops ipv4_dst_blackhole_ops = {
+ .family = AF_INET,
+ .protocol = __constant_htons(ETH_P_IP),
+ .destroy = ipv4_dst_destroy,
+ .check = ipv4_dst_check,
+ .update_pmtu = ipv4_rt_blackhole_update_pmtu,
+ .entry_size = sizeof(struct rtable),
+};
+
+
+static int ipv4_blackhole_output(struct sk_buff *skb)
+{
+ kfree_skb(skb);
+ return 0;
+}
+
+static int ipv4_dst_blackhole(struct rtable **rp, struct flowi *flp, struct sock *sk)
+{
+ struct rtable *ort = *rp;
+ struct rtable *rt = (struct rtable *)
+ dst_alloc(&ipv4_dst_blackhole_ops);
+
+ if (rt) {
+ struct dst_entry *new = &rt->u.dst;
+
+ atomic_set(&new->__refcnt, 1);
+ new->__use = 1;
+ new->input = ipv4_blackhole_output;
+ new->output = ipv4_blackhole_output;
+ memcpy(new->metrics, ort->u.dst.metrics, RTAX_MAX*sizeof(u32));
+
+ new->dev = ort->u.dst.dev;
+ if (new->dev)
+ dev_hold(new->dev);
+
+ rt->fl = ort->fl;
+
+ rt->idev = ort->idev;
+ if (rt->idev)
+ in_dev_hold(rt->idev);
+ rt->rt_flags = ort->rt_flags;
+ rt->rt_type = ort->rt_type;
+ rt->rt_dst = ort->rt_dst;
+ rt->rt_src = ort->rt_src;
+ rt->rt_iif = ort->rt_iif;
+ rt->rt_gateway = ort->rt_gateway;
+ rt->rt_spec_dst = ort->rt_spec_dst;
+ rt->peer = ort->peer;
+ if (rt->peer)
+ atomic_inc(&rt->peer->refcnt);
+
+ dst_free(new);
+ }
+
+ dst_release(&(*rp)->u.dst);
+ *rp = rt;
+ return (rt ? 0 : -ENOMEM);
+}
+
int ip_route_output_flow(struct rtable **rp, struct flowi *flp, struct sock *sk, int flags)
{
int err;
flp->fl4_src = (*rp)->rt_src;
if (!flp->fl4_dst)
flp->fl4_dst = (*rp)->rt_dst;
- return xfrm_lookup((struct dst_entry **)rp, flp, sk, flags);
+ err = __xfrm_lookup((struct dst_entry **)rp, flp, sk, flags);
+ if (err == -EREMOTE)
+ err = ipv4_dst_blackhole(rp, flp, sk);
+
+ return err;
}
return 0;
#ifdef CONFIG_NET_CLS_ROUTE
if (rt->u.dst.tclassid)
NLA_PUT_U32(skb, RTA_FLOW, rt->u.dst.tclassid);
-#endif
-#ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
- if (rt->rt_multipath_alg != IP_MP_ALG_NONE)
- NLA_PUT_U32(skb, RTA_MP_ALGO, rt->rt_multipath_alg);
#endif
if (rt->fl.iif)
NLA_PUT_BE32(skb, RTA_PREFSRC, rt->rt_spec_dst);
__be32 dst = rt->rt_dst;
if (MULTICAST(dst) && !LOCAL_MCAST(dst) &&
- ipv4_devconf.mc_forwarding) {
+ IPV4_DEVCONF_ALL(MC_FORWARDING)) {
int err = ipmr_get_route(skb, r, nowait);
if (err <= 0) {
if (!nowait) {
kmem_cache_create("ip_dst_cache", sizeof(struct rtable), 0,
SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL, NULL);
+ ipv4_dst_blackhole_ops.kmem_cachep = ipv4_dst_ops.kmem_cachep;
+
rt_hash_table = (struct rt_hash_bucket *)
alloc_large_system_hash("IP route cache",
sizeof(struct rt_hash_bucket),
git.openpandora.org / pandora-kernel.git / blobdiff