net/ipv6: Cleanup exception and cache route handling

IPv6 FIB will only contain FIB entries with exception routes added to
the FIB entry. Once this transformation is complete, FIB lookups will
return a fib6_info with the lookup functions still returning a dst
based rt6_info. The current code uses rt6_info for both paths and
overloads the rt6_info variable usually called 'rt'.

This patch introduces a new 'f6i' variable name for the result of the FIB
lookup and keeps 'rt' as the dst based return variable. 'f6i' becomes a
fib6_info in a later patch which is why it is introduced as f6i now;
avoids the additional churn in the later patch.

In addition, remove RTF_CACHE and dst checks from fib6 add and delete
since they can not happen now and will never happen after the data
type flip.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

include/net/ip6_route.h

@@ -106,7 +106,6 @@ int ip6_ins_rt(struct net *net, struct rt6_info *rt);
 int ip6_del_rt(struct net *net, struct rt6_info *rt);
 
 void rt6_flush_exceptions(struct rt6_info *rt);
-int rt6_remove_exception_rt(struct rt6_info *rt);
 void rt6_age_exceptions(struct rt6_info *rt, struct fib6_gc_args *gc_args,
 			unsigned long now);
 

net/ipv6/ip6_fib.c

@@ -1074,7 +1074,7 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt,
 static void fib6_start_gc(struct net *net, struct rt6_info *rt)
 {
 	if (!timer_pending(&net->ipv6.ip6_fib_timer) &&
-	    (rt->rt6i_flags & (RTF_EXPIRES | RTF_CACHE)))
+	    (rt->rt6i_flags & RTF_EXPIRES))
 		mod_timer(&net->ipv6.ip6_fib_timer,
 			  jiffies + net->ipv6.sysctl.ip6_rt_gc_interval);
 }
@@ -1125,8 +1125,6 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt,
 
 	if (WARN_ON_ONCE(!atomic_read(&rt->dst.__refcnt)))
 		return -EINVAL;
-	if (WARN_ON_ONCE(rt->rt6i_flags & RTF_CACHE))
-		return -EINVAL;
 
 	if (info->nlh) {
 		if (!(info->nlh->nlmsg_flags & NLM_F_CREATE))
@@ -1650,8 +1648,6 @@ static void fib6_del_route(struct fib6_table *table, struct fib6_node *fn,
 
 	RT6_TRACE("fib6_del_route\n");
 
-	WARN_ON_ONCE(rt->rt6i_flags & RTF_CACHE);
-
 	/* Unlink it */
 	*rtp = rt->rt6_next;
 	rt->rt6i_node = NULL;
@@ -1720,21 +1716,11 @@ int fib6_del(struct rt6_info *rt, struct nl_info *info)
 	struct rt6_info __rcu **rtp;
 	struct rt6_info __rcu **rtp_next;
 
-#if RT6_DEBUG >= 2
-	if (rt->dst.obsolete > 0) {
-		WARN_ON(fn);
-		return -ENOENT;
-	}
-#endif
 	if (!fn || rt == net->ipv6.fib6_null_entry)
 		return -ENOENT;
 
 	WARN_ON(!(fn->fn_flags & RTN_RTINFO));
 
-	/* remove cached dst from exception table */
-	if (rt->rt6i_flags & RTF_CACHE)
-		return rt6_remove_exception_rt(rt);
-
 	/*
 	 *	Walk the leaf entries looking for ourself
 	 */

net/ipv6/route.c

@@ -1013,8 +1013,8 @@ static void rt6_set_from(struct rt6_info *rt, struct rt6_info *from)
 	BUG_ON(from->from);
 
 	rt->rt6i_flags &= ~RTF_EXPIRES;
-	dst_hold(&from->dst);
+	if (dst_hold_safe(&from->dst))
-	rt->from = from;
+		rt->from = from;
 	dst_init_metrics(&rt->dst, from->fib6_metrics->metrics, true);
 	if (from->fib6_metrics != &dst_default_metrics) {
 		rt->dst._metrics |= DST_METRICS_REFCOUNTED;
@@ -1097,8 +1097,9 @@ static struct rt6_info *ip6_pol_route_lookup(struct net *net,
 					     const struct sk_buff *skb,
 					     int flags)
 {
-	struct rt6_info *rt, *rt_cache;
+	struct rt6_info *f6i;
 	struct fib6_node *fn;
+	struct rt6_info *rt;
 
 	if (fl6->flowi6_flags & FLOWI_FLAG_SKIP_NH_OIF)
 		flags &= ~RT6_LOOKUP_F_IFACE;
@@ -1106,36 +1107,36 @@ static struct rt6_info *ip6_pol_route_lookup(struct net *net,
 	rcu_read_lock();
 	fn = fib6_lookup(&table->tb6_root, &fl6->daddr, &fl6->saddr);
 restart:
-	rt = rcu_dereference(fn->leaf);
+	f6i = rcu_dereference(fn->leaf);
-	if (!rt) {
+	if (!f6i) {
-		rt = net->ipv6.fib6_null_entry;
+		f6i = net->ipv6.fib6_null_entry;
 	} else {
-		rt = rt6_device_match(net, rt, &fl6->saddr,
+		f6i = rt6_device_match(net, f6i, &fl6->saddr,
 				      fl6->flowi6_oif, flags);
-		if (rt->rt6i_nsiblings && fl6->flowi6_oif == 0)
+		if (f6i->rt6i_nsiblings && fl6->flowi6_oif == 0)
-			rt = rt6_multipath_select(net, rt, fl6, fl6->flowi6_oif,
+			f6i = rt6_multipath_select(net, f6i, fl6,
-						  skb, flags);
+						   fl6->flowi6_oif, skb, flags);
 	}
-	if (rt == net->ipv6.fib6_null_entry) {
+	if (f6i == net->ipv6.fib6_null_entry) {
 		fn = fib6_backtrack(fn, &fl6->saddr);
 		if (fn)
 			goto restart;
 	}
+
 	/* Search through exception table */
-	rt_cache = rt6_find_cached_rt(rt, &fl6->daddr, &fl6->saddr);
+	rt = rt6_find_cached_rt(f6i, &fl6->daddr, &fl6->saddr);
-	if (rt_cache) {
+	if (rt) {
-		rt = rt_cache;
 		if (ip6_hold_safe(net, &rt, true))
 			dst_use_noref(&rt->dst, jiffies);
-	} else if (dst_hold_safe(&rt->dst)) {
+	} else if (f6i == net->ipv6.fib6_null_entry) {
-		struct rt6_info *nrt;
-
-		nrt = ip6_create_rt_rcu(rt);
-		dst_release(&rt->dst);
-		rt = nrt;
-	} else {
 		rt = net->ipv6.ip6_null_entry;
 		dst_hold(&rt->dst);
+	} else {
+		rt = ip6_create_rt_rcu(f6i);
+		if (!rt) {
+			rt = net->ipv6.ip6_null_entry;
+			dst_hold(&rt->dst);
+		}
 	}
 
 	rcu_read_unlock();
@@ -1218,9 +1219,6 @@ static struct rt6_info *ip6_rt_cache_alloc(struct rt6_info *ort,
 	 *	Clone the route.
 	 */
 
-	if (ort->rt6i_flags & (RTF_CACHE | RTF_PCPU))
-		ort = ort->from;
-
 	rcu_read_lock();
 	dev = ip6_rt_get_dev_rcu(ort);
 	rt = __ip6_dst_alloc(dev_net(dev), dev, 0);
@@ -1446,11 +1444,6 @@ static int rt6_insert_exception(struct rt6_info *nrt,
 	struct rt6_exception *rt6_ex;
 	int err = 0;
 
-	/* ort can't be a cache or pcpu route */
-	if (ort->rt6i_flags & (RTF_CACHE | RTF_PCPU))
-		ort = ort->from;
-	WARN_ON_ONCE(ort->rt6i_flags & (RTF_CACHE | RTF_PCPU));
-
 	spin_lock_bh(&rt6_exception_lock);
 
 	if (ort->exception_bucket_flushed) {
@@ -1589,7 +1582,7 @@ static struct rt6_info *rt6_find_cached_rt(struct rt6_info *rt,
 }
 
 /* Remove the passed in cached rt from the hash table that contains it */
-int rt6_remove_exception_rt(struct rt6_info *rt)
+static int rt6_remove_exception_rt(struct rt6_info *rt)
 {
 	struct rt6_exception_bucket *bucket;
 	struct rt6_info *from = rt->from;
@@ -1854,7 +1847,8 @@ struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table,
 			       const struct sk_buff *skb, int flags)
 {
 	struct fib6_node *fn, *saved_fn;
-	struct rt6_info *rt, *rt_cache;
+	struct rt6_info *f6i;
+	struct rt6_info *rt;
 	int strict = 0;
 
 	strict |= flags & RT6_LOOKUP_F_IFACE;
@@ -1871,10 +1865,10 @@ struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table,
 		oif = 0;
 
 redo_rt6_select:
-	rt = rt6_select(net, fn, oif, strict);
+	f6i = rt6_select(net, fn, oif, strict);
-	if (rt->rt6i_nsiblings)
+	if (f6i->rt6i_nsiblings)
-		rt = rt6_multipath_select(net, rt, fl6, oif, skb, strict);
+		f6i = rt6_multipath_select(net, f6i, fl6, oif, skb, strict);
-	if (rt == net->ipv6.fib6_null_entry) {
+	if (f6i == net->ipv6.fib6_null_entry) {
 		fn = fib6_backtrack(fn, &fl6->saddr);
 		if (fn)
 			goto redo_rt6_select;
@@ -1886,18 +1880,17 @@ struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table,
 		}
 	}
 
-	/*Search through exception table */
+	if (f6i == net->ipv6.fib6_null_entry) {
-	rt_cache = rt6_find_cached_rt(rt, &fl6->daddr, &fl6->saddr);
-	if (rt_cache)
-		rt = rt_cache;
-
-	if (rt == net->ipv6.fib6_null_entry) {
 		rt = net->ipv6.ip6_null_entry;
 		rcu_read_unlock();
 		dst_hold(&rt->dst);
 		trace_fib6_table_lookup(net, rt, table, fl6);
 		return rt;
-	} else if (rt->rt6i_flags & RTF_CACHE) {
+	}
+
+	/*Search through exception table */
+	rt = rt6_find_cached_rt(f6i, &fl6->daddr, &fl6->saddr);
+	if (rt) {
 		if (ip6_hold_safe(net, &rt, true))
 			dst_use_noref(&rt->dst, jiffies);
 
@@ -1905,7 +1898,7 @@ struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table,
 		trace_fib6_table_lookup(net, rt, table, fl6);
 		return rt;
 	} else if (unlikely((fl6->flowi6_flags & FLOWI_FLAG_KNOWN_NH) &&
-			    !(rt->rt6i_flags & RTF_GATEWAY))) {
+			    !(f6i->rt6i_flags & RTF_GATEWAY))) {
 		/* Create a RTF_CACHE clone which will not be
 		 * owned by the fib6 tree.  It is for the special case where
 		 * the daddr in the skb during the neighbor look-up is different
@@ -1914,16 +1907,16 @@ struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table,
 
 		struct rt6_info *uncached_rt;
 
-		if (ip6_hold_safe(net, &rt, true)) {
+		if (ip6_hold_safe(net, &f6i, true)) {
-			dst_use_noref(&rt->dst, jiffies);
+			dst_use_noref(&f6i->dst, jiffies);
 		} else {
 			rcu_read_unlock();
-			uncached_rt = rt;
+			uncached_rt = f6i;
 			goto uncached_rt_out;
 		}
 		rcu_read_unlock();
 
-		uncached_rt = ip6_rt_cache_alloc(rt, &fl6->daddr, NULL);
+		uncached_rt = ip6_rt_cache_alloc(f6i, &fl6->daddr, NULL);
 		dst_release(&rt->dst);
 
 		if (uncached_rt) {
@@ -1946,18 +1939,18 @@ struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table,
 
 		struct rt6_info *pcpu_rt;
 
-		dst_use_noref(&rt->dst, jiffies);
+		dst_use_noref(&f6i->dst, jiffies);
 		local_bh_disable();
-		pcpu_rt = rt6_get_pcpu_route(rt);
+		pcpu_rt = rt6_get_pcpu_route(f6i);
 
 		if (!pcpu_rt) {
 			/* atomic_inc_not_zero() is needed when using rcu */
-			if (atomic_inc_not_zero(&rt->rt6i_ref)) {
+			if (atomic_inc_not_zero(&f6i->rt6i_ref)) {
 				/* No dst_hold() on rt is needed because grabbing
 				 * rt->rt6i_ref makes sure rt can't be released.
 				 */
-				pcpu_rt = rt6_make_pcpu_route(net, rt);
+				pcpu_rt = rt6_make_pcpu_route(net, f6i);
-				rt6_release(rt);
+				rt6_release(f6i);
 			} else {
 				/* rt is already removed from tree */
 				pcpu_rt = net->ipv6.ip6_null_entry;
@@ -2419,7 +2412,8 @@ static struct rt6_info *__ip6_route_redirect(struct net *net,
 					     int flags)
 {
 	struct ip6rd_flowi *rdfl = (struct ip6rd_flowi *)fl6;
-	struct rt6_info *rt, *rt_cache;
+	struct rt6_info *ret = NULL, *rt_cache;
+	struct rt6_info *rt;
 	struct fib6_node *fn;
 
 	/* Get the "current" route for this destination and
@@ -2458,7 +2452,7 @@ static struct rt6_info *__ip6_route_redirect(struct net *net,
 			if (rt_cache &&
 			    ipv6_addr_equal(&rdfl->gateway,
 					    &rt_cache->rt6i_gateway)) {
-				rt = rt_cache;
+				ret = rt_cache;
 				break;
 			}
 			continue;
@@ -2469,7 +2463,7 @@ static struct rt6_info *__ip6_route_redirect(struct net *net,
 	if (!rt)
 		rt = net->ipv6.fib6_null_entry;
 	else if (rt->rt6i_flags & RTF_REJECT) {
-		rt = net->ipv6.ip6_null_entry;
+		ret = net->ipv6.ip6_null_entry;
 		goto out;
 	}
 
@@ -2480,12 +2474,15 @@ static struct rt6_info *__ip6_route_redirect(struct net *net,
 	}
 
 out:
-	ip6_hold_safe(net, &rt, true);
+	if (ret)
+		dst_hold(&ret->dst);
+	else
+		ret = ip6_create_rt_rcu(rt);
 
 	rcu_read_unlock();
 
-	trace_fib6_table_lookup(net, rt, table, fl6);
+	trace_fib6_table_lookup(net, ret, table, fl6);
-	return rt;
+	return ret;
 };
 
 static struct dst_entry *ip6_route_redirect(struct net *net,
@@ -3182,6 +3179,22 @@ static int __ip6_del_rt_siblings(struct rt6_info *rt, struct fib6_config *cfg)
 	return err;
 }
 
+static int ip6_del_cached_rt(struct rt6_info *rt, struct fib6_config *cfg)
+{
+	int rc = -ESRCH;
+
+	if (cfg->fc_ifindex && rt->dst.dev->ifindex != cfg->fc_ifindex)
+		goto out;
+
+	if (cfg->fc_flags & RTF_GATEWAY &&
+	    !ipv6_addr_equal(&cfg->fc_gateway, &rt->rt6i_gateway))
+		goto out;
+	if (dst_hold_safe(&rt->dst))
+		rc = rt6_remove_exception_rt(rt);
+out:
+	return rc;
+}
+
 static int ip6_route_del(struct fib6_config *cfg,
 			 struct netlink_ext_ack *extack)
 {
@@ -3206,11 +3219,16 @@ static int ip6_route_del(struct fib6_config *cfg,
 	if (fn) {
 		for_each_fib6_node_rt_rcu(fn) {
 			if (cfg->fc_flags & RTF_CACHE) {
+				int rc;
+
 				rt_cache = rt6_find_cached_rt(rt, &cfg->fc_dst,
 							      &cfg->fc_src);
-				if (!rt_cache)
+				if (rt_cache) {
-					continue;
+					rc = ip6_del_cached_rt(rt_cache, cfg);
-				rt = rt_cache;
+					if (rc != -ESRCH)
+						return rc;
+				}
+				continue;
 			}
 			if (cfg->fc_ifindex &&
 			    (!rt->fib6_nh.nh_dev ||
@@ -3327,7 +3345,7 @@ static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_bu
 				     NEIGH_UPDATE_F_ISROUTER)),
 		     NDISC_REDIRECT, &ndopts);
 
-	nrt = ip6_rt_cache_alloc(rt, &msg->dest, NULL);
+	nrt = ip6_rt_cache_alloc(rt->from, &msg->dest, NULL);
 	if (!nrt)
 		goto out;