ipv4: properly refresh rtable entries on pmtu/redirect events

This reverts commit 05ab86c5 (xfrm4: Invalidate all ipv4 routes on
IPsec pmtu events). Flushing all cached entries is not needed.

Instead, invalidate only the related next hop dsts to recheck for
the added next hop exception where needed. This also fixes a subtle
race due to bumping generation id's before updating the pmtu.

Cc: Steffen Klassert <[e-mail]>
Signed-off-by: Timo Teräs <[e-mail]>
Signed-off-by: David S. Miller <[e-mail]>

net/ipv4/ah4.c

···
 	if (!x)
 		return;
 
-	if (icmp_hdr(skb)->type == ICMP_DEST_UNREACH) {
-		atomic_inc(&flow_cache_genid);
-		rt_genid_bump(net);
-
+	if (icmp_hdr(skb)->type == ICMP_DEST_UNREACH)
 		ipv4_update_pmtu(skb, net, info, 0, 0, IPPROTO_AH, 0);
-	} else
+	else
 		ipv4_redirect(skb, net, 0, 0, IPPROTO_AH, 0);
 	xfrm_state_put(x);
 }

net/ipv4/esp4.c

···
 	if (!x)
 		return;
 
-	if (icmp_hdr(skb)->type == ICMP_DEST_UNREACH) {
-		atomic_inc(&flow_cache_genid);
-		rt_genid_bump(net);
-
+	if (icmp_hdr(skb)->type == ICMP_DEST_UNREACH)
 		ipv4_update_pmtu(skb, net, info, 0, 0, IPPROTO_ESP, 0);
-	} else
+	else
 		ipv4_redirect(skb, net, 0, 0, IPPROTO_ESP, 0);
 	xfrm_state_put(x);
 }

net/ipv4/ipcomp.c

···
 	if (!x)
 		return;
 
-	if (icmp_hdr(skb)->type == ICMP_DEST_UNREACH) {
-		atomic_inc(&flow_cache_genid);
-		rt_genid_bump(net);
-
+	if (icmp_hdr(skb)->type == ICMP_DEST_UNREACH)
 		ipv4_update_pmtu(skb, net, info, 0, 0, IPPROTO_COMP, 0);
-	} else
+	else
 		ipv4_redirect(skb, net, 0, 0, IPPROTO_COMP, 0);
 	xfrm_state_put(x);
 }

net/ipv4/route.c

···
 	return hval & (FNHE_HASH_SIZE - 1);
 }
 
+static void fill_route_from_fnhe(struct rtable *rt, struct fib_nh_exception *fnhe)
+{
+	rt->rt_pmtu = fnhe->fnhe_pmtu;
+	rt->dst.expires = fnhe->fnhe_expires;
+
+	if (fnhe->fnhe_gw) {
+		rt->rt_flags |= RTCF_REDIRECTED;
+		rt->rt_gateway = fnhe->fnhe_gw;
+		rt->rt_uses_gateway = 1;
+	}
+}
+
 static void update_or_create_fnhe(struct fib_nh *nh, __be32 daddr, __be32 gw,
 				  u32 pmtu, unsigned long expires)
 {
 	struct fnhe_hash_bucket *hash;
 	struct fib_nh_exception *fnhe;
+	struct rtable *rt;
+	unsigned int i;
 	int depth;
 	u32 hval = fnhe_hashfun(daddr);
 
···
 			fnhe->fnhe_gw = gw;
 		if (pmtu) {
 			fnhe->fnhe_pmtu = pmtu;
-			fnhe->fnhe_expires = expires;
+			fnhe->fnhe_expires = max(1UL, expires);
 		}
+		/* Update all cached dsts too */
+		rt = rcu_dereference(fnhe->fnhe_rth);
+		if (rt)
+			fill_route_from_fnhe(rt, fnhe);
 	} else {
 		if (depth > FNHE_RECLAIM_DEPTH)
 			fnhe = fnhe_oldest(hash);
···
 		fnhe->fnhe_gw = gw;
 		fnhe->fnhe_pmtu = pmtu;
 		fnhe->fnhe_expires = expires;
+
+		/* Exception created; mark the cached routes for the nexthop
+		 * stale, so anyone caching it rechecks if this exception
+		 * applies to them.
+		 */
+		for_each_possible_cpu(i) {
+			struct rtable __rcu **prt;
+			prt = per_cpu_ptr(nh->nh_pcpu_rth_output, i);
+			rt = rcu_dereference(*prt);
+			if (rt)
+				rt->dst.obsolete = DST_OBSOLETE_KILL;
+		}
 	}
 
 	fnhe->fnhe_stamp = jiffies;
···
 	if (mtu < ip_rt_min_pmtu)
 		mtu = ip_rt_min_pmtu;
 
-	if (!rt->rt_pmtu) {
-		dst->obsolete = DST_OBSOLETE_KILL;
-	} else {
-		rt->rt_pmtu = mtu;
-		dst->expires = max(1UL, jiffies + ip_rt_mtu_expires);
-	}
-
 	rcu_read_lock();
 	if (fib_lookup(dev_net(dst->dev), fl4, &res) == 0) {
 		struct fib_nh *nh = &FIB_RES_NH(res);
···
 	 * DST_OBSOLETE_FORCE_CHK which forces validation calls down
 	 * into this function always.
 	 *
-	 * When a PMTU/redirect information update invalidates a
-	 * route, this is indicated by setting obsolete to
-	 * DST_OBSOLETE_KILL.
+	 * When a PMTU/redirect information update invalidates a route,
+	 * this is indicated by setting obsolete to DST_OBSOLETE_KILL or
+	 * DST_OBSOLETE_DEAD by dst_free().
 	 */
-	if (dst->obsolete == DST_OBSOLETE_KILL || rt_is_expired(rt))
+	if (dst->obsolete != DST_OBSOLETE_FORCE_CHK || rt_is_expired(rt))
 		return NULL;
 	return dst;
 }
···
 			fnhe->fnhe_pmtu = 0;
 			fnhe->fnhe_expires = 0;
 		}
-		if (fnhe->fnhe_pmtu) {
-			unsigned long expires = fnhe->fnhe_expires;
-			unsigned long diff = expires - jiffies;
-
-			if (time_before(jiffies, expires)) {
-				rt->rt_pmtu = fnhe->fnhe_pmtu;
-				dst_set_expires(&rt->dst, diff);
-			}
-		}
-		if (fnhe->fnhe_gw) {
-			rt->rt_flags |= RTCF_REDIRECTED;
-			rt->rt_gateway = fnhe->fnhe_gw;
-			rt->rt_uses_gateway = 1;
-		} else if (!rt->rt_gateway)
+		fill_route_from_fnhe(rt, fnhe);
+		if (!rt->rt_gateway)
 			rt->rt_gateway = daddr;
 
 		rcu_assign_pointer(fnhe->fnhe_rth, rt);