ipv4: make exception cache less predictible

Even after commit 6457378f ("ipv4: use siphash instead of Jenkins in
fnhe_hashfun()"), an attacker can still use brute force to learn
some secrets from a victim linux host.

One way to defeat these attacks is to make the max depth of the hash
table bucket a random value.

Before this patch, each bucket of the hash table used to store exceptions
could contain 6 items under attack.

After the patch, each bucket would contains a random number of items,
between 6 and 10. The attacker can no longer infer secrets.

This is slightly increasing memory size used by the hash table,
by 50% in average, we do not expect this to be a problem.

This patch is more complex than the prior one (IPv6 equivalent),
because IPv4 was reusing the oldest entry.
Since we need to be able to evict more than one entry per
update_or_create_fnhe() call, I had to replace
fnhe_oldest() with fnhe_remove_oldest().

Also note that we will queue extra kfree_rcu() calls under stress,
which hopefully wont be a too big issue.

Fixes: 4895c771 ("ipv4: Add FIB nexthop exceptions.")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Keyu Man <kman001@ucr.edu>
Cc: Willy Tarreau <w@1wt.eu>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reviewed-by: David Ahern <dsahern@kernel.org>
Tested-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/ipv4/route.c

@@ -587,18 +587,25 @@ static void fnhe_flush_routes(struct fib_nh_exception *fnhe)
 	}
 }
 
-static struct fib_nh_exception *fnhe_oldest(struct fnhe_hash_bucket *hash)
+static void fnhe_remove_oldest(struct fnhe_hash_bucket *hash)
 {
-	struct fib_nh_exception *fnhe, *oldest;
+	struct fib_nh_exception __rcu **fnhe_p, **oldest_p;
+	struct fib_nh_exception *fnhe, *oldest = NULL;
 
-	oldest = rcu_dereference(hash->chain);
-	for (fnhe = rcu_dereference(oldest->fnhe_next); fnhe;
-	     fnhe = rcu_dereference(fnhe->fnhe_next)) {
-		if (time_before(fnhe->fnhe_stamp, oldest->fnhe_stamp))
+	for (fnhe_p = &hash->chain; ; fnhe_p = &fnhe->fnhe_next) {
+		fnhe = rcu_dereference_protected(*fnhe_p,
+						 lockdep_is_held(&fnhe_lock));
+		if (!fnhe)
+			break;
+		if (!oldest ||
+		    time_before(fnhe->fnhe_stamp, oldest->fnhe_stamp)) {
 			oldest = fnhe;
+			oldest_p = fnhe_p;
+		}
 	}
 	fnhe_flush_routes(oldest);
-	return oldest;
+	*oldest_p = oldest->fnhe_next;
+	kfree_rcu(oldest, rcu);
 }
 
 static u32 fnhe_hashfun(__be32 daddr)
@@ -677,16 +684,21 @@ static void update_or_create_fnhe(struct fib_nh_common *nhc, __be32 daddr,
 		if (rt)
 			fill_route_from_fnhe(rt, fnhe);
 	} else {
-		if (depth > FNHE_RECLAIM_DEPTH)
-			fnhe = fnhe_oldest(hash);
-		else {
+		/* Randomize max depth to avoid some side channels attacks. */
+		int max_depth = FNHE_RECLAIM_DEPTH +
+				prandom_u32_max(FNHE_RECLAIM_DEPTH);
+
+		while (depth > max_depth) {
+			fnhe_remove_oldest(hash);
+			depth--;
+		}
+
 		fnhe = kzalloc(sizeof(*fnhe), GFP_ATOMIC);
 		if (!fnhe)
 			goto out_unlock;
 
 		fnhe->fnhe_next = hash->chain;
-			rcu_assign_pointer(hash->chain, fnhe);
-		}
+
 		fnhe->fnhe_genid = genid;
 		fnhe->fnhe_daddr = daddr;
 		fnhe->fnhe_gw = gw;
@@ -694,6 +706,8 @@ static void update_or_create_fnhe(struct fib_nh_common *nhc, __be32 daddr,
 		fnhe->fnhe_mtu_locked = lock;
 		fnhe->fnhe_expires = max(1UL, expires);
 
+		rcu_assign_pointer(hash->chain, fnhe);
+
 		/* Exception created; mark the cached routes for the nexthop
 		 * stale, so anyone caching it rechecks if this exception
 		 * applies to them.