Revert "tcp/dccp: get rid of inet_twsk_purge()"

This reverts commits:

0dad4087 ("tcp/dccp: get rid of inet_twsk_purge()")
d507204d ("tcp/dccp: add tw->tw_bslot")

As Leonard pointed out, a newly allocated netns can happen
to reuse a freed 'struct net'.

While TCP TW timers were covered by my patches, other things were not:

1) Lookups in rx path (INET_MATCH() and INET6_MATCH()), as they look
  at 4-tuple plus the 'struct net' pointer.

2) /proc/net/tcp[6] and inet_diag, same reason.

3) hashinfo->bhash[], same reason.

Fixing all this seems risky, lets instead revert.

In the future, we might have a per netns tcp hash table, or
a per netns list of timewait sockets...

Fixes: 0dad4087 ("tcp/dccp: get rid of inet_twsk_purge()")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Leonard Crestez <cdleonard@gmail.com>
Tested-by: Leonard Crestez <cdleonard@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

include/net/inet_timewait_sock.h

@@ -71,7 +71,6 @@ struct inet_timewait_sock {
 				tw_tos		: 8;
 	u32			tw_txhash;
 	u32			tw_priority;
-	u32			tw_bslot; /* bind bucket slot */
 	struct timer_list	tw_timer;
 	struct inet_bind_bucket	*tw_tb;
 };
@@ -110,6 +109,8 @@ static inline void inet_twsk_reschedule(struct inet_timewait_sock *tw, int timeo
 
 void inet_twsk_deschedule_put(struct inet_timewait_sock *tw);
 
+void inet_twsk_purge(struct inet_hashinfo *hashinfo, int family);
+
 static inline
 struct net *twsk_net(const struct inet_timewait_sock *twsk)
 {

net/dccp/ipv4.c

@@ -1030,9 +1030,15 @@ static void __net_exit dccp_v4_exit_net(struct net *net)
 	inet_ctl_sock_destroy(pn->v4_ctl_sk);
 }
 
+static void __net_exit dccp_v4_exit_batch(struct list_head *net_exit_list)
+{
+	inet_twsk_purge(&dccp_hashinfo, AF_INET);
+}
+
 static struct pernet_operations dccp_v4_ops = {
 	.init	= dccp_v4_init_net,
 	.exit	= dccp_v4_exit_net,
+	.exit_batch = dccp_v4_exit_batch,
 	.id	= &dccp_v4_pernet_id,
 	.size   = sizeof(struct dccp_v4_pernet),
 };

net/dccp/ipv6.c

@@ -1115,9 +1115,15 @@ static void __net_exit dccp_v6_exit_net(struct net *net)
 	inet_ctl_sock_destroy(pn->v6_ctl_sk);
 }
 
+static void __net_exit dccp_v6_exit_batch(struct list_head *net_exit_list)
+{
+	inet_twsk_purge(&dccp_hashinfo, AF_INET6);
+}
+
 static struct pernet_operations dccp_v6_ops = {
 	.init   = dccp_v6_init_net,
 	.exit   = dccp_v6_exit_net,
+	.exit_batch = dccp_v6_exit_batch,
 	.id	= &dccp_v6_pernet_id,
 	.size   = sizeof(struct dccp_v6_pernet),
 };

net/ipv4/inet_timewait_sock.c

@@ -52,7 +52,8 @@ static void inet_twsk_kill(struct inet_timewait_sock *tw)
 	spin_unlock(lock);
 
 	/* Disassociate with bind bucket. */
-	bhead = &hashinfo->bhash[tw->tw_bslot];
+	bhead = &hashinfo->bhash[inet_bhashfn(twsk_net(tw), tw->tw_num,
+			hashinfo->bhash_size)];
 
 	spin_lock(&bhead->lock);
 	inet_twsk_bind_unhash(tw, hashinfo);
@@ -111,12 +112,8 @@ void inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk,
 	   Note, that any socket with inet->num != 0 MUST be bound in
 	   binding cache, even if it is closed.
 	 */
-	/* Cache inet_bhashfn(), because 'struct net' might be no longer
-	 * available later in inet_twsk_kill().
-	 */
-	tw->tw_bslot = inet_bhashfn(twsk_net(tw), inet->inet_num,
-				    hashinfo->bhash_size);
-	bhead = &hashinfo->bhash[tw->tw_bslot];
+	bhead = &hashinfo->bhash[inet_bhashfn(twsk_net(tw), inet->inet_num,
+			hashinfo->bhash_size)];
 	spin_lock(&bhead->lock);
 	tw->tw_tb = icsk->icsk_bind_hash;
 	WARN_ON(!icsk->icsk_bind_hash);
@@ -257,3 +254,50 @@ void __inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo, bool rearm)
 	}
 }
 EXPORT_SYMBOL_GPL(__inet_twsk_schedule);
+
+void inet_twsk_purge(struct inet_hashinfo *hashinfo, int family)
+{
+	struct inet_timewait_sock *tw;
+	struct sock *sk;
+	struct hlist_nulls_node *node;
+	unsigned int slot;
+
+	for (slot = 0; slot <= hashinfo->ehash_mask; slot++) {
+		struct inet_ehash_bucket *head = &hashinfo->ehash[slot];
+restart_rcu:
+		cond_resched();
+		rcu_read_lock();
+restart:
+		sk_nulls_for_each_rcu(sk, node, &head->chain) {
+			if (sk->sk_state != TCP_TIME_WAIT)
+				continue;
+			tw = inet_twsk(sk);
+			if ((tw->tw_family != family) ||
+				refcount_read(&twsk_net(tw)->ns.count))
+				continue;
+
+			if (unlikely(!refcount_inc_not_zero(&tw->tw_refcnt)))
+				continue;
+
+			if (unlikely((tw->tw_family != family) ||
+				     refcount_read(&twsk_net(tw)->ns.count))) {
+				inet_twsk_put(tw);
+				goto restart;
+			}
+
+			rcu_read_unlock();
+			local_bh_disable();
+			inet_twsk_deschedule_put(tw);
+			local_bh_enable();
+			goto restart_rcu;
+		}
+		/* If the nulls value we got at the end of this lookup is
+		 * not the expected one, we must restart lookup.
+		 * We probably met an item that was moved to another chain.
+		 */
+		if (get_nulls_value(node) != slot)
+			goto restart;
+		rcu_read_unlock();
+	}
+}
+EXPORT_SYMBOL_GPL(inet_twsk_purge);

net/ipv4/tcp_ipv4.c

@@ -3173,6 +3173,8 @@ static void __net_exit tcp_sk_exit_batch(struct list_head *net_exit_list)
 {
 	struct net *net;
 
+	inet_twsk_purge(&tcp_hashinfo, AF_INET);
+
 	list_for_each_entry(net, net_exit_list, exit_list)
 		tcp_fastopen_ctx_destroy(net);
 }

net/ipv6/tcp_ipv6.c

@@ -2207,9 +2207,15 @@ static void __net_exit tcpv6_net_exit(struct net *net)
 	inet_ctl_sock_destroy(net->ipv6.tcp_sk);
 }
 
+static void __net_exit tcpv6_net_exit_batch(struct list_head *net_exit_list)
+{
+	inet_twsk_purge(&tcp_hashinfo, AF_INET6);
+}
+
 static struct pernet_operations tcpv6_net_ops = {
 	.init	    = tcpv6_net_init,
 	.exit	    = tcpv6_net_exit,
+	.exit_batch = tcpv6_net_exit_batch,
 };
 
 int __init tcpv6_init(void)