From: Felix Fietkau <nbd@nbd.name>
Subject: netfilter: match bypass default table

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 net/ipv4/netfilter/ip_tables.c | 79 +++++++++++++++++++++++++++++++-----------
 1 file changed, 58 insertions(+), 21 deletions(-)

--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -246,6 +246,33 @@ struct ipt_entry *ipt_next_entry(const s
 	return (void *)entry + entry->next_offset;
 }
 
+static bool
+ipt_handle_default_rule(struct ipt_entry *e, unsigned int *verdict)
+{
+	struct xt_entry_target *t;
+	struct xt_standard_target *st;
+
+	if (e->target_offset != sizeof(struct ipt_entry))
+		return false;
+
+	if (!(e->ip.flags & IPT_F_NO_DEF_MATCH))
+		return false;
+
+	t = ipt_get_target(e);
+	if (t->u.kernel.target->target)
+		return false;
+
+	st = (struct xt_standard_target *) t;
+	if (st->verdict == XT_RETURN)
+		return false;
+
+	if (st->verdict >= 0)
+		return false;
+
+	*verdict = (unsigned)(-st->verdict) - 1;
+	return true;
+}
+
 /* Returns one of the generic firewall policies, like NF_ACCEPT. */
 unsigned int
 ipt_do_table(struct sk_buff *skb,
@@ -266,27 +293,28 @@ ipt_do_table(struct sk_buff *skb,
 	unsigned int addend;
 
 	/* Initialization */
+	WARN_ON(!(table->valid_hooks & (1 << hook)));
+	local_bh_disable();
+	private = READ_ONCE(table->private); /* Address dependency. */
+	cpu        = smp_processor_id();
+	table_base = private->entries;
+
+	e = get_entry(table_base, private->hook_entry[hook]);
+	if (ipt_handle_default_rule(e, &verdict)) {
+		struct xt_counters *counter;
+
+		counter = xt_get_this_cpu_counter(&e->counters);
+		ADD_COUNTER(*counter, skb->len, 1);
+		local_bh_enable();
+		return verdict;
+	}
+
 	stackidx = 0;
 	ip = ip_hdr(skb);
 	indev = state->in ? state->in->name : nulldevname;
 	outdev = state->out ? state->out->name : nulldevname;
-	/* We handle fragments by dealing with the first fragment as
-	 * if it was a normal packet.  All other fragments are treated
-	 * normally, except that they will NEVER match rules that ask
-	 * things we don't know, ie. tcp syn flag or ports).  If the
-	 * rule is also a fragment-specific rule, non-fragments won't
-	 * match it. */
-	acpar.fragoff = ntohs(ip->frag_off) & IP_OFFSET;
-	acpar.thoff   = ip_hdrlen(skb);
-	acpar.hotdrop = false;
-	acpar.state   = state;
 
-	WARN_ON(!(table->valid_hooks & (1 << hook)));
-	local_bh_disable();
 	addend = xt_write_recseq_begin();
-	private = READ_ONCE(table->private); /* Address dependency. */
-	cpu        = smp_processor_id();
-	table_base = private->entries;
 	jumpstack  = (struct ipt_entry **)private->jumpstack[cpu];
 
 	/* Switch to alternate jumpstack if we're being invoked via TEE.
@@ -299,7 +327,16 @@ ipt_do_table(struct sk_buff *skb,
 	if (static_key_false(&xt_tee_enabled))
 		jumpstack += private->stacksize * __this_cpu_read(nf_skb_duplicated);
 
-	e = get_entry(table_base, private->hook_entry[hook]);
+	/* We handle fragments by dealing with the first fragment as
+	 * if it was a normal packet.  All other fragments are treated
+	 * normally, except that they will NEVER match rules that ask
+	 * things we don't know, ie. tcp syn flag or ports).  If the
+	 * rule is also a fragment-specific rule, non-fragments won't
+	 * match it. */
+	acpar.fragoff = ntohs(ip->frag_off) & IP_OFFSET;
+	acpar.thoff   = ip_hdrlen(skb);
+	acpar.hotdrop = false;
+	acpar.state   = state;
 
 	do {
 		const struct xt_entry_target *t;