netfilter: nftables: add catch-all set element support

This patch extends the set infrastructure to add a special catch-all set
element. If the lookup fails to find an element (or range) in the set,
then the catch-all element is selected. Users can specify a mapping,
expression(s) and timeout to be attached to the catch-all element.

This patch adds a catchall list to the set, this list might contain more
than one single catch-all element (e.g. in case that the catch-all
element is removed and a new one is added in the same transaction).
However, most of the time, there will be either one element or no
elements at all in this list.

The catch-all element is identified via NFT_SET_ELEM_CATCHALL flag and
such special element has no NFTA_SET_ELEM_KEY attribute. There is a new
nft_set_elem_catchall object that stores a reference to the dummy
catch-all element (catchall->elem) whose layout is the same of the set
element type to reuse the existing set element codebase.

The set size does not apply to the catch-all element, users can define a
catch-all element even if the set is full.

The check for valid set element flags hava been updates to report
EOPNOTSUPP in case userspace requests flags that are not supported when
using new userspace nftables and old kernel.

Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

include/net/netfilter/nf_tables.h

@@ -497,6 +497,7 @@ struct nft_set {
 	u8				dlen;
 	u8				num_exprs;
 	struct nft_expr			*exprs[NFT_SET_EXPR_MAX];
+	struct list_head		catchall_list;
 	unsigned char			data[]
 		__attribute__((aligned(__alignof__(u64))));
 };
@@ -522,6 +523,10 @@ struct nft_set *nft_set_lookup_global(const struct net *net,
 				      const struct nlattr *nla_set_id,
 				      u8 genmask);
 
+struct nft_set_ext *nft_set_catchall_lookup(const struct net *net,
+					    const struct nft_set *set);
+void *nft_set_catchall_gc(const struct nft_set *set);
+
 static inline unsigned long nft_set_gc_interval(const struct nft_set *set)
 {
 	return set->gc_int ? msecs_to_jiffies(set->gc_int) : HZ;

include/uapi/linux/netfilter/nf_tables.h

@@ -398,9 +398,11 @@ enum nft_set_attributes {
  * enum nft_set_elem_flags - nf_tables set element flags
  *
  * @NFT_SET_ELEM_INTERVAL_END: element ends the previous interval
+ * @NFT_SET_ELEM_CATCHALL: special catch-all element
  */
 enum nft_set_elem_flags {
 	NFT_SET_ELEM_INTERVAL_END	= 0x1,
+	NFT_SET_ELEM_CATCHALL		= 0x2,
 };
 
 /**

net/netfilter/nft_lookup.c

@@ -30,13 +30,17 @@ void nft_lookup_eval(const struct nft_expr *expr,
 	const struct nft_lookup *priv = nft_expr_priv(expr);
 	const struct nft_set *set = priv->set;
 	const struct nft_set_ext *ext = NULL;
+	const struct net *net = nft_net(pkt);
 	bool found;
 
-	found = set->ops->lookup(nft_net(pkt), set, &regs->data[priv->sreg],
-				 &ext) ^ priv->invert;
+	found = set->ops->lookup(net, set, &regs->data[priv->sreg], &ext) ^
+				 priv->invert;
 	if (!found) {
-		regs->verdict.code = NFT_BREAK;
-		return;
+		ext = nft_set_catchall_lookup(net, set);
+		if (!ext) {
+			regs->verdict.code = NFT_BREAK;
+			return;
+		}
 	}
 
 	if (ext) {

net/netfilter/nft_objref.c

@@ -105,15 +105,18 @@ static void nft_objref_map_eval(const struct nft_expr *expr,
 {
 	struct nft_objref_map *priv = nft_expr_priv(expr);
 	const struct nft_set *set = priv->set;
+	struct net *net = nft_net(pkt);
 	const struct nft_set_ext *ext;
 	struct nft_object *obj;
 	bool found;
 
-	found = set->ops->lookup(nft_net(pkt), set, &regs->data[priv->sreg],
-				 &ext);
+	found = set->ops->lookup(net, set, &regs->data[priv->sreg], &ext);
 	if (!found) {
-		regs->verdict.code = NFT_BREAK;
-		return;
+		ext = nft_set_catchall_lookup(net, set);
+		if (!ext) {
+			regs->verdict.code = NFT_BREAK;
+			return;
+		}
 	}
 	obj = *nft_set_ext_obj(ext);
 	obj->ops->eval(obj, regs, pkt);

net/netfilter/nft_set_hash.c

@@ -350,6 +350,12 @@ needs_gc_run:
 	rhashtable_walk_stop(&hti);
 	rhashtable_walk_exit(&hti);
 
+	he = nft_set_catchall_gc(set);
+	if (he) {
+		gcb = nft_set_gc_batch_check(set, gcb, GFP_ATOMIC);
+		if (gcb)
+			nft_set_gc_batch_add(gcb, he);
+	}
 	nft_set_gc_batch_complete(gcb);
 	queue_delayed_work(system_power_efficient_wq, &priv->gc_work,
 			   nft_set_gc_interval(set));

net/netfilter/nft_set_pipapo.c

@@ -1529,11 +1529,11 @@ static void pipapo_gc(const struct nft_set *set, struct nft_pipapo_match *m)
 {
 	struct nft_pipapo *priv = nft_set_priv(set);
 	int rules_f0, first_rule = 0;
+	struct nft_pipapo_elem *e;
 
 	while ((rules_f0 = pipapo_rules_same_key(m->f, first_rule))) {
 		union nft_pipapo_map_bucket rulemap[NFT_PIPAPO_MAX_FIELDS];
 		struct nft_pipapo_field *f;
-		struct nft_pipapo_elem *e;
 		int i, start, rules_fx;
 
 		start = first_rule;
@@ -1569,6 +1569,10 @@ static void pipapo_gc(const struct nft_set *set, struct nft_pipapo_match *m)
 		}
 	}
 
+	e = nft_set_catchall_gc(set);
+	if (e)
+		nft_set_elem_destroy(set, e, true);
+
 	priv->last_gc = jiffies;
 }
 

net/netfilter/nft_set_rbtree.c

@@ -541,6 +541,12 @@ static void nft_rbtree_gc(struct work_struct *work)
 	write_seqcount_end(&priv->count);
 	write_unlock_bh(&priv->lock);
 
+	rbe = nft_set_catchall_gc(set);
+	if (rbe) {
+		gcb = nft_set_gc_batch_check(set, gcb, GFP_ATOMIC);
+		if (gcb)
+			nft_set_gc_batch_add(gcb, rbe);
+	}
 	nft_set_gc_batch_complete(gcb);
 
 	queue_delayed_work(system_power_efficient_wq, &priv->gc_work,