Merge tag 'nf-next-23-06-26' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next

Pablo Neira Ayuso says:

====================
Netfilter/IPVS updates for net-next

1) Allow slightly larger IPVS connection table size from Kconfig for
   64-bit arch, from Abhijeet Rastogi.

2) Since IPVS connection table might be larger than 2^20 after previous
   patch, allow to limit it depending on the available memory.
   Moreover, use kvmalloc. From Julian Anastasov.

3) Do not rebuild VLAN header in nft_payload when matching source and
   destination MAC address.

4) Remove nested rcu read lock side in ip_set_test(), from Florian Westphal.

5) Allow to update set size, also from Florian.

6) Improve NAT tuple selection when connection is closing,
   from Florian Westphal.

7) Support for resetting set element stateful expression, from Phil Sutter.

8) Use NLA_POLICY_MAX to narrow down maximum attribute value in nf_tables,
   from Florian Westphal.

* tag 'nf-next-23-06-26' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next:
  netfilter: nf_tables: limit allowed range via nla_policy
  netfilter: nf_tables: Introduce NFT_MSG_GETSETELEM_RESET
  netfilter: snat: evict closing tcp entries on reply tuple collision
  netfilter: nf_tables: permit update of set size
  netfilter: ipset: remove rcu_read_lock_bh pair from ip_set_test
  netfilter: nft_payload: rebuild vlan header when needed
  ipvs: dynamically limit the connection hash table
  ipvs: increase ip_vs_conn_tab_bits range for 64BIT
====================

Link: https://lore.kernel.org/r/20230626064749.75525-1-pablo@netfilter.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

1 files changed, 52 insertions, 20 deletions

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index d543787fc8517..0911e6d5f5d2c 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -567,6 +567,7 @@ static int __nft_trans_set_add(const struct nft_ctx *ctx, int msg_type,
 		nft_trans_set_update(trans) = true;
 		nft_trans_set_gc_int(trans) = desc->gc_int;
 		nft_trans_set_timeout(trans) = desc->timeout;
+		nft_trans_set_size(trans) = desc->size;
 	}
 	nft_trans_commit_list_add_tail(ctx->net, trans);
 
@@ -5442,7 +5443,8 @@ static const struct nla_policy nft_set_elem_list_policy[NFTA_SET_ELEM_LIST_MAX +
 
 static int nft_set_elem_expr_dump(struct sk_buff *skb,
 				  const struct nft_set *set,
-				  const struct nft_set_ext *ext)
+				  const struct nft_set_ext *ext,
+				  bool reset)
 {
 	struct nft_set_elem_expr *elem_expr;
 	u32 size, num_exprs = 0;
@@ -5455,7 +5457,7 @@ static int nft_set_elem_expr_dump(struct sk_buff *skb,
 
 	if (num_exprs == 1) {
 		expr = nft_setelem_expr_at(elem_expr, 0);
-		if (nft_expr_dump(skb, NFTA_SET_ELEM_EXPR, expr, false) < 0)
+		if (nft_expr_dump(skb, NFTA_SET_ELEM_EXPR, expr, reset) < 0)
 			return -1;
 
 		return 0;
@@ -5466,7 +5468,7 @@ static int nft_set_elem_expr_dump(struct sk_buff *skb,
 
 		nft_setelem_expr_foreach(expr, elem_expr, size) {
 			expr = nft_setelem_expr_at(elem_expr, size);
-			if (nft_expr_dump(skb, NFTA_LIST_ELEM, expr, false) < 0)
+			if (nft_expr_dump(skb, NFTA_LIST_ELEM, expr, reset) < 0)
 				goto nla_put_failure;
 		}
 		nla_nest_end(skb, nest);
@@ -5479,11 +5481,13 @@ nla_put_failure:
 
 static int nf_tables_fill_setelem(struct sk_buff *skb,
 				  const struct nft_set *set,
-				  const struct nft_set_elem *elem)
+				  const struct nft_set_elem *elem,
+				  bool reset)
 {
 	const struct nft_set_ext *ext = nft_set_elem_ext(set, elem->priv);
 	unsigned char *b = skb_tail_pointer(skb);
 	struct nlattr *nest;
+	u64 timeout = 0;
 
 	nest = nla_nest_start_noflag(skb, NFTA_LIST_ELEM);
 	if (nest == NULL)
@@ -5506,7 +5510,7 @@ static int nf_tables_fill_setelem(struct sk_buff *skb,
 		goto nla_put_failure;
 
 	if (nft_set_ext_exists(ext, NFT_SET_EXT_EXPRESSIONS) &&
-	    nft_set_elem_expr_dump(skb, set, ext))
+	    nft_set_elem_expr_dump(skb, set, ext, reset))
 		goto nla_put_failure;
 
 	if (nft_set_ext_exists(ext, NFT_SET_EXT_OBJREF) &&
@@ -5519,11 +5523,15 @@ static int nf_tables_fill_setelem(struct sk_buff *skb,
 		         htonl(*nft_set_ext_flags(ext))))
 		goto nla_put_failure;
 
-	if (nft_set_ext_exists(ext, NFT_SET_EXT_TIMEOUT) &&
-	    nla_put_be64(skb, NFTA_SET_ELEM_TIMEOUT,
-			 nf_jiffies64_to_msecs(*nft_set_ext_timeout(ext)),
-			 NFTA_SET_ELEM_PAD))
-		goto nla_put_failure;
+	if (nft_set_ext_exists(ext, NFT_SET_EXT_TIMEOUT)) {
+		timeout = *nft_set_ext_timeout(ext);
+		if (nla_put_be64(skb, NFTA_SET_ELEM_TIMEOUT,
+				 nf_jiffies64_to_msecs(timeout),
+				 NFTA_SET_ELEM_PAD))
+			goto nla_put_failure;
+	} else if (set->flags & NFT_SET_TIMEOUT) {
+		timeout = READ_ONCE(set->timeout);
+	}
 
 	if (nft_set_ext_exists(ext, NFT_SET_EXT_EXPIRATION)) {
 		u64 expires, now = get_jiffies_64();
@@ -5538,6 +5546,9 @@ static int nf_tables_fill_setelem(struct sk_buff *skb,
 				 nf_jiffies64_to_msecs(expires),
 				 NFTA_SET_ELEM_PAD))
 			goto nla_put_failure;
+
+		if (reset)
+			*nft_set_ext_expiration(ext) = now + timeout;
 	}
 
 	if (nft_set_ext_exists(ext, NFT_SET_EXT_USERDATA)) {
@@ -5561,6 +5572,7 @@ struct nft_set_dump_args {
 	const struct netlink_callback	*cb;
 	struct nft_set_iter		iter;
 	struct sk_buff			*skb;
+	bool				reset;
 };
 
 static int nf_tables_dump_setelem(const struct nft_ctx *ctx,
@@ -5571,7 +5583,7 @@ static int nf_tables_dump_setelem(const struct nft_ctx *ctx,
 	struct nft_set_dump_args *args;
 
 	args = container_of(iter, struct nft_set_dump_args, iter);
-	return nf_tables_fill_setelem(args->skb, set, elem);
+	return nf_tables_fill_setelem(args->skb, set, elem, args->reset);
 }
 
 struct nft_set_dump_ctx {
@@ -5580,7 +5592,7 @@ struct nft_set_dump_ctx {
 };
 
 static int nft_set_catchall_dump(struct net *net, struct sk_buff *skb,
-				 const struct nft_set *set)
+				 const struct nft_set *set, bool reset)
 {
 	struct nft_set_elem_catchall *catchall;
 	u8 genmask = nft_genmask_cur(net);
@@ -5595,7 +5607,7 @@ static int nft_set_catchall_dump(struct net *net, struct sk_buff *skb,
 			continue;
 
 		elem.priv = catchall->elem;
-		ret = nf_tables_fill_setelem(skb, set, &elem);
+		ret = nf_tables_fill_setelem(skb, set, &elem, reset);
 		break;
 	}
 
@@ -5613,6 +5625,7 @@ static int nf_tables_dump_set(struct sk_buff *skb, struct netlink_callback *cb)
 	bool set_found = false;
 	struct nlmsghdr *nlh;
 	struct nlattr *nest;
+	bool reset = false;
 	u32 portid, seq;
 	int event;
 
@@ -5660,8 +5673,12 @@ static int nf_tables_dump_set(struct sk_buff *skb, struct netlink_callback *cb)
 	if (nest == NULL)
 		goto nla_put_failure;
 
+	if (NFNL_MSG_TYPE(cb->nlh->nlmsg_type) == NFT_MSG_GETSETELEM_RESET)
+		reset = true;
+
 	args.cb			= cb;
 	args.skb		= skb;
+	args.reset		= reset;
 	args.iter.genmask	= nft_genmask_cur(net);
 	args.iter.skip		= cb->args[0];
 	args.iter.count		= 0;
@@ -5670,7 +5687,7 @@ static int nf_tables_dump_set(struct sk_buff *skb, struct netlink_callback *cb)
 	set->ops->walk(&dump_ctx->ctx, set, &args.iter);
 
 	if (!args.iter.err && args.iter.count == cb->args[0])
-		args.iter.err = nft_set_catchall_dump(net, skb, set);
+		args.iter.err = nft_set_catchall_dump(net, skb, set, reset);
 	rcu_read_unlock();
 
 	nla_nest_end(skb, nest);
@@ -5708,7 +5725,8 @@ static int nf_tables_fill_setelem_info(struct sk_buff *skb,
 				       const struct nft_ctx *ctx, u32 seq,
 				       u32 portid, int event, u16 flags,
 				       const struct nft_set *set,
-				       const struct nft_set_elem *elem)
+				       const struct nft_set_elem *elem,
+				       bool reset)
 {
 	struct nlmsghdr *nlh;
 	struct nlattr *nest;
@@ -5729,7 +5747,7 @@ static int nf_tables_fill_setelem_info(struct sk_buff *skb,
 	if (nest == NULL)
 		goto nla_put_failure;
 
-	err = nf_tables_fill_setelem(skb, set, elem);
+	err = nf_tables_fill_setelem(skb, set, elem, reset);
 	if (err < 0)
 		goto nla_put_failure;
 
@@ -5835,7 +5853,7 @@ static int nft_setelem_get(struct nft_ctx *ctx, struct nft_set *set,
 }
 
 static int nft_get_set_elem(struct nft_ctx *ctx, struct nft_set *set,
-			    const struct nlattr *attr)
+			    const struct nlattr *attr, bool reset)
 {
 	struct nlattr *nla[NFTA_SET_ELEM_MAX + 1];
 	struct nft_set_elem elem;
@@ -5879,7 +5897,8 @@ static int nft_get_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 		return err;
 
 	err = nf_tables_fill_setelem_info(skb, ctx, ctx->seq, ctx->portid,
-					  NFT_MSG_NEWSETELEM, 0, set, &elem);
+					  NFT_MSG_NEWSETELEM, 0, set, &elem,
+					  reset);
 	if (err < 0)
 		goto err_fill_setelem;
 
@@ -5903,6 +5922,7 @@ static int nf_tables_getsetelem(struct sk_buff *skb,
 	struct nft_set *set;
 	struct nlattr *attr;
 	struct nft_ctx ctx;
+	bool reset = false;
 	int rem, err = 0;
 
 	table = nft_table_lookup(net, nla[NFTA_SET_ELEM_LIST_TABLE], family,
@@ -5937,8 +5957,11 @@ static int nf_tables_getsetelem(struct sk_buff *skb,
 	if (!nla[NFTA_SET_ELEM_LIST_ELEMENTS])
 		return -EINVAL;
 
+	if (NFNL_MSG_TYPE(info->nlh->nlmsg_type) == NFT_MSG_GETSETELEM_RESET)
+		reset = true;
+
 	nla_for_each_nested(attr, nla[NFTA_SET_ELEM_LIST_ELEMENTS], rem) {
-		err = nft_get_set_elem(&ctx, set, attr);
+		err = nft_get_set_elem(&ctx, set, attr, reset);
 		if (err < 0) {
 			NL_SET_BAD_ATTR(extack, attr);
 			break;
@@ -5971,7 +5994,7 @@ static void nf_tables_setelem_notify(const struct nft_ctx *ctx,
 		flags |= ctx->flags & (NLM_F_CREATE | NLM_F_EXCL);
 
 	err = nf_tables_fill_setelem_info(skb, ctx, 0, portid, event, flags,
-					  set, elem);
+					  set, elem, false);
 	if (err < 0) {
 		kfree_skb(skb);
 		goto err;
@@ -8920,6 +8943,12 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = {
 		.attr_count	= NFTA_SET_ELEM_LIST_MAX,
 		.policy		= nft_set_elem_list_policy,
 	},
+	[NFT_MSG_GETSETELEM_RESET] = {
+		.call		= nf_tables_getsetelem,
+		.type		= NFNL_CB_RCU,
+		.attr_count	= NFTA_SET_ELEM_LIST_MAX,
+		.policy		= nft_set_elem_list_policy,
+	},
 	[NFT_MSG_DELSETELEM] = {
 		.call		= nf_tables_delsetelem,
 		.type		= NFNL_CB_BATCH,
@@ -9670,6 +9699,9 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 
 				WRITE_ONCE(set->timeout, nft_trans_set_timeout(trans));
 				WRITE_ONCE(set->gc_int, nft_trans_set_gc_int(trans));
+
+				if (nft_trans_set_size(trans))
+					WRITE_ONCE(set->size, nft_trans_set_size(trans));
 			} else {
 				nft_clear(net, nft_trans_set(trans));
 				/* This avoids hitting -EBUSY when deleting the table