Merge tag 'nf-next-25-03-23' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next

Pablo Neira Ayuso says:

====================
Netfilter updates for net-next

The following batch contains Netfilter updates for net-next:

1) Use kvmalloc in xt_hashlimit, from Denis Kirjanov.

2) Tighten nf_conntrack sysctl accepted values for nf_conntrack_max
   and nf_ct_expect_max, from Nicolas Bouchinet.

3) Avoid lookup in nft_fib if socket is available, from Florian Westphal.

4) Initialize struct lsm_context in nfnetlink_queue to avoid
   hypothetical ENOMEM errors, Chenyuan Yang.

5) Use strscpy() instead of _pad when initializing xtables table name,
   kzalloc is already used to initialized the table memory area.
   From Thorsten Blum.

6) Missing socket lookup by conntrack information for IPv6 traffic
   in nft_socket, there is a similar chunk in IPv4, this was never
   added when IPv6 NAT was introduced. From Maxim Mikityanskiy.

7) Fix clang issues with nf_tables CONFIG_MITIGATION_RETPOLINE,
   from WangYuli.

* tag 'nf-next-25-03-23' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next:
  netfilter: nf_tables: Only use nf_skip_indirect_calls() when MITIGATION_RETPOLINE
  netfilter: socket: Lookup orig tuple for IPv6 SNAT
  netfilter: xtables: Use strscpy() instead of strscpy_pad()
  netfilter: nfnetlink_queue: Initialize ctx to avoid memory allocation error
  netfilter: fib: avoid lookup if socket is available
  netfilter: conntrack: Bound nf_conntrack sysctl writes
  netfilter: xt_hashlimit: replace vmalloc calls with kvmalloc
====================

Link: https://patch.msgid.link/20250323100922.59983-1-pablo@netfilter.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

include/net/netfilter/nft_fib.h

@@ -18,6 +18,27 @@ nft_fib_is_loopback(const struct sk_buff *skb, const struct net_device *in)
 	return skb->pkt_type == PACKET_LOOPBACK || in->flags & IFF_LOOPBACK;
 }
 
+static inline bool nft_fib_can_skip(const struct nft_pktinfo *pkt)
+{
+	const struct net_device *indev = nft_in(pkt);
+	const struct sock *sk;
+
+	switch (nft_hook(pkt)) {
+	case NF_INET_PRE_ROUTING:
+	case NF_INET_INGRESS:
+	case NF_INET_LOCAL_IN:
+		break;
+	default:
+		return false;
+	}
+
+	sk = pkt->skb->sk;
+	if (sk && sk_fullsock(sk))
+	       return sk->sk_rx_dst_ifindex == indev->ifindex;
+
+	return nft_fib_is_loopback(pkt->skb, indev);
+}
+
 int nft_fib_dump(struct sk_buff *skb, const struct nft_expr *expr, bool reset);
 int nft_fib_init(const struct nft_ctx *ctx, const struct nft_expr *expr,
 		 const struct nlattr * const tb[]);

net/ipv4/netfilter/nft_fib_ipv4.c

@@ -71,6 +71,11 @@ void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs,
 	const struct net_device *oif;
 	const struct net_device *found;
 
+	if (nft_fib_can_skip(pkt)) {
+		nft_fib_store_result(dest, priv, nft_in(pkt));
+		return;
+	}
+
 	/*
 	 * Do not set flowi4_oif, it restricts results (for example, asking
 	 * for oif 3 will get RTN_UNICAST result even if the daddr exits
@@ -85,12 +90,6 @@ void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs,
 	else
 		oif = NULL;
 
-	if (nft_hook(pkt) == NF_INET_PRE_ROUTING &&
-	    nft_fib_is_loopback(pkt->skb, nft_in(pkt))) {
-		nft_fib_store_result(dest, priv, nft_in(pkt));
-		return;
-	}
-
 	iph = skb_header_pointer(pkt->skb, noff, sizeof(_iph), &_iph);
 	if (!iph) {
 		regs->verdict.code = NFT_BREAK;

net/ipv6/netfilter/nf_socket_ipv6.c

@@ -103,6 +103,10 @@ struct sock *nf_sk_lookup_slow_v6(struct net *net, const struct sk_buff *skb,
 	struct sk_buff *data_skb = NULL;
 	int doff = 0;
 	int thoff = 0, tproto;
+#if IS_ENABLED(CONFIG_NF_CONNTRACK)
+	enum ip_conntrack_info ctinfo;
+	struct nf_conn const *ct;
+#endif
 
 	tproto = ipv6_find_hdr(skb, &thoff, -1, NULL, NULL);
 	if (tproto < 0) {
@@ -136,6 +140,25 @@ struct sock *nf_sk_lookup_slow_v6(struct net *net, const struct sk_buff *skb,
 		return NULL;
 	}
 
+#if IS_ENABLED(CONFIG_NF_CONNTRACK)
+	/* Do the lookup with the original socket address in
+	 * case this is a reply packet of an established
+	 * SNAT-ted connection.
+	 */
+	ct = nf_ct_get(skb, &ctinfo);
+	if (ct &&
+	    ((tproto != IPPROTO_ICMPV6 &&
+	      ctinfo == IP_CT_ESTABLISHED_REPLY) ||
+	     (tproto == IPPROTO_ICMPV6 &&
+	      ctinfo == IP_CT_RELATED_REPLY)) &&
+	    (ct->status & IPS_SRC_NAT_DONE)) {
+		daddr = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.in6;
+		dport = (tproto == IPPROTO_TCP) ?
+			ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u.tcp.port :
+			ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u.udp.port;
+	}
+#endif
+
 	return nf_socket_get_sock_v6(net, data_skb, doff, tproto, saddr, daddr,
 				     sport, dport, indev);
 }

net/ipv6/netfilter/nft_fib_ipv6.c

@@ -170,6 +170,11 @@ void nft_fib6_eval(const struct nft_expr *expr, struct nft_regs *regs,
 	struct rt6_info *rt;
 	int lookup_flags;
 
+	if (nft_fib_can_skip(pkt)) {
+		nft_fib_store_result(dest, priv, nft_in(pkt));
+		return;
+	}
+
 	if (priv->flags & NFTA_FIB_F_IIF)
 		oif = nft_in(pkt);
 	else if (priv->flags & NFTA_FIB_F_OIF)
@@ -181,17 +186,13 @@ void nft_fib6_eval(const struct nft_expr *expr, struct nft_regs *regs,
 		return;
 	}
 
-	lookup_flags = nft_fib6_flowi_init(&fl6, priv, pkt, oif, iph);
-
-	if (nft_hook(pkt) == NF_INET_PRE_ROUTING ||
-	    nft_hook(pkt) == NF_INET_INGRESS) {
-		if (nft_fib_is_loopback(pkt->skb, nft_in(pkt)) ||
-		    nft_fib_v6_skip_icmpv6(pkt->skb, pkt->tprot, iph)) {
-			nft_fib_store_result(dest, priv, nft_in(pkt));
-			return;
-		}
+	if (nft_fib_v6_skip_icmpv6(pkt->skb, pkt->tprot, iph)) {
+		nft_fib_store_result(dest, priv, nft_in(pkt));
+		return;
 	}
 
+	lookup_flags = nft_fib6_flowi_init(&fl6, priv, pkt, oif, iph);
+
 	*dest = 0;
 	rt = (void *)ip6_route_lookup(nft_net(pkt), &fl6, pkt->skb,
 				      lookup_flags);

net/netfilter/nf_conntrack_standalone.c

@@ -618,7 +618,9 @@ static struct ctl_table nf_ct_sysctl_table[] = {
 		.data		= &nf_conntrack_max,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= proc_dointvec,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= SYSCTL_ZERO,
+		.extra2		= SYSCTL_INT_MAX,
 	},
 	[NF_SYSCTL_CT_COUNT] = {
 		.procname	= "nf_conntrack_count",
@@ -654,7 +656,9 @@ static struct ctl_table nf_ct_sysctl_table[] = {
 		.data		= &nf_ct_expect_max,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= proc_dointvec,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= SYSCTL_ONE,
+		.extra2		= SYSCTL_INT_MAX,
 	},
 	[NF_SYSCTL_CT_ACCT] = {
 		.procname	= "nf_conntrack_acct",
@@ -947,7 +951,9 @@ static struct ctl_table nf_ct_netfilter_table[] = {
 		.data		= &nf_conntrack_max,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
-		.proc_handler	= proc_dointvec,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= SYSCTL_ZERO,
+		.extra2		= SYSCTL_INT_MAX,
 	},
 };
 

net/netfilter/nf_tables_core.c

@@ -21,25 +21,22 @@
 #include <net/netfilter/nf_log.h>
 #include <net/netfilter/nft_meta.h>
 
-#if defined(CONFIG_MITIGATION_RETPOLINE) && defined(CONFIG_X86)
-
+#ifdef CONFIG_MITIGATION_RETPOLINE
 static struct static_key_false nf_tables_skip_direct_calls;
 
-static bool nf_skip_indirect_calls(void)
+static inline bool nf_skip_indirect_calls(void)
 {
 	return static_branch_likely(&nf_tables_skip_direct_calls);
 }
 
-static void __init nf_skip_indirect_calls_enable(void)
+static inline void __init nf_skip_indirect_calls_enable(void)
 {
 	if (!cpu_feature_enabled(X86_FEATURE_RETPOLINE))
 		static_branch_enable(&nf_tables_skip_direct_calls);
 }
 #else
-static inline bool nf_skip_indirect_calls(void) { return false; }
-
 static inline void nf_skip_indirect_calls_enable(void) { }
-#endif
+#endif /* CONFIG_MITIGATION_RETPOLINE */
 
 static noinline void __nft_trace_packet(const struct nft_pktinfo *pkt,
 					const struct nft_verdict *verdict,

net/netfilter/nfnetlink_queue.c

@@ -567,7 +567,7 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue,
 	enum ip_conntrack_info ctinfo = 0;
 	const struct nfnl_ct_hook *nfnl_ct;
 	bool csum_verify;
-	struct lsm_context ctx;
+	struct lsm_context ctx = { NULL, 0, 0 };
 	int seclen = 0;
 	ktime_t tstamp;
 

net/netfilter/xt_hashlimit.c

@@ -15,7 +15,6 @@
 #include <linux/random.h>
 #include <linux/jhash.h>
 #include <linux/slab.h>
-#include <linux/vmalloc.h>
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
 #include <linux/list.h>
@@ -294,8 +293,7 @@ static int htable_create(struct net *net, struct hashlimit_cfg3 *cfg,
 		if (size < 16)
 			size = 16;
 	}
-	/* FIXME: don't use vmalloc() here or anywhere else -HW */
-	hinfo = vmalloc(struct_size(hinfo, hash, size));
+	hinfo = kvmalloc(struct_size(hinfo, hash, size), GFP_KERNEL);
 	if (hinfo == NULL)
 		return -ENOMEM;
 	*out_hinfo = hinfo;
@@ -303,7 +301,7 @@ static int htable_create(struct net *net, struct hashlimit_cfg3 *cfg,
 	/* copy match config into hashtable config */
 	ret = cfg_copy(&hinfo->cfg, (void *)cfg, 3);
 	if (ret) {
-		vfree(hinfo);
+		kvfree(hinfo);
 		return ret;
 	}
 
@@ -322,7 +320,7 @@ static int htable_create(struct net *net, struct hashlimit_cfg3 *cfg,
 	hinfo->rnd_initialized = false;
 	hinfo->name = kstrdup(name, GFP_KERNEL);
 	if (!hinfo->name) {
-		vfree(hinfo);
+		kvfree(hinfo);
 		return -ENOMEM;
 	}
 	spin_lock_init(&hinfo->lock);
@@ -344,7 +342,7 @@ static int htable_create(struct net *net, struct hashlimit_cfg3 *cfg,
 		ops, hinfo);
 	if (hinfo->pde == NULL) {
 		kfree(hinfo->name);
-		vfree(hinfo);
+		kvfree(hinfo);
 		return -ENOMEM;
 	}
 	hinfo->net = net;
@@ -433,7 +431,7 @@ static void htable_put(struct xt_hashlimit_htable *hinfo)
 		cancel_delayed_work_sync(&hinfo->gc_work);
 		htable_selective_cleanup(hinfo, true);
 		kfree(hinfo->name);
-		vfree(hinfo);
+		kvfree(hinfo);
 	}
 }
 

net/netfilter/xt_repldata.h

@@ -29,7 +29,7 @@
 	if (tbl == NULL) \
 		return NULL; \
 	term = (struct type##_error *)&(((char *)tbl)[term_offset]); \
-	strscpy_pad(tbl->repl.name, info->name, sizeof(tbl->repl.name)); \
+	strscpy(tbl->repl.name, info->name); \
 	*term = (struct type##_error)typ2##_ERROR_INIT;  \
 	tbl->repl.valid_hooks = hook_mask; \
 	tbl->repl.num_entries = nhooks + 1; \