netfilter: nf_tables: add inet ingress support

This patch adds a new ingress hook for the inet family. The inet ingress
hook emulates the IP receive path code, therefore, unclean packets are
drop before walking over the ruleset in this basechain.

This patch also introduces the nft_base_chain_netdev() helper function
to check if this hook is bound to one or more devices (through the hook
list infrastructure). This check allows to perform the same handling for
the inet ingress as it would be a netdev ingress chain from the control
plane.

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

diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h
index 0bd2a08..3965ce1 100644 (file)
--- a/include/net/netfilter/nf_tables.h
+++ b/include/net/netfilter/nf_tables.h
@@ -1081,6 +1081,12 @@ struct nft_table {
        u8                              *udata;
 };
 
+static inline bool nft_base_chain_netdev(int family, u32 hooknum)
+{
+       return family == NFPROTO_NETDEV ||
+              (family == NFPROTO_INET && hooknum == NF_INET_INGRESS);
+}
+
 void nft_register_chain_type(const struct nft_chain_type *);
 void nft_unregister_chain_type(const struct nft_chain_type *);
 

diff --git a/include/net/netfilter/nf_tables_ipv4.h b/include/net/netfilter/nf_tables_ipv4.h
index ed7b511..1f7bea3 100644 (file)
--- a/include/net/netfilter/nf_tables_ipv4.h
+++ b/include/net/netfilter/nf_tables_ipv4.h
@@ -53,4 +53,37 @@ static inline void nft_set_pktinfo_ipv4_validate(struct nft_pktinfo *pkt,
                nft_set_pktinfo_unspec(pkt, skb);
 }
 
+static inline int nft_set_pktinfo_ipv4_ingress(struct nft_pktinfo *pkt,
+                                              struct sk_buff *skb)
+{
+       struct iphdr *iph;
+       u32 len, thoff;
+
+       if (!pskb_may_pull(skb, sizeof(*iph)))
+               return -1;
+
+       iph = ip_hdr(skb);
+       if (iph->ihl < 5 || iph->version != 4)
+               goto inhdr_error;
+
+       len = ntohs(iph->tot_len);
+       thoff = iph->ihl * 4;
+       if (skb->len < len) {
+               __IP_INC_STATS(nft_net(pkt), IPSTATS_MIB_INTRUNCATEDPKTS);
+               return -1;
+       } else if (len < thoff) {
+               goto inhdr_error;
+       }
+
+       pkt->tprot_set = true;
+       pkt->tprot = iph->protocol;
+       pkt->xt.thoff = thoff;
+       pkt->xt.fragoff = ntohs(iph->frag_off) & IP_OFFSET;
+
+       return 0;
+
+inhdr_error:
+       __IP_INC_STATS(nft_net(pkt), IPSTATS_MIB_INHDRERRORS);
+       return -1;
+}
 #endif

diff --git a/include/net/netfilter/nf_tables_ipv6.h b/include/net/netfilter/nf_tables_ipv6.h
index d0f1c53..867de29 100644 (file)
--- a/include/net/netfilter/nf_tables_ipv6.h
+++ b/include/net/netfilter/nf_tables_ipv6.h
@@ -70,4 +70,50 @@ static inline void nft_set_pktinfo_ipv6_validate(struct nft_pktinfo *pkt,
                nft_set_pktinfo_unspec(pkt, skb);
 }
 
+static inline int nft_set_pktinfo_ipv6_ingress(struct nft_pktinfo *pkt,
+                                              struct sk_buff *skb)
+{
+#if IS_ENABLED(CONFIG_IPV6)
+       unsigned int flags = IP6_FH_F_AUTH;
+       unsigned short frag_off;
+       unsigned int thoff = 0;
+       struct inet6_dev *idev;
+       struct ipv6hdr *ip6h;
+       int protohdr;
+       u32 pkt_len;
+
+       if (!pskb_may_pull(skb, sizeof(*ip6h)))
+               return -1;
+
+       ip6h = ipv6_hdr(skb);
+       if (ip6h->version != 6)
+               goto inhdr_error;
+
+       pkt_len = ntohs(ip6h->payload_len);
+       if (pkt_len + sizeof(*ip6h) > skb->len) {
+               idev = __in6_dev_get(nft_in(pkt));
+               __IP6_INC_STATS(nft_net(pkt), idev, IPSTATS_MIB_INTRUNCATEDPKTS);
+               return -1;
+       }
+
+       protohdr = ipv6_find_hdr(pkt->skb, &thoff, -1, &frag_off, &flags);
+       if (protohdr < 0)
+               goto inhdr_error;
+
+       pkt->tprot_set = true;
+       pkt->tprot = protohdr;
+       pkt->xt.thoff = thoff;
+       pkt->xt.fragoff = frag_off;
+
+       return 0;
+
+inhdr_error:
+       idev = __in6_dev_get(nft_in(pkt));
+       __IP6_INC_STATS(nft_net(pkt), idev, IPSTATS_MIB_INHDRERRORS);
+       return -1;
+#else
+       return -1;
+#endif
+}
+
 #endif

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index ae2c04d..f22ad21 100644 (file)
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -206,7 +206,7 @@ static int nf_tables_register_hook(struct net *net,
        if (basechain->type->ops_register)
                return basechain->type->ops_register(net, ops);
 
-       if (table->family == NFPROTO_NETDEV)
+       if (nft_base_chain_netdev(table->family, basechain->ops.hooknum))
                return nft_netdev_register_hooks(net, &basechain->hook_list);
 
        return nf_register_net_hook(net, &basechain->ops);
@@ -228,7 +228,7 @@ static void nf_tables_unregister_hook(struct net *net,
        if (basechain->type->ops_unregister)
                return basechain->type->ops_unregister(net, ops);
 
-       if (table->family == NFPROTO_NETDEV)
+       if (nft_base_chain_netdev(table->family, basechain->ops.hooknum))
                nft_netdev_unregister_hooks(net, &basechain->hook_list);
        else
                nf_unregister_net_hook(net, &basechain->ops);
@@ -1381,7 +1381,7 @@ static int nft_dump_basechain_hook(struct sk_buff *skb, int family,
        if (nla_put_be32(skb, NFTA_HOOK_PRIORITY, htonl(ops->priority)))
                goto nla_put_failure;
 
-       if (family == NFPROTO_NETDEV) {
+       if (nft_base_chain_netdev(family, ops->hooknum)) {
                nest_devs = nla_nest_start_noflag(skb, NFTA_HOOK_DEVS);
                list_for_each_entry(hook, &basechain->hook_list, list) {
                        if (!first)
@@ -1685,7 +1685,7 @@ void nf_tables_chain_destroy(struct nft_ctx *ctx)
        if (nft_is_base_chain(chain)) {
                struct nft_base_chain *basechain = nft_base_chain(chain);
 
-               if (ctx->family == NFPROTO_NETDEV) {
+               if (nft_base_chain_netdev(ctx->family, basechain->ops.hooknum)) {
                        list_for_each_entry_safe(hook, next,
                                                 &basechain->hook_list, list) {
                                list_del_rcu(&hook->list);
@@ -1877,7 +1877,7 @@ static int nft_chain_parse_hook(struct net *net,
        hook->type = type;
 
        INIT_LIST_HEAD(&hook->list);
-       if (family == NFPROTO_NETDEV) {
+       if (nft_base_chain_netdev(family, hook->num)) {
                err = nft_chain_parse_netdev(net, ha, &hook->list);
                if (err < 0) {
                        module_put(type->owner);
@@ -1944,7 +1944,7 @@ static int nft_basechain_init(struct nft_base_chain *basechain, u8 family,
        INIT_LIST_HEAD(&basechain->hook_list);
        chain = &basechain->chain;
 
-       if (family == NFPROTO_NETDEV) {
+       if (nft_base_chain_netdev(family, hook->num)) {
                list_splice_init(&hook->list, &basechain->hook_list);
                list_for_each_entry(h, &basechain->hook_list, list)
                        nft_basechain_hook_init(&h->ops, family, hook, chain);
@@ -2168,7 +2168,7 @@ static int nf_tables_updchain(struct nft_ctx *ctx, u8 genmask, u8 policy,
                        return -EEXIST;
                }
 
-               if (ctx->family == NFPROTO_NETDEV) {
+               if (nft_base_chain_netdev(ctx->family, hook.num)) {
                        if (!nft_hook_list_equal(&basechain->hook_list,
                                                 &hook.list)) {
                                nft_chain_release_hook(&hook);

diff --git a/net/netfilter/nft_chain_filter.c b/net/netfilter/nft_chain_filter.c
index c78d01b..ff8528a 100644 (file)
--- a/net/netfilter/nft_chain_filter.c
+++ b/net/netfilter/nft_chain_filter.c
@@ -161,16 +161,49 @@ static unsigned int nft_do_chain_inet(void *priv, struct sk_buff *skb,
        return nft_do_chain(&pkt, priv);
 }
 
+static unsigned int nft_do_chain_inet_ingress(void *priv, struct sk_buff *skb,
+                                             const struct nf_hook_state *state)
+{
+       struct nf_hook_state ingress_state = *state;
+       struct nft_pktinfo pkt;
+
+       switch (skb->protocol) {
+       case htons(ETH_P_IP):
+               /* Original hook is NFPROTO_NETDEV and NF_NETDEV_INGRESS. */
+               ingress_state.pf = NFPROTO_IPV4;
+               ingress_state.hook = NF_INET_INGRESS;
+               nft_set_pktinfo(&pkt, skb, &ingress_state);
+
+               if (nft_set_pktinfo_ipv4_ingress(&pkt, skb) < 0)
+                       return NF_DROP;
+               break;
+       case htons(ETH_P_IPV6):
+               ingress_state.pf = NFPROTO_IPV6;
+               ingress_state.hook = NF_INET_INGRESS;
+               nft_set_pktinfo(&pkt, skb, &ingress_state);
+
+               if (nft_set_pktinfo_ipv6_ingress(&pkt, skb) < 0)
+                       return NF_DROP;
+               break;
+       default:
+               return NF_ACCEPT;
+       }
+
+       return nft_do_chain(&pkt, priv);
+}
+
 static const struct nft_chain_type nft_chain_filter_inet = {
        .name           = "filter",
        .type           = NFT_CHAIN_T_DEFAULT,
        .family         = NFPROTO_INET,
-       .hook_mask      = (1 << NF_INET_LOCAL_IN) |
+       .hook_mask      = (1 << NF_INET_INGRESS) |
+                         (1 << NF_INET_LOCAL_IN) |
                          (1 << NF_INET_LOCAL_OUT) |
                          (1 << NF_INET_FORWARD) |
                          (1 << NF_INET_PRE_ROUTING) |
                          (1 << NF_INET_POST_ROUTING),
        .hooks          = {
+               [NF_INET_INGRESS]       = nft_do_chain_inet_ingress,
                [NF_INET_LOCAL_IN]      = nft_do_chain_inet,
                [NF_INET_LOCAL_OUT]     = nft_do_chain_inet,
                [NF_INET_FORWARD]       = nft_do_chain_inet,