diff --git a/include/net/netfilter/ipv4/nf_reject.h b/include/net/netfilter/ipv4/nf_reject.h new file mode 100644 index 000000000000..931fbf812171 --- /dev/null +++ b/include/net/netfilter/ipv4/nf_reject.h @@ -0,0 +1,128 @@ +#ifndef _IPV4_NF_REJECT_H +#define _IPV4_NF_REJECT_H + +#include +#include +#include +#include + +static inline void nf_send_unreach(struct sk_buff *skb_in, int code) +{ + icmp_send(skb_in, ICMP_DEST_UNREACH, code, 0); +} + +/* Send RST reply */ +static void nf_send_reset(struct sk_buff *oldskb, int hook) +{ + struct sk_buff *nskb; + const struct iphdr *oiph; + struct iphdr *niph; + const struct tcphdr *oth; + struct tcphdr _otcph, *tcph; + + /* IP header checks: fragment. */ + if (ip_hdr(oldskb)->frag_off & htons(IP_OFFSET)) + return; + + oth = skb_header_pointer(oldskb, ip_hdrlen(oldskb), + sizeof(_otcph), &_otcph); + if (oth == NULL) + return; + + /* No RST for RST. */ + if (oth->rst) + return; + + if (skb_rtable(oldskb)->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST)) + return; + + /* Check checksum */ + if (nf_ip_checksum(oldskb, hook, ip_hdrlen(oldskb), IPPROTO_TCP)) + return; + oiph = ip_hdr(oldskb); + + nskb = alloc_skb(sizeof(struct iphdr) + sizeof(struct tcphdr) + + LL_MAX_HEADER, GFP_ATOMIC); + if (!nskb) + return; + + skb_reserve(nskb, LL_MAX_HEADER); + + skb_reset_network_header(nskb); + niph = (struct iphdr *)skb_put(nskb, sizeof(struct iphdr)); + niph->version = 4; + niph->ihl = sizeof(struct iphdr) / 4; + niph->tos = 0; + niph->id = 0; + niph->frag_off = htons(IP_DF); + niph->protocol = IPPROTO_TCP; + niph->check = 0; + niph->saddr = oiph->daddr; + niph->daddr = oiph->saddr; + + skb_reset_transport_header(nskb); + tcph = (struct tcphdr *)skb_put(nskb, sizeof(struct tcphdr)); + memset(tcph, 0, sizeof(*tcph)); + tcph->source = oth->dest; + tcph->dest = oth->source; + tcph->doff = sizeof(struct tcphdr) / 4; + + if (oth->ack) + tcph->seq = oth->ack_seq; + else { + tcph->ack_seq = htonl(ntohl(oth->seq) + oth->syn + oth->fin + + oldskb->len - ip_hdrlen(oldskb) - + (oth->doff << 2)); + tcph->ack = 1; + } + + tcph->rst = 1; + tcph->check = ~tcp_v4_check(sizeof(struct tcphdr), niph->saddr, + niph->daddr, 0); + nskb->ip_summed = CHECKSUM_PARTIAL; + nskb->csum_start = (unsigned char *)tcph - nskb->head; + nskb->csum_offset = offsetof(struct tcphdr, check); + + /* ip_route_me_harder expects skb->dst to be set */ + skb_dst_set_noref(nskb, skb_dst(oldskb)); + + nskb->protocol = htons(ETH_P_IP); + if (ip_route_me_harder(nskb, RTN_UNSPEC)) + goto free_nskb; + + niph->ttl = ip4_dst_hoplimit(skb_dst(nskb)); + + /* "Never happens" */ + if (nskb->len > dst_mtu(skb_dst(nskb))) + goto free_nskb; + + nf_ct_attach(nskb, oldskb); + +#ifdef CONFIG_BRIDGE_NETFILTER + /* If we use ip_local_out for bridged traffic, the MAC source on + * the RST will be ours, instead of the destination's. This confuses + * some routers/firewalls, and they drop the packet. So we need to + * build the eth header using the original destination's MAC as the + * source, and send the RST packet directly. + */ + if (oldskb->nf_bridge) { + struct ethhdr *oeth = eth_hdr(oldskb); + nskb->dev = oldskb->nf_bridge->physindev; + niph->tot_len = htons(nskb->len); + ip_send_check(niph); + if (dev_hard_header(nskb, nskb->dev, ntohs(nskb->protocol), + oeth->h_source, oeth->h_dest, nskb->len) < 0) + goto free_nskb; + dev_queue_xmit(nskb); + } else +#endif + ip_local_out(nskb); + + return; + + free_nskb: + kfree_skb(nskb); +} + + +#endif /* _IPV4_NF_REJECT_H */ diff --git a/include/net/netfilter/ipv6/nf_reject.h b/include/net/netfilter/ipv6/nf_reject.h new file mode 100644 index 000000000000..710d17ed70b4 --- /dev/null +++ b/include/net/netfilter/ipv6/nf_reject.h @@ -0,0 +1,171 @@ +#ifndef _IPV6_NF_REJECT_H +#define _IPV6_NF_REJECT_H + +#include +#include +#include +#include +#include + +static inline void +nf_send_unreach6(struct net *net, struct sk_buff *skb_in, unsigned char code, + unsigned int hooknum) +{ + if (hooknum == NF_INET_LOCAL_OUT && skb_in->dev == NULL) + skb_in->dev = net->loopback_dev; + + icmpv6_send(skb_in, ICMPV6_DEST_UNREACH, code, 0); +} + +/* Send RST reply */ +static void nf_send_reset6(struct net *net, struct sk_buff *oldskb, int hook) +{ + struct sk_buff *nskb; + struct tcphdr otcph, *tcph; + unsigned int otcplen, hh_len; + int tcphoff, needs_ack; + const struct ipv6hdr *oip6h = ipv6_hdr(oldskb); + struct ipv6hdr *ip6h; +#define DEFAULT_TOS_VALUE 0x0U + const __u8 tclass = DEFAULT_TOS_VALUE; + struct dst_entry *dst = NULL; + u8 proto; + __be16 frag_off; + struct flowi6 fl6; + + if ((!(ipv6_addr_type(&oip6h->saddr) & IPV6_ADDR_UNICAST)) || + (!(ipv6_addr_type(&oip6h->daddr) & IPV6_ADDR_UNICAST))) { + pr_debug("addr is not unicast.\n"); + return; + } + + proto = oip6h->nexthdr; + tcphoff = ipv6_skip_exthdr(oldskb, ((u8*)(oip6h+1) - oldskb->data), &proto, &frag_off); + + if ((tcphoff < 0) || (tcphoff > oldskb->len)) { + pr_debug("Cannot get TCP header.\n"); + return; + } + + otcplen = oldskb->len - tcphoff; + + /* IP header checks: fragment, too short. */ + if (proto != IPPROTO_TCP || otcplen < sizeof(struct tcphdr)) { + pr_debug("proto(%d) != IPPROTO_TCP, " + "or too short. otcplen = %d\n", + proto, otcplen); + return; + } + + if (skb_copy_bits(oldskb, tcphoff, &otcph, sizeof(struct tcphdr))) + BUG(); + + /* No RST for RST. */ + if (otcph.rst) { + pr_debug("RST is set\n"); + return; + } + + /* Check checksum. */ + if (nf_ip6_checksum(oldskb, hook, tcphoff, IPPROTO_TCP)) { + pr_debug("TCP checksum is invalid\n"); + return; + } + + memset(&fl6, 0, sizeof(fl6)); + fl6.flowi6_proto = IPPROTO_TCP; + fl6.saddr = oip6h->daddr; + fl6.daddr = oip6h->saddr; + fl6.fl6_sport = otcph.dest; + fl6.fl6_dport = otcph.source; + security_skb_classify_flow(oldskb, flowi6_to_flowi(&fl6)); + dst = ip6_route_output(net, NULL, &fl6); + if (dst == NULL || dst->error) { + dst_release(dst); + return; + } + dst = xfrm_lookup(net, dst, flowi6_to_flowi(&fl6), NULL, 0); + if (IS_ERR(dst)) + return; + + hh_len = (dst->dev->hard_header_len + 15)&~15; + nskb = alloc_skb(hh_len + 15 + dst->header_len + sizeof(struct ipv6hdr) + + sizeof(struct tcphdr) + dst->trailer_len, + GFP_ATOMIC); + + if (!nskb) { + net_dbg_ratelimited("cannot alloc skb\n"); + dst_release(dst); + return; + } + + skb_dst_set(nskb, dst); + + skb_reserve(nskb, hh_len + dst->header_len); + + skb_put(nskb, sizeof(struct ipv6hdr)); + skb_reset_network_header(nskb); + ip6h = ipv6_hdr(nskb); + ip6_flow_hdr(ip6h, tclass, 0); + ip6h->hop_limit = ip6_dst_hoplimit(dst); + ip6h->nexthdr = IPPROTO_TCP; + ip6h->saddr = oip6h->daddr; + ip6h->daddr = oip6h->saddr; + + skb_reset_transport_header(nskb); + tcph = (struct tcphdr *)skb_put(nskb, sizeof(struct tcphdr)); + /* Truncate to length (no data) */ + tcph->doff = sizeof(struct tcphdr)/4; + tcph->source = otcph.dest; + tcph->dest = otcph.source; + + if (otcph.ack) { + needs_ack = 0; + tcph->seq = otcph.ack_seq; + tcph->ack_seq = 0; + } else { + needs_ack = 1; + tcph->ack_seq = htonl(ntohl(otcph.seq) + otcph.syn + otcph.fin + + otcplen - (otcph.doff<<2)); + tcph->seq = 0; + } + + /* Reset flags */ + ((u_int8_t *)tcph)[13] = 0; + tcph->rst = 1; + tcph->ack = needs_ack; + tcph->window = 0; + tcph->urg_ptr = 0; + tcph->check = 0; + + /* Adjust TCP checksum */ + tcph->check = csum_ipv6_magic(&ipv6_hdr(nskb)->saddr, + &ipv6_hdr(nskb)->daddr, + sizeof(struct tcphdr), IPPROTO_TCP, + csum_partial(tcph, + sizeof(struct tcphdr), 0)); + + nf_ct_attach(nskb, oldskb); + +#ifdef CONFIG_BRIDGE_NETFILTER + /* If we use ip6_local_out for bridged traffic, the MAC source on + * the RST will be ours, instead of the destination's. This confuses + * some routers/firewalls, and they drop the packet. So we need to + * build the eth header using the original destination's MAC as the + * source, and send the RST packet directly. + */ + if (oldskb->nf_bridge) { + struct ethhdr *oeth = eth_hdr(oldskb); + nskb->dev = oldskb->nf_bridge->physindev; + nskb->protocol = htons(ETH_P_IPV6); + ip6h->payload_len = htons(sizeof(struct tcphdr)); + if (dev_hard_header(nskb, nskb->dev, ntohs(nskb->protocol), + oeth->h_source, oeth->h_dest, nskb->len) < 0) + return; + dev_queue_xmit(nskb); + } else +#endif + ip6_local_out(nskb); +} + +#endif /* _IPV6_NF_REJECT_H */ diff --git a/include/net/netfilter/nf_queue.h b/include/net/netfilter/nf_queue.h index c1d5b3e34a21..84a53d780306 100644 --- a/include/net/netfilter/nf_queue.h +++ b/include/net/netfilter/nf_queue.h @@ -1,6 +1,10 @@ #ifndef _NF_QUEUE_H #define _NF_QUEUE_H +#include +#include +#include + /* Each queued (to userspace) skbuff has one of these. */ struct nf_queue_entry { struct list_head list; @@ -33,4 +37,62 @@ void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict); bool nf_queue_entry_get_refs(struct nf_queue_entry *entry); void nf_queue_entry_release_refs(struct nf_queue_entry *entry); +static inline void init_hashrandom(u32 *jhash_initval) +{ + while (*jhash_initval == 0) + *jhash_initval = prandom_u32(); +} + +static inline u32 hash_v4(const struct sk_buff *skb, u32 jhash_initval) +{ + const struct iphdr *iph = ip_hdr(skb); + + /* packets in either direction go into same queue */ + if ((__force u32)iph->saddr < (__force u32)iph->daddr) + return jhash_3words((__force u32)iph->saddr, + (__force u32)iph->daddr, iph->protocol, jhash_initval); + + return jhash_3words((__force u32)iph->daddr, + (__force u32)iph->saddr, iph->protocol, jhash_initval); +} + +#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES) +static inline u32 hash_v6(const struct sk_buff *skb, u32 jhash_initval) +{ + const struct ipv6hdr *ip6h = ipv6_hdr(skb); + u32 a, b, c; + + if ((__force u32)ip6h->saddr.s6_addr32[3] < + (__force u32)ip6h->daddr.s6_addr32[3]) { + a = (__force u32) ip6h->saddr.s6_addr32[3]; + b = (__force u32) ip6h->daddr.s6_addr32[3]; + } else { + b = (__force u32) ip6h->saddr.s6_addr32[3]; + a = (__force u32) ip6h->daddr.s6_addr32[3]; + } + + if ((__force u32)ip6h->saddr.s6_addr32[1] < + (__force u32)ip6h->daddr.s6_addr32[1]) + c = (__force u32) ip6h->saddr.s6_addr32[1]; + else + c = (__force u32) ip6h->daddr.s6_addr32[1]; + + return jhash_3words(a, b, c, jhash_initval); +} +#endif + +static inline u32 +nfqueue_hash(const struct sk_buff *skb, u16 queue, u16 queues_total, u8 family, + u32 jhash_initval) +{ + if (family == NFPROTO_IPV4) + queue += ((u64) hash_v4(skb, jhash_initval) * queues_total) >> 32; +#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES) + else if (family == NFPROTO_IPV6) + queue += ((u64) hash_v6(skb, jhash_initval) * queues_total) >> 32; +#endif + + return queue; +} + #endif /* _NF_QUEUE_H */ diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index fbfd229a8e99..aa86a15293e1 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -110,11 +110,13 @@ enum nft_table_flags { * * @NFTA_TABLE_NAME: name of the table (NLA_STRING) * @NFTA_TABLE_FLAGS: bitmask of enum nft_table_flags (NLA_U32) + * @NFTA_TABLE_USE: number of chains in this table (NLA_U32) */ enum nft_table_attributes { NFTA_TABLE_UNSPEC, NFTA_TABLE_NAME, NFTA_TABLE_FLAGS, + NFTA_TABLE_USE, __NFTA_TABLE_MAX }; #define NFTA_TABLE_MAX (__NFTA_TABLE_MAX - 1) @@ -553,11 +555,13 @@ enum nft_meta_keys { * * @NFTA_META_DREG: destination register (NLA_U32) * @NFTA_META_KEY: meta data item to load (NLA_U32: nft_meta_keys) + * @NFTA_META_SREG: source register (NLA_U32) */ enum nft_meta_attributes { NFTA_META_UNSPEC, NFTA_META_DREG, NFTA_META_KEY, + NFTA_META_SREG, __NFTA_META_MAX }; #define NFTA_META_MAX (__NFTA_META_MAX - 1) @@ -657,6 +661,26 @@ enum nft_log_attributes { }; #define NFTA_LOG_MAX (__NFTA_LOG_MAX - 1) +/** + * enum nft_queue_attributes - nf_tables queue expression netlink attributes + * + * @NFTA_QUEUE_NUM: netlink queue to send messages to (NLA_U16) + * @NFTA_QUEUE_TOTAL: number of queues to load balance packets on (NLA_U16) + * @NFTA_QUEUE_FLAGS: various flags (NLA_U16) + */ +enum nft_queue_attributes { + NFTA_QUEUE_UNSPEC, + NFTA_QUEUE_NUM, + NFTA_QUEUE_TOTAL, + NFTA_QUEUE_FLAGS, + __NFTA_QUEUE_MAX +}; +#define NFTA_QUEUE_MAX (__NFTA_QUEUE_MAX - 1) + +#define NFT_QUEUE_FLAG_BYPASS 0x01 /* for compatibility with v2 */ +#define NFT_QUEUE_FLAG_CPU_FANOUT 0x02 /* use current CPU (no hashing) */ +#define NFT_QUEUE_FLAG_MASK 0x03 + /** * enum nft_reject_types - nf_tables reject expression reject types * diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 40d56073cd19..81c6910cfa92 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -39,23 +39,33 @@ config NF_CONNTRACK_PROC_COMPAT config NF_TABLES_IPV4 depends on NF_TABLES tristate "IPv4 nf_tables support" - -config NFT_REJECT_IPV4 - depends on NF_TABLES_IPV4 - tristate "nf_tables IPv4 reject support" + help + This option enables the IPv4 support for nf_tables. config NFT_CHAIN_ROUTE_IPV4 depends on NF_TABLES_IPV4 tristate "IPv4 nf_tables route chain support" + help + This option enables the "route" chain for IPv4 in nf_tables. This + chain type is used to force packet re-routing after mangling header + fields such as the source, destination, type of service and + the packet mark. config NFT_CHAIN_NAT_IPV4 depends on NF_TABLES_IPV4 depends on NF_NAT_IPV4 && NFT_NAT tristate "IPv4 nf_tables nat chain support" + help + This option enables the "nat" chain for IPv4 in nf_tables. This + chain type is used to perform Network Address Translation (NAT) + packet transformations such as the source, destination address and + source and destination ports. config NF_TABLES_ARP depends on NF_TABLES tristate "ARP nf_tables support" + help + This option enables the ARP support for nf_tables. config IP_NF_IPTABLES tristate "IP tables support (required for filtering/masq/NAT)" diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 19df72b7ba88..c16be9d58420 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -28,7 +28,6 @@ obj-$(CONFIG_NF_NAT_SNMP_BASIC) += nf_nat_snmp_basic.o obj-$(CONFIG_NF_NAT_PROTO_GRE) += nf_nat_proto_gre.o obj-$(CONFIG_NF_TABLES_IPV4) += nf_tables_ipv4.o -obj-$(CONFIG_NFT_REJECT_IPV4) += nft_reject_ipv4.o obj-$(CONFIG_NFT_CHAIN_ROUTE_IPV4) += nft_chain_route_ipv4.o obj-$(CONFIG_NFT_CHAIN_NAT_IPV4) += nft_chain_nat_ipv4.o obj-$(CONFIG_NF_TABLES_ARP) += nf_tables_arp.o diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c index b969131ad1c1..5b6e0df4ccff 100644 --- a/net/ipv4/netfilter/ipt_REJECT.c +++ b/net/ipv4/netfilter/ipt_REJECT.c @@ -17,10 +17,6 @@ #include #include #include -#include -#include -#include -#include #include #include #include @@ -28,128 +24,12 @@ #include #endif +#include + MODULE_LICENSE("GPL"); MODULE_AUTHOR("Netfilter Core Team "); MODULE_DESCRIPTION("Xtables: packet \"rejection\" target for IPv4"); -/* Send RST reply */ -static void send_reset(struct sk_buff *oldskb, int hook) -{ - struct sk_buff *nskb; - const struct iphdr *oiph; - struct iphdr *niph; - const struct tcphdr *oth; - struct tcphdr _otcph, *tcph; - - /* IP header checks: fragment. */ - if (ip_hdr(oldskb)->frag_off & htons(IP_OFFSET)) - return; - - oth = skb_header_pointer(oldskb, ip_hdrlen(oldskb), - sizeof(_otcph), &_otcph); - if (oth == NULL) - return; - - /* No RST for RST. */ - if (oth->rst) - return; - - if (skb_rtable(oldskb)->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST)) - return; - - /* Check checksum */ - if (nf_ip_checksum(oldskb, hook, ip_hdrlen(oldskb), IPPROTO_TCP)) - return; - oiph = ip_hdr(oldskb); - - nskb = alloc_skb(sizeof(struct iphdr) + sizeof(struct tcphdr) + - LL_MAX_HEADER, GFP_ATOMIC); - if (!nskb) - return; - - skb_reserve(nskb, LL_MAX_HEADER); - - skb_reset_network_header(nskb); - niph = (struct iphdr *)skb_put(nskb, sizeof(struct iphdr)); - niph->version = 4; - niph->ihl = sizeof(struct iphdr) / 4; - niph->tos = 0; - niph->id = 0; - niph->frag_off = htons(IP_DF); - niph->protocol = IPPROTO_TCP; - niph->check = 0; - niph->saddr = oiph->daddr; - niph->daddr = oiph->saddr; - - skb_reset_transport_header(nskb); - tcph = (struct tcphdr *)skb_put(nskb, sizeof(struct tcphdr)); - memset(tcph, 0, sizeof(*tcph)); - tcph->source = oth->dest; - tcph->dest = oth->source; - tcph->doff = sizeof(struct tcphdr) / 4; - - if (oth->ack) - tcph->seq = oth->ack_seq; - else { - tcph->ack_seq = htonl(ntohl(oth->seq) + oth->syn + oth->fin + - oldskb->len - ip_hdrlen(oldskb) - - (oth->doff << 2)); - tcph->ack = 1; - } - - tcph->rst = 1; - tcph->check = ~tcp_v4_check(sizeof(struct tcphdr), niph->saddr, - niph->daddr, 0); - nskb->ip_summed = CHECKSUM_PARTIAL; - nskb->csum_start = (unsigned char *)tcph - nskb->head; - nskb->csum_offset = offsetof(struct tcphdr, check); - - /* ip_route_me_harder expects skb->dst to be set */ - skb_dst_set_noref(nskb, skb_dst(oldskb)); - - nskb->protocol = htons(ETH_P_IP); - if (ip_route_me_harder(nskb, RTN_UNSPEC)) - goto free_nskb; - - niph->ttl = ip4_dst_hoplimit(skb_dst(nskb)); - - /* "Never happens" */ - if (nskb->len > dst_mtu(skb_dst(nskb))) - goto free_nskb; - - nf_ct_attach(nskb, oldskb); - -#ifdef CONFIG_BRIDGE_NETFILTER - /* If we use ip_local_out for bridged traffic, the MAC source on - * the RST will be ours, instead of the destination's. This confuses - * some routers/firewalls, and they drop the packet. So we need to - * build the eth header using the original destination's MAC as the - * source, and send the RST packet directly. - */ - if (oldskb->nf_bridge) { - struct ethhdr *oeth = eth_hdr(oldskb); - nskb->dev = oldskb->nf_bridge->physindev; - niph->tot_len = htons(nskb->len); - ip_send_check(niph); - if (dev_hard_header(nskb, nskb->dev, ntohs(nskb->protocol), - oeth->h_source, oeth->h_dest, nskb->len) < 0) - goto free_nskb; - dev_queue_xmit(nskb); - } else -#endif - ip_local_out(nskb); - - return; - - free_nskb: - kfree_skb(nskb); -} - -static inline void send_unreach(struct sk_buff *skb_in, int code) -{ - icmp_send(skb_in, ICMP_DEST_UNREACH, code, 0); -} - static unsigned int reject_tg(struct sk_buff *skb, const struct xt_action_param *par) { @@ -157,28 +37,28 @@ reject_tg(struct sk_buff *skb, const struct xt_action_param *par) switch (reject->with) { case IPT_ICMP_NET_UNREACHABLE: - send_unreach(skb, ICMP_NET_UNREACH); + nf_send_unreach(skb, ICMP_NET_UNREACH); break; case IPT_ICMP_HOST_UNREACHABLE: - send_unreach(skb, ICMP_HOST_UNREACH); + nf_send_unreach(skb, ICMP_HOST_UNREACH); break; case IPT_ICMP_PROT_UNREACHABLE: - send_unreach(skb, ICMP_PROT_UNREACH); + nf_send_unreach(skb, ICMP_PROT_UNREACH); break; case IPT_ICMP_PORT_UNREACHABLE: - send_unreach(skb, ICMP_PORT_UNREACH); + nf_send_unreach(skb, ICMP_PORT_UNREACH); break; case IPT_ICMP_NET_PROHIBITED: - send_unreach(skb, ICMP_NET_ANO); + nf_send_unreach(skb, ICMP_NET_ANO); break; case IPT_ICMP_HOST_PROHIBITED: - send_unreach(skb, ICMP_HOST_ANO); + nf_send_unreach(skb, ICMP_HOST_ANO); break; case IPT_ICMP_ADMIN_PROHIBITED: - send_unreach(skb, ICMP_PKT_FILTERED); + nf_send_unreach(skb, ICMP_PKT_FILTERED); break; case IPT_TCP_RESET: - send_reset(skb, par->hooknum); + nf_send_reset(skb, par->hooknum); case IPT_ICMP_ECHOREPLY: /* Doesn't happen. */ break; diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig index 7702f9e90a04..35750df744dc 100644 --- a/net/ipv6/netfilter/Kconfig +++ b/net/ipv6/netfilter/Kconfig @@ -28,15 +28,27 @@ config NF_CONNTRACK_IPV6 config NF_TABLES_IPV6 depends on NF_TABLES tristate "IPv6 nf_tables support" + help + This option enables the IPv6 support for nf_tables. config NFT_CHAIN_ROUTE_IPV6 depends on NF_TABLES_IPV6 tristate "IPv6 nf_tables route chain support" + help + This option enables the "route" chain for IPv6 in nf_tables. This + chain type is used to force packet re-routing after mangling header + fields such as the source, destination, flowlabel, hop-limit and + the packet mark. config NFT_CHAIN_NAT_IPV6 depends on NF_TABLES_IPV6 depends on NF_NAT_IPV6 && NFT_NAT tristate "IPv6 nf_tables nat chain support" + help + This option enables the "nat" chain for IPv6 in nf_tables. This + chain type is used to perform Network Address Translation (NAT) + packet transformations such as the source, destination address and + source and destination ports. config IP6_NF_IPTABLES tristate "IP6 tables support (required for filtering)" diff --git a/net/ipv6/netfilter/ip6t_REJECT.c b/net/ipv6/netfilter/ip6t_REJECT.c index da00a2ecde55..544b0a9da1b5 100644 --- a/net/ipv6/netfilter/ip6t_REJECT.c +++ b/net/ipv6/netfilter/ip6t_REJECT.c @@ -23,181 +23,18 @@ #include #include #include -#include -#include #include -#include -#include -#include #include #include #include #include +#include + MODULE_AUTHOR("Yasuyuki KOZAKAI "); MODULE_DESCRIPTION("Xtables: packet \"rejection\" target for IPv6"); MODULE_LICENSE("GPL"); -/* Send RST reply */ -static void send_reset(struct net *net, struct sk_buff *oldskb, int hook) -{ - struct sk_buff *nskb; - struct tcphdr otcph, *tcph; - unsigned int otcplen, hh_len; - int tcphoff, needs_ack; - const struct ipv6hdr *oip6h = ipv6_hdr(oldskb); - struct ipv6hdr *ip6h; -#define DEFAULT_TOS_VALUE 0x0U - const __u8 tclass = DEFAULT_TOS_VALUE; - struct dst_entry *dst = NULL; - u8 proto; - __be16 frag_off; - struct flowi6 fl6; - - if ((!(ipv6_addr_type(&oip6h->saddr) & IPV6_ADDR_UNICAST)) || - (!(ipv6_addr_type(&oip6h->daddr) & IPV6_ADDR_UNICAST))) { - pr_debug("addr is not unicast.\n"); - return; - } - - proto = oip6h->nexthdr; - tcphoff = ipv6_skip_exthdr(oldskb, ((u8*)(oip6h+1) - oldskb->data), &proto, &frag_off); - - if ((tcphoff < 0) || (tcphoff > oldskb->len)) { - pr_debug("Cannot get TCP header.\n"); - return; - } - - otcplen = oldskb->len - tcphoff; - - /* IP header checks: fragment, too short. */ - if (proto != IPPROTO_TCP || otcplen < sizeof(struct tcphdr)) { - pr_debug("proto(%d) != IPPROTO_TCP, " - "or too short. otcplen = %d\n", - proto, otcplen); - return; - } - - if (skb_copy_bits(oldskb, tcphoff, &otcph, sizeof(struct tcphdr))) - BUG(); - - /* No RST for RST. */ - if (otcph.rst) { - pr_debug("RST is set\n"); - return; - } - - /* Check checksum. */ - if (nf_ip6_checksum(oldskb, hook, tcphoff, IPPROTO_TCP)) { - pr_debug("TCP checksum is invalid\n"); - return; - } - - memset(&fl6, 0, sizeof(fl6)); - fl6.flowi6_proto = IPPROTO_TCP; - fl6.saddr = oip6h->daddr; - fl6.daddr = oip6h->saddr; - fl6.fl6_sport = otcph.dest; - fl6.fl6_dport = otcph.source; - security_skb_classify_flow(oldskb, flowi6_to_flowi(&fl6)); - dst = ip6_route_output(net, NULL, &fl6); - if (dst == NULL || dst->error) { - dst_release(dst); - return; - } - dst = xfrm_lookup(net, dst, flowi6_to_flowi(&fl6), NULL, 0); - if (IS_ERR(dst)) - return; - - hh_len = (dst->dev->hard_header_len + 15)&~15; - nskb = alloc_skb(hh_len + 15 + dst->header_len + sizeof(struct ipv6hdr) - + sizeof(struct tcphdr) + dst->trailer_len, - GFP_ATOMIC); - - if (!nskb) { - net_dbg_ratelimited("cannot alloc skb\n"); - dst_release(dst); - return; - } - - skb_dst_set(nskb, dst); - - skb_reserve(nskb, hh_len + dst->header_len); - - skb_put(nskb, sizeof(struct ipv6hdr)); - skb_reset_network_header(nskb); - ip6h = ipv6_hdr(nskb); - ip6_flow_hdr(ip6h, tclass, 0); - ip6h->hop_limit = ip6_dst_hoplimit(dst); - ip6h->nexthdr = IPPROTO_TCP; - ip6h->saddr = oip6h->daddr; - ip6h->daddr = oip6h->saddr; - - skb_reset_transport_header(nskb); - tcph = (struct tcphdr *)skb_put(nskb, sizeof(struct tcphdr)); - /* Truncate to length (no data) */ - tcph->doff = sizeof(struct tcphdr)/4; - tcph->source = otcph.dest; - tcph->dest = otcph.source; - - if (otcph.ack) { - needs_ack = 0; - tcph->seq = otcph.ack_seq; - tcph->ack_seq = 0; - } else { - needs_ack = 1; - tcph->ack_seq = htonl(ntohl(otcph.seq) + otcph.syn + otcph.fin - + otcplen - (otcph.doff<<2)); - tcph->seq = 0; - } - - /* Reset flags */ - ((u_int8_t *)tcph)[13] = 0; - tcph->rst = 1; - tcph->ack = needs_ack; - tcph->window = 0; - tcph->urg_ptr = 0; - tcph->check = 0; - - /* Adjust TCP checksum */ - tcph->check = csum_ipv6_magic(&ipv6_hdr(nskb)->saddr, - &ipv6_hdr(nskb)->daddr, - sizeof(struct tcphdr), IPPROTO_TCP, - csum_partial(tcph, - sizeof(struct tcphdr), 0)); - - nf_ct_attach(nskb, oldskb); - -#ifdef CONFIG_BRIDGE_NETFILTER - /* If we use ip6_local_out for bridged traffic, the MAC source on - * the RST will be ours, instead of the destination's. This confuses - * some routers/firewalls, and they drop the packet. So we need to - * build the eth header using the original destination's MAC as the - * source, and send the RST packet directly. - */ - if (oldskb->nf_bridge) { - struct ethhdr *oeth = eth_hdr(oldskb); - nskb->dev = oldskb->nf_bridge->physindev; - nskb->protocol = htons(ETH_P_IPV6); - ip6h->payload_len = htons(sizeof(struct tcphdr)); - if (dev_hard_header(nskb, nskb->dev, ntohs(nskb->protocol), - oeth->h_source, oeth->h_dest, nskb->len) < 0) - return; - dev_queue_xmit(nskb); - } else -#endif - ip6_local_out(nskb); -} - -static inline void -send_unreach(struct net *net, struct sk_buff *skb_in, unsigned char code, - unsigned int hooknum) -{ - if (hooknum == NF_INET_LOCAL_OUT && skb_in->dev == NULL) - skb_in->dev = net->loopback_dev; - - icmpv6_send(skb_in, ICMPV6_DEST_UNREACH, code, 0); -} static unsigned int reject_tg6(struct sk_buff *skb, const struct xt_action_param *par) @@ -208,25 +45,25 @@ reject_tg6(struct sk_buff *skb, const struct xt_action_param *par) pr_debug("%s: medium point\n", __func__); switch (reject->with) { case IP6T_ICMP6_NO_ROUTE: - send_unreach(net, skb, ICMPV6_NOROUTE, par->hooknum); + nf_send_unreach6(net, skb, ICMPV6_NOROUTE, par->hooknum); break; case IP6T_ICMP6_ADM_PROHIBITED: - send_unreach(net, skb, ICMPV6_ADM_PROHIBITED, par->hooknum); + nf_send_unreach6(net, skb, ICMPV6_ADM_PROHIBITED, par->hooknum); break; case IP6T_ICMP6_NOT_NEIGHBOUR: - send_unreach(net, skb, ICMPV6_NOT_NEIGHBOUR, par->hooknum); + nf_send_unreach6(net, skb, ICMPV6_NOT_NEIGHBOUR, par->hooknum); break; case IP6T_ICMP6_ADDR_UNREACH: - send_unreach(net, skb, ICMPV6_ADDR_UNREACH, par->hooknum); + nf_send_unreach6(net, skb, ICMPV6_ADDR_UNREACH, par->hooknum); break; case IP6T_ICMP6_PORT_UNREACH: - send_unreach(net, skb, ICMPV6_PORT_UNREACH, par->hooknum); + nf_send_unreach6(net, skb, ICMPV6_PORT_UNREACH, par->hooknum); break; case IP6T_ICMP6_ECHOREPLY: /* Do nothing */ break; case IP6T_TCP_RESET: - send_reset(net, skb, par->hooknum); + nf_send_reset6(net, skb, par->hooknum); break; default: net_info_ratelimited("case %u not handled yet\n", reject->with); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index c17902cb5df9..c3b3b26c4c4e 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -414,47 +414,104 @@ config NETFILTER_SYNPROXY endif # NF_CONNTRACK config NF_TABLES - depends on NETFILTER_NETLINK + select NETFILTER_NETLINK tristate "Netfilter nf_tables support" + help + nftables is the new packet classification framework that intends to + replace the existing {ip,ip6,arp,eb}_tables infrastructure. It + provides a pseudo-state machine with an extensible instruction-set + (also known as expressions) that the userspace 'nft' utility + (http://www.netfilter.org/projects/nftables) uses to build the + rule-set. It also comes with the generic set infrastructure that + allows you to construct mappings between matchings and actions + for performance lookups. + + To compile it as a module, choose M here. config NFT_EXTHDR depends on NF_TABLES tristate "Netfilter nf_tables IPv6 exthdr module" + help + This option adds the "exthdr" expression that you can use to match + IPv6 extension headers. config NFT_META depends on NF_TABLES tristate "Netfilter nf_tables meta module" + help + This option adds the "meta" expression that you can use to match and + to set packet metainformation such as the packet mark. config NFT_CT depends on NF_TABLES depends on NF_CONNTRACK tristate "Netfilter nf_tables conntrack module" + help + This option adds the "meta" expression that you can use to match + connection tracking information such as the flow state. config NFT_RBTREE depends on NF_TABLES tristate "Netfilter nf_tables rbtree set module" + help + This option adds the "rbtree" set type (Red Black tree) that is used + to build interval-based sets. config NFT_HASH depends on NF_TABLES tristate "Netfilter nf_tables hash set module" + help + This option adds the "hash" set type that is used to build one-way + mappings between matchings and actions. config NFT_COUNTER depends on NF_TABLES tristate "Netfilter nf_tables counter module" + help + This option adds the "counter" expression that you can use to + include packet and byte counters in a rule. config NFT_LOG depends on NF_TABLES tristate "Netfilter nf_tables log module" + help + This option adds the "log" expression that you can use to log + packets matching some criteria. config NFT_LIMIT depends on NF_TABLES tristate "Netfilter nf_tables limit module" + help + This option adds the "limit" expression that you can use to + ratelimit rule matchings. config NFT_NAT depends on NF_TABLES depends on NF_CONNTRACK depends on NF_NAT tristate "Netfilter nf_tables nat module" + help + This option adds the "nat" expression that you can use to perform + typical Network Address Translation (NAT) packet transformations. + +config NFT_QUEUE + depends on NF_TABLES + depends on NETFILTER_XTABLES + depends on NETFILTER_NETLINK_QUEUE + tristate "Netfilter nf_tables queue module" + help + This is required if you intend to use the userspace queueing + infrastructure (also known as NFQUEUE) from nftables. + +config NFT_REJECT + depends on NF_TABLES + depends on NF_TABLES_IPV6 || !NF_TABLES_IPV6 + default m if NETFILTER_ADVANCED=n + tristate "Netfilter nf_tables reject support" + help + This option adds the "reject" expression that you can use to + explicitly deny and notify via TCP reset/ICMP informational errors + unallowed traffic. config NFT_COMPAT depends on NF_TABLES diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 407fc232f625..78b4e1c9c595 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -76,7 +76,8 @@ obj-$(CONFIG_NFT_META) += nft_meta.o obj-$(CONFIG_NFT_CT) += nft_ct.o obj-$(CONFIG_NFT_LIMIT) += nft_limit.o obj-$(CONFIG_NFT_NAT) += nft_nat.o -#nf_tables-objs += nft_meta_target.o +obj-$(CONFIG_NFT_QUEUE) += nft_queue.o +obj-$(CONFIG_NFT_REJECT) += nft_reject.o obj-$(CONFIG_NFT_RBTREE) += nft_rbtree.o obj-$(CONFIG_NFT_HASH) += nft_hash.o obj-$(CONFIG_NFT_COUNTER) += nft_counter.o diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index f93b7d06f4be..629b6da98318 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -180,7 +180,8 @@ static int nf_tables_fill_table_info(struct sk_buff *skb, u32 portid, u32 seq, nfmsg->res_id = 0; if (nla_put_string(skb, NFTA_TABLE_NAME, table->name) || - nla_put_be32(skb, NFTA_TABLE_FLAGS, htonl(table->flags))) + nla_put_be32(skb, NFTA_TABLE_FLAGS, htonl(table->flags)) || + nla_put_be32(skb, NFTA_TABLE_USE, htonl(table->use))) goto nla_put_failure; return nlmsg_end(skb, nlh); @@ -1923,12 +1924,14 @@ static int nft_ctx_init_from_setattr(struct nft_ctx *ctx, { struct net *net = sock_net(skb->sk); const struct nfgenmsg *nfmsg = nlmsg_data(nlh); - const struct nft_af_info *afi; + const struct nft_af_info *afi = NULL; const struct nft_table *table = NULL; - afi = nf_tables_afinfo_lookup(net, nfmsg->nfgen_family, false); - if (IS_ERR(afi)) - return PTR_ERR(afi); + if (nfmsg->nfgen_family != NFPROTO_UNSPEC) { + afi = nf_tables_afinfo_lookup(net, nfmsg->nfgen_family, false); + if (IS_ERR(afi)) + return PTR_ERR(afi); + } if (nla[NFTA_SET_TABLE] != NULL) { table = nf_tables_table_lookup(afi, nla[NFTA_SET_TABLE]); @@ -1973,11 +1976,14 @@ static int nf_tables_set_alloc_name(struct nft_ctx *ctx, struct nft_set *set, return -ENOMEM; list_for_each_entry(i, &ctx->table->sets, list) { - if (!sscanf(i->name, name, &n)) + int tmp; + + if (!sscanf(i->name, name, &tmp)) continue; - if (n < 0 || n > BITS_PER_LONG * PAGE_SIZE) + if (tmp < 0 || tmp > BITS_PER_LONG * PAGE_SIZE) continue; - set_bit(n, inuse); + + set_bit(tmp, inuse); } n = find_first_zero_bit(inuse, BITS_PER_LONG * PAGE_SIZE); @@ -2094,8 +2100,8 @@ done: return skb->len; } -static int nf_tables_dump_sets_all(struct nft_ctx *ctx, struct sk_buff *skb, - struct netlink_callback *cb) +static int nf_tables_dump_sets_family(struct nft_ctx *ctx, struct sk_buff *skb, + struct netlink_callback *cb) { const struct nft_set *set; unsigned int idx = 0, s_idx = cb->args[0]; @@ -2127,6 +2133,61 @@ done: return skb->len; } +static int nf_tables_dump_sets_all(struct nft_ctx *ctx, struct sk_buff *skb, + struct netlink_callback *cb) +{ + const struct nft_set *set; + unsigned int idx, s_idx = cb->args[0]; + const struct nft_af_info *afi; + struct nft_table *table, *cur_table = (struct nft_table *)cb->args[2]; + struct net *net = sock_net(skb->sk); + int cur_family = cb->args[3]; + + if (cb->args[1]) + return skb->len; + + list_for_each_entry(afi, &net->nft.af_info, list) { + if (cur_family) { + if (afi->family != cur_family) + continue; + + cur_family = 0; + } + + list_for_each_entry(table, &afi->tables, list) { + if (cur_table) { + if (cur_table != table) + continue; + + cur_table = NULL; + } + + ctx->table = table; + ctx->afi = afi; + idx = 0; + list_for_each_entry(set, &ctx->table->sets, list) { + if (idx < s_idx) + goto cont; + if (nf_tables_fill_set(skb, ctx, set, + NFT_MSG_NEWSET, + NLM_F_MULTI) < 0) { + cb->args[0] = idx; + cb->args[2] = (unsigned long) table; + cb->args[3] = afi->family; + goto done; + } +cont: + idx++; + } + if (s_idx) + s_idx = 0; + } + } + cb->args[1] = 1; +done: + return skb->len; +} + static int nf_tables_dump_sets(struct sk_buff *skb, struct netlink_callback *cb) { const struct nfgenmsg *nfmsg = nlmsg_data(cb->nlh); @@ -2143,9 +2204,12 @@ static int nf_tables_dump_sets(struct sk_buff *skb, struct netlink_callback *cb) if (err < 0) return err; - if (ctx.table == NULL) - ret = nf_tables_dump_sets_all(&ctx, skb, cb); - else + if (ctx.table == NULL) { + if (ctx.afi == NULL) + ret = nf_tables_dump_sets_all(&ctx, skb, cb); + else + ret = nf_tables_dump_sets_family(&ctx, skb, cb); + } else ret = nf_tables_dump_sets_table(&ctx, skb, cb); return ret; @@ -2158,6 +2222,7 @@ static int nf_tables_getset(struct sock *nlsk, struct sk_buff *skb, const struct nft_set *set; struct nft_ctx ctx; struct sk_buff *skb2; + const struct nfgenmsg *nfmsg = nlmsg_data(nlh); int err; /* Verify existance before starting dump */ @@ -2172,6 +2237,10 @@ static int nf_tables_getset(struct sock *nlsk, struct sk_buff *skb, return netlink_dump_start(nlsk, skb, nlh, &c); } + /* Only accept unspec with dump */ + if (nfmsg->nfgen_family == NFPROTO_UNSPEC) + return -EAFNOSUPPORT; + set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_NAME]); if (IS_ERR(set)) return PTR_ERR(set); @@ -2341,6 +2410,7 @@ static int nf_tables_delset(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { + const struct nfgenmsg *nfmsg = nlmsg_data(nlh); struct nft_set *set; struct nft_ctx ctx; int err; @@ -2352,6 +2422,9 @@ static int nf_tables_delset(struct sock *nlsk, struct sk_buff *skb, if (err < 0) return err; + if (nfmsg->nfgen_family == NFPROTO_UNSPEC) + return -EAFNOSUPPORT; + set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_NAME]); if (IS_ERR(set)) return PTR_ERR(set); @@ -2521,9 +2594,8 @@ static int nf_tables_dump_set(struct sk_buff *skb, struct netlink_callback *cb) u32 portid, seq; int event, err; - nfmsg = nlmsg_data(cb->nlh); - err = nlmsg_parse(cb->nlh, sizeof(*nfmsg), nla, NFTA_SET_ELEM_LIST_MAX, - nft_set_elem_list_policy); + err = nlmsg_parse(cb->nlh, sizeof(struct nfgenmsg), nla, + NFTA_SET_ELEM_LIST_MAX, nft_set_elem_list_policy); if (err < 0) return err; diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c index cb9e685caae1..e8fcc343c2b9 100644 --- a/net/netfilter/nf_tables_core.c +++ b/net/netfilter/nf_tables_core.c @@ -164,7 +164,7 @@ next_rule: break; } - switch (data[NFT_REG_VERDICT].verdict) { + switch (data[NFT_REG_VERDICT].verdict & NF_VERDICT_MASK) { case NF_ACCEPT: case NF_DROP: case NF_QUEUE: @@ -172,6 +172,9 @@ next_rule: nft_trace_packet(pkt, chain, rulenum, NFT_TRACE_RULE); return data[NFT_REG_VERDICT].verdict; + } + + switch (data[NFT_REG_VERDICT].verdict) { case NFT_JUMP: if (unlikely(pkt->skb->nf_trace)) nft_trace_packet(pkt, chain, rulenum, NFT_TRACE_RULE); diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c index 8c28220a90b3..1ceaaa6dfe72 100644 --- a/net/netfilter/nft_meta.c +++ b/net/netfilter/nft_meta.c @@ -21,12 +21,15 @@ struct nft_meta { enum nft_meta_keys key:8; - enum nft_registers dreg:8; + union { + enum nft_registers dreg:8; + enum nft_registers sreg:8; + }; }; -static void nft_meta_eval(const struct nft_expr *expr, - struct nft_data data[NFT_REG_MAX + 1], - const struct nft_pktinfo *pkt) +static void nft_meta_get_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt) { const struct nft_meta *priv = nft_expr_priv(expr); const struct sk_buff *skb = pkt->skb; @@ -132,23 +135,50 @@ err: data[NFT_REG_VERDICT].verdict = NFT_BREAK; } +static void nft_meta_set_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt) +{ + const struct nft_meta *meta = nft_expr_priv(expr); + struct sk_buff *skb = pkt->skb; + u32 value = data[meta->sreg].data[0]; + + switch (meta->key) { + case NFT_META_MARK: + skb->mark = value; + break; + case NFT_META_PRIORITY: + skb->priority = value; + break; + case NFT_META_NFTRACE: + skb->nf_trace = 1; + break; + default: + WARN_ON(1); + } +} + static const struct nla_policy nft_meta_policy[NFTA_META_MAX + 1] = { [NFTA_META_DREG] = { .type = NLA_U32 }, [NFTA_META_KEY] = { .type = NLA_U32 }, + [NFTA_META_SREG] = { .type = NLA_U32 }, }; -static int nft_meta_init(const struct nft_ctx *ctx, const struct nft_expr *expr, - const struct nlattr * const tb[]) +static int nft_meta_init_validate_set(uint32_t key) { - struct nft_meta *priv = nft_expr_priv(expr); - int err; + switch (key) { + case NFT_META_MARK: + case NFT_META_PRIORITY: + case NFT_META_NFTRACE: + return 0; + default: + return -EOPNOTSUPP; + } +} - if (tb[NFTA_META_DREG] == NULL || - tb[NFTA_META_KEY] == NULL) - return -EINVAL; - - priv->key = ntohl(nla_get_be32(tb[NFTA_META_KEY])); - switch (priv->key) { +static int nft_meta_init_validate_get(uint32_t key) +{ + switch (key) { case NFT_META_LEN: case NFT_META_PROTOCOL: case NFT_META_PRIORITY: @@ -167,26 +197,69 @@ static int nft_meta_init(const struct nft_ctx *ctx, const struct nft_expr *expr, #ifdef CONFIG_NETWORK_SECMARK case NFT_META_SECMARK: #endif - break; + return 0; default: return -EOPNOTSUPP; } - priv->dreg = ntohl(nla_get_be32(tb[NFTA_META_DREG])); - err = nft_validate_output_register(priv->dreg); - if (err < 0) - return err; - return nft_validate_data_load(ctx, priv->dreg, NULL, NFT_DATA_VALUE); } -static int nft_meta_dump(struct sk_buff *skb, const struct nft_expr *expr) +static int nft_meta_init(const struct nft_ctx *ctx, const struct nft_expr *expr, + const struct nlattr * const tb[]) +{ + struct nft_meta *priv = nft_expr_priv(expr); + int err; + + priv->key = ntohl(nla_get_be32(tb[NFTA_META_KEY])); + + if (tb[NFTA_META_DREG]) { + err = nft_meta_init_validate_get(priv->key); + if (err < 0) + return err; + + priv->dreg = ntohl(nla_get_be32(tb[NFTA_META_DREG])); + err = nft_validate_output_register(priv->dreg); + if (err < 0) + return err; + + return nft_validate_data_load(ctx, priv->dreg, NULL, + NFT_DATA_VALUE); + } + + err = nft_meta_init_validate_set(priv->key); + if (err < 0) + return err; + + priv->sreg = ntohl(nla_get_be32(tb[NFTA_META_SREG])); + + return 0; +} + +static int nft_meta_get_dump(struct sk_buff *skb, + const struct nft_expr *expr) { const struct nft_meta *priv = nft_expr_priv(expr); - if (nla_put_be32(skb, NFTA_META_DREG, htonl(priv->dreg))) - goto nla_put_failure; if (nla_put_be32(skb, NFTA_META_KEY, htonl(priv->key))) goto nla_put_failure; + if (nla_put_be32(skb, NFTA_META_DREG, htonl(priv->dreg))) + goto nla_put_failure; + return 0; + +nla_put_failure: + return -1; +} + +static int nft_meta_set_dump(struct sk_buff *skb, + const struct nft_expr *expr) +{ + const struct nft_meta *priv = nft_expr_priv(expr); + + if (nla_put_be32(skb, NFTA_META_KEY, htonl(priv->key))) + goto nla_put_failure; + if (nla_put_be32(skb, NFTA_META_SREG, htonl(priv->sreg))) + goto nla_put_failure; + return 0; nla_put_failure: @@ -194,17 +267,44 @@ nla_put_failure: } static struct nft_expr_type nft_meta_type; -static const struct nft_expr_ops nft_meta_ops = { +static const struct nft_expr_ops nft_meta_get_ops = { .type = &nft_meta_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_meta)), - .eval = nft_meta_eval, + .eval = nft_meta_get_eval, .init = nft_meta_init, - .dump = nft_meta_dump, + .dump = nft_meta_get_dump, }; +static const struct nft_expr_ops nft_meta_set_ops = { + .type = &nft_meta_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_meta)), + .eval = nft_meta_set_eval, + .init = nft_meta_init, + .dump = nft_meta_set_dump, +}; + +static const struct nft_expr_ops * +nft_meta_select_ops(const struct nft_ctx *ctx, + const struct nlattr * const tb[]) +{ + if (tb[NFTA_META_KEY] == NULL) + return ERR_PTR(-EINVAL); + + if (tb[NFTA_META_DREG] && tb[NFTA_META_SREG]) + return ERR_PTR(-EINVAL); + + if (tb[NFTA_META_DREG]) + return &nft_meta_get_ops; + + if (tb[NFTA_META_SREG]) + return &nft_meta_set_ops; + + return ERR_PTR(-EINVAL); +} + static struct nft_expr_type nft_meta_type __read_mostly = { .name = "meta", - .ops = &nft_meta_ops, + .select_ops = &nft_meta_select_ops, .policy = nft_meta_policy, .maxattr = NFTA_META_MAX, .owner = THIS_MODULE, diff --git a/net/netfilter/nft_meta_target.c b/net/netfilter/nft_meta_target.c deleted file mode 100644 index 71177df75ffb..000000000000 --- a/net/netfilter/nft_meta_target.c +++ /dev/null @@ -1,117 +0,0 @@ -/* - * Copyright (c) 2008 Patrick McHardy - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - * - * Development of this code funded by Astaro AG (http://www.astaro.com/) - */ - -#include -#include -#include -#include -#include -#include -#include -#include - -struct nft_meta { - enum nft_meta_keys key; -}; - -static void nft_meta_eval(const struct nft_expr *expr, - struct nft_data *nfres, - struct nft_data *data, - const struct nft_pktinfo *pkt) -{ - const struct nft_meta *meta = nft_expr_priv(expr); - struct sk_buff *skb = pkt->skb; - u32 val = data->data[0]; - - switch (meta->key) { - case NFT_META_MARK: - skb->mark = val; - break; - case NFT_META_PRIORITY: - skb->priority = val; - break; - case NFT_META_NFTRACE: - skb->nf_trace = val; - break; -#ifdef CONFIG_NETWORK_SECMARK - case NFT_META_SECMARK: - skb->secmark = val; - break; -#endif - default: - WARN_ON(1); - } -} - -static const struct nla_policy nft_meta_policy[NFTA_META_MAX + 1] = { - [NFTA_META_KEY] = { .type = NLA_U32 }, -}; - -static int nft_meta_init(const struct nft_expr *expr, struct nlattr *tb[]) -{ - struct nft_meta *meta = nft_expr_priv(expr); - - if (tb[NFTA_META_KEY] == NULL) - return -EINVAL; - - meta->key = ntohl(nla_get_be32(tb[NFTA_META_KEY])); - switch (meta->key) { - case NFT_META_MARK: - case NFT_META_PRIORITY: - case NFT_META_NFTRACE: -#ifdef CONFIG_NETWORK_SECMARK - case NFT_META_SECMARK: -#endif - break; - default: - return -EINVAL; - } - - return 0; -} - -static int nft_meta_dump(struct sk_buff *skb, const struct nft_expr *expr) -{ - struct nft_meta *meta = nft_expr_priv(expr); - - NLA_PUT_BE32(skb, NFTA_META_KEY, htonl(meta->key)); - return 0; - -nla_put_failure: - return -1; -} - -static struct nft_expr_ops meta_target __read_mostly = { - .name = "meta", - .size = NFT_EXPR_SIZE(sizeof(struct nft_meta)), - .owner = THIS_MODULE, - .eval = nft_meta_eval, - .init = nft_meta_init, - .dump = nft_meta_dump, - .policy = nft_meta_policy, - .maxattr = NFTA_META_MAX, -}; - -static int __init nft_meta_target_init(void) -{ - return nft_register_expr(&meta_target); -} - -static void __exit nft_meta_target_exit(void) -{ - nft_unregister_expr(&meta_target); -} - -module_init(nft_meta_target_init); -module_exit(nft_meta_target_exit); - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Patrick McHardy "); -MODULE_ALIAS_NFT_EXPR("meta"); diff --git a/net/netfilter/nft_queue.c b/net/netfilter/nft_queue.c new file mode 100644 index 000000000000..cbea473d69e9 --- /dev/null +++ b/net/netfilter/nft_queue.c @@ -0,0 +1,134 @@ +/* + * Copyright (c) 2013 Eric Leblond + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Development of this code partly funded by OISF + * (http://www.openinfosecfoundation.org/) + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +static u32 jhash_initval __read_mostly; + +struct nft_queue { + u16 queuenum; + u16 queues_total; + u16 flags; + u8 family; +}; + +static void nft_queue_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt) +{ + struct nft_queue *priv = nft_expr_priv(expr); + u32 queue = priv->queuenum; + u32 ret; + + if (priv->queues_total > 1) { + if (priv->flags & NFT_QUEUE_FLAG_CPU_FANOUT) { + int cpu = smp_processor_id(); + + queue = priv->queuenum + cpu % priv->queues_total; + } else { + queue = nfqueue_hash(pkt->skb, queue, + priv->queues_total, priv->family, + jhash_initval); + } + } + + ret = NF_QUEUE_NR(queue); + if (priv->flags & NFT_QUEUE_FLAG_BYPASS) + ret |= NF_VERDICT_FLAG_QUEUE_BYPASS; + + data[NFT_REG_VERDICT].verdict = ret; +} + +static const struct nla_policy nft_queue_policy[NFTA_QUEUE_MAX + 1] = { + [NFTA_QUEUE_NUM] = { .type = NLA_U16 }, + [NFTA_QUEUE_TOTAL] = { .type = NLA_U16 }, + [NFTA_QUEUE_FLAGS] = { .type = NLA_U16 }, +}; + +static int nft_queue_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) +{ + struct nft_queue *priv = nft_expr_priv(expr); + + if (tb[NFTA_QUEUE_NUM] == NULL) + return -EINVAL; + + init_hashrandom(&jhash_initval); + priv->family = ctx->afi->family; + priv->queuenum = ntohs(nla_get_be16(tb[NFTA_QUEUE_NUM])); + + if (tb[NFTA_QUEUE_TOTAL] != NULL) + priv->queues_total = ntohs(nla_get_be16(tb[NFTA_QUEUE_TOTAL])); + if (tb[NFTA_QUEUE_FLAGS] != NULL) { + priv->flags = ntohs(nla_get_be16(tb[NFTA_QUEUE_FLAGS])); + if (priv->flags & ~NFT_QUEUE_FLAG_MASK) + return -EINVAL; + } + return 0; +} + +static int nft_queue_dump(struct sk_buff *skb, const struct nft_expr *expr) +{ + const struct nft_queue *priv = nft_expr_priv(expr); + + if (nla_put_be16(skb, NFTA_QUEUE_NUM, htons(priv->queuenum)) || + nla_put_be16(skb, NFTA_QUEUE_TOTAL, htons(priv->queues_total)) || + nla_put_be16(skb, NFTA_QUEUE_FLAGS, htons(priv->flags))) + goto nla_put_failure; + + return 0; + +nla_put_failure: + return -1; +} + +static struct nft_expr_type nft_queue_type; +static const struct nft_expr_ops nft_queue_ops = { + .type = &nft_queue_type, + .size = NFT_EXPR_SIZE(sizeof(struct nft_queue)), + .eval = nft_queue_eval, + .init = nft_queue_init, + .dump = nft_queue_dump, +}; + +static struct nft_expr_type nft_queue_type __read_mostly = { + .name = "queue", + .ops = &nft_queue_ops, + .policy = nft_queue_policy, + .maxattr = NFTA_QUEUE_MAX, + .owner = THIS_MODULE, +}; + +static int __init nft_queue_module_init(void) +{ + return nft_register_expr(&nft_queue_type); +} + +static void __exit nft_queue_module_exit(void) +{ + nft_unregister_expr(&nft_queue_type); +} + +module_init(nft_queue_module_init); +module_exit(nft_queue_module_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Eric Leblond "); +MODULE_ALIAS_NFT_EXPR("queue"); diff --git a/net/ipv4/netfilter/nft_reject_ipv4.c b/net/netfilter/nft_reject.c similarity index 79% rename from net/ipv4/netfilter/nft_reject_ipv4.c rename to net/netfilter/nft_reject.c index 4a5e94ac314a..0d690d4101d2 100644 --- a/net/ipv4/netfilter/nft_reject_ipv4.c +++ b/net/netfilter/nft_reject.c @@ -1,5 +1,6 @@ /* * Copyright (c) 2008-2009 Patrick McHardy + * Copyright (c) 2013 Eric Leblond * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -16,10 +17,16 @@ #include #include #include +#include + +#if IS_ENABLED(CONFIG_NF_TABLES_IPV6) +#include +#endif struct nft_reject { enum nft_reject_types type:8; u8 icmp_code; + u8 family; }; static void nft_reject_eval(const struct nft_expr *expr, @@ -27,12 +34,25 @@ static void nft_reject_eval(const struct nft_expr *expr, const struct nft_pktinfo *pkt) { struct nft_reject *priv = nft_expr_priv(expr); + struct net *net = dev_net((pkt->in != NULL) ? pkt->in : pkt->out); switch (priv->type) { case NFT_REJECT_ICMP_UNREACH: - icmp_send(pkt->skb, ICMP_DEST_UNREACH, priv->icmp_code, 0); + if (priv->family == NFPROTO_IPV4) + nf_send_unreach(pkt->skb, priv->icmp_code); +#if IS_ENABLED(CONFIG_NF_TABLES_IPV6) + else if (priv->family == NFPROTO_IPV6) + nf_send_unreach6(net, pkt->skb, priv->icmp_code, + pkt->hooknum); +#endif break; case NFT_REJECT_TCP_RST: + if (priv->family == NFPROTO_IPV4) + nf_send_reset(pkt->skb, pkt->hooknum); +#if IS_ENABLED(CONFIG_NF_TABLES_IPV6) + else if (priv->family == NFPROTO_IPV6) + nf_send_reset6(net, pkt->skb, pkt->hooknum); +#endif break; } @@ -53,6 +73,7 @@ static int nft_reject_init(const struct nft_ctx *ctx, if (tb[NFTA_REJECT_TYPE] == NULL) return -EINVAL; + priv->family = ctx->afi->family; priv->type = ntohl(nla_get_be32(tb[NFTA_REJECT_TYPE])); switch (priv->type) { case NFT_REJECT_ICMP_UNREACH: diff --git a/net/netfilter/xt_NFQUEUE.c b/net/netfilter/xt_NFQUEUE.c index ed00fef58996..8f1779ff7e30 100644 --- a/net/netfilter/xt_NFQUEUE.c +++ b/net/netfilter/xt_NFQUEUE.c @@ -11,15 +11,13 @@ #include #include -#include -#include -#include - #include #include #include #include +#include + MODULE_AUTHOR("Harald Welte "); MODULE_DESCRIPTION("Xtables: packet forwarding to netlink"); MODULE_LICENSE("GPL"); @@ -28,7 +26,6 @@ MODULE_ALIAS("ip6t_NFQUEUE"); MODULE_ALIAS("arpt_NFQUEUE"); static u32 jhash_initval __read_mostly; -static bool rnd_inited __read_mostly; static unsigned int nfqueue_tg(struct sk_buff *skb, const struct xt_action_param *par) @@ -38,69 +35,16 @@ nfqueue_tg(struct sk_buff *skb, const struct xt_action_param *par) return NF_QUEUE_NR(tinfo->queuenum); } -static u32 hash_v4(const struct sk_buff *skb) -{ - const struct iphdr *iph = ip_hdr(skb); - - /* packets in either direction go into same queue */ - if ((__force u32)iph->saddr < (__force u32)iph->daddr) - return jhash_3words((__force u32)iph->saddr, - (__force u32)iph->daddr, iph->protocol, jhash_initval); - - return jhash_3words((__force u32)iph->daddr, - (__force u32)iph->saddr, iph->protocol, jhash_initval); -} - -#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES) -static u32 hash_v6(const struct sk_buff *skb) -{ - const struct ipv6hdr *ip6h = ipv6_hdr(skb); - u32 a, b, c; - - if ((__force u32)ip6h->saddr.s6_addr32[3] < - (__force u32)ip6h->daddr.s6_addr32[3]) { - a = (__force u32) ip6h->saddr.s6_addr32[3]; - b = (__force u32) ip6h->daddr.s6_addr32[3]; - } else { - b = (__force u32) ip6h->saddr.s6_addr32[3]; - a = (__force u32) ip6h->daddr.s6_addr32[3]; - } - - if ((__force u32)ip6h->saddr.s6_addr32[1] < - (__force u32)ip6h->daddr.s6_addr32[1]) - c = (__force u32) ip6h->saddr.s6_addr32[1]; - else - c = (__force u32) ip6h->daddr.s6_addr32[1]; - - return jhash_3words(a, b, c, jhash_initval); -} -#endif - -static u32 -nfqueue_hash(const struct sk_buff *skb, const struct xt_action_param *par) -{ - const struct xt_NFQ_info_v1 *info = par->targinfo; - u32 queue = info->queuenum; - - if (par->family == NFPROTO_IPV4) - queue += ((u64) hash_v4(skb) * info->queues_total) >> 32; -#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES) - else if (par->family == NFPROTO_IPV6) - queue += ((u64) hash_v6(skb) * info->queues_total) >> 32; -#endif - - return queue; -} - static unsigned int nfqueue_tg_v1(struct sk_buff *skb, const struct xt_action_param *par) { const struct xt_NFQ_info_v1 *info = par->targinfo; u32 queue = info->queuenum; - if (info->queues_total > 1) - queue = nfqueue_hash(skb, par); - + if (info->queues_total > 1) { + queue = nfqueue_hash(skb, queue, info->queues_total, + par->family, jhash_initval); + } return NF_QUEUE_NR(queue); } @@ -120,10 +64,8 @@ static int nfqueue_tg_check(const struct xt_tgchk_param *par) const struct xt_NFQ_info_v3 *info = par->targinfo; u32 maxid; - if (unlikely(!rnd_inited)) { - get_random_bytes(&jhash_initval, sizeof(jhash_initval)); - rnd_inited = true; - } + init_hashrandom(&jhash_initval); + if (info->queues_total == 0) { pr_err("NFQUEUE: number of total queues is 0\n"); return -EINVAL; @@ -154,8 +96,10 @@ nfqueue_tg_v3(struct sk_buff *skb, const struct xt_action_param *par) int cpu = smp_processor_id(); queue = info->queuenum + cpu % info->queues_total; - } else - queue = nfqueue_hash(skb, par); + } else { + queue = nfqueue_hash(skb, queue, info->queues_total, + par->family, jhash_initval); + } } ret = NF_QUEUE_NR(queue);