diff options
45 files changed, 1972 insertions, 1718 deletions
diff --git a/include/linux/netfilter/ipset/ip_set.h b/include/linux/netfilter/ipset/ip_set.h index ffdfdc24952a..48bb01edcf30 100644 --- a/include/linux/netfilter/ipset/ip_set.h +++ b/include/linux/netfilter/ipset/ip_set.h @@ -108,8 +108,13 @@ struct ip_set_counter { atomic64_t packets; }; +struct ip_set_comment_rcu { + struct rcu_head rcu; + char str[0]; +}; + struct ip_set_comment { - char *str; + struct ip_set_comment_rcu __rcu *c; }; struct ip_set_skbinfo { @@ -176,6 +181,9 @@ struct ip_set_type_variant { /* List elements */ int (*list)(const struct ip_set *set, struct sk_buff *skb, struct netlink_callback *cb); + /* Keep listing private when resizing runs parallel */ + void (*uref)(struct ip_set *set, struct netlink_callback *cb, + bool start); /* Return true if "b" set is the same as "a" * according to the create set parameters */ @@ -223,7 +231,7 @@ struct ip_set { /* The name of the set */ char name[IPSET_MAXNAMELEN]; /* Lock protecting the set data */ - rwlock_t lock; + spinlock_t lock; /* References to the set */ u32 ref; /* The core set type */ @@ -341,12 +349,11 @@ ip_set_put_skbinfo(struct sk_buff *skb, struct ip_set_skbinfo *skbinfo) cpu_to_be64((u64)skbinfo->skbmark << 32 | skbinfo->skbmarkmask))) || (skbinfo->skbprio && - nla_put_net32(skb, IPSET_ATTR_SKBPRIO, + nla_put_net32(skb, IPSET_ATTR_SKBPRIO, cpu_to_be32(skbinfo->skbprio))) || (skbinfo->skbqueue && - nla_put_net16(skb, IPSET_ATTR_SKBQUEUE, + nla_put_net16(skb, IPSET_ATTR_SKBQUEUE, cpu_to_be16(skbinfo->skbqueue))); - } static inline void @@ -380,12 +387,12 @@ ip_set_init_counter(struct ip_set_counter *counter, /* Netlink CB args */ enum { - IPSET_CB_NET = 0, - IPSET_CB_DUMP, - IPSET_CB_INDEX, - IPSET_CB_ARG0, + IPSET_CB_NET = 0, /* net namespace */ + IPSET_CB_DUMP, /* dump single set/all sets */ + IPSET_CB_INDEX, /* set index */ + IPSET_CB_PRIVATE, /* set private data */ + IPSET_CB_ARG0, /* type specific */ IPSET_CB_ARG1, - IPSET_CB_ARG2, }; /* register and unregister set references */ @@ -545,8 +552,6 @@ ip_set_put_extensions(struct sk_buff *skb, const struct ip_set *set, { .bytes = ULLONG_MAX, .packets = ULLONG_MAX, \ .timeout = (set)->timeout } -#define IP_SET_INIT_CIDR(a, b) ((a) ? (a) : (b)) - #define IPSET_CONCAT(a, b) a##b #define IPSET_TOKEN(a, b) IPSET_CONCAT(a, b) diff --git a/include/linux/netfilter/ipset/ip_set_comment.h b/include/linux/netfilter/ipset/ip_set_comment.h index 21217ea008d7..8d0248525957 100644 --- a/include/linux/netfilter/ipset/ip_set_comment.h +++ b/include/linux/netfilter/ipset/ip_set_comment.h @@ -16,41 +16,57 @@ ip_set_comment_uget(struct nlattr *tb) return nla_data(tb); } +/* Called from uadd only, protected by the set spinlock. + * The kadt functions don't use the comment extensions in any way. + */ static inline void ip_set_init_comment(struct ip_set_comment *comment, const struct ip_set_ext *ext) { + struct ip_set_comment_rcu *c = rcu_dereference_protected(comment->c, 1); size_t len = ext->comment ? strlen(ext->comment) : 0; - if (unlikely(comment->str)) { - kfree(comment->str); - comment->str = NULL; + if (unlikely(c)) { + kfree_rcu(c, rcu); + rcu_assign_pointer(comment->c, NULL); } if (!len) return; if (unlikely(len > IPSET_MAX_COMMENT_SIZE)) len = IPSET_MAX_COMMENT_SIZE; - comment->str = kzalloc(len + 1, GFP_ATOMIC); - if (unlikely(!comment->str)) + c = kzalloc(sizeof(*c) + len + 1, GFP_ATOMIC); + if (unlikely(!c)) return; - strlcpy(comment->str, ext->comment, len + 1); + strlcpy(c->str, ext->comment, len + 1); + rcu_assign_pointer(comment->c, c); } +/* Used only when dumping a set, protected by rcu_read_lock_bh() */ static inline int ip_set_put_comment(struct sk_buff *skb, struct ip_set_comment *comment) { - if (!comment->str) + struct ip_set_comment_rcu *c = rcu_dereference_bh(comment->c); + + if (!c) return 0; - return nla_put_string(skb, IPSET_ATTR_COMMENT, comment->str); + return nla_put_string(skb, IPSET_ATTR_COMMENT, c->str); } +/* Called from uadd/udel, flush or the garbage collectors protected + * by the set spinlock. + * Called when the set is destroyed and when there can't be any user + * of the set data anymore. + */ static inline void ip_set_comment_free(struct ip_set_comment *comment) { - if (unlikely(!comment->str)) + struct ip_set_comment_rcu *c; + + c = rcu_dereference_protected(comment->c, 1); + if (unlikely(!c)) return; - kfree(comment->str); - comment->str = NULL; + kfree_rcu(c, rcu); + rcu_assign_pointer(comment->c, NULL); } #endif diff --git a/include/linux/netfilter/ipset/ip_set_timeout.h b/include/linux/netfilter/ipset/ip_set_timeout.h index 83c2f9e0886c..1d6a935c1ac5 100644 --- a/include/linux/netfilter/ipset/ip_set_timeout.h +++ b/include/linux/netfilter/ipset/ip_set_timeout.h @@ -40,38 +40,33 @@ ip_set_timeout_uget(struct nlattr *tb) } static inline bool -ip_set_timeout_test(unsigned long timeout) +ip_set_timeout_expired(unsigned long *t) { - return timeout == IPSET_ELEM_PERMANENT || - time_is_after_jiffies(timeout); -} - -static inline bool -ip_set_timeout_expired(unsigned long *timeout) -{ - return *timeout != IPSET_ELEM_PERMANENT && - time_is_before_jiffies(*timeout); + return *t != IPSET_ELEM_PERMANENT && time_is_before_jiffies(*t); } static inline void -ip_set_timeout_set(unsigned long *timeout, u32 t) +ip_set_timeout_set(unsigned long *timeout, u32 value) { - if (!t) { + unsigned long t; + + if (!value) { *timeout = IPSET_ELEM_PERMANENT; return; } - *timeout = msecs_to_jiffies(t * 1000) + jiffies; - if (*timeout == IPSET_ELEM_PERMANENT) + t = msecs_to_jiffies(value * MSEC_PER_SEC) + jiffies; + if (t == IPSET_ELEM_PERMANENT) /* Bingo! :-) */ - (*timeout)--; + t--; + *timeout = t; } static inline u32 ip_set_timeout_get(unsigned long *timeout) { return *timeout == IPSET_ELEM_PERMANENT ? 0 : - jiffies_to_msecs(*timeout - jiffies)/1000; + jiffies_to_msecs(*timeout - jiffies)/MSEC_PER_SEC; } #endif /* __KERNEL__ */ diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 09f38206c18f..95693c4cebdd 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -224,13 +224,10 @@ struct xt_table_info { unsigned int stacksize; unsigned int __percpu *stackptr; void ***jumpstack; - /* ipt_entry tables: one per CPU */ - /* Note : this field MUST be the last one, see XT_TABLE_INFO_SZ */ - void *entries[1]; + + unsigned char entries[0] __aligned(8); }; -#define XT_TABLE_INFO_SZ (offsetof(struct xt_table_info, entries) \ - + nr_cpu_ids * sizeof(char *)) int xt_register_target(struct xt_target *target); void xt_unregister_target(struct xt_target *target); int xt_register_targets(struct xt_target *target, unsigned int n); @@ -353,6 +350,55 @@ static inline unsigned long ifname_compare_aligned(const char *_a, return ret; } + +/* On SMP, ip(6)t_entry->counters.pcnt holds address of the + * real (percpu) counter. On !SMP, its just the packet count, + * so nothing needs to be done there. + * + * xt_percpu_counter_alloc returns the address of the percpu + * counter, or 0 on !SMP. + * + * Hence caller must use IS_ERR_VALUE to check for error, this + * allows us to return 0 for single core systems without forcing + * callers to deal with SMP vs. NONSMP issues. + */ +static inline u64 xt_percpu_counter_alloc(void) +{ + if (nr_cpu_ids > 1) { + void __percpu *res = alloc_percpu(struct xt_counters); + + if (res == NULL) + return (u64) -ENOMEM; + + return (__force u64) res; + } + + return 0; +} +static inline void xt_percpu_counter_free(u64 pcnt) +{ + if (nr_cpu_ids > 1) + free_percpu((void __percpu *) pcnt); +} + +static inline struct xt_counters * +xt_get_this_cpu_counter(struct xt_counters *cnt) +{ + if (nr_cpu_ids > 1) + return this_cpu_ptr((void __percpu *) cnt->pcnt); + + return cnt; +} + +static inline struct xt_counters * +xt_get_per_cpu_counter(struct xt_counters *cnt, unsigned int cpu) +{ + if (nr_cpu_ids > 1) + return per_cpu_ptr((void __percpu *) cnt->pcnt, cpu); + + return cnt; +} + struct nf_hook_ops *xt_hook_link(const struct xt_table *, nf_hookfn *); void xt_hook_unlink(const struct xt_table *, struct nf_hook_ops *); diff --git a/include/linux/netfilter_bridge.h b/include/linux/netfilter_bridge.h index f2fdb5a52070..6d80fc686323 100644 --- a/include/linux/netfilter_bridge.h +++ b/include/linux/netfilter_bridge.h @@ -20,13 +20,6 @@ enum nf_br_hook_priorities { #define BRNF_BRIDGED_DNAT 0x02 #define BRNF_NF_BRIDGE_PREROUTING 0x08 -static inline unsigned int nf_bridge_mtu_reduction(const struct sk_buff *skb) -{ - if (skb->nf_bridge->orig_proto == BRNF_PROTO_PPPOE) - return PPPOE_SES_HLEN; - return 0; -} - int br_handle_frame_finish(struct sock *sk, struct sk_buff *skb); static inline void br_drop_fake_rtable(struct sk_buff *skb) diff --git a/include/linux/netfilter_ipv6.h b/include/linux/netfilter_ipv6.h index 64dad1cc1a4b..8b7d28f3aada 100644 --- a/include/linux/netfilter_ipv6.h +++ b/include/linux/netfilter_ipv6.h @@ -25,6 +25,9 @@ void ipv6_netfilter_fini(void); struct nf_ipv6_ops { int (*chk_addr)(struct net *net, const struct in6_addr *addr, const struct net_device *dev, int strict); + void (*route_input)(struct sk_buff *skb); + int (*fragment)(struct sock *sk, struct sk_buff *skb, + int (*output)(struct sock *, struct sk_buff *)); }; extern const struct nf_ipv6_ops __rcu *nf_ipv6_ops; diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index a7acc92aa668..d6cdd6e87d53 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -36,6 +36,7 @@ #include <linux/sched.h> #include <net/flow_dissector.h> #include <linux/splice.h> +#include <linux/in6.h> /* A. Checksumming of received packets by device. * @@ -173,13 +174,17 @@ struct nf_bridge_info { BRNF_PROTO_PPPOE } orig_proto:8; bool pkt_otherhost; + __u16 frag_max_size; unsigned int mask; struct net_device *physindev; union { struct net_device *physoutdev; char neigh_header[8]; }; - __be32 ipv4_daddr; + union { + __be32 ipv4_daddr; + struct in6_addr ipv6_daddr; + }; }; #endif diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 3d6f48ca40a7..2a246680a6c3 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -781,6 +781,7 @@ struct nft_stats { }; #define NFT_HOOK_OPS_MAX 2 +#define NFT_BASECHAIN_DISABLED (1 << 0) /** * struct nft_base_chain - nf_tables base chain @@ -791,14 +792,17 @@ struct nft_stats { * @policy: default policy * @stats: per-cpu chain stats * @chain: the chain + * @dev_name: device name that this base chain is attached to (if any) */ struct nft_base_chain { struct nf_hook_ops ops[NFT_HOOK_OPS_MAX]; possible_net_t pnet; const struct nf_chain_type *type; u8 policy; + u8 flags; struct nft_stats __percpu *stats; struct nft_chain chain; + char dev_name[IFNAMSIZ]; }; static inline struct nft_base_chain *nft_base_chain(const struct nft_chain *chain) @@ -806,6 +810,11 @@ static inline struct nft_base_chain *nft_base_chain(const struct nft_chain *chai return container_of(chain, struct nft_base_chain, chain); } +int nft_register_basechain(struct nft_base_chain *basechain, + unsigned int hook_nops); +void nft_unregister_basechain(struct nft_base_chain *basechain, + unsigned int hook_nops); + unsigned int nft_do_chain(struct nft_pktinfo *pkt, const struct nf_hook_ops *ops); @@ -819,7 +828,6 @@ unsigned int nft_do_chain(struct nft_pktinfo *pkt, * @use: number of chain references to this table * @flags: table flag (see enum nft_table_flags) * @name: name of the table - * @dev: this table is bound to this device (if any) */ struct nft_table { struct list_head list; @@ -829,7 +837,6 @@ struct nft_table { u32 use; u16 flags; char name[NFT_TABLE_MAXNAMELEN]; - struct net_device *dev; }; enum nft_af_flags { diff --git a/include/uapi/linux/netfilter/ipset/ip_set.h b/include/uapi/linux/netfilter/ipset/ip_set.h index 5ab4e60894cf..63b2e34f1b60 100644 --- a/include/uapi/linux/netfilter/ipset/ip_set.h +++ b/include/uapi/linux/netfilter/ipset/ip_set.h @@ -15,12 +15,12 @@ /* The protocol version */ #define IPSET_PROTOCOL 6 -/* The maximum permissible comment length we will accept over netlink */ -#define IPSET_MAX_COMMENT_SIZE 255 - /* The max length of strings including NUL: set and type identifiers */ #define IPSET_MAXNAMELEN 32 +/* The maximum permissible comment length we will accept over netlink */ +#define IPSET_MAX_COMMENT_SIZE 255 + /* Message types and commands */ enum ipset_cmd { IPSET_CMD_NONE, diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index 89a671e0f5e7..a99e6a997140 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -122,11 +122,13 @@ enum nft_list_attributes { * * @NFTA_HOOK_HOOKNUM: netfilter hook number (NLA_U32) * @NFTA_HOOK_PRIORITY: netfilter hook priority (NLA_U32) + * @NFTA_HOOK_DEV: netdevice name (NLA_STRING) */ enum nft_hook_attributes { NFTA_HOOK_UNSPEC, NFTA_HOOK_HOOKNUM, NFTA_HOOK_PRIORITY, + NFTA_HOOK_DEV, __NFTA_HOOK_MAX }; #define NFTA_HOOK_MAX (__NFTA_HOOK_MAX - 1) @@ -146,14 +148,12 @@ 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) - * @NFTA_TABLE_DEV: net device name (NLA_STRING) */ enum nft_table_attributes { NFTA_TABLE_UNSPEC, NFTA_TABLE_NAME, NFTA_TABLE_FLAGS, NFTA_TABLE_USE, - NFTA_TABLE_DEV, __NFTA_TABLE_MAX }; #define NFTA_TABLE_MAX (__NFTA_TABLE_MAX - 1) diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index 46660a28feef..e4e5f2f29173 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -34,6 +34,7 @@ #include <net/ip.h> #include <net/ipv6.h> +#include <net/addrconf.h> #include <net/route.h> #include <net/netfilter/br_netfilter.h> @@ -115,6 +116,8 @@ struct brnf_frag_data { char mac[NF_BRIDGE_MAX_MAC_HEADER_LENGTH]; u8 encap_size; u8 size; + u16 vlan_tci; + __be16 vlan_proto; }; static DEFINE_PER_CPU(struct brnf_frag_data, brnf_frag_data_storage); @@ -216,7 +219,7 @@ static inline void nf_bridge_pull_encap_header_rcsum(struct sk_buff *skb) * expected format */ -static int br_parse_ip_options(struct sk_buff *skb) +static int br_validate_ipv4(struct sk_buff *skb) { const struct iphdr *iph; struct net_device *dev = skb->dev; @@ -264,6 +267,111 @@ drop: return -1; } +/* We only check the length. A bridge shouldn't do any hop-by-hop stuff + * anyway + */ +static int check_hbh_len(struct sk_buff *skb) +{ + unsigned char *raw = (u8 *)(ipv6_hdr(skb) + 1); + u32 pkt_len; + const unsigned char *nh = skb_network_header(skb); + int off = raw - nh; + int len = (raw[1] + 1) << 3; + + if ((raw + len) - skb->data > skb_headlen(skb)) + goto bad; + + off += 2; + len -= 2; + + while (len > 0) { + int optlen = nh[off + 1] + 2; + + switch (nh[off]) { + case IPV6_TLV_PAD1: + optlen = 1; + break; + + case IPV6_TLV_PADN: + break; + + case IPV6_TLV_JUMBO: + if (nh[off + 1] != 4 || (off & 3) != 2) + goto bad; + pkt_len = ntohl(*(__be32 *)(nh + off + 2)); + if (pkt_len <= IPV6_MAXPLEN || + ipv6_hdr(skb)->payload_len) + goto bad; + if (pkt_len > skb->len - sizeof(struct ipv6hdr)) + goto bad; + if (pskb_trim_rcsum(skb, + pkt_len + sizeof(struct ipv6hdr))) + goto bad; + nh = skb_network_header(skb); + break; + default: + if (optlen > len) + goto bad; + break; + } + off += optlen; + len -= optlen; + } + if (len == 0) + return 0; +bad: + return -1; +} + +/* Equivalent to br_validate_ipv4 for IPv6 */ +static int br_validate_ipv6(struct sk_buff *skb) +{ + const struct ipv6hdr *hdr; + struct net_device *dev = skb->dev; + struct inet6_dev *idev = in6_dev_get(skb->dev); + u32 pkt_len; + u8 ip6h_len = sizeof(struct ipv6hdr); + + if (!pskb_may_pull(skb, ip6h_len)) + goto inhdr_error; + + if (skb->len < ip6h_len) + goto drop; + + hdr = ipv6_hdr(skb); + + if (hdr->version != 6) + goto inhdr_error; + + pkt_len = ntohs(hdr->payload_len); + + if (pkt_len || hdr->nexthdr != NEXTHDR_HOP) { + if (pkt_len + ip6h_len > skb->len) { + IP6_INC_STATS_BH(dev_net(dev), idev, + IPSTATS_MIB_INTRUNCATEDPKTS); + goto drop; + } + if (pskb_trim_rcsum(skb, pkt_len + ip6h_len)) { + IP6_INC_STATS_BH(dev_net(dev), idev, + IPSTATS_MIB_INDISCARDS); + goto drop; + } + } + if (hdr->nexthdr == NEXTHDR_HOP && check_hbh_len(skb)) + goto drop; + + memset(IP6CB(skb), 0, sizeof(struct inet6_skb_parm)); + /* No IP options in IPv6 header; however it should be + * checked if some next headers need special treatment + */ + return 0; + +inhdr_error: + IP6_INC_STATS_BH(dev_net(dev), idev, IPSTATS_MIB_INHDRERRORS); +drop: + return -1; +} + static void nf_bridge_update_protocol(struct sk_buff *skb) { switch (skb->nf_bridge->orig_proto) { @@ -278,37 +386,6 @@ static void nf_bridge_update_protocol(struct sk_buff *skb) } } -/* PF_BRIDGE/PRE_ROUTING *********************************************/ -/* Undo the changes made for ip6tables PREROUTING and continue the - * bridge PRE_ROUTING hook. */ -static int br_nf_pre_routing_finish_ipv6(struct sock *sk, struct sk_buff *skb) -{ - struct nf_bridge_info *nf_bridge = nf_bridge_info_get(skb); - struct rtable *rt; - - if (nf_bridge->pkt_otherhost) { - skb->pkt_type = PACKET_OTHERHOST; - nf_bridge->pkt_otherhost = false; - } - nf_bridge->mask ^= BRNF_NF_BRIDGE_PREROUTING; - - rt = bridge_parent_rtable(nf_bridge->physindev); - if (!rt) { - kfree_skb(skb); - return 0; - } - skb_dst_set_noref(skb, &rt->dst); - - skb->dev = nf_bridge->physindev; - nf_bridge_update_protocol(skb); - nf_bridge_push_encap_header(skb); - NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, sk, skb, - skb->dev, NULL, - br_handle_frame_finish, 1); - - return 0; -} - /* Obtain the correct destination MAC address, while preserving the original * source MAC address. If we already know this address, we just copy it. If we * don't, we use the neighbour framework to find out. In both cases, we make @@ -357,7 +434,74 @@ free_skb: static bool daddr_was_changed(const struct sk_buff *skb, const struct nf_bridge_info *nf_bridge) { - return ip_hdr(skb)->daddr != nf_bridge->ipv4_daddr; + switch (skb->protocol) { + case htons(ETH_P_IP): + return ip_hdr(skb)->daddr != nf_bridge->ipv4_daddr; + case htons(ETH_P_IPV6): + return memcmp(&nf_bridge->ipv6_daddr, &ipv6_hdr(skb)->daddr, + sizeof(ipv6_hdr(skb)->daddr)) != 0; + default: + return false; + } +} + +/* PF_BRIDGE/PRE_ROUTING: Undo the changes made for ip6tables + * PREROUTING and continue the bridge PRE_ROUTING hook. See comment + * for br_nf_pre_routing_finish(), same logic is used here but + * equivalent IPv6 function ip6_route_input() called indirectly. + */ +static int br_nf_pre_routing_finish_ipv6(struct sock *sk, struct sk_buff *skb) +{ + struct nf_bridge_info *nf_bridge = nf_bridge_info_get(skb); + struct rtable *rt; + struct net_device *dev = skb->dev; + const struct nf_ipv6_ops *v6ops = nf_get_ipv6_ops(); + + nf_bridge->frag_max_size = IP6CB(skb)->frag_max_size; + + if (nf_bridge->pkt_otherhost) { + skb->pkt_type = PACKET_OTHERHOST; + nf_bridge->pkt_otherhost = false; + } + nf_bridge->mask &= ~BRNF_NF_BRIDGE_PREROUTING; + if (daddr_was_changed(skb, nf_bridge)) { + skb_dst_drop(skb); + v6ops->route_input(skb); + + if (skb_dst(skb)->error) { + kfree_skb(skb); + return 0; + } + + if (skb_dst(skb)->dev == dev) { + skb->dev = nf_bridge->physindev; + nf_bridge_update_protocol(skb); + nf_bridge_push_encap_header(skb); + NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, + sk, skb, skb->dev, NULL, + br_nf_pre_routing_finish_bridge, + 1); + return 0; + } + ether_addr_copy(eth_hdr(skb)->h_dest, dev->dev_addr); + skb->pkt_type = PACKET_HOST; + } else { + rt = bridge_parent_rtable(nf_bridge->physindev); + if (!rt) { + kfree_skb(skb); + return 0; + } + skb_dst_set_noref(skb, &rt->dst); + } + + skb->dev = nf_bridge->physindev; + nf_bridge_update_protocol(skb); + nf_bridge_push_encap_header(skb); + NF_HOOK_THRESH(NFPROTO_BRIDGE, NF_BR_PRE_ROUTING, sk, skb, + skb->dev, NULL, + br_handle_frame_finish, 1); + + return 0; } /* This requires some explaining. If DNAT has taken place, @@ -406,16 +550,14 @@ static int br_nf_pre_routing_finish(struct sock *sk, struct sk_buff *skb) struct nf_bridge_info *nf_bridge = nf_bridge_info_get(skb); struct rtable *rt; int err; - int frag_max_size; - frag_max_size = IPCB(skb)->frag_max_size; - BR_INPUT_SKB_CB(skb)->frag_max_size = frag_max_size; + nf_bridge->frag_max_size = IPCB(skb)->frag_max_size; if (nf_bridge->pkt_otherhost) { skb->pkt_type = PACKET_OTHERHOST; nf_bridge->pkt_otherhost = false; } - nf_bridge->mask ^= BRNF_NF_BRIDGE_PREROUTING; + nf_bridge->mask &= ~BRNF_NF_BRIDGE_PREROUTING; if (daddr_was_changed(skb, nf_bridge)) { if ((err = ip_route_input(skb, iph->daddr, iph->saddr, iph->tos, dev))) { struct in_device *in_dev = __in_dev_get_rcu(dev); @@ -517,90 +659,16 @@ static struct net_device *setup_pre_routing(struct sk_buff *skb) return skb->dev; } -/* We only check the length. A bridge shouldn't do any hop-by-hop stuff anyway */ -static int check_hbh_len(struct sk_buff *skb) -{ - unsigned char *raw = (u8 *)(ipv6_hdr(skb) + 1); - u32 pkt_len; - const unsigned char *nh = skb_network_header(skb); - int off = raw - nh; - int len = (raw[1] + 1) << 3; - - if ((raw + len) - skb->data > skb_headlen(skb)) - goto bad; - - off += 2; - len -= 2; - - while (len > 0) { - int optlen = nh[off + 1] + 2; - - switch (nh[off]) { - case IPV6_TLV_PAD1: - optlen = 1; - break; - - case IPV6_TLV_PADN: - break; - - case IPV6_TLV_JUMBO: - if (nh[off + 1] != 4 || (off & 3) != 2) - goto bad; - pkt_len = ntohl(*(__be32 *) (nh + off + 2)); - if (pkt_len <= IPV6_MAXPLEN || - ipv6_hdr(skb)->payload_len) - goto bad; - if (pkt_len > skb->len - sizeof(struct ipv6hdr)) - goto bad; - if (pskb_trim_rcsum(skb, - pkt_len + sizeof(struct ipv6hdr))) - goto bad; - nh = skb_network_header(skb); - break; - default: - if (optlen > len) - goto bad; - break; - } - off += optlen; - len -= optlen; - } - if (len == 0) - return 0; -bad: - return -1; - -} - /* Replicate the checks that IPv6 does on packet reception and pass the packet - * to ip6tables, which doesn't support NAT, so things are fairly simple. */ + * to ip6tables. + */ static unsigned int br_nf_pre_routing_ipv6(const struct nf_hook_ops *ops, struct sk_buff *skb, const struct nf_hook_state *state) { - const struct ipv6hdr *hdr; - u32 pkt_len; - - if (skb->len < sizeof(struct ipv6hdr)) - return NF_DROP; - - if (!pskb_may_pull(skb, sizeof(struct ipv6hdr))) - return NF_DROP; - - hdr = ipv6_hdr(skb); + struct nf_bridge_info *nf_bridge; - if (hdr->version != 6) - return NF_DROP; - - pkt_len = ntohs(hdr->payload_len); - - if (pkt_len || hdr->nexthdr != NEXTHDR_HOP) { - if (pkt_len + sizeof(struct ipv6hdr) > skb->len) - return NF_DROP; - if (pskb_trim_rcsum(skb, pkt_len + sizeof(struct ipv6hdr))) - return NF_DROP; - } - if (hdr->nexthdr == NEXTHDR_HOP && check_hbh_len(skb)) + if (br_validate_ipv6(skb)) return NF_DROP; nf_bridge_put(skb->nf_bridge); @@ -609,6 +677,9 @@ static unsigned int br_nf_pre_routing_ipv6(const struct nf_hook_ops *ops, if (!setup_pre_routing(skb)) return NF_DROP; + nf_bridge = nf_bridge_info_get(skb); + nf_bridge->ipv6_daddr = ipv6_hdr(skb)->daddr; + skb->protocol = htons(ETH_P_IPV6); NF_HOOK(NFPROTO_IPV6, NF_INET_PRE_ROUTING, state->sk, skb, skb->dev, NULL, @@ -656,7 +727,7 @@ static unsigned int br_nf_pre_routing(const struct nf_hook_ops *ops, nf_bridge_pull_encap_header_rcsum(skb); - if (br_parse_ip_options(skb)) + if (br_validate_ipv4(skb)) return NF_DROP; nf_bridge_put(skb->nf_bridge); @@ -700,12 +771,12 @@ static int br_nf_forward_finish(struct sock *sk, struct sk_buff *skb) struct net_device *in; if (!IS_ARP(skb) && !IS_VLAN_ARP(skb)) { - int frag_max_size; - if (skb->protocol == htons(ETH_P_IP)) { - frag_max_size = IPCB(skb)->frag_max_size; - BR_INPUT_SKB_CB(skb)->frag_max_size = frag_max_size; - } + if (skb->protocol == htons(ETH_P_IP)) + nf_bridge->frag_max_size = IPCB(skb)->frag_max_size; + + if (skb->protocol == htons(ETH_P_IPV6)) + nf_bridge->frag_max_size = IP6CB(skb)->frag_max_size; in = nf_bridge->physindev; if (nf_bridge->pkt_otherhost) { @@ -768,12 +839,15 @@ static unsigned int br_nf_forward_ip(const struct nf_hook_ops *ops, } if (pf == NFPROTO_IPV4) { - int frag_max = BR_INPUT_SKB_CB(skb)->frag_max_size; - - if (br_parse_ip_options(skb)) + if (br_validate_ipv4(skb)) return NF_DROP; + IPCB(skb)->frag_max_size = nf_bridge->frag_max_size; + } - IPCB(skb)->frag_max_size = frag_max; + if (pf == NFPROTO_IPV6) { + if (br_validate_ipv6(skb)) + return NF_DROP; + IP6CB(skb)->frag_max_size = nf_bridge->frag_max_size; } nf_bridge->physoutdev = skb->dev; @@ -823,7 +897,7 @@ static unsigned int br_nf_forward_arp(const struct nf_hook_ops *ops, return NF_STOLEN; } -#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV4) +#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV4) || IS_ENABLED(CONFIG_NF_DEFRAG_IPV6) static int br_nf_push_frag_xmit(struct sock *sk, struct sk_buff *skb) { struct brnf_frag_data *data; @@ -837,12 +911,18 @@ static int br_nf_push_frag_xmit(struct sock *sk, struct sk_buff *skb) return 0; } + if (data->vlan_tci) { + skb->vlan_tci = data->vlan_tci; + skb->vlan_proto = data->vlan_proto; + } + skb_copy_to_linear_data_offset(skb, -data->size, data->mac, data->size); __skb_push(skb, data->encap_size); nf_bridge_info_free(skb); return br_dev_queue_push_xmit(sk, skb); } +#endif static int br_nf_ip_fragment(struct sock *sk, struct sk_buff *skb, int (*output)(struct sock *, struct sk_buff *)) @@ -863,54 +943,82 @@ static int br_nf_ip_fragment(struct sock *sk, struct sk_buff *skb, return ip_do_fragment(sk, skb, output); } +static unsigned int nf_bridge_mtu_reduction(const struct sk_buff *skb) +{ + if (skb->nf_bridge->orig_proto == BRNF_PROTO_PPPOE) + return PPPOE_SES_HLEN; + return 0; +} + static int br_nf_dev_queue_xmit(struct sock *sk, struct sk_buff *skb) { - int ret; - int frag_max_size; + struct nf_bridge_info *nf_bridge; unsigned int mtu_reserved; - if (skb_is_gso(skb) || skb->protocol != htons(ETH_P_IP)) { + mtu_reserved = nf_bridge_mtu_reduction(skb); + + if (skb_is_gso(skb) || skb->len + mtu_reserved <= skb->dev->mtu) { nf_bridge_info_free(skb); return br_dev_queue_push_xmit(sk, skb); } - mtu_reserved = nf_bridge_mtu_reduction(skb); + nf_bridge = nf_bridge_info_get(skb); + +#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV4) /* This is wrong! We should preserve the original fragment * boundaries by preserving frag_list rather than refragmenting. */ - if (skb->len + mtu_reserved > skb->dev->mtu) { + if (skb->protocol == htons(ETH_P_IP)) { struct brnf_frag_data *data; - frag_max_size = BR_INPUT_SKB_CB(skb)->frag_max_size; - if (br_parse_ip_options(skb)) - /* Drop invalid packet */ + if (br_validate_ipv4(skb)) return NF_DROP; - IPCB(skb)->frag_max_size = frag_max_size; + + IPCB(skb)->frag_max_size = nf_bridge->frag_max_size; nf_bridge_update_protocol(skb); data = this_cpu_ptr(&brnf_frag_data_storage); + + data->vlan_tci = skb->vlan_tci; + data->vlan_proto = skb->vlan_proto; data->encap_size = nf_bridge_encap_header_len(skb); data->size = ETH_HLEN + data->encap_size; skb_copy_from_linear_data_offset(skb, -data->size, data->mac, data->size); - ret = br_nf_ip_fragment(sk, skb, br_nf_push_frag_xmit); - } else { - nf_bridge_info_free(skb); - ret = br_dev_queue_push_xmit(sk, skb); + return br_nf_ip_fragment(sk, skb, br_nf_push_frag_xmit); } +#endif +#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6) + if (skb->protocol == htons(ETH_P_IPV6)) { + const struct nf_ipv6_ops *v6ops = nf_get_ipv6_ops(); + struct brnf_frag_data *data; - return ret; -} -#else -static int br_nf_dev_queue_xmit(struct sock *sk, struct sk_buff *skb) -{ + if (br_validate_ipv6(skb)) + return NF_DROP; + + IP6CB(skb)->frag_max_size = nf_bridge->frag_max_size; + + nf_bridge_update_protocol(skb); + + data = this_cpu_ptr(&brnf_frag_data_storage); + data->encap_size = nf_bridge_encap_header_len(skb); + data->size = ETH_HLEN + data->encap_size; + + skb_copy_from_linear_data_offset(skb, -data->size, data->mac, + data->size); + + if (v6ops) + return v6ops->fragment(sk, skb, br_nf_push_frag_xmit); + else + return -EMSGSIZE; + } +#endif nf_bridge_info_free(skb); return br_dev_queue_push_xmit(sk, skb); } -#endif /* PF_BRIDGE/POST_ROUTING ********************************************/ static unsigned int br_nf_post_routing(const struct nf_hook_ops *ops, diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index 1f36fa70639b..5dccced71269 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -18,6 +18,7 @@ #include <linux/netpoll.h> #include <linux/u64_stats_sync.h> #include <net/route.h> +#include <net/ip6_fib.h> #include <linux/if_vlan.h> #define BR_HASH_BITS 8 @@ -214,7 +215,10 @@ struct net_bridge spinlock_t hash_lock; struct hlist_head hash[BR_HASH_SIZE]; #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) - struct rtable fake_rtable; + union { + struct rtable fake_rtable; + struct rt6_info fake_rt6_info; + }; bool nf_call_iptables; bool nf_call_ip6tables; bool nf_call_arptables; @@ -304,7 +308,6 @@ struct br_input_skb_cb { int mrouters_only; #endif - u16 frag_max_size; bool proxyarp_replied; #ifdef CONFIG_BRIDGE_VLAN_FILTERING diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 55f3c2e94357..6bf89a6312bc 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -549,10 +549,6 @@ int ip_do_fragment(struct sock *sk, struct sk_buff *skb, hlen = iph->ihl * 4; mtu = mtu - hlen; /* Size of data space */ -#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) - if (skb->nf_bridge) - mtu -= nf_bridge_mtu_reduction(skb); -#endif IPCB(skb)->flags |= IPSKB_FRAG_COMPLETE; /* When frag_list is given, use it. First, check its validity: diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index fb20f363151f..2199a5db25e6 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -195,7 +195,8 @@ config IP_NF_MATCH_ECN config IP_NF_MATCH_RPFILTER tristate '"rpfilter" reverse path filter match support' - depends on NETFILTER_ADVANCED && (IP_NF_MANGLE || IP_NF_RAW) + depends on NETFILTER_ADVANCED + depends on IP_NF_MANGLE || IP_NF_RAW ---help--- This option allows you to match packets whose replies would go out via the interface the packet came in. diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index a61200754f4b..95c9b6eece25 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -256,7 +256,7 @@ unsigned int arpt_do_table(struct sk_buff *skb, const struct arphdr *arp; struct arpt_entry *e, *back; const char *indev, *outdev; - void *table_base; + const void *table_base; const struct xt_table_info *private; struct xt_action_param acpar; unsigned int addend; @@ -275,7 +275,7 @@ unsigned int arpt_do_table(struct sk_buff *skb, * pointer. */ smp_read_barrier_depends(); - table_base = private->entries[smp_processor_id()]; + table_base = private->entries; e = get_entry(table_base, private->hook_entry[hook]); back = get_entry(table_base, private->underflow[hook]); @@ -289,13 +289,15 @@ unsigned int arpt_do_table(struct sk_buff *skb, arp = arp_hdr(skb); do { const struct xt_entry_target *t; + struct xt_counters *counter; if (!arp_packet_match(arp, skb->dev, indev, outdev, &e->arp)) { e = arpt_next_entry(e); continue; } - ADD_COUNTER(e->counters, arp_hdr_len(skb->dev), 1); + counter = xt_get_this_cpu_counter(&e->counters); + ADD_COUNTER(*counter, arp_hdr_len(skb->dev), 1); t = arpt_get_target_c(e); @@ -521,6 +523,10 @@ find_check_entry(struct arpt_entry *e, const char *name, unsigned int size) if (ret) return ret; + e->counters.pcnt = xt_percpu_counter_alloc(); + if (IS_ERR_VALUE(e->counters.pcnt)) + return -ENOMEM; + t = arpt_get_target(e); target = xt_request_find_target(NFPROTO_ARP, t->u.user.name, t->u.user.revision); @@ -538,6 +544,8 @@ find_check_entry(struct arpt_entry *e, const char *name, unsigned int size) err: module_put(t->u.kernel.target->me); out: + xt_percpu_counter_free(e->counters.pcnt); + return ret; } @@ -614,6 +622,7 @@ static inline void cleanup_entry(struct arpt_entry *e) if (par.target->destroy != NULL) par.target->destroy(&par); module_put(par.target->me); + xt_percpu_counter_free(e->counters.pcnt); } /* Checks and translates the user-supplied table segment (held in @@ -702,12 +711,6 @@ static int translate_table(struct xt_table_info *newinfo, void *entry0, return ret; } - /* And one copy for every other CPU */ - for_each_possible_cpu(i) { - if (newinfo->entries[i] && newinfo->entries[i] != entry0) - memcpy(newinfo->entries[i], entry0, newinfo->size); - } - return ret; } @@ -722,14 +725,16 @@ static void get_counters(const struct xt_table_info *t, seqcount_t *s = &per_cpu(xt_recseq, cpu); i = 0; - xt_entry_foreach(iter, t->entries[cpu], t->size) { + xt_entry_foreach(iter, t->entries, t->size) { + struct xt_counters *tmp; u64 bcnt, pcnt; unsigned int start; + tmp = xt_get_per_cpu_counter(&iter->counters, cpu); do { start = read_seqcount_begin(s); - bcnt = iter->counters.bcnt; - pcnt = iter->counters.pcnt; + bcnt = tmp->bcnt; + pcnt = tmp->pcnt; } while (read_seqcount_retry(s, start)); ADD_COUNTER(counters[i], bcnt, pcnt); @@ -774,7 +779,7 @@ static int copy_entries_to_user(unsigned int total_size, if (IS_ERR(counters)) return PTR_ERR(counters); - loc_cpu_entry = private->entries[raw_smp_processor_id()]; + loc_cpu_entry = private->entries; /* ... then copy entire thing ... */ if (copy_to_user(userptr, loc_cpu_entry, total_size) != 0) { ret = -EFAULT; @@ -863,16 +868,16 @@ static int compat_table_info(const struct xt_table_info *info, struct xt_table_info *newinfo) { struct arpt_entry *iter; - void *loc_cpu_entry; + const void *loc_cpu_entry; int ret; if (!newinfo || !info) return -EINVAL; - /* we dont care about newinfo->entries[] */ + /* we dont care about newinfo->entries */ memcpy(newinfo, info, offsetof(struct xt_table_info, entries)); newinfo->initial_entries = 0; - loc_cpu_entry = info->entries[raw_smp_processor_id()]; + loc_cpu_entry = info->entries; xt_compat_init_offsets(NFPROTO_ARP, info->number); xt_entry_foreach(iter, loc_cpu_entry, info->size) { ret = compat_calc_entry(iter, info, loc_cpu_entry, newinfo); @@ -1037,7 +1042,7 @@ static int __do_replace(struct net *net, const char *name, get_counters(oldinfo, counters); /* Decrease module usage counts and free resource */ - loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()]; + loc_cpu_old_entry = oldinfo->entries; xt_entry_foreach(iter, loc_cpu_old_entry, oldinfo->size) cleanup_entry(iter); @@ -1084,8 +1089,7 @@ static int do_replace(struct net *net, const void __user *user, if (!newinfo) return -ENOMEM; - /* choose the copy that is on our node/cpu */ - loc_cpu_entry = newinfo->entries[raw_smp_processor_id()]; + loc_cpu_entry = newinfo->entries; if (copy_from_user(loc_cpu_entry, user + sizeof(tmp), tmp.size) != 0) { ret = -EFAULT; @@ -1115,7 +1119,7 @@ static int do_replace(struct net *net, const void __user *user, static int do_add_counters(struct net *net, const void __user *user, unsigned int len, int compat) { - unsigned int i, curcpu; + unsigned int i; struct xt_counters_info tmp; struct xt_counters *paddc; unsigned int num_counters; @@ -1125,7 +1129,6 @@ static int do_add_counters(struct net *net, const void __user *user, struct xt_table *t; const struct xt_table_info *private; int ret = 0; - void *loc_cpu_entry; struct arpt_entry *iter; unsigned int addend; #ifdef CONFIG_COMPAT @@ -1181,12 +1184,13 @@ static int do_add_counters(struct net *net, const void __user *user, } i = 0; - /* Choose the copy that is on our node */ - curcpu = smp_processor_id(); - loc_cpu_entry = private->entries[curcpu]; + addend = xt_write_recseq_begin(); - xt_entry_foreach(iter, loc_cpu_entry, private->size) { - ADD_COUNTER(iter->counters, paddc[i].bcnt, paddc[i].pcnt); + xt_entry_foreach(iter, private->entries, private->size) { + struct xt_counters *tmp; + + tmp = xt_get_this_cpu_counter(&iter->counters); + ADD_COUNTER(*tmp, paddc[i].bcnt, paddc[i].pcnt); ++i; } xt_write_recseq_end(addend); @@ -1396,7 +1400,7 @@ static int translate_compat_table(const char *name, newinfo->hook_entry[i] = info->hook_entry[i]; newinfo->underflow[i] = info->underflow[i]; } - entry1 = newinfo->entries[raw_smp_processor_id()]; + entry1 = newinfo->entries; pos = entry1; size = total_size; xt_entry_foreach(iter0, entry0, total_size) { @@ -1416,9 +1420,17 @@ static int translate_compat_table(const char *name, i = 0; xt_entry_foreach(iter1, entry1, newinfo->size) { + iter1->counters.pcnt = xt_percpu_counter_alloc(); + if (IS_ERR_VALUE(iter1->counters.pcnt)) { + ret = -ENOMEM; + break; + } + ret = check_target(iter1, name); - if (ret != 0) + if (ret != 0) { + xt_percpu_counter_free(iter1->counters.pcnt); break; + } ++i; if (strcmp(arpt_get_target(iter1)->u.user.name, XT_ERROR_TARGET) == 0) @@ -1448,11 +1460,6 @@ static int translate_compat_table(const char *name, return ret; } - /* And one copy for every other CPU */ - for_each_possible_cpu(i) - if (newinfo->entries[i] && newinfo->entries[i] != entry1) - memcpy(newinfo->entries[i], entry1, newinfo->size); - *pinfo = newinfo; *pentry0 = entry1; xt_free_table_info(info); @@ -1511,8 +1518,7 @@ static int compat_do_replace(struct net *net, void __user *user, if (!newinfo) return -ENOMEM; - /* choose the copy that is on our node/cpu */ - loc_cpu_entry = newinfo->entries[raw_smp_processor_id()]; + loc_cpu_entry = newinfo->entries; if (copy_from_user(loc_cpu_entry, user + sizeof(tmp), tmp.size) != 0) { ret = -EFAULT; goto free_newinfo; @@ -1609,7 +1615,6 @@ static int compat_copy_entries_to_user(unsigned int total_size, void __user *pos; unsigned int size; int ret = 0; - void *loc_cpu_entry; unsigned int i = 0; struct arpt_entry *iter; @@ -1617,11 +1622,9 @@ static int compat_copy_entries_to_user(unsigned int total_size, if (IS_ERR(counters)) return PTR_ERR(counters); - /* choose the copy on our node/cpu */ - loc_cpu_entry = private->entries[raw_smp_processor_id()]; pos = userptr; size = total_size; - xt_entry_foreach(iter, loc_cpu_entry, total_size) { + xt_entry_foreach(iter, private->entries, total_size) { ret = compat_copy_entry_to_user(iter, &pos, &size, counters, i++); if (ret != 0) @@ -1790,8 +1793,7 @@ struct xt_table *arpt_register_table(struct net *net, goto out; } - /* choose the copy on our node/cpu */ - loc_cpu_entry = newinfo->entries[raw_smp_processor_id()]; + loc_cpu_entry = newinfo->entries; memcpy(loc_cpu_entry, repl->entries, repl->size); ret = translate_table(newinfo, loc_cpu_entry, repl); @@ -1822,7 +1824,7 @@ void arpt_unregister_table(struct xt_table *table) private = xt_unregister_table(table); /* Decrease module usage counts and free resources */ - loc_cpu_entry = private->entries[raw_smp_processor_id()]; + loc_cpu_entry = private->entries; xt_entry_foreach(iter, loc_cpu_entry, private->size) cleanup_entry(iter); if (private->number > private->initial_entries) diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index e7abf5145edc..6c72fbb7b49e 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -254,15 +254,13 @@ static void trace_packet(const struct sk_buff *skb, const struct xt_table_info *private, const struct ipt_entry *e) { - const void *table_base; const struct ipt_entry *root; const char *hookname, *chainname, *comment; const struct ipt_entry *iter; unsigned int rulenum = 0; struct net *net = dev_net(in ? in : out); - table_base = private->entries[smp_processor_id()]; - root = get_entry(table_base, private->hook_entry[hook]); + root = get_entry(private->entries, private->hook_entry[hook]); hookname = chainname = hooknames[hook]; comment = comments[NF_IP_TRACE_COMMENT_RULE]; @@ -331,7 +329,7 @@ ipt_do_table(struct sk_buff *skb, * pointer. */ smp_read_barrier_depends(); - table_base = private->entries[cpu]; + table_base = private->entries; jumpstack = (struct ipt_entry **)private->jumpstack[cpu]; stackptr = per_cpu_ptr(private->stackptr, cpu); origptr = *stackptr; @@ -345,6 +343,7 @@ ipt_do_table(struct sk_buff *skb, do { const struct xt_entry_target *t; const struct xt_entry_match *ematch; + struct xt_counters *counter; IP_NF_ASSERT(e); if (!ip_packet_match(ip, indev, outdev, @@ -361,7 +360,8 @@ ipt_do_table(struct sk_buff *skb, goto no_match; } - ADD_COUNTER(e->counters, skb->len, 1); + counter = xt_get_this_cpu_counter(&e->counters); + ADD_COUNTER(*counter, skb->len, 1); t = ipt_get_target(e); IP_NF_ASSERT(t->u.kernel.target); @@ -665,6 +665,10 @@ find_check_entry(struct ipt_entry *e, struct net *net, const char *name, if (ret) return ret; + e->counters.pcnt = xt_percpu_counter_alloc(); + if (IS_ERR_VALUE(e->counters.pcnt)) + return -ENOMEM; + j = 0; mtpar.net = net; mtpar.table = name; @@ -691,6 +695,7 @@ find_check_entry(struct ipt_entry *e, struct net *net, const char *name, ret = check_target(e, net, name); if (ret) goto err; + return 0; err: module_put(t->u.kernel.target->me); @@ -700,6 +705,9 @@ find_check_entry(struct ipt_entry *e, struct net *net, const char *name, break; cleanup_match(ematch, net); } + + xt_percpu_counter_free(e->counters.pcnt); + return ret; } @@ -784,6 +792,7 @@ cleanup_entry(struct ipt_entry *e, struct net *net) if (par.target->destroy != NULL) par.target->destroy(&par); module_put(par.target->me); + xt_percpu_counter_free(e->counters.pcnt); } /* Checks and translates the user-supplied table segment (held in @@ -866,12 +875,6 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, return ret; } - /* And one copy for every other CPU */ - for_each_possible_cpu(i) { - if (newinfo->entries[i] && newinfo->entries[i] != entry0) - memcpy(newinfo->entries[i], entry0, newinfo->size); - } - return ret; } @@ -887,14 +890,16 @@ get_counters(const struct xt_table_info *t, seqcount_t *s = &per_cpu(xt_recseq, cpu); i = 0; - xt_entry_foreach(iter, t->entries[cpu], t->size) { + xt_entry_foreach(iter, t->entries, t->size) { + struct xt_counters *tmp; u64 bcnt, pcnt; unsigned int start; + tmp = xt_get_per_cpu_counter(&iter->counters, cpu); do { start = read_seqcount_begin(s); - bcnt = iter->counters.bcnt; - pcnt = iter->counters.pcnt; + bcnt = tmp->bcnt; + pcnt = tmp->pcnt; } while (read_seqcount_retry(s, start)); ADD_COUNTER(counters[i], bcnt, pcnt); @@ -939,11 +944,7 @@ copy_entries_to_user(unsigned int total_size, if (IS_ERR(counters)) return PTR_ERR(counters); - /* choose the copy that is on our node/cpu, ... - * This choice is lazy (because current thread is - * allowed to migrate to another cpu) - */ - loc_cpu_entry = private->entries[raw_smp_processor_id()]; + loc_cpu_entry = private->entries; if (copy_to_user(userptr, loc_cpu_entry, total_size) != 0) { ret = -EFAULT; goto free_counters; @@ -1051,16 +1052,16 @@ static int compat_table_info(const struct xt_table_info *info, struct xt_table_info *newinfo) { struct ipt_entry *iter; - void *loc_cpu_entry; + const void *loc_cpu_entry; int ret; if (!newinfo || !info) return -EINVAL; - /* we dont care about newinfo->entries[] */ + /* we dont care about newinfo->entries */ memcpy(newinfo, info, offsetof(struct xt_table_info, entries)); newinfo->initial_entries = 0; - loc_cpu_entry = info->entries[raw_smp_processor_id()]; + loc_cpu_entry = info->entries; xt_compat_init_offsets(AF_INET, info->number); xt_entry_foreach(iter, loc_cpu_entry, info->size) { ret = compat_calc_entry(iter, info, loc_cpu_entry, newinfo); @@ -1181,7 +1182,6 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks, struct xt_table *t; struct xt_table_info *oldinfo; struct xt_counters *counters; - void *loc_cpu_old_entry; struct ipt_entry *iter; ret = 0; @@ -1224,8 +1224,7 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks, get_counters(oldinfo, counters); /* Decrease module usage counts and free resource */ - loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()]; - xt_entry_foreach(iter, loc_cpu_old_entry, oldinfo->size) + xt_entry_foreach(iter, oldinfo->entries, oldinfo->size) cleanup_entry(iter, net); xt_free_table_info(oldinfo); @@ -1271,8 +1270,7 @@ do_replace(struct net *net, const void __user *user, unsigned int len) if (!newinfo) return -ENOMEM; - /* choose the copy that is on our node/cpu */ - loc_cpu_entry = newinfo->entries[raw_smp_processor_id()]; + loc_cpu_entry = newinfo->entries; if (copy_from_user(loc_cpu_entry, user + sizeof(tmp), tmp.size) != 0) { ret = -EFAULT; @@ -1303,7 +1301,7 @@ static int do_add_counters(struct net *net, const void __user *user, unsigned int len, int compat) { - unsigned int i, curcpu; + unsigned int i; struct xt_counters_info tmp; struct xt_counters *paddc; unsigned int num_counters; @@ -1313,7 +1311,6 @@ do_add_counters(struct net *net, const void __user *user, struct xt_table *t; const struct xt_table_info *private; int ret = 0; - void *loc_cpu_entry; struct ipt_entry *iter; unsigned int addend; #ifdef CONFIG_COMPAT @@ -1369,12 +1366,12 @@ do_add_counters(struct net *net, const void __user *user, } i = 0; - /* Choose the copy that is on our node */ - curcpu = smp_processor_id(); - loc_cpu_entry = private->entries[curcpu]; addend = xt_write_recseq_begin(); - xt_entry_foreach(iter, loc_cpu_entry, private->size) { - ADD_COUNTER(iter->counters, paddc[i].bcnt, paddc[i].pcnt); + xt_entry_foreach(iter, private->entries, private->size) { + struct xt_counters *tmp; + + tmp = xt_get_this_cpu_counter(&iter->counters); + ADD_COUNTER(*tmp, paddc[i].bcnt, paddc[i].pcnt); ++i; } xt_write_recseq_end(addend); @@ -1608,6 +1605,10 @@ compat_check_entry(struct ipt_entry *e, struct net *net, const char *name) unsigned int j; int ret = 0; + e->counters.pcnt = xt_percpu_counter_alloc(); + if (IS_ERR_VALUE(e->counters.pcnt)) + return -ENOMEM; + j = 0; mtpar.net = net; mtpar.table = name; @@ -1632,6 +1633,9 @@ compat_check_entry(struct ipt_entry *e, struct net *net, const char *name) break; cleanup_match(ematch, net); } + + xt_percpu_counter_free(e->counters.pcnt); + return ret; } @@ -1716,7 +1720,7 @@ translate_compat_table(struct net *net, newinfo->hook_entry[i] = info->hook_entry[i]; newinfo->underflow[i] = info->underflow[i]; } - entry1 = newinfo->entries[raw_smp_processor_id()]; + entry1 = newinfo->entries; pos = entry1; size = total_size; xt_entry_foreach(iter0, entry0, total_size) { @@ -1768,11 +1772,6 @@ translate_compat_table(struct net *net, return ret; } - /* And one copy for every other CPU */ - for_each_possible_cpu(i) - if (newinfo->entries[i] && newinfo->entries[i] != entry1) - memcpy(newinfo->entries[i], entry1, newinfo->size); - *pinfo = newinfo; *pentry0 = entry1; xt_free_table_info(info); @@ -1819,8 +1818,7 @@ compat_do_replace(struct net *net, void __user *user, unsigned int len) if (!newinfo) return -ENOMEM; - /* choose the copy that is on our node/cpu */ - loc_cpu_entry = newinfo->entries[raw_smp_processor_id()]; + loc_cpu_entry = newinfo->entries; if (copy_from_user(loc_cpu_entry, user + sizeof(tmp), tmp.size) != 0) { ret = -EFAULT; @@ -1891,7 +1889,6 @@ compat_copy_entries_to_user(unsigned int total_size, struct xt_table *table, void __user *pos; unsigned int size; int ret = 0; - const void *loc_cpu_entry; unsigned int i = 0; struct ipt_entry *iter; @@ -1899,14 +1896,9 @@ compat_copy_entries_to_user(unsigned int total_size, struct xt_table *table, if (IS_ERR(counters)) return PTR_ERR(counters); - /* choose the copy that is on our node/cpu, ... - * This choice is lazy (because current thread is - * allowed to migrate to another cpu) - */ - loc_cpu_entry = private->entries[raw_smp_processor_id()]; pos = userptr; size = total_size; - xt_entry_foreach(iter, loc_cpu_entry, total_size) { + xt_entry_foreach(iter, private->entries, total_size) { ret = compat_copy_entry_to_user(iter, &pos, &size, counters, i++); if (ret != 0) @@ -2081,8 +2073,7 @@ struct xt_table *ipt_register_table(struct net *net, goto out; } - /* choose the copy on our node/cpu, but dont care about preemption */ - loc_cpu_entry = newinfo->entries[raw_smp_processor_id()]; + loc_cpu_entry = newinfo->entries; memcpy(loc_cpu_entry, repl->entries, repl->size); ret = translate_table(net, newinfo, loc_cpu_entry, repl); @@ -2113,7 +2104,7 @@ void ipt_unregister_table(struct net *net, struct xt_table *table) private = xt_unregister_table(table); /* Decrease module usage counts and free resources */ - loc_cpu_entry = private->entries[raw_smp_processor_id()]; + loc_cpu_entry = private->entries; xt_entry_foreach(iter, loc_cpu_entry, private->size) cleanup_entry(iter, net); if (private->number > private->initial_entries) diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index d958718b5031..b4de08a83e0b 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -191,6 +191,8 @@ static __sum16 nf_ip6_checksum_partial(struct sk_buff *skb, unsigned int hook, static const struct nf_ipv6_ops ipv6ops = { .chk_addr = ipv6_chk_addr, + .route_input = ip6_route_input, + .fragment = ip6_fragment }; static const struct nf_afinfo nf_ip6_afinfo = { diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig index ca6998345b42..b552cf0d6198 100644 --- a/net/ipv6/netfilter/Kconfig +++ b/net/ipv6/netfilter/Kconfig @@ -186,7 +186,8 @@ config IP6_NF_MATCH_MH config IP6_NF_MATCH_RPFILTER tristate '"rpfilter" reverse path filter match support' - depends on NETFILTER_ADVANCED && (IP6_NF_MANGLE || IP6_NF_RAW) + depends on NETFILTER_ADVANCED + depends on IP6_NF_MANGLE || IP6_NF_RAW ---help--- This option allows you to match packets whose replies would go out via the interface the packet came in. diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index cdd085f8b770..3c35ced39b42 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -283,15 +283,13 @@ static void trace_packet(const struct sk_buff *skb, const struct xt_table_info *private, const struct ip6t_entry *e) { - const void *table_base; const struct ip6t_entry *root; const char *hookname, *chainname, *comment; const struct ip6t_entry *iter; unsigned int rulenum = 0; struct net *net = dev_net(in ? in : out); - table_base = private->entries[smp_processor_id()]; - root = get_entry(table_base, private->hook_entry[hook]); + root = get_entry(private->entries, private->hook_entry[hook]); hookname = chainname = hooknames[hook]; comment = comments[NF_IP6_TRACE_COMMENT_RULE]; @@ -357,7 +355,7 @@ ip6t_do_table(struct sk_buff *skb, */ smp_read_barrier_depends(); cpu = smp_processor_id(); - table_base = private->entries[cpu]; + table_base = private->entries; jumpstack = (struct ip6t_entry **)private->jumpstack[cpu]; stackptr = per_cpu_ptr(private->stackptr, cpu); origptr = *stackptr; @@ -367,6 +365,7 @@ ip6t_do_table(struct sk_buff *skb, do { const struct xt_entry_target *t; const struct xt_entry_match *ematch; + struct xt_counters *counter; IP_NF_ASSERT(e); acpar.thoff = 0; @@ -384,7 +383,8 @@ ip6t_do_table(struct sk_buff *skb, goto no_match; } - ADD_COUNTER(e->counters, skb->len, 1); + counter = xt_get_this_cpu_counter(&e->counters); + ADD_COUNTER(*counter, skb->len, 1); t = ip6t_get_target_c(e); IP_NF_ASSERT(t->u.kernel.target); @@ -679,6 +679,10 @@ find_check_entry(struct ip6t_entry *e, struct net *net, const char *name, if (ret) return ret; + e->counters.pcnt = xt_percpu_counter_alloc(); + if (IS_ERR_VALUE(e->counters.pcnt)) + return -ENOMEM; + j = 0; mtpar.net = net; mtpar.table = name; @@ -714,6 +718,9 @@ find_check_entry(struct ip6t_entry *e, struct net *net, const char *name, break; cleanup_match(ematch, net); } + + xt_percpu_counter_free(e->counters.pcnt); + return ret; } @@ -797,6 +804,8 @@ static void cleanup_entry(struct ip6t_entry *e, struct net *net) if (par.target->destroy != NULL) par.target->destroy(&par); module_put(par.target->me); + + xt_percpu_counter_free(e->counters.pcnt); } /* Checks and translates the user-supplied table segment (held in @@ -879,12 +888,6 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, return ret; } - /* And one copy for every other CPU */ - for_each_possible_cpu(i) { - if (newinfo->entries[i] && newinfo->entries[i] != entry0) - memcpy(newinfo->entries[i], entry0, newinfo->size); - } - return ret; } @@ -900,14 +903,16 @@ get_counters(const struct xt_table_info *t, seqcount_t *s = &per_cpu(xt_recseq, cpu); i = 0; - xt_entry_foreach(iter, t->entries[cpu], t->size) { + xt_entry_foreach(iter, t->entries, t->size) { + struct xt_counters *tmp; u64 bcnt, pcnt; unsigned int start; + tmp = xt_get_per_cpu_counter(&iter->counters, cpu); do { start = read_seqcount_begin(s); - bcnt = iter->counters.bcnt; - pcnt = iter->counters.pcnt; + bcnt = tmp->bcnt; + pcnt = tmp->pcnt; } while (read_seqcount_retry(s, start)); ADD_COUNTER(counters[i], bcnt, pcnt); @@ -952,11 +957,7 @@ copy_entries_to_user(unsigned int total_size, if (IS_ERR(counters)) return PTR_ERR(counters); - /* choose the copy that is on our node/cpu, ... - * This choice is lazy (because current thread is - * allowed to migrate to another cpu) - */ - loc_cpu_entry = private->entries[raw_smp_processor_id()]; + loc_cpu_entry = private->entries; if (copy_to_user(userptr, loc_cpu_entry, total_size) != 0) { ret = -EFAULT; goto free_counters; @@ -1064,16 +1065,16 @@ static int compat_table_info(const struct xt_table_info *info, struct xt_table_info *newinfo) { struct ip6t_entry *iter; - void *loc_cpu_entry; + const void *loc_cpu_entry; int ret; if (!newinfo || !info) return -EINVAL; - /* we dont care about newinfo->entries[] */ + /* we dont care about newinfo->entries */ memcpy(newinfo, info, offsetof(struct xt_table_info, entries)); newinfo->initial_entries = 0; - loc_cpu_entry = info->entries[raw_smp_processor_id()]; + loc_cpu_entry = info->entries; xt_compat_init_offsets(AF_INET6, info->number); xt_entry_foreach(iter, loc_cpu_entry, info->size) { ret = compat_calc_entry(iter, info, loc_cpu_entry, newinfo); @@ -1194,7 +1195,6 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks, struct xt_table *t; struct xt_table_info *oldinfo; struct xt_counters *counters; - const void *loc_cpu_old_entry; struct ip6t_entry *iter; ret = 0; @@ -1237,8 +1237,7 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks, get_counters(oldinfo, counters); /* Decrease module usage counts and free resource */ - loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()]; - xt_entry_foreach(iter, loc_cpu_old_entry, oldinfo->size) + xt_entry_foreach(iter, oldinfo->entries, oldinfo->size) cleanup_entry(iter, net); xt_free_table_info(oldinfo); @@ -1284,8 +1283,7 @@ do_replace(struct net *net, const void __user *user, unsigned int len) if (!newinfo) return -ENOMEM; - /* choose the copy that is on our node/cpu */ - loc_cpu_entry = newinfo->entries[raw_smp_processor_id()]; + loc_cpu_entry = newinfo->entries; if (copy_from_user(loc_cpu_entry, user + sizeof(tmp), tmp.size) != 0) { ret = -EFAULT; @@ -1316,7 +1314,7 @@ static int do_add_counters(struct net *net, const void __user *user, unsigned int len, int compat) { - unsigned int i, curcpu; + unsigned int i; struct xt_counters_info tmp; struct xt_counters *paddc; unsigned int num_counters; @@ -1326,7 +1324,6 @@ do_add_counters(struct net *net, const void __user *user, unsigned int len, struct xt_table *t; const struct xt_table_info *private; int ret = 0; - const void *loc_cpu_entry; struct ip6t_entry *iter; unsigned int addend; #ifdef CONFIG_COMPAT @@ -1374,7 +1371,6 @@ do_add_counters(struct net *net, const void __user *user, unsigned int len, goto free; } - local_bh_disable(); private = t->private; if (private->number != num_counters) { @@ -1383,16 +1379,15 @@ do_add_counters(struct net *net, const void __user *user, unsigned int len, } i = 0; - /* Choose the copy that is on our node */ - curcpu = smp_processor_id(); addend = xt_write_recseq_begin(); - loc_cpu_entry = private->entries[curcpu]; - xt_entry_foreach(iter, loc_cpu_entry, private->size) { - ADD_COUNTER(iter->counters, paddc[i].bcnt, paddc[i].pcnt); + xt_entry_foreach(iter, private->entries, private->size) { + struct xt_counters *tmp; + + tmp = xt_get_this_cpu_counter(&iter->counters); + ADD_COUNTER(*tmp, paddc[i].bcnt, paddc[i].pcnt); ++i; } xt_write_recseq_end(addend); - unlock_up_free: local_bh_enable(); xt_table_unlock(t); @@ -1621,6 +1616,9 @@ static int compat_check_entry(struct ip6t_entry *e, struct net *net, struct xt_mtchk_param mtpar; struct xt_entry_match *ematch; + e->counters.pcnt = xt_percpu_counter_alloc(); + if (IS_ERR_VALUE(e->counters.pcnt)) + return -ENOMEM; j = 0; mtpar.net = net; mtpar.table = name; @@ -1645,6 +1643,9 @@ static int compat_check_entry(struct ip6t_entry *e, struct net *net, break; cleanup_match(ematch, net); } + + xt_percpu_counter_free(e->counters.pcnt); + return ret; } @@ -1729,7 +1730,7 @@ translate_compat_table(struct net *net, newinfo->hook_entry[i] = info->hook_entry[i]; newinfo->underflow[i] = info->underflow[i]; } - entry1 = newinfo->entries[raw_smp_processor_id()]; + entry1 = newinfo->entries; pos = entry1; size = total_size; xt_entry_foreach(iter0, entry0, total_size) { @@ -1781,11 +1782,6 @@ translate_compat_table(struct net *net, return ret; } - /* And one copy for every other CPU */ - for_each_possible_cpu(i) - if (newinfo->entries[i] && newinfo->entries[i] != entry1) - memcpy(newinfo->entries[i], entry1, newinfo->size); - *pinfo = newinfo; *pentry0 = entry1; xt_free_table_info(info); @@ -1832,8 +1828,7 @@ compat_do_replace(struct net *net, void __user *user, unsigned int len) if (!newinfo) return -ENOMEM; - /* choose the copy that is on our node/cpu */ - loc_cpu_entry = newinfo->entries[raw_smp_processor_id()]; + loc_cpu_entry = newinfo->entries; if (copy_from_user(loc_cpu_entry, user + sizeof(tmp), tmp.size) != 0) { ret = -EFAULT; @@ -1904,7 +1899,6 @@ compat_copy_entries_to_user(unsigned int total_size, struct xt_table *table, void __user *pos; unsigned int size; int ret = 0; - const void *loc_cpu_entry; unsigned int i = 0; struct ip6t_entry *iter; @@ -1912,14 +1906,9 @@ compat_copy_entries_to_user(unsigned int total_size, struct xt_table *table, if (IS_ERR(counters)) return PTR_ERR(counters); - /* choose the copy that is on our node/cpu, ... - * This choice is lazy (because current thread is - * allowed to migrate to another cpu) - */ - loc_cpu_entry = private->entries[raw_smp_processor_id()]; pos = userptr; size = total_size; - xt_entry_foreach(iter, loc_cpu_entry, total_size) { + xt_entry_foreach(iter, private->entries, total_size) { ret = compat_copy_entry_to_user(iter, &pos, &size, counters, i++); if (ret != 0) @@ -2094,8 +2083,7 @@ struct xt_table *ip6t_register_table(struct net *net, goto out; } - /* choose the copy on our node/cpu, but dont care about preemption */ - loc_cpu_entry = newinfo->entries[raw_smp_processor_id()]; + loc_cpu_entry = newinfo->entries; memcpy(loc_cpu_entry, repl->entries, repl->size); ret = translate_table(net, newinfo, loc_cpu_entry, repl); @@ -2125,7 +2113,7 @@ void ip6t_unregister_table(struct net *net, struct xt_table *table) private = xt_unregister_table(table); /* Decrease module usage counts and free resources */ - loc_cpu_entry = private->entries[raw_smp_processor_id()]; + loc_cpu_entry = private->entries; xt_entry_foreach(iter, loc_cpu_entry, private->size) cleanup_entry(iter, net); if (private->number > private->initial_entries) diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index fbc8d15c7fda..6eae69a698ed 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -206,7 +206,7 @@ config NF_CONNTRACK_FTP config NF_CONNTRACK_H323 tristate "H.323 protocol support" - depends on (IPV6 || IPV6=n) + depends on IPV6 || IPV6=n depends on NETFILTER_ADVANCED help H.323 is a VoIP signalling protocol from ITU-T. As one of the most @@ -723,7 +723,7 @@ config NETFILTER_XT_TARGET_HL config NETFILTER_XT_TARGET_HMARK tristate '"HMARK" target support' - depends on (IP6_NF_IPTABLES || IP6_NF_IPTABLES=n) + depends on IP6_NF_IPTABLES || IP6_NF_IPTABLES=n depends on NETFILTER_ADVANCED ---help--- This option adds the "HMARK" target. @@ -865,7 +865,7 @@ config NETFILTER_XT_TARGET_REDIRECT config NETFILTER_XT_TARGET_TEE tristate '"TEE" - packet cloning to alternate destination' depends on NETFILTER_ADVANCED - depends on (IPV6 || IPV6=n) + depends on IPV6 || IPV6=n depends on !NF_CONNTRACK || NF_CONNTRACK ---help--- This option adds a "TEE" target with which a packet can be cloned and @@ -875,8 +875,8 @@ config NETFILTER_XT_TARGET_TPROXY tristate '"TPROXY" target transparent proxying support' depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED - depends on (IPV6 || IPV6=n) - depends on (IP6_NF_IPTABLES || IP6_NF_IPTABLES=n) + depends on IPV6 || IPV6=n + depends on IP6_NF_IPTABLES || IP6_NF_IPTABLES=n depends on IP_NF_MANGLE select NF_DEFRAG_IPV4 select NF_DEFRAG_IPV6 if IP6_NF_IPTABLES @@ -915,7 +915,7 @@ config NETFILTER_XT_TARGET_SECMARK config NETFILTER_XT_TARGET_TCPMSS tristate '"TCPMSS" target support' - depends on (IPV6 || IPV6=n) + depends on IPV6 || IPV6=n default m if NETFILTER_ADVANCED=n ---help--- This option adds a `TCPMSS' target, which allows you to alter the @@ -1127,7 +1127,7 @@ config NETFILTER_XT_MATCH_ESP config NETFILTER_XT_MATCH_HASHLIMIT tristate '"hashlimit" match support' - depends on (IP6_NF_IPTABLES || IP6_NF_IPTABLES=n) + depends on IP6_NF_IPTABLES || IP6_NF_IPTABLES=n depends on NETFILTER_ADVANCED help This option adds a `hashlimit' match. @@ -1369,8 +1369,8 @@ config NETFILTER_XT_MATCH_SOCKET depends on NETFILTER_XTABLES depends on NETFILTER_ADVANCED depends on !NF_CONNTRACK || NF_CONNTRACK - depends on (IPV6 || IPV6=n) - depends on (IP6_NF_IPTABLES || IP6_NF_IPTABLES=n) + depends on IPV6 || IPV6=n + depends on IP6_NF_IPTABLES || IP6_NF_IPTABLES=n select NF_DEFRAG_IPV4 select NF_DEFRAG_IPV6 if IP6_NF_IPTABLES help diff --git a/net/netfilter/ipset/ip_set_bitmap_gen.h b/net/netfilter/ipset/ip_set_bitmap_gen.h index 6f024a8a1534..d05e759ed0fa 100644 --- a/net/netfilter/ipset/ip_set_bitmap_gen.h +++ b/net/netfilter/ipset/ip_set_bitmap_gen.h @@ -41,7 +41,7 @@ mtype_gc_init(struct ip_set *set, void (*gc)(unsigned long ul_set)) struct mtype *map = set->data; init_timer(&map->gc); - map->gc.data = (unsigned long) set; + map->gc.data = (unsigned long)set; map->gc.function = gc; map->gc.expires = jiffies + IPSET_GC_PERIOD(set->timeout) * HZ; add_timer(&map->gc); @@ -144,10 +144,12 @@ mtype_add(struct ip_set *set, void *value, const struct ip_set_ext *ext, if (ret == IPSET_ADD_FAILED) { if (SET_WITH_TIMEOUT(set) && - ip_set_timeout_expired(ext_timeout(x, set))) + ip_set_timeout_expired(ext_timeout(x, set))) { ret = 0; - else if (!(flags & IPSET_FLAG_EXIST)) + } else if (!(flags & IPSET_FLAG_EXIST)) { + set_bit(e->id, map->members); return -IPSET_ERR_EXIST; + } /* Element is re-added, cleanup extensions */ ip_set_ext_destroy(set, x); } @@ -165,6 +167,10 @@ mtype_add(struct ip_set *set, void *value, const struct ip_set_ext *ext, ip_set_init_comment(ext_comment(x, set), ext); if (SET_WITH_SKBINFO(set)) ip_set_init_skbinfo(ext_skbinfo(x, set), ext); + + /* Activate element */ + set_bit(e->id, map->members); + return 0; } @@ -203,10 +209,13 @@ mtype_list(const struct ip_set *set, struct nlattr *adt, *nested; void *x; u32 id, first = cb->args[IPSET_CB_ARG0]; + int ret = 0; adt = ipset_nest_start(skb, IPSET_ATTR_ADT); if (!adt) return -EMSGSIZE; + /* Extensions may be replaced */ + rcu_read_lock(); for (; cb->args[IPSET_CB_ARG0] < map->elements; cb->args[IPSET_CB_ARG0]++) { id = cb->args[IPSET_CB_ARG0]; @@ -214,7 +223,7 @@ mtype_list(const struct ip_set *set, if (!test_bit(id, map->members) || (SET_WITH_TIMEOUT(set) && #ifdef IP_SET_BITMAP_STORED_TIMEOUT - mtype_is_filled((const struct mtype_elem *) x) && + mtype_is_filled((const struct mtype_elem *)x) && #endif ip_set_timeout_expired(ext_timeout(x, set)))) continue; @@ -222,14 +231,16 @@ mtype_list(const struct ip_set *set, if (!nested) { if (id == first) { nla_nest_cancel(skb, adt); - return -EMSGSIZE; - } else - goto nla_put_failure; + ret = -EMSGSIZE; + goto out; + } + + goto nla_put_failure; } if (mtype_do_list(skb, map, id, set->dsize)) goto nla_put_failure; if (ip_set_put_extensions(skb, set, x, - mtype_is_filled((const struct mtype_elem *) x))) + mtype_is_filled((const struct mtype_elem *)x))) goto nla_put_failure; ipset_nest_end(skb, nested); } @@ -238,29 +249,32 @@ mtype_list(const struct ip_set *set, /* Set listing finished */ cb->args[IPSET_CB_ARG0] = 0; - return 0; + goto out; nla_put_failure: nla_nest_cancel(skb, nested); if (unlikely(id == first)) { cb->args[IPSET_CB_ARG0] = 0; - return -EMSGSIZE; + ret = -EMSGSIZE; } ipset_nest_end(skb, adt); - return 0; +out: + rcu_read_unlock(); + return ret; } static void mtype_gc(unsigned long ul_set) { - struct ip_set *set = (struct ip_set *) ul_set; + struct ip_set *set = (struct ip_set *)ul_set; struct mtype *map = set->data; void *x; u32 id; /* We run parallel with other readers (test element) - * but adding/deleting new entries is locked out */ - read_lock_bh(&set->lock); + * but adding/deleting new entries is locked out + */ + spin_lock_bh(&set->lock); for (id = 0; id < map->elements; id++) if (mtype_gc_test(id, map, set->dsize)) { x = get_ext(set, map, id); @@ -269,7 +283,7 @@ mtype_gc(unsigned long ul_set) ip_set_ext_destroy(set, x); } } - read_unlock_bh(&set->lock); + spin_unlock_bh(&set->lock); map->gc.expires = jiffies + IPSET_GC_PERIOD(set->timeout) * HZ; add_timer(&map->gc); diff --git a/net/netfilter/ipset/ip_set_bitmap_ip.c b/net/netfilter/ipset/ip_set_bitmap_ip.c index 2fe6de46f6d0..64a564334418 100644 --- a/net/netfilter/ipset/ip_set_bitmap_ip.c +++ b/net/netfilter/ipset/ip_set_bitmap_ip.c @@ -59,7 +59,7 @@ struct bitmap_ip_adt_elem { static inline u32 ip_to_id(const struct bitmap_ip *m, u32 ip) { - return ((ip & ip_set_hostmask(m->netmask)) - m->first_ip)/m->hosts; + return ((ip & ip_set_hostmask(m->netmask)) - m->first_ip) / m->hosts; } /* Common functions */ @@ -81,7 +81,7 @@ static inline int bitmap_ip_do_add(const struct bitmap_ip_adt_elem *e, struct bitmap_ip *map, u32 flags, size_t dsize) { - return !!test_and_set_bit(e->id, map->members); + return !!test_bit(e->id, map->members); } static inline int @@ -138,18 +138,12 @@ bitmap_ip_uadt(struct ip_set *set, struct nlattr *tb[], struct ip_set_ext ext = IP_SET_INIT_UEXT(set); int ret = 0; - if (unlikely(!tb[IPSET_ATTR_IP] || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) - return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP])) + return -IPSET_ERR_PROTOCOL; + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP], &ip); if (ret) return ret; @@ -181,8 +175,9 @@ bitmap_ip_uadt(struct ip_set *set, struct nlattr *tb[], if (!cidr || cidr > HOST_MASK) return -IPSET_ERR_INVALID_CIDR; ip_set_mask_from_to(ip, ip_to, cidr); - } else + } else { ip_to = ip; + } if (ip_to > map->last_ip) return -IPSET_ERR_BITMAP_RANGE; @@ -193,8 +188,8 @@ bitmap_ip_uadt(struct ip_set *set, struct nlattr *tb[], if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; } return ret; } @@ -284,8 +279,9 @@ bitmap_ip_create(struct net *net, struct ip_set *set, struct nlattr *tb[], if (cidr >= HOST_MASK) return -IPSET_ERR_INVALID_CIDR; ip_set_mask_from_to(first_ip, last_ip, cidr); - } else + } else { return -IPSET_ERR_PROTOCOL; + } if (tb[IPSET_ATTR_NETMASK]) { netmask = nla_get_u8(tb[IPSET_ATTR_NETMASK]); @@ -382,6 +378,7 @@ bitmap_ip_init(void) static void __exit bitmap_ip_fini(void) { + rcu_barrier(); ip_set_type_unregister(&bitmap_ip_type); } diff --git a/net/netfilter/ipset/ip_set_bitmap_ipmac.c b/net/netfilter/ipset/ip_set_bitmap_ipmac.c index eb188561d65f..1430535118fb 100644 --- a/net/netfilter/ipset/ip_set_bitmap_ipmac.c +++ b/net/netfilter/ipset/ip_set_bitmap_ipmac.c @@ -90,7 +90,7 @@ bitmap_ipmac_do_test(const struct bitmap_ipmac_adt_elem *e, return 0; elem = get_elem(map->extensions, e->id, dsize); if (elem->filled == MAC_FILLED) - return e->ether == NULL || + return !e->ether || ether_addr_equal(e->ether, elem->ether); /* Trigger kernel to fill out the ethernet address */ return -EAGAIN; @@ -131,7 +131,8 @@ bitmap_ipmac_add_timeout(unsigned long *timeout, /* If MAC is unset yet, we store plain timeout value * because the timer is not activated yet * and we can reuse it later when MAC is filled out, - * possibly by the kernel */ + * possibly by the kernel + */ if (e->ether) ip_set_timeout_set(timeout, t); else @@ -147,28 +148,35 @@ bitmap_ipmac_do_add(const struct bitmap_ipmac_adt_elem *e, struct bitmap_ipmac_elem *elem; elem = get_elem(map->extensions, e->id, dsize); - if (test_and_set_bit(e->id, map->members)) { + if (test_bit(e->id, map->members)) { if (elem->filled == MAC_FILLED) { - if (e->ether && (flags & IPSET_FLAG_EXIST)) - memcpy(elem->ether, e->ether, ETH_ALEN); + if (e->ether && + (flags & IPSET_FLAG_EXIST) && + !ether_addr_equal(e->ether, elem->ether)) { + /* memcpy isn't atomic */ + clear_bit(e->id, map->members); + smp_mb__after_atomic(); + ether_addr_copy(elem->ether, e->ether); + } return IPSET_ADD_FAILED; } else if (!e->ether) /* Already added without ethernet address */ return IPSET_ADD_FAILED; /* Fill the MAC address and trigger the timer activation */ - memcpy(elem->ether, e->ether, ETH_ALEN); + clear_bit(e->id, map->members); + smp_mb__after_atomic(); + ether_addr_copy(elem->ether, e->ether); elem->filled = MAC_FILLED; return IPSET_ADD_START_STORED_TIMEOUT; } else if (e->ether) { /* We can store MAC too */ - memcpy(elem->ether, e->ether, ETH_ALEN); + ether_addr_copy(elem->ether, e->ether); elem->filled = MAC_FILLED; return 0; - } else { - elem->filled = MAC_UNSET; - /* MAC is not stored yet, don't start timer */ - return IPSET_ADD_STORE_PLAIN_TIMEOUT; } + elem->filled = MAC_UNSET; + /* MAC is not stored yet, don't start timer */ + return IPSET_ADD_STORE_PLAIN_TIMEOUT; } static inline int @@ -239,18 +247,12 @@ bitmap_ipmac_uadt(struct ip_set *set, struct nlattr *tb[], u32 ip = 0; int ret = 0; - if (unlikely(!tb[IPSET_ATTR_IP] || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) - return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP])) + return -IPSET_ERR_PROTOCOL; + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP], &ip); if (ret) return ret; @@ -350,8 +352,9 @@ bitmap_ipmac_create(struct net *net, struct ip_set *set, struct nlattr *tb[], if (cidr >= HOST_MASK) return -IPSET_ERR_INVALID_CIDR; ip_set_mask_from_to(first_ip, last_ip, cidr); - } else + } else { return -IPSET_ERR_PROTOCOL; + } elements = (u64)last_ip - first_ip + 1; @@ -419,6 +422,7 @@ bitmap_ipmac_init(void) static void __exit bitmap_ipmac_fini(void) { + rcu_barrier(); ip_set_type_unregister(&bitmap_ipmac_type); } diff --git a/net/netfilter/ipset/ip_set_bitmap_port.c b/net/netfilter/ipset/ip_set_bitmap_port.c index 898edb693b3f..5338ccd5da46 100644 --- a/net/netfilter/ipset/ip_set_bitmap_port.c +++ b/net/netfilter/ipset/ip_set_bitmap_port.c @@ -73,7 +73,7 @@ static inline int bitmap_port_do_add(const struct bitmap_port_adt_elem *e, struct bitmap_port *map, u32 flags, size_t dsize) { - return !!test_and_set_bit(e->id, map->members); + return !!test_bit(e->id, map->members); } static inline int @@ -136,19 +136,13 @@ bitmap_port_uadt(struct ip_set *set, struct nlattr *tb[], u16 port_to; int ret = 0; - if (unlikely(!ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) - return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO))) + return -IPSET_ERR_PROTOCOL; + port = ip_set_get_h16(tb[IPSET_ATTR_PORT]); if (port < map->first_port || port > map->last_port) return -IPSET_ERR_BITMAP_RANGE; @@ -168,8 +162,9 @@ bitmap_port_uadt(struct ip_set *set, struct nlattr *tb[], if (port < map->first_port) return -IPSET_ERR_BITMAP_RANGE; } - } else + } else { port_to = port; + } if (port_to > map->last_port) return -IPSET_ERR_BITMAP_RANGE; @@ -180,8 +175,8 @@ bitmap_port_uadt(struct ip_set *set, struct nlattr *tb[], if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; } return ret; } @@ -312,6 +307,7 @@ bitmap_port_init(void) static void __exit bitmap_port_fini(void) { + rcu_barrier(); ip_set_type_unregister(&bitmap_port_type); } diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c index 475e4960a164..338b4047776f 100644 --- a/net/netfilter/ipset/ip_set_core.c +++ b/net/netfilter/ipset/ip_set_core.c @@ -32,8 +32,10 @@ static DEFINE_RWLOCK(ip_set_ref_lock); /* protects the set refs */ struct ip_set_net { struct ip_set * __rcu *ip_set_list; /* all individual sets */ ip_set_id_t ip_set_max; /* max number of sets */ - int is_deleted; /* deleted by ip_set_net_exit */ + bool is_deleted; /* deleted by ip_set_net_exit */ + bool is_destroyed; /* all sets are destroyed */ }; + static int ip_set_net_id __read_mostly; static inline struct ip_set_net *ip_set_pernet(struct net *net) @@ -59,8 +61,7 @@ MODULE_ALIAS_NFNL_SUBSYS(NFNL_SUBSYS_IPSET); #define ip_set(inst, id) \ ip_set_dereference((inst)->ip_set_list)[id] -/* - * The set types are implemented in modules and registered set types +/* The set types are implemented in modules and registered set types * can be found in ip_set_type_list. Adding/deleting types is * serialized by ip_set_type_mutex. */ @@ -130,7 +131,8 @@ __find_set_type_get(const char *name, u8 family, u8 revision, goto unlock; } /* Make sure the type is already loaded - * but we don't support the revision */ + * but we don't support the revision + */ list_for_each_entry_rcu(type, &ip_set_type_list, list) if (STRNCMP(type->name, name)) { err = -IPSET_ERR_FIND_TYPE; @@ -208,15 +210,15 @@ ip_set_type_register(struct ip_set_type *type) pr_warn("ip_set type %s, family %s with revision min %u already registered!\n", type->name, family_name(type->family), type->revision_min); - ret = -EINVAL; - goto unlock; + ip_set_type_unlock(); + return -EINVAL; } list_add_rcu(&type->list, &ip_set_type_list); pr_debug("type %s, family %s, revision %u:%u registered.\n", type->name, family_name(type->family), type->revision_min, type->revision_max); -unlock: ip_set_type_unlock(); + return ret; } EXPORT_SYMBOL_GPL(ip_set_type_register); @@ -230,12 +232,12 @@ ip_set_type_unregister(struct ip_set_type *type) pr_warn("ip_set type %s, family %s with revision min %u not registered\n", type->name, family_name(type->family), type->revision_min); - goto unlock; + ip_set_type_unlock(); + return; } list_del_rcu(&type->list); pr_debug("type %s, family %s with revision min %u unregistered.\n", type->name, family_name(type->family), type->revision_min); -unlock: ip_set_type_unlock(); synchronize_rcu(); @@ -289,7 +291,7 @@ static const struct nla_policy ipaddr_policy[IPSET_ATTR_IPADDR_MAX + 1] = { int ip_set_get_ipaddr4(struct nlattr *nla, __be32 *ipaddr) { - struct nlattr *tb[IPSET_ATTR_IPADDR_MAX+1]; + struct nlattr *tb[IPSET_ATTR_IPADDR_MAX + 1]; if (unlikely(!flag_nested(nla))) return -IPSET_ERR_PROTOCOL; @@ -306,7 +308,7 @@ EXPORT_SYMBOL_GPL(ip_set_get_ipaddr4); int ip_set_get_ipaddr6(struct nlattr *nla, union nf_inet_addr *ipaddr) { - struct nlattr *tb[IPSET_ATTR_IPADDR_MAX+1]; + struct nlattr *tb[IPSET_ATTR_IPADDR_MAX + 1]; if (unlikely(!flag_nested(nla))) return -IPSET_ERR_PROTOCOL; @@ -317,7 +319,7 @@ ip_set_get_ipaddr6(struct nlattr *nla, union nf_inet_addr *ipaddr) return -IPSET_ERR_PROTOCOL; memcpy(ipaddr, nla_data(tb[IPSET_ATTR_IPADDR_IPV6]), - sizeof(struct in6_addr)); + sizeof(struct in6_addr)); return 0; } EXPORT_SYMBOL_GPL(ip_set_get_ipaddr6); @@ -389,13 +391,22 @@ ip_set_get_extensions(struct ip_set *set, struct nlattr *tb[], struct ip_set_ext *ext) { u64 fullmark; + + if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) + return -IPSET_ERR_PROTOCOL; + if (tb[IPSET_ATTR_TIMEOUT]) { - if (!(set->extensions & IPSET_EXT_TIMEOUT)) + if (!SET_WITH_TIMEOUT(set)) return -IPSET_ERR_TIMEOUT; ext->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); } if (tb[IPSET_ATTR_BYTES] || tb[IPSET_ATTR_PACKETS]) { - if (!(set->extensions & IPSET_EXT_COUNTER)) + if (!SET_WITH_COUNTER(set)) return -IPSET_ERR_COUNTER; if (tb[IPSET_ATTR_BYTES]) ext->bytes = be64_to_cpu(nla_get_be64( @@ -405,25 +416,25 @@ ip_set_get_extensions(struct ip_set *set, struct nlattr *tb[], tb[IPSET_ATTR_PACKETS])); } if (tb[IPSET_ATTR_COMMENT]) { - if (!(set->extensions & IPSET_EXT_COMMENT)) + if (!SET_WITH_COMMENT(set)) return -IPSET_ERR_COMMENT; ext->comment = ip_set_comment_uget(tb[IPSET_ATTR_COMMENT]); } if (tb[IPSET_ATTR_SKBMARK]) { - if (!(set->extensions & IPSET_EXT_SKBINFO)) + if (!SET_WITH_SKBINFO(set)) return -IPSET_ERR_SKBINFO; fullmark = be64_to_cpu(nla_get_be64(tb[IPSET_ATTR_SKBMARK])); ext->skbmark = fullmark >> 32; ext->skbmarkmask = fullmark & 0xffffffff; } if (tb[IPSET_ATTR_SKBPRIO]) { - if (!(set->extensions & IPSET_EXT_SKBINFO)) + if (!SET_WITH_SKBINFO(set)) return -IPSET_ERR_SKBINFO; ext->skbprio = be32_to_cpu(nla_get_be32( tb[IPSET_ATTR_SKBPRIO])); } if (tb[IPSET_ATTR_SKBQUEUE]) { - if (!(set->extensions & IPSET_EXT_SKBINFO)) + if (!SET_WITH_SKBINFO(set)) return -IPSET_ERR_SKBINFO; ext->skbqueue = be16_to_cpu(nla_get_be16( tb[IPSET_ATTR_SKBQUEUE])); @@ -457,8 +468,7 @@ ip_set_put_extensions(struct sk_buff *skb, const struct ip_set *set, } EXPORT_SYMBOL_GPL(ip_set_put_extensions); -/* - * Creating/destroying/renaming/swapping affect the existence and +/* Creating/destroying/renaming/swapping affect the existence and * the properties of a set. All of these can be executed from userspace * only and serialized by the nfnl mutex indirectly from nfnetlink. * @@ -485,8 +495,7 @@ __ip_set_put(struct ip_set *set) write_unlock_bh(&ip_set_ref_lock); } -/* - * Add, del and test set entries from kernel. +/* Add, del and test set entries from kernel. * * The set behind the index must exist and must be referenced * so it can't be destroyed (or changed) under our foot. @@ -514,23 +523,23 @@ ip_set_test(ip_set_id_t index, const struct sk_buff *skb, dev_net(par->in ? par->in : par->out), index); int ret = 0; - BUG_ON(set == NULL); + BUG_ON(!set); pr_debug("set %s, index %u\n", set->name, index); if (opt->dim < set->type->dimension || !(opt->family == set->family || set->family == NFPROTO_UNSPEC)) return 0; - read_lock_bh(&set->lock); + rcu_read_lock_bh(); ret = set->variant->kadt(set, skb, par, IPSET_TEST, opt); - read_unlock_bh(&set->lock); + rcu_read_unlock_bh(); if (ret == -EAGAIN) { /* Type requests element to be completed */ pr_debug("element must be completed, ADD is triggered\n"); - write_lock_bh(&set->lock); + spin_lock_bh(&set->lock); set->variant->kadt(set, skb, par, IPSET_ADD, opt); - write_unlock_bh(&set->lock); + spin_unlock_bh(&set->lock); ret = 1; } else { /* --return-nomatch: invert matched element */ @@ -553,16 +562,16 @@ ip_set_add(ip_set_id_t index, const struct sk_buff *skb, dev_net(par->in ? par->in : par->out), index); int ret; - BUG_ON(set == NULL); + BUG_ON(!set); pr_debug("set %s, index %u\n", set->name, index); if (opt->dim < set->type->dimension || !(opt->family == set->family || set->family == NFPROTO_UNSPEC)) return -IPSET_ERR_TYPE_MISMATCH; - write_lock_bh(&set->lock); + spin_lock_bh(&set->lock); ret = set->variant->kadt(set, skb, par, IPSET_ADD, opt); - write_unlock_bh(&set->lock); + spin_unlock_bh(&set->lock); return ret; } @@ -576,23 +585,22 @@ ip_set_del(ip_set_id_t index, const struct sk_buff *skb, dev_net(par->in ? par->in : par->out), index); int ret = 0; - BUG_ON(set == NULL); + BUG_ON(!set); pr_debug("set %s, index %u\n", set->name, index); if (opt->dim < set->type->dimension || !(opt->family == set->family || set->family == NFPROTO_UNSPEC)) return -IPSET_ERR_TYPE_MISMATCH; - write_lock_bh(&set->lock); + spin_lock_bh(&set->lock); ret = set->variant->kadt(set, skb, par, IPSET_DEL, opt); - write_unlock_bh(&set->lock); + spin_unlock_bh(&set->lock); return ret; } EXPORT_SYMBOL_GPL(ip_set_del); -/* - * Find set by name, reference it once. The reference makes sure the +/* Find set by name, reference it once. The reference makes sure the * thing pointed to, does not go away under our feet. * */ @@ -606,7 +614,7 @@ ip_set_get_byname(struct net *net, const char *name, struct ip_set **set) rcu_read_lock(); for (i = 0; i < inst->ip_set_max; i++) { s = rcu_dereference(inst->ip_set_list)[i]; - if (s != NULL && STRNCMP(s->name, name)) { + if (s && STRNCMP(s->name, name)) { __ip_set_get(s); index = i; *set = s; @@ -619,8 +627,7 @@ ip_set_get_byname(struct net *net, const char *name, struct ip_set **set) } EXPORT_SYMBOL_GPL(ip_set_get_byname); -/* - * If the given set pointer points to a valid set, decrement +/* If the given set pointer points to a valid set, decrement * reference count by 1. The caller shall not assume the index * to be valid, after calling this function. * @@ -633,7 +640,7 @@ __ip_set_put_byindex(struct ip_set_net *inst, ip_set_id_t index) rcu_read_lock(); set = rcu_dereference(inst->ip_set_list)[index]; - if (set != NULL) + if (set) __ip_set_put(set); rcu_read_unlock(); } @@ -647,8 +654,7 @@ ip_set_put_byindex(struct net *net, ip_set_id_t index) } EXPORT_SYMBOL_GPL(ip_set_put_byindex); -/* - * Get the name of a set behind a set index. +/* Get the name of a set behind a set index. * We assume the set is referenced, so it does exist and * can't be destroyed. The set cannot be renamed due to * the referencing either. @@ -659,7 +665,7 @@ ip_set_name_byindex(struct net *net, ip_set_id_t index) { const struct ip_set *set = ip_set_rcu_get(net, index); - BUG_ON(set == NULL); + BUG_ON(!set); BUG_ON(set->ref == 0); /* Referenced, so it's safe */ @@ -667,13 +673,11 @@ ip_set_name_byindex(struct net *net, ip_set_id_t index) } EXPORT_SYMBOL_GPL(ip_set_name_byindex); -/* - * Routines to call by external subsystems, which do not +/* Routines to call by external subsystems, which do not * call nfnl_lock for us. */ -/* - * Find set by index, reference it once. The reference makes sure the +/* Find set by index, reference it once. The reference makes sure the * thing pointed to, does not go away under our feet. * * The nfnl mutex is used in the function. @@ -699,8 +703,7 @@ ip_set_nfnl_get_byindex(struct net *net, ip_set_id_t index) } EXPORT_SYMBOL_GPL(ip_set_nfnl_get_byindex); -/* - * If the given set pointer points to a valid set, decrement +/* If the given set pointer points to a valid set, decrement * reference count by 1. The caller shall not assume the index * to be valid, after calling this function. * @@ -715,15 +718,14 @@ ip_set_nfnl_put(struct net *net, ip_set_id_t index) nfnl_lock(NFNL_SUBSYS_IPSET); if (!inst->is_deleted) { /* already deleted from ip_set_net_exit() */ set = ip_set(inst, index); - if (set != NULL) + if (set) __ip_set_put(set); } nfnl_unlock(NFNL_SUBSYS_IPSET); } EXPORT_SYMBOL_GPL(ip_set_nfnl_put); -/* - * Communication protocol with userspace over netlink. +/* Communication protocol with userspace over netlink. * * The commands are serialized by the nfnl mutex. */ @@ -750,7 +752,7 @@ start_msg(struct sk_buff *skb, u32 portid, u32 seq, unsigned int flags, nlh = nlmsg_put(skb, portid, seq, cmd | (NFNL_SUBSYS_IPSET << 8), sizeof(*nfmsg), flags); - if (nlh == NULL) + if (!nlh) return NULL; nfmsg = nlmsg_data(nlh); @@ -783,7 +785,7 @@ find_set_and_id(struct ip_set_net *inst, const char *name, ip_set_id_t *id) *id = IPSET_INVALID_ID; for (i = 0; i < inst->ip_set_max; i++) { set = ip_set(inst, i); - if (set != NULL && STRNCMP(set->name, name)) { + if (set && STRNCMP(set->name, name)) { *id = i; break; } @@ -809,7 +811,7 @@ find_free_id(struct ip_set_net *inst, const char *name, ip_set_id_t *index, *index = IPSET_INVALID_ID; for (i = 0; i < inst->ip_set_max; i++) { s = ip_set(inst, i); - if (s == NULL) { + if (!s) { if (*index == IPSET_INVALID_ID) *index = i; } else if (STRNCMP(name, s->name)) { @@ -841,18 +843,18 @@ ip_set_create(struct sock *ctnl, struct sk_buff *skb, struct ip_set_net *inst = ip_set_pernet(net); struct ip_set *set, *clash = NULL; ip_set_id_t index = IPSET_INVALID_ID; - struct nlattr *tb[IPSET_ATTR_CREATE_MAX+1] = {}; + struct nlattr *tb[IPSET_ATTR_CREATE_MAX + 1] = {}; const char *name, *typename; u8 family, revision; u32 flags = flag_exist(nlh); int ret = 0; if (unlikely(protocol_failed(attr) || - attr[IPSET_ATTR_SETNAME] == NULL || - attr[IPSET_ATTR_TYPENAME] == NULL || - attr[IPSET_ATTR_REVISION] == NULL || - attr[IPSET_ATTR_FAMILY] == NULL || - (attr[IPSET_ATTR_DATA] != NULL && + !attr[IPSET_ATTR_SETNAME] || + !attr[IPSET_ATTR_TYPENAME] || + !attr[IPSET_ATTR_REVISION] || + !attr[IPSET_ATTR_FAMILY] || + (attr[IPSET_ATTR_DATA] && !flag_nested(attr[IPSET_ATTR_DATA])))) return -IPSET_ERR_PROTOCOL; @@ -863,33 +865,29 @@ ip_set_create(struct sock *ctnl, struct sk_buff *skb, pr_debug("setname: %s, typename: %s, family: %s, revision: %u\n", name, typename, family_name(family), revision); - /* - * First, and without any locks, allocate and initialize + /* First, and without any locks, allocate and initialize * a normal base set structure. */ - set = kzalloc(sizeof(struct ip_set), GFP_KERNEL); + set = kzalloc(sizeof(*set), GFP_KERNEL); if (!set) return -ENOMEM; - rwlock_init(&set->lock); + spin_lock_init(&set->lock); strlcpy(set->name, name, IPSET_MAXNAMELEN); set->family = family; set->revision = revision; - /* - * Next, check that we know the type, and take + /* Next, check that we know the type, and take * a reference on the type, to make sure it stays available * while constructing our new set. * * After referencing the type, we try to create the type * specific part of the set without holding any locks. */ - ret = find_set_type_get(typename, family, revision, &(set->type)); + ret = find_set_type_get(typename, family, revision, &set->type); if (ret) goto out; - /* - * Without holding any locks, create private part. - */ + /* Without holding any locks, create private part. */ if (attr[IPSET_ATTR_DATA] && nla_parse_nested(tb, IPSET_ATTR_CREATE_MAX, attr[IPSET_ATTR_DATA], set->type->create_policy)) { @@ -903,8 +901,7 @@ ip_set_create(struct sock *ctnl, struct sk_buff *skb, /* BTW, ret==0 here. */ - /* - * Here, we have a valid, constructed set and we are protected + /* Here, we have a valid, constructed set and we are protected * by the nfnl mutex. Find the first free index in ip_set_list * and check clashing. */ @@ -927,7 +924,7 @@ ip_set_create(struct sock *ctnl, struct sk_buff *skb, /* Wraparound */ goto cleanup; - list = kzalloc(sizeof(struct ip_set *) * i, GFP_KERNEL); + list = kcalloc(i, sizeof(struct ip_set *), GFP_KERNEL); if (!list) goto cleanup; /* nfnl mutex is held, both lists are valid */ @@ -941,12 +938,11 @@ ip_set_create(struct sock *ctnl, struct sk_buff *skb, inst->ip_set_max = i; kfree(tmp); ret = 0; - } else if (ret) + } else if (ret) { goto cleanup; + } - /* - * Finally! Add our shiny new set to the list, and be done. - */ + /* Finally! Add our shiny new set to the list, and be done. */ pr_debug("create: '%s' created with index %u!\n", set->name, index); ip_set(inst, index) = set; @@ -971,12 +967,9 @@ ip_set_setname_policy[IPSET_ATTR_CMD_MAX + 1] = { }; static void -ip_set_destroy_set(struct ip_set_net *inst, ip_set_id_t index) +ip_set_destroy_set(struct ip_set *set) { - struct ip_set *set = ip_set(inst, index); - pr_debug("set: %s\n", set->name); - ip_set(inst, index) = NULL; /* Must call it without holding any lock */ set->variant->destroy(set); @@ -1011,30 +1004,36 @@ ip_set_destroy(struct sock *ctnl, struct sk_buff *skb, if (!attr[IPSET_ATTR_SETNAME]) { for (i = 0; i < inst->ip_set_max; i++) { s = ip_set(inst, i); - if (s != NULL && s->ref) { + if (s && s->ref) { ret = -IPSET_ERR_BUSY; goto out; } } + inst->is_destroyed = true; read_unlock_bh(&ip_set_ref_lock); for (i = 0; i < inst->ip_set_max; i++) { s = ip_set(inst, i); - if (s != NULL) - ip_set_destroy_set(inst, i); + if (s) { + ip_set(inst, i) = NULL; + ip_set_destroy_set(s); + } } + /* Modified by ip_set_destroy() only, which is serialized */ + inst->is_destroyed = false; } else { s = find_set_and_id(inst, nla_data(attr[IPSET_ATTR_SETNAME]), &i); - if (s == NULL) { + if (!s) { ret = -ENOENT; goto out; } else if (s->ref) { ret = -IPSET_ERR_BUSY; goto out; } + ip_set(inst, i) = NULL; read_unlock_bh(&ip_set_ref_lock); - ip_set_destroy_set(inst, i); + ip_set_destroy_set(s); } return 0; out: @@ -1049,9 +1048,9 @@ ip_set_flush_set(struct ip_set *set) { pr_debug("set: %s\n", set->name); - write_lock_bh(&set->lock); + spin_lock_bh(&set->lock); set->variant->flush(set); - write_unlock_bh(&set->lock); + spin_unlock_bh(&set->lock); } static int @@ -1069,12 +1068,12 @@ ip_set_flush(struct sock *ctnl, struct sk_buff *skb, if (!attr[IPSET_ATTR_SETNAME]) { for (i = 0; i < inst->ip_set_max; i++) { s = ip_set(inst, i); - if (s != NULL) + if (s) ip_set_flush_set(s); } } else { s = find_set(inst, nla_data(attr[IPSET_ATTR_SETNAME])); - if (s == NULL) + if (!s) return -ENOENT; ip_set_flush_set(s); @@ -1106,12 +1105,12 @@ ip_set_rename(struct sock *ctnl, struct sk_buff *skb, int ret = 0; if (unlikely(protocol_failed(attr) || - attr[IPSET_ATTR_SETNAME] == NULL || - attr[IPSET_ATTR_SETNAME2] == NULL)) + !attr[IPSET_ATTR_SETNAME] || + !attr[IPSET_ATTR_SETNAME2])) return -IPSET_ERR_PROTOCOL; set = find_set(inst, nla_data(attr[IPSET_ATTR_SETNAME])); - if (set == NULL) + if (!set) return -ENOENT; read_lock_bh(&ip_set_ref_lock); @@ -1123,7 +1122,7 @@ ip_set_rename(struct sock *ctnl, struct sk_buff *skb, name2 = nla_data(attr[IPSET_ATTR_SETNAME2]); for (i = 0; i < inst->ip_set_max; i++) { s = ip_set(inst, i); - if (s != NULL && STRNCMP(s->name, name2)) { + if (s && STRNCMP(s->name, name2)) { ret = -IPSET_ERR_EXIST_SETNAME2; goto out; } @@ -1155,23 +1154,24 @@ ip_set_swap(struct sock *ctnl, struct sk_buff *skb, char from_name[IPSET_MAXNAMELEN]; if (unlikely(protocol_failed(attr) || - attr[IPSET_ATTR_SETNAME] == NULL || - attr[IPSET_ATTR_SETNAME2] == NULL)) + !attr[IPSET_ATTR_SETNAME] || + !attr[IPSET_ATTR_SETNAME2])) return -IPSET_ERR_PROTOCOL; from = find_set_and_id(inst, nla_data(attr[IPSET_ATTR_SETNAME]), &from_id); - if (from == NULL) + if (!from) return -ENOENT; to = find_set_and_id(inst, nla_data(attr[IPSET_ATTR_SETNAME2]), &to_id); - if (to == NULL) + if (!to) return -IPSET_ERR_EXIST_SETNAME2; /* Features must not change. - * Not an artificial restriction anymore, as we must prevent - * possible loops created by swapping in setlist type of sets. */ + * Not an artifical restriction anymore, as we must prevent + * possible loops created by swapping in setlist type of sets. + */ if (!(from->type->features == to->type->features && from->family == to->family)) return -IPSET_ERR_TYPE_MISMATCH; @@ -1202,12 +1202,16 @@ ip_set_swap(struct sock *ctnl, struct sk_buff *skb, static int ip_set_dump_done(struct netlink_callback *cb) { - struct ip_set_net *inst = (struct ip_set_net *)cb->args[IPSET_CB_NET]; if (cb->args[IPSET_CB_ARG0]) { - pr_debug("release set %s\n", - ip_set(inst, cb->args[IPSET_CB_INDEX])->name); - __ip_set_put_byindex(inst, - (ip_set_id_t) cb->args[IPSET_CB_INDEX]); + struct ip_set_net *inst = + (struct ip_set_net *)cb->args[IPSET_CB_NET]; + ip_set_id_t index = (ip_set_id_t)cb->args[IPSET_CB_INDEX]; + struct ip_set *set = ip_set(inst, index); + + if (set->variant->uref) + set->variant->uref(set, cb, false); + pr_debug("release set %s\n", set->name); + __ip_set_put_byindex(inst, index); } return 0; } @@ -1229,7 +1233,7 @@ dump_init(struct netlink_callback *cb, struct ip_set_net *inst) { struct nlmsghdr *nlh = nlmsg_hdr(cb->skb); int min_len = nlmsg_total_size(sizeof(struct nfgenmsg)); - struct nlattr *cda[IPSET_ATTR_CMD_MAX+1]; + struct nlattr *cda[IPSET_ATTR_CMD_MAX + 1]; struct nlattr *attr = (void *)nlh + min_len; u32 dump_type; ip_set_id_t index; @@ -1238,27 +1242,23 @@ dump_init(struct netlink_callback *cb, struct ip_set_net *inst) nla_parse(cda, IPSET_ATTR_CMD_MAX, attr, nlh->nlmsg_len - min_len, ip_set_setname_policy); - /* cb->args[IPSET_CB_NET]: net namespace - * [IPSET_CB_DUMP]: dump single set/all sets - * [IPSET_CB_INDEX]: set index - * [IPSET_CB_ARG0]: type specific - */ - if (cda[IPSET_ATTR_SETNAME]) { struct ip_set *set; set = find_set_and_id(inst, nla_data(cda[IPSET_ATTR_SETNAME]), &index); - if (set == NULL) + if (!set) return -ENOENT; dump_type = DUMP_ONE; cb->args[IPSET_CB_INDEX] = index; - } else + } else { dump_type = DUMP_ALL; + } if (cda[IPSET_ATTR_FLAGS]) { u32 f = ip_set_get_h32(cda[IPSET_ATTR_FLAGS]); + dump_type |= (f << 16); } cb->args[IPSET_CB_NET] = (unsigned long)inst; @@ -1276,6 +1276,7 @@ ip_set_dump_start(struct sk_buff *skb, struct netlink_callback *cb) unsigned int flags = NETLINK_CB(cb->skb).portid ? NLM_F_MULTI : 0; struct ip_set_net *inst = ip_set_pernet(sock_net(skb->sk)); u32 dump_type, dump_flags; + bool is_destroyed; int ret = 0; if (!cb->args[IPSET_CB_DUMP]) { @@ -1283,7 +1284,8 @@ ip_set_dump_start(struct sk_buff *skb, struct netlink_callback *cb) if (ret < 0) { nlh = nlmsg_hdr(cb->skb); /* We have to create and send the error message - * manually :-( */ + * manually :-( + */ if (nlh->nlmsg_flags & NLM_F_ACK) netlink_ack(cb->skb, nlh, ret); return ret; @@ -1301,13 +1303,21 @@ dump_last: pr_debug("dump type, flag: %u %u index: %ld\n", dump_type, dump_flags, cb->args[IPSET_CB_INDEX]); for (; cb->args[IPSET_CB_INDEX] < max; cb->args[IPSET_CB_INDEX]++) { - index = (ip_set_id_t) cb->args[IPSET_CB_INDEX]; + index = (ip_set_id_t)cb->args[IPSET_CB_INDEX]; + write_lock_bh(&ip_set_ref_lock); set = ip_set(inst, index); - if (set == NULL) { + is_destroyed = inst->is_destroyed; + if (!set || is_destroyed) { + write_unlock_bh(&ip_set_ref_lock); if (dump_type == DUMP_ONE) { ret = -ENOENT; goto out; } + if (is_destroyed) { + /* All sets are just being destroyed */ + ret = 0; + goto out; + } continue; } /* When dumping all sets, we must dump "sorted" @@ -1315,14 +1325,17 @@ dump_last: */ if (dump_type != DUMP_ONE && ((dump_type == DUMP_ALL) == - !!(set->type->features & IPSET_DUMP_LAST))) + !!(set->type->features & IPSET_DUMP_LAST))) { + write_unlock_bh(&ip_set_ref_lock); continue; + } pr_debug("List set: %s\n", set->name); if (!cb->args[IPSET_CB_ARG0]) { /* Start listing: make sure set won't be destroyed */ pr_debug("reference set\n"); - __ip_set_get(set); + set->ref++; } + write_unlock_bh(&ip_set_ref_lock); nlh = start_msg(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, flags, IPSET_CMD_LIST); @@ -1350,11 +1363,13 @@ dump_last: goto release_refcount; if (dump_flags & IPSET_FLAG_LIST_HEADER) goto next_set; + if (set->variant->uref) + set->variant->uref(set, cb, true); /* Fall through and add elements */ default: - read_lock_bh(&set->lock); + rcu_read_lock_bh(); ret = set->variant->list(set, skb, cb); - read_unlock_bh(&set->lock); + rcu_read_unlock_bh(); if (!cb->args[IPSET_CB_ARG0]) /* Set is done, proceed with next one */ goto next_set; @@ -1366,6 +1381,8 @@ dump_last: dump_type = DUMP_LAST; cb->args[IPSET_CB_DUMP] = dump_type | (dump_flags << 16); cb->args[IPSET_CB_INDEX] = 0; + if (set && set->variant->uref) + set->variant->uref(set, cb, false); goto dump_last; } goto out; @@ -1380,7 +1397,10 @@ next_set: release_refcount: /* If there was an error or set is done, release set */ if (ret || !cb->args[IPSET_CB_ARG0]) { - pr_debug("release set %s\n", ip_set(inst, index)->name); + set = ip_set(inst, index); + if (set->variant->uref) + set->variant->uref(set, cb, false); + pr_debug("release set %s\n", set->name); __ip_set_put_byindex(inst, index); cb->args[IPSET_CB_ARG0] = 0; } @@ -1432,9 +1452,9 @@ call_ad(struct sock *ctnl, struct sk_buff *skb, struct ip_set *set, bool eexist = flags & IPSET_FLAG_EXIST, retried = false; do { - write_lock_bh(&set->lock); + spin_lock_bh(&set->lock); ret = set->variant->uadt(set, tb, adt, &lineno, flags, retried); - write_unlock_bh(&set->lock); + spin_unlock_bh(&set->lock); retried = true; } while (ret == -EAGAIN && set->variant->resize && @@ -1450,12 +1470,12 @@ call_ad(struct sock *ctnl, struct sk_buff *skb, struct ip_set *set, size_t payload = min(SIZE_MAX, sizeof(*errmsg) + nlmsg_len(nlh)); int min_len = nlmsg_total_size(sizeof(struct nfgenmsg)); - struct nlattr *cda[IPSET_ATTR_CMD_MAX+1]; + struct nlattr *cda[IPSET_ATTR_CMD_MAX + 1]; struct nlattr *cmdattr; u32 *errline; skb2 = nlmsg_new(payload, GFP_KERNEL); - if (skb2 == NULL) + if (!skb2) return -ENOMEM; rep = __nlmsg_put(skb2, NETLINK_CB(skb).portid, nlh->nlmsg_seq, NLMSG_ERROR, payload, 0); @@ -1472,7 +1492,8 @@ call_ad(struct sock *ctnl, struct sk_buff *skb, struct ip_set *set, *errline = lineno; - netlink_unicast(ctnl, skb2, NETLINK_CB(skb).portid, MSG_DONTWAIT); + netlink_unicast(ctnl, skb2, NETLINK_CB(skb).portid, + MSG_DONTWAIT); /* Signal netlink not to send its ACK/errmsg. */ return -EINTR; } @@ -1487,25 +1508,25 @@ ip_set_uadd(struct sock *ctnl, struct sk_buff *skb, { struct ip_set_net *inst = ip_set_pernet(sock_net(ctnl)); struct ip_set *set; - struct nlattr *tb[IPSET_ATTR_ADT_MAX+1] = {}; + struct nlattr *tb[IPSET_ATTR_ADT_MAX + 1] = {}; const struct nlattr *nla; u32 flags = flag_exist(nlh); bool use_lineno; int ret = 0; if (unlikely(protocol_failed(attr) || - attr[IPSET_ATTR_SETNAME] == NULL || + !attr[IPSET_ATTR_SETNAME] || !((attr[IPSET_ATTR_DATA] != NULL) ^ (attr[IPSET_ATTR_ADT] != NULL)) || - (attr[IPSET_ATTR_DATA] != NULL && + (attr[IPSET_ATTR_DATA] && !flag_nested(attr[IPSET_ATTR_DATA])) || - (attr[IPSET_ATTR_ADT] != NULL && + (attr[IPSET_ATTR_ADT] && (!flag_nested(attr[IPSET_ATTR_ADT]) || - attr[IPSET_ATTR_LINENO] == NULL)))) + !attr[IPSET_ATTR_LINENO])))) return -IPSET_ERR_PROTOCOL; set = find_set(inst, nla_data(attr[IPSET_ATTR_SETNAME])); - if (set == NULL) + if (!set) return -ENOENT; use_lineno = !!attr[IPSET_ATTR_LINENO]; @@ -1542,25 +1563,25 @@ ip_set_udel(struct sock *ctnl, struct sk_buff *skb, { struct ip_set_net *inst = ip_set_pernet(sock_net(ctnl)); struct ip_set *set; - struct nlattr *tb[IPSET_ATTR_ADT_MAX+1] = {}; + struct nlattr *tb[IPSET_ATTR_ADT_MAX + 1] = {}; const struct nlattr *nla; u32 flags = flag_exist(nlh); bool use_lineno; int ret = 0; if (unlikely(protocol_failed(attr) || - attr[IPSET_ATTR_SETNAME] == NULL || + !attr[IPSET_ATTR_SETNAME] || !((attr[IPSET_ATTR_DATA] != NULL) ^ (attr[IPSET_ATTR_ADT] != NULL)) || - (attr[IPSET_ATTR_DATA] != NULL && + (attr[IPSET_ATTR_DATA] && !flag_nested(attr[IPSET_ATTR_DATA])) || - (attr[IPSET_ATTR_ADT] != NULL && + (attr[IPSET_ATTR_ADT] && (!flag_nested(attr[IPSET_ATTR_ADT]) || - attr[IPSET_ATTR_LINENO] == NULL)))) + !attr[IPSET_ATTR_LINENO])))) return -IPSET_ERR_PROTOCOL; set = find_set(inst, nla_data(attr[IPSET_ATTR_SETNAME])); - if (set == NULL) + if (!set) return -ENOENT; use_lineno = !!attr[IPSET_ATTR_LINENO]; @@ -1597,26 +1618,26 @@ ip_set_utest(struct sock *ctnl, struct sk_buff *skb, { struct ip_set_net *inst = ip_set_pernet(sock_net(ctnl)); struct ip_set *set; - struct nlattr *tb[IPSET_ATTR_ADT_MAX+1] = {}; + struct nlattr *tb[IPSET_ATTR_ADT_MAX + 1] = {}; int ret = 0; if (unlikely(protocol_failed(attr) || - attr[IPSET_ATTR_SETNAME] == NULL || - attr[IPSET_ATTR_DATA] == NULL || + !attr[IPSET_ATTR_SETNAME] || + !attr[IPSET_ATTR_DATA] || !flag_nested(attr[IPSET_ATTR_DATA]))) return -IPSET_ERR_PROTOCOL; set = find_set(inst, nla_data(attr[IPSET_ATTR_SETNAME])); - if (set == NULL) + if (!set) return -ENOENT; if (nla_parse_nested(tb, IPSET_ATTR_ADT_MAX, attr[IPSET_ATTR_DATA], set->type->adt_policy)) return -IPSET_ERR_PROTOCOL; - read_lock_bh(&set->lock); + rcu_read_lock_bh(); ret = set->variant->uadt(set, tb, IPSET_TEST, NULL, 0, 0); - read_unlock_bh(&set->lock); + rcu_read_unlock_bh(); /* Userspace can't trigger element to be re-added */ if (ret == -EAGAIN) ret = 1; @@ -1638,15 +1659,15 @@ ip_set_header(struct sock *ctnl, struct sk_buff *skb, int ret = 0; if (unlikely(protocol_failed(attr) || - attr[IPSET_ATTR_SETNAME] == NULL)) + !attr[IPSET_ATTR_SETNAME])) return -IPSET_ERR_PROTOCOL; set = find_set(inst, nla_data(attr[IPSET_ATTR_SETNAME])); - if (set == NULL) + if (!set) return -ENOENT; skb2 = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); - if (skb2 == NULL) + if (!skb2) return -ENOMEM; nlh2 = start_msg(skb2, NETLINK_CB(skb).portid, nlh->nlmsg_seq, 0, @@ -1695,8 +1716,8 @@ ip_set_type(struct sock *ctnl, struct sk_buff *skb, int ret = 0; if (unlikely(protocol_failed(attr) || - attr[IPSET_ATTR_TYPENAME] == NULL || - attr[IPSET_ATTR_FAMILY] == NULL)) + !attr[IPSET_ATTR_TYPENAME] || + !attr[IPSET_ATTR_FAMILY])) return -IPSET_ERR_PROTOCOL; family = nla_get_u8(attr[IPSET_ATTR_FAMILY]); @@ -1706,7 +1727,7 @@ ip_set_type(struct sock *ctnl, struct sk_buff *skb, return ret; skb2 = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); - if (skb2 == NULL) + if (!skb2) return -ENOMEM; nlh2 = start_msg(skb2, NETLINK_CB(skb).portid, nlh->nlmsg_seq, 0, @@ -1751,11 +1772,11 @@ ip_set_protocol(struct sock *ctnl, struct sk_buff *skb, struct nlmsghdr *nlh2; int ret = 0; - if (unlikely(attr[IPSET_ATTR_PROTOCOL] == NULL)) + if (unlikely(!attr[IPSET_ATTR_PROTOCOL])) return -IPSET_ERR_PROTOCOL; skb2 = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); - if (skb2 == NULL) + if (!skb2) return -ENOMEM; nlh2 = start_msg(skb2, NETLINK_CB(skb).portid, nlh->nlmsg_seq, 0, @@ -1883,7 +1904,7 @@ ip_set_sockfn_get(struct sock *sk, int optval, void __user *user, int *len) ret = -EFAULT; goto done; } - op = (unsigned int *) data; + op = (unsigned int *)data; if (*op < IP_SET_OP_VERSION) { /* Check the version at the beginning of operations */ @@ -1995,10 +2016,11 @@ ip_set_net_init(struct net *net) if (inst->ip_set_max >= IPSET_INVALID_ID) inst->ip_set_max = IPSET_INVALID_ID - 1; - list = kzalloc(sizeof(struct ip_set *) * inst->ip_set_max, GFP_KERNEL); + list = kcalloc(inst->ip_set_max, sizeof(struct ip_set *), GFP_KERNEL); if (!list) return -ENOMEM; - inst->is_deleted = 0; + inst->is_deleted = false; + inst->is_destroyed = false; rcu_assign_pointer(inst->ip_set_list, list); return 0; } @@ -2011,12 +2033,14 @@ ip_set_net_exit(struct net *net) struct ip_set *set = NULL; ip_set_id_t i; - inst->is_deleted = 1; /* flag for ip_set_nfnl_put */ + inst->is_deleted = true; /* flag for ip_set_nfnl_put */ for (i = 0; i < inst->ip_set_max; i++) { set = ip_set(inst, i); - if (set != NULL) - ip_set_destroy_set(inst, i); + if (set) { + ip_set(inst, i) = NULL; + ip_set_destroy_set(set); + } } kfree(rcu_dereference_protected(inst->ip_set_list, 1)); } @@ -2028,11 +2052,11 @@ static struct pernet_operations ip_set_net_ops = { .size = sizeof(struct ip_set_net) }; - static int __init ip_set_init(void) { int ret = nfnetlink_subsys_register(&ip_set_netlink_subsys); + if (ret != 0) { pr_err("ip_set: cannot register with nfnetlink.\n"); return ret; diff --git a/net/netfilter/ipset/ip_set_getport.c b/net/netfilter/ipset/ip_set_getport.c index 1981f021cc60..42c3e3ba1b94 100644 --- a/net/netfilter/ipset/ip_set_getport.c +++ b/net/netfilter/ipset/ip_set_getport.c @@ -30,7 +30,7 @@ get_port(const struct sk_buff *skb, int protocol, unsigned int protooff, const struct tcphdr *th; th = skb_header_pointer(skb, protooff, sizeof(_tcph), &_tcph); - if (th == NULL) + if (!th) /* No choice either */ return false; @@ -42,7 +42,7 @@ get_port(const struct sk_buff *skb, int protocol, unsigned int protooff, const sctp_sctphdr_t *sh; sh = skb_header_pointer(skb, protooff, sizeof(_sh), &_sh); - if (sh == NULL) + if (!sh) /* No choice either */ return false; @@ -55,7 +55,7 @@ get_port(const struct sk_buff *skb, int protocol, unsigned int protooff, const struct udphdr *uh; uh = skb_header_pointer(skb, protooff, sizeof(_udph), &_udph); - if (uh == NULL) + if (!uh) /* No choice either */ return false; @@ -67,7 +67,7 @@ get_port(const struct sk_buff *skb, int protocol, unsigned int protooff, const struct icmphdr *ic; ic = skb_header_pointer(skb, protooff, sizeof(_ich), &_ich); - if (ic == NULL) + if (!ic) return false; *port = (__force __be16)htons((ic->type << 8) | ic->code); @@ -78,7 +78,7 @@ get_port(const struct sk_buff *skb, int protocol, unsigned int protooff, const struct icmp6hdr *ic; ic = skb_header_pointer(skb, protooff, sizeof(_ich), &_ich); - if (ic == NULL) + if (!ic) return false; *port = (__force __be16) @@ -116,7 +116,8 @@ ip_set_get_ip4_port(const struct sk_buff *skb, bool src, return false; default: /* Other protocols doesn't have ports, - so we can match fragments */ + * so we can match fragments. + */ *proto = protocol; return true; } diff --git a/net/netfilter/ipset/ip_set_hash_gen.h b/net/netfilter/ipset/ip_set_hash_gen.h index 7952869c8023..afe905c208af 100644 --- a/net/netfilter/ipset/ip_set_hash_gen.h +++ b/net/netfilter/ipset/ip_set_hash_gen.h @@ -10,19 +10,19 @@ #include <linux/rcupdate.h> #include <linux/jhash.h> +#include <linux/types.h> #include <linux/netfilter/ipset/ip_set_timeout.h> -#ifndef rcu_dereference_bh -#define rcu_dereference_bh(p) rcu_dereference(p) -#endif + +#define __ipset_dereference_protected(p, c) rcu_dereference_protected(p, c) +#define ipset_dereference_protected(p, set) \ + __ipset_dereference_protected(p, spin_is_locked(&(set)->lock)) #define rcu_dereference_bh_nfnl(p) rcu_dereference_bh_check(p, 1) /* Hashing which uses arrays to resolve clashing. The hash table is resized * (doubled) when searching becomes too long. * Internally jhash is used with the assumption that the size of the - * stored data is a multiple of sizeof(u32). If storage supports timeout, - * the timeout field must be the last one in the data structure - that field - * is ignored when computing the hash key. + * stored data is a multiple of sizeof(u32). * * Readers and resizing * @@ -35,7 +35,9 @@ /* Number of elements to store in an initial array block */ #define AHASH_INIT_SIZE 4 /* Max number of elements to store in an array block */ -#define AHASH_MAX_SIZE (3*AHASH_INIT_SIZE) +#define AHASH_MAX_SIZE (3 * AHASH_INIT_SIZE) +/* Max muber of elements in the array block when tuned */ +#define AHASH_MAX_TUNED 64 /* Max number of elements can be tuned */ #ifdef IP_SET_HASH_WITH_MULTI @@ -53,8 +55,9 @@ tune_ahash_max(u8 curr, u32 multi) /* Currently, at listing one hash bucket must fit into a message. * Therefore we have a hard limit here. */ - return n > curr && n <= 64 ? n : curr; + return n > curr && n <= AHASH_MAX_TUNED ? n : curr; } + #define TUNE_AHASH_MAX(h, multi) \ ((h)->ahash_max = tune_ahash_max((h)->ahash_max, multi)) #else @@ -64,18 +67,23 @@ tune_ahash_max(u8 curr, u32 multi) /* A hash bucket */ struct hbucket { - void *value; /* the array of the values */ + struct rcu_head rcu; /* for call_rcu_bh */ + /* Which positions are used in the array */ + DECLARE_BITMAP(used, AHASH_MAX_TUNED); u8 size; /* size of the array */ u8 pos; /* position of the first free entry */ -}; + unsigned char value[0]; /* the array of the values */ +} __attribute__ ((aligned)); /* The hash table: the table size stored here in order to make resizing easy */ struct htable { + atomic_t ref; /* References for resizing */ + atomic_t uref; /* References for dumping */ u8 htable_bits; /* size of hash table == 2^htable_bits */ - struct hbucket bucket[0]; /* hashtable buckets */ + struct hbucket __rcu *bucket[0]; /* hashtable buckets */ }; -#define hbucket(h, i) (&((h)->bucket[i])) +#define hbucket(h, i) ((h)->bucket[i]) #ifndef IPSET_NET_COUNT #define IPSET_NET_COUNT 1 @@ -83,8 +91,8 @@ struct htable { /* Book-keeping of the prefixes added to the set */ struct net_prefixes { - u32 nets[IPSET_NET_COUNT]; /* number of elements per cidr */ - u8 cidr[IPSET_NET_COUNT]; /* the different cidr values in the set */ + u32 nets[IPSET_NET_COUNT]; /* number of elements for this cidr */ + u8 cidr[IPSET_NET_COUNT]; /* the cidr value */ }; /* Compute the hash table size */ @@ -97,11 +105,11 @@ htable_size(u8 hbits) if (hbits > 31) return 0; hsize = jhash_size(hbits); - if ((((size_t)-1) - sizeof(struct htable))/sizeof(struct hbucket) + if ((((size_t)-1) - sizeof(struct htable)) / sizeof(struct hbucket *) < hsize) return 0; - return hsize * sizeof(struct hbucket) + sizeof(struct htable); + return hsize * sizeof(struct hbucket *) + sizeof(struct htable); } /* Compute htable_bits from the user input parameter hashsize */ @@ -110,6 +118,7 @@ htable_bits(u32 hashsize) { /* Assume that hashsize == 2^htable_bits */ u8 bits = fls(hashsize - 1); + if (jhash_size(bits) != hashsize) /* Round up to the first 2^n value */ bits = fls(hashsize); @@ -117,30 +126,6 @@ htable_bits(u32 hashsize) return bits; } -static int -hbucket_elem_add(struct hbucket *n, u8 ahash_max, size_t dsize) -{ - if (n->pos >= n->size) { - void *tmp; - - if (n->size >= ahash_max) - /* Trigger rehashing */ - return -EAGAIN; - - tmp = kzalloc((n->size + AHASH_INIT_SIZE) * dsize, - GFP_ATOMIC); - if (!tmp) - return -ENOMEM; - if (n->size) { - memcpy(tmp, n->value, n->size * dsize); - kfree(n->value); - } - n->value = tmp; - n->size += AHASH_INIT_SIZE; - } - return 0; -} - #ifdef IP_SET_HASH_WITH_NETS #if IPSET_NET_COUNT > 1 #define __CIDR(cidr, i) (cidr[i]) @@ -149,17 +134,21 @@ hbucket_elem_add(struct hbucket *n, u8 ahash_max, size_t dsize) #endif /* cidr + 1 is stored in net_prefixes to support /0 */ -#define SCIDR(cidr, i) (__CIDR(cidr, i) + 1) +#define NCIDR_PUT(cidr) ((cidr) + 1) +#define NCIDR_GET(cidr) ((cidr) - 1) #ifdef IP_SET_HASH_WITH_NETS_PACKED /* When cidr is packed with nomatch, cidr - 1 is stored in the data entry */ -#define GCIDR(cidr, i) (__CIDR(cidr, i) + 1) -#define NCIDR(cidr) (cidr) +#define DCIDR_PUT(cidr) ((cidr) - 1) +#define DCIDR_GET(cidr, i) (__CIDR(cidr, i) + 1) #else -#define GCIDR(cidr, i) (__CIDR(cidr, i)) -#define NCIDR(cidr) (cidr - 1) +#define DCIDR_PUT(cidr) (cidr) +#define DCIDR_GET(cidr, i) __CIDR(cidr, i) #endif +#define INIT_CIDR(cidr, host_mask) \ + DCIDR_PUT(((cidr) ? NCIDR_GET(cidr) : host_mask)) + #define SET_HOST_MASK(family) (family == AF_INET ? 32 : 128) #ifdef IP_SET_HASH_WITH_NET0 @@ -203,6 +192,7 @@ hbucket_elem_add(struct hbucket *n, u8 ahash_max, size_t dsize) #undef mtype_del #undef mtype_test_cidrs #undef mtype_test +#undef mtype_uref #undef mtype_expire #undef mtype_resize #undef mtype_head @@ -244,6 +234,7 @@ hbucket_elem_add(struct hbucket *n, u8 ahash_max, size_t dsize) #define mtype_del IPSET_TOKEN(MTYPE, _del) #define mtype_test_cidrs IPSET_TOKEN(MTYPE, _test_cidrs) #define mtype_test IPSET_TOKEN(MTYPE, _test) +#define mtype_uref IPSET_TOKEN(MTYPE, _uref) #define mtype_expire IPSET_TOKEN(MTYPE, _expire) #define mtype_resize IPSET_TOKEN(MTYPE, _resize) #define mtype_head IPSET_TOKEN(MTYPE, _head) @@ -266,7 +257,7 @@ hbucket_elem_add(struct hbucket *n, u8 ahash_max, size_t dsize) #endif #define HKEY(data, initval, htable_bits) \ -(jhash2((u32 *)(data), HKEY_DATALEN/sizeof(u32), initval) \ +(jhash2((u32 *)(data), HKEY_DATALEN / sizeof(u32), initval) \ & jhash_mask(htable_bits)) #ifndef htype @@ -292,9 +283,6 @@ struct htype { #ifdef IP_SET_HASH_WITH_NETMASK u8 netmask; /* netmask value for subnets to store */ #endif -#ifdef IP_SET_HASH_WITH_RBTREE - struct rb_root rbtree; -#endif #ifdef IP_SET_HASH_WITH_NETS struct net_prefixes nets[0]; /* book-keeping of prefixes */ #endif @@ -303,7 +291,8 @@ struct htype { #ifdef IP_SET_HASH_WITH_NETS /* Network cidr size book keeping when the hash stores different - * sized networks */ + * sized networks. cidr == real cidr + 1 to support /0. + */ static void mtype_add_cidr(struct htype *h, u8 cidr, u8 nets_length, u8 n) { @@ -311,11 +300,11 @@ mtype_add_cidr(struct htype *h, u8 cidr, u8 nets_length, u8 n) /* Add in increasing prefix order, so larger cidr first */ for (i = 0, j = -1; i < nets_length && h->nets[i].cidr[n]; i++) { - if (j != -1) + if (j != -1) { continue; - else if (h->nets[i].cidr[n] < cidr) + } else if (h->nets[i].cidr[n] < cidr) { j = i; - else if (h->nets[i].cidr[n] == cidr) { + } else if (h->nets[i].cidr[n] == cidr) { h->nets[cidr - 1].nets[n]++; return; } @@ -334,15 +323,15 @@ mtype_del_cidr(struct htype *h, u8 cidr, u8 nets_length, u8 n) u8 i, j, net_end = nets_length - 1; for (i = 0; i < nets_length; i++) { - if (h->nets[i].cidr[n] != cidr) - continue; - h->nets[cidr -1].nets[n]--; - if (h->nets[cidr -1].nets[n] > 0) - return; + if (h->nets[i].cidr[n] != cidr) + continue; + h->nets[cidr - 1].nets[n]--; + if (h->nets[cidr - 1].nets[n] > 0) + return; for (j = i; j < net_end && h->nets[j].cidr[n]; j++) - h->nets[j].cidr[n] = h->nets[j + 1].cidr[n]; + h->nets[j].cidr[n] = h->nets[j + 1].cidr[n]; h->nets[j].cidr[n] = 0; - return; + return; } } #endif @@ -353,15 +342,18 @@ mtype_ahash_memsize(const struct htype *h, const struct htable *t, u8 nets_length, size_t dsize) { u32 i; - size_t memsize = sizeof(*h) - + sizeof(*t) + struct hbucket *n; + size_t memsize = sizeof(*h) + sizeof(*t); + #ifdef IP_SET_HASH_WITH_NETS - + sizeof(struct net_prefixes) * nets_length + memsize += sizeof(struct net_prefixes) * nets_length; #endif - + jhash_size(t->htable_bits) * sizeof(struct hbucket); - - for (i = 0; i < jhash_size(t->htable_bits); i++) - memsize += t->bucket[i].size * dsize; + for (i = 0; i < jhash_size(t->htable_bits); i++) { + n = rcu_dereference_bh(hbucket(t, i)); + if (!n) + continue; + memsize += sizeof(struct hbucket) + n->size * dsize; + } return memsize; } @@ -376,7 +368,8 @@ mtype_ext_cleanup(struct ip_set *set, struct hbucket *n) int i; for (i = 0; i < n->pos; i++) - ip_set_ext_destroy(set, ahash_data(n, i, set->dsize)); + if (test_bit(i, n->used)) + ip_set_ext_destroy(set, ahash_data(n, i, set->dsize)); } /* Flush a hash type of set: destroy all elements */ @@ -388,16 +381,16 @@ mtype_flush(struct ip_set *set) struct hbucket *n; u32 i; - t = rcu_dereference_bh_nfnl(h->table); + t = ipset_dereference_protected(h->table, set); for (i = 0; i < jhash_size(t->htable_bits); i++) { - n = hbucket(t, i); - if (n->size) { - if (set->extensions & IPSET_EXT_DESTROY) - mtype_ext_cleanup(set, n); - n->size = n->pos = 0; - /* FIXME: use slab cache */ - kfree(n->value); - } + n = __ipset_dereference_protected(hbucket(t, i), 1); + if (!n) + continue; + if (set->extensions & IPSET_EXT_DESTROY) + mtype_ext_cleanup(set, n); + /* FIXME: use slab cache */ + rcu_assign_pointer(hbucket(t, i), NULL); + kfree_rcu(n, rcu); } #ifdef IP_SET_HASH_WITH_NETS memset(h->nets, 0, sizeof(struct net_prefixes) * NLEN(set->family)); @@ -413,13 +406,13 @@ mtype_ahash_destroy(struct ip_set *set, struct htable *t, bool ext_destroy) u32 i; for (i = 0; i < jhash_size(t->htable_bits); i++) { - n = hbucket(t, i); - if (n->size) { - if (set->extensions & IPSET_EXT_DESTROY && ext_destroy) - mtype_ext_cleanup(set, n); - /* FIXME: use slab cache */ - kfree(n->value); - } + n = __ipset_dereference_protected(hbucket(t, i), 1); + if (!n) + continue; + if (set->extensions & IPSET_EXT_DESTROY && ext_destroy) + mtype_ext_cleanup(set, n); + /* FIXME: use slab cache */ + kfree(n); } ip_set_free(t); @@ -431,13 +424,11 @@ mtype_destroy(struct ip_set *set) { struct htype *h = set->data; - if (set->extensions & IPSET_EXT_TIMEOUT) + if (SET_WITH_TIMEOUT(set)) del_timer_sync(&h->gc); - mtype_ahash_destroy(set, rcu_dereference_bh_nfnl(h->table), true); -#ifdef IP_SET_HASH_WITH_RBTREE - rbtree_destroy(&h->rbtree); -#endif + mtype_ahash_destroy(set, + __ipset_dereference_protected(h->table, 1), true); kfree(h); set->data = NULL; @@ -449,7 +440,7 @@ mtype_gc_init(struct ip_set *set, void (*gc)(unsigned long ul_set)) struct htype *h = set->data; init_timer(&h->gc); - h->gc.data = (unsigned long) set; + h->gc.data = (unsigned long)set; h->gc.function = gc; h->gc.expires = jiffies + IPSET_GC_PERIOD(set->timeout) * HZ; add_timer(&h->gc); @@ -482,61 +473,71 @@ mtype_expire(struct ip_set *set, struct htype *h, u8 nets_length, size_t dsize) struct htable *t; struct hbucket *n; struct mtype_elem *data; - u32 i; - int j; + u32 i, j, d; #ifdef IP_SET_HASH_WITH_NETS u8 k; #endif - rcu_read_lock_bh(); - t = rcu_dereference_bh(h->table); + t = ipset_dereference_protected(h->table, set); for (i = 0; i < jhash_size(t->htable_bits); i++) { - n = hbucket(t, i); - for (j = 0; j < n->pos; j++) { + n = __ipset_dereference_protected(hbucket(t, i), 1); + if (!n) + continue; + for (j = 0, d = 0; j < n->pos; j++) { + if (!test_bit(j, n->used)) { + d++; + continue; + } data = ahash_data(n, j, dsize); if (ip_set_timeout_expired(ext_timeout(data, set))) { pr_debug("expired %u/%u\n", i, j); + clear_bit(j, n->used); + smp_mb__after_atomic(); #ifdef IP_SET_HASH_WITH_NETS for (k = 0; k < IPSET_NET_COUNT; k++) - mtype_del_cidr(h, SCIDR(data->cidr, k), - nets_length, k); + mtype_del_cidr(h, + NCIDR_PUT(DCIDR_GET(data->cidr, + k)), + nets_length, k); #endif ip_set_ext_destroy(set, data); - if (j != n->pos - 1) - /* Not last one */ - memcpy(data, - ahash_data(n, n->pos - 1, dsize), - dsize); - n->pos--; h->elements--; + d++; } } - if (n->pos + AHASH_INIT_SIZE < n->size) { - void *tmp = kzalloc((n->size - AHASH_INIT_SIZE) - * dsize, - GFP_ATOMIC); + if (d >= AHASH_INIT_SIZE) { + struct hbucket *tmp = kzalloc(sizeof(*tmp) + + (n->size - AHASH_INIT_SIZE) * dsize, + GFP_ATOMIC); if (!tmp) /* Still try to delete expired elements */ continue; - n->size -= AHASH_INIT_SIZE; - memcpy(tmp, n->value, n->size * dsize); - kfree(n->value); - n->value = tmp; + tmp->size = n->size - AHASH_INIT_SIZE; + for (j = 0, d = 0; j < n->pos; j++) { + if (!test_bit(j, n->used)) + continue; + data = ahash_data(n, j, dsize); + memcpy(tmp->value + d * dsize, data, dsize); + set_bit(j, tmp->used); + d++; + } + tmp->pos = d; + rcu_assign_pointer(hbucket(t, i), tmp); + kfree_rcu(n, rcu); } } - rcu_read_unlock_bh(); } static void mtype_gc(unsigned long ul_set) { - struct ip_set *set = (struct ip_set *) ul_set; + struct ip_set *set = (struct ip_set *)ul_set; struct htype *h = set->data; pr_debug("called\n"); - write_lock_bh(&set->lock); + spin_lock_bh(&set->lock); mtype_expire(set, h, NLEN(set->family), set->dsize); - write_unlock_bh(&set->lock); + spin_unlock_bh(&set->lock); h->gc.expires = jiffies + IPSET_GC_PERIOD(set->timeout) * HZ; add_timer(&h->gc); @@ -544,93 +545,152 @@ mtype_gc(unsigned long ul_set) /* Resize a hash: create a new hash table with doubling the hashsize * and inserting the elements to it. Repeat until we succeed or - * fail due to memory pressures. */ + * fail due to memory pressures. + */ static int mtype_resize(struct ip_set *set, bool retried) { struct htype *h = set->data; - struct htable *t, *orig = rcu_dereference_bh_nfnl(h->table); - u8 htable_bits = orig->htable_bits; + struct htable *t, *orig; + u8 htable_bits; + size_t dsize = set->dsize; #ifdef IP_SET_HASH_WITH_NETS u8 flags; + struct mtype_elem *tmp; #endif struct mtype_elem *data; struct mtype_elem *d; struct hbucket *n, *m; - u32 i, j; + u32 i, j, key; int ret; - /* Try to cleanup once */ - if (SET_WITH_TIMEOUT(set) && !retried) { - i = h->elements; - write_lock_bh(&set->lock); - mtype_expire(set, set->data, NLEN(set->family), set->dsize); - write_unlock_bh(&set->lock); - if (h->elements < i) - return 0; - } +#ifdef IP_SET_HASH_WITH_NETS + tmp = kmalloc(dsize, GFP_KERNEL); + if (!tmp) + return -ENOMEM; +#endif + rcu_read_lock_bh(); + orig = rcu_dereference_bh_nfnl(h->table); + htable_bits = orig->htable_bits; + rcu_read_unlock_bh(); retry: ret = 0; htable_bits++; - pr_debug("attempt to resize set %s from %u to %u, t %p\n", - set->name, orig->htable_bits, htable_bits, orig); if (!htable_bits) { /* In case we have plenty of memory :-) */ pr_warn("Cannot increase the hashsize of set %s further\n", set->name); - return -IPSET_ERR_HASH_FULL; + ret = -IPSET_ERR_HASH_FULL; + goto out; + } + t = ip_set_alloc(htable_size(htable_bits)); + if (!t) { + ret = -ENOMEM; + goto out; } - t = ip_set_alloc(sizeof(*t) - + jhash_size(htable_bits) * sizeof(struct hbucket)); - if (!t) - return -ENOMEM; t->htable_bits = htable_bits; - read_lock_bh(&set->lock); + spin_lock_bh(&set->lock); + orig = __ipset_dereference_protected(h->table, 1); + /* There can't be another parallel resizing, but dumping is possible */ + atomic_set(&orig->ref, 1); + atomic_inc(&orig->uref); + pr_debug("attempt to resize set %s from %u to %u, t %p\n", + set->name, orig->htable_bits, htable_bits, orig); for (i = 0; i < jhash_size(orig->htable_bits); i++) { - n = hbucket(orig, i); + n = __ipset_dereference_protected(hbucket(orig, i), 1); + if (!n) + continue; for (j = 0; j < n->pos; j++) { - data = ahash_data(n, j, set->dsize); + if (!test_bit(j, n->used)) + continue; + data = ahash_data(n, j, dsize); #ifdef IP_SET_HASH_WITH_NETS + /* We have readers running parallel with us, + * so the live data cannot be modified. + */ flags = 0; + memcpy(tmp, data, dsize); + data = tmp; mtype_data_reset_flags(data, &flags); #endif - m = hbucket(t, HKEY(data, h->initval, htable_bits)); - ret = hbucket_elem_add(m, AHASH_MAX(h), set->dsize); - if (ret < 0) { -#ifdef IP_SET_HASH_WITH_NETS - mtype_data_reset_flags(data, &flags); -#endif - read_unlock_bh(&set->lock); - mtype_ahash_destroy(set, t, false); - if (ret == -EAGAIN) - goto retry; - return ret; + key = HKEY(data, h->initval, htable_bits); + m = __ipset_dereference_protected(hbucket(t, key), 1); + if (!m) { + m = kzalloc(sizeof(*m) + + AHASH_INIT_SIZE * dsize, + GFP_ATOMIC); + if (!m) { + ret = -ENOMEM; + goto cleanup; + } + m->size = AHASH_INIT_SIZE; + RCU_INIT_POINTER(hbucket(t, key), m); + } else if (m->pos >= m->size) { + struct hbucket *ht; + + if (m->size >= AHASH_MAX(h)) { + ret = -EAGAIN; + } else { + ht = kzalloc(sizeof(*ht) + + (m->size + AHASH_INIT_SIZE) + * dsize, + GFP_ATOMIC); + if (!ht) + ret = -ENOMEM; + } + if (ret < 0) + goto cleanup; + memcpy(ht, m, sizeof(struct hbucket) + + m->size * dsize); + ht->size = m->size + AHASH_INIT_SIZE; + kfree(m); + m = ht; + RCU_INIT_POINTER(hbucket(t, key), ht); } - d = ahash_data(m, m->pos++, set->dsize); - memcpy(d, data, set->dsize); + d = ahash_data(m, m->pos, dsize); + memcpy(d, data, dsize); + set_bit(m->pos++, m->used); #ifdef IP_SET_HASH_WITH_NETS mtype_data_reset_flags(d, &flags); #endif } } - rcu_assign_pointer(h->table, t); - read_unlock_bh(&set->lock); + + spin_unlock_bh(&set->lock); /* Give time to other readers of the set */ synchronize_rcu_bh(); pr_debug("set %s resized from %u (%p) to %u (%p)\n", set->name, orig->htable_bits, orig, t->htable_bits, t); - mtype_ahash_destroy(set, orig, false); + /* If there's nobody else dumping the table, destroy it */ + if (atomic_dec_and_test(&orig->uref)) { + pr_debug("Table destroy by resize %p\n", orig); + mtype_ahash_destroy(set, orig, false); + } - return 0; +out: +#ifdef IP_SET_HASH_WITH_NETS + kfree(tmp); +#endif + return ret; + +cleanup: + atomic_set(&orig->ref, 0); + atomic_dec(&orig->uref); + spin_unlock_bh(&set->lock); + mtype_ahash_destroy(set, t, false); + if (ret == -EAGAIN) + goto retry; + goto out; } /* Add an element to a hash and update the internal counters when succeeded, - * otherwise report the proper error code. */ + * otherwise report the proper error code. + */ static int mtype_add(struct ip_set *set, void *value, const struct ip_set_ext *ext, struct ip_set_ext *mext, u32 flags) @@ -639,17 +699,49 @@ mtype_add(struct ip_set *set, void *value, const struct ip_set_ext *ext, struct htable *t; const struct mtype_elem *d = value; struct mtype_elem *data; - struct hbucket *n; - int i, ret = 0; - int j = AHASH_MAX(h) + 1; + struct hbucket *n, *old = ERR_PTR(-ENOENT); + int i, j = -1; bool flag_exist = flags & IPSET_FLAG_EXIST; + bool deleted = false, forceadd = false, reuse = false; u32 key, multi = 0; - rcu_read_lock_bh(); - t = rcu_dereference_bh(h->table); + if (h->elements >= h->maxelem) { + if (SET_WITH_TIMEOUT(set)) + /* FIXME: when set is full, we slow down here */ + mtype_expire(set, h, NLEN(set->family), set->dsize); + if (h->elements >= h->maxelem && SET_WITH_FORCEADD(set)) + forceadd = true; + } + + t = ipset_dereference_protected(h->table, set); key = HKEY(value, h->initval, t->htable_bits); - n = hbucket(t, key); + n = __ipset_dereference_protected(hbucket(t, key), 1); + if (!n) { + if (forceadd) { + if (net_ratelimit()) + pr_warn("Set %s is full, maxelem %u reached\n", + set->name, h->maxelem); + return -IPSET_ERR_HASH_FULL; + } else if (h->elements >= h->maxelem) { + goto set_full; + } + old = NULL; + n = kzalloc(sizeof(*n) + AHASH_INIT_SIZE * set->dsize, + GFP_ATOMIC); + if (!n) + return -ENOMEM; + n->size = AHASH_INIT_SIZE; + goto copy_elem; + } for (i = 0; i < n->pos; i++) { + if (!test_bit(i, n->used)) { + /* Reuse first deleted entry */ + if (j == -1) { + deleted = reuse = true; + j = i; + } + continue; + } data = ahash_data(n, i, set->dsize); if (mtype_data_equal(data, d, &multi)) { if (flag_exist || @@ -657,85 +749,94 @@ mtype_add(struct ip_set *set, void *value, const struct ip_set_ext *ext, ip_set_timeout_expired(ext_timeout(data, set)))) { /* Just the extensions could be overwritten */ j = i; - goto reuse_slot; - } else { - ret = -IPSET_ERR_EXIST; - goto out; + goto overwrite_extensions; } + return -IPSET_ERR_EXIST; } /* Reuse first timed out entry */ if (SET_WITH_TIMEOUT(set) && ip_set_timeout_expired(ext_timeout(data, set)) && - j != AHASH_MAX(h) + 1) + j == -1) { j = i; + reuse = true; + } } - if (h->elements >= h->maxelem && SET_WITH_FORCEADD(set) && n->pos) { - /* Choosing the first entry in the array to replace */ - j = 0; - goto reuse_slot; - } - if (SET_WITH_TIMEOUT(set) && h->elements >= h->maxelem) - /* FIXME: when set is full, we slow down here */ - mtype_expire(set, h, NLEN(set->family), set->dsize); - - if (h->elements >= h->maxelem) { - if (net_ratelimit()) - pr_warn("Set %s is full, maxelem %u reached\n", - set->name, h->maxelem); - ret = -IPSET_ERR_HASH_FULL; - goto out; - } - -reuse_slot: - if (j != AHASH_MAX(h) + 1) { - /* Fill out reused slot */ + if (reuse || forceadd) { data = ahash_data(n, j, set->dsize); + if (!deleted) { #ifdef IP_SET_HASH_WITH_NETS - for (i = 0; i < IPSET_NET_COUNT; i++) { - mtype_del_cidr(h, SCIDR(data->cidr, i), - NLEN(set->family), i); - mtype_add_cidr(h, SCIDR(d->cidr, i), - NLEN(set->family), i); - } + for (i = 0; i < IPSET_NET_COUNT; i++) + mtype_del_cidr(h, + NCIDR_PUT(DCIDR_GET(data->cidr, i)), + NLEN(set->family), i); #endif - ip_set_ext_destroy(set, data); - } else { - /* Use/create a new slot */ + ip_set_ext_destroy(set, data); + h->elements--; + } + goto copy_data; + } + if (h->elements >= h->maxelem) + goto set_full; + /* Create a new slot */ + if (n->pos >= n->size) { TUNE_AHASH_MAX(h, multi); - ret = hbucket_elem_add(n, AHASH_MAX(h), set->dsize); - if (ret != 0) { - if (ret == -EAGAIN) - mtype_data_next(&h->next, d); - goto out; + if (n->size >= AHASH_MAX(h)) { + /* Trigger rehashing */ + mtype_data_next(&h->next, d); + return -EAGAIN; } - data = ahash_data(n, n->pos++, set->dsize); + old = n; + n = kzalloc(sizeof(*n) + + (old->size + AHASH_INIT_SIZE) * set->dsize, + GFP_ATOMIC); + if (!n) + return -ENOMEM; + memcpy(n, old, sizeof(struct hbucket) + + old->size * set->dsize); + n->size = old->size + AHASH_INIT_SIZE; + } + +copy_elem: + j = n->pos++; + data = ahash_data(n, j, set->dsize); +copy_data: + h->elements++; #ifdef IP_SET_HASH_WITH_NETS - for (i = 0; i < IPSET_NET_COUNT; i++) - mtype_add_cidr(h, SCIDR(d->cidr, i), NLEN(set->family), - i); + for (i = 0; i < IPSET_NET_COUNT; i++) + mtype_add_cidr(h, NCIDR_PUT(DCIDR_GET(d->cidr, i)), + NLEN(set->family), i); #endif - h->elements++; - } memcpy(data, d, sizeof(struct mtype_elem)); +overwrite_extensions: #ifdef IP_SET_HASH_WITH_NETS mtype_data_set_flags(data, flags); #endif - if (SET_WITH_TIMEOUT(set)) - ip_set_timeout_set(ext_timeout(data, set), ext->timeout); if (SET_WITH_COUNTER(set)) ip_set_init_counter(ext_counter(data, set), ext); if (SET_WITH_COMMENT(set)) ip_set_init_comment(ext_comment(data, set), ext); if (SET_WITH_SKBINFO(set)) ip_set_init_skbinfo(ext_skbinfo(data, set), ext); + /* Must come last for the case when timed out entry is reused */ + if (SET_WITH_TIMEOUT(set)) + ip_set_timeout_set(ext_timeout(data, set), ext->timeout); + smp_mb__before_atomic(); + set_bit(j, n->used); + if (old != ERR_PTR(-ENOENT)) { + rcu_assign_pointer(hbucket(t, key), n); + if (old) + kfree_rcu(old, rcu); + } -out: - rcu_read_unlock_bh(); - return ret; + return 0; +set_full: + if (net_ratelimit()) + pr_warn("Set %s is full, maxelem %u reached\n", + set->name, h->maxelem); + return -IPSET_ERR_HASH_FULL; } -/* Delete an element from the hash: swap it with the last element - * and free up space if possible. +/* Delete an element from the hash and free up space if possible. */ static int mtype_del(struct ip_set *set, void *value, const struct ip_set_ext *ext, @@ -746,55 +847,70 @@ mtype_del(struct ip_set *set, void *value, const struct ip_set_ext *ext, const struct mtype_elem *d = value; struct mtype_elem *data; struct hbucket *n; - int i, ret = -IPSET_ERR_EXIST; -#ifdef IP_SET_HASH_WITH_NETS - u8 j; -#endif + int i, j, k, ret = -IPSET_ERR_EXIST; u32 key, multi = 0; + size_t dsize = set->dsize; - rcu_read_lock_bh(); - t = rcu_dereference_bh(h->table); + t = ipset_dereference_protected(h->table, set); key = HKEY(value, h->initval, t->htable_bits); - n = hbucket(t, key); - for (i = 0; i < n->pos; i++) { - data = ahash_data(n, i, set->dsize); + n = __ipset_dereference_protected(hbucket(t, key), 1); + if (!n) + goto out; + for (i = 0, k = 0; i < n->pos; i++) { + if (!test_bit(i, n->used)) { + k++; + continue; + } + data = ahash_data(n, i, dsize); if (!mtype_data_equal(data, d, &multi)) continue; if (SET_WITH_TIMEOUT(set) && ip_set_timeout_expired(ext_timeout(data, set))) goto out; - if (i != n->pos - 1) - /* Not last one */ - memcpy(data, ahash_data(n, n->pos - 1, set->dsize), - set->dsize); - n->pos--; + ret = 0; + clear_bit(i, n->used); + smp_mb__after_atomic(); + if (i + 1 == n->pos) + n->pos--; h->elements--; #ifdef IP_SET_HASH_WITH_NETS for (j = 0; j < IPSET_NET_COUNT; j++) - mtype_del_cidr(h, SCIDR(d->cidr, j), NLEN(set->family), - j); + mtype_del_cidr(h, NCIDR_PUT(DCIDR_GET(d->cidr, j)), + NLEN(set->family), j); #endif ip_set_ext_destroy(set, data); - if (n->pos + AHASH_INIT_SIZE < n->size) { - void *tmp = kzalloc((n->size - AHASH_INIT_SIZE) - * set->dsize, - GFP_ATOMIC); - if (!tmp) { - ret = 0; + + for (; i < n->pos; i++) { + if (!test_bit(i, n->used)) + k++; + } + if (n->pos == 0 && k == 0) { + rcu_assign_pointer(hbucket(t, key), NULL); + kfree_rcu(n, rcu); + } else if (k >= AHASH_INIT_SIZE) { + struct hbucket *tmp = kzalloc(sizeof(*tmp) + + (n->size - AHASH_INIT_SIZE) * dsize, + GFP_ATOMIC); + if (!tmp) goto out; + tmp->size = n->size - AHASH_INIT_SIZE; + for (j = 0, k = 0; j < n->pos; j++) { + if (!test_bit(j, n->used)) + continue; + data = ahash_data(n, j, dsize); + memcpy(tmp->value + k * dsize, data, dsize); + set_bit(j, tmp->used); + k++; } - n->size -= AHASH_INIT_SIZE; - memcpy(tmp, n->value, n->size * set->dsize); - kfree(n->value); - n->value = tmp; + tmp->pos = k; + rcu_assign_pointer(hbucket(t, key), tmp); + kfree_rcu(n, rcu); } - ret = 0; goto out; } out: - rcu_read_unlock_bh(); return ret; } @@ -813,7 +929,8 @@ mtype_data_match(struct mtype_elem *data, const struct ip_set_ext *ext, #ifdef IP_SET_HASH_WITH_NETS /* Special test function which takes into account the different network - * sizes added to the set */ + * sizes added to the set + */ static int mtype_test_cidrs(struct ip_set *set, struct mtype_elem *d, const struct ip_set_ext *ext, @@ -836,16 +953,21 @@ mtype_test_cidrs(struct ip_set *set, struct mtype_elem *d, for (; j < nets_length && h->nets[j].cidr[0] && !multi; j++) { #if IPSET_NET_COUNT == 2 mtype_data_reset_elem(d, &orig); - mtype_data_netmask(d, NCIDR(h->nets[j].cidr[0]), false); + mtype_data_netmask(d, NCIDR_GET(h->nets[j].cidr[0]), false); for (k = 0; k < nets_length && h->nets[k].cidr[1] && !multi; k++) { - mtype_data_netmask(d, NCIDR(h->nets[k].cidr[1]), true); + mtype_data_netmask(d, NCIDR_GET(h->nets[k].cidr[1]), + true); #else - mtype_data_netmask(d, NCIDR(h->nets[j].cidr[0])); + mtype_data_netmask(d, NCIDR_GET(h->nets[j].cidr[0])); #endif key = HKEY(d, h->initval, t->htable_bits); - n = hbucket(t, key); + n = rcu_dereference_bh(hbucket(t, key)); + if (!n) + continue; for (i = 0; i < n->pos; i++) { + if (!test_bit(i, n->used)) + continue; data = ahash_data(n, i, set->dsize); if (!mtype_data_equal(data, d, &multi)) continue; @@ -883,13 +1005,13 @@ mtype_test(struct ip_set *set, void *value, const struct ip_set_ext *ext, int i, ret = 0; u32 key, multi = 0; - rcu_read_lock_bh(); t = rcu_dereference_bh(h->table); #ifdef IP_SET_HASH_WITH_NETS /* If we test an IP address and not a network address, - * try all possible network sizes */ + * try all possible network sizes + */ for (i = 0; i < IPSET_NET_COUNT; i++) - if (GCIDR(d->cidr, i) != SET_HOST_MASK(set->family)) + if (DCIDR_GET(d->cidr, i) != SET_HOST_MASK(set->family)) break; if (i == IPSET_NET_COUNT) { ret = mtype_test_cidrs(set, d, ext, mext, flags); @@ -898,8 +1020,14 @@ mtype_test(struct ip_set *set, void *value, const struct ip_set_ext *ext, #endif key = HKEY(d, h->initval, t->htable_bits); - n = hbucket(t, key); + n = rcu_dereference_bh(hbucket(t, key)); + if (!n) { + ret = 0; + goto out; + } for (i = 0; i < n->pos; i++) { + if (!test_bit(i, n->used)) + continue; data = ahash_data(n, i, set->dsize); if (mtype_data_equal(data, d, &multi) && !(SET_WITH_TIMEOUT(set) && @@ -909,7 +1037,6 @@ mtype_test(struct ip_set *set, void *value, const struct ip_set_ext *ext, } } out: - rcu_read_unlock_bh(); return ret; } @@ -921,15 +1048,19 @@ mtype_head(struct ip_set *set, struct sk_buff *skb) const struct htable *t; struct nlattr *nested; size_t memsize; + u8 htable_bits; + rcu_read_lock_bh(); t = rcu_dereference_bh_nfnl(h->table); memsize = mtype_ahash_memsize(h, t, NLEN(set->family), set->dsize); + htable_bits = t->htable_bits; + rcu_read_unlock_bh(); nested = ipset_nest_start(skb, IPSET_ATTR_DATA); if (!nested) goto nla_put_failure; if (nla_put_net32(skb, IPSET_ATTR_HASHSIZE, - htonl(jhash_size(t->htable_bits))) || + htonl(jhash_size(htable_bits))) || nla_put_net32(skb, IPSET_ATTR_MAXELEM, htonl(h->maxelem))) goto nla_put_failure; #ifdef IP_SET_HASH_WITH_NETMASK @@ -953,32 +1084,63 @@ nla_put_failure: return -EMSGSIZE; } +/* Make possible to run dumping parallel with resizing */ +static void +mtype_uref(struct ip_set *set, struct netlink_callback *cb, bool start) +{ + struct htype *h = set->data; + struct htable *t; + + if (start) { + rcu_read_lock_bh(); + t = rcu_dereference_bh_nfnl(h->table); + atomic_inc(&t->uref); + cb->args[IPSET_CB_PRIVATE] = (unsigned long)t; + rcu_read_unlock_bh(); + } else if (cb->args[IPSET_CB_PRIVATE]) { + t = (struct htable *)cb->args[IPSET_CB_PRIVATE]; + if (atomic_dec_and_test(&t->uref) && atomic_read(&t->ref)) { + /* Resizing didn't destroy the hash table */ + pr_debug("Table destroy by dump: %p\n", t); + mtype_ahash_destroy(set, t, false); + } + cb->args[IPSET_CB_PRIVATE] = 0; + } +} + /* Reply a LIST/SAVE request: dump the elements of the specified set */ static int mtype_list(const struct ip_set *set, struct sk_buff *skb, struct netlink_callback *cb) { - const struct htype *h = set->data; - const struct htable *t = rcu_dereference_bh_nfnl(h->table); + const struct htable *t; struct nlattr *atd, *nested; const struct hbucket *n; const struct mtype_elem *e; u32 first = cb->args[IPSET_CB_ARG0]; /* We assume that one hash bucket fills into one page */ void *incomplete; - int i; + int i, ret = 0; atd = ipset_nest_start(skb, IPSET_ATTR_ADT); if (!atd) return -EMSGSIZE; + pr_debug("list hash set %s\n", set->name); + t = (const struct htable *)cb->args[IPSET_CB_PRIVATE]; + /* Expire may replace a hbucket with another one */ + rcu_read_lock(); for (; cb->args[IPSET_CB_ARG0] < jhash_size(t->htable_bits); cb->args[IPSET_CB_ARG0]++) { incomplete = skb_tail_pointer(skb); - n = hbucket(t, cb->args[IPSET_CB_ARG0]); + n = rcu_dereference(hbucket(t, cb->args[IPSET_CB_ARG0])); pr_debug("cb->arg bucket: %lu, t %p n %p\n", cb->args[IPSET_CB_ARG0], t, n); + if (!n) + continue; for (i = 0; i < n->pos; i++) { + if (!test_bit(i, n->used)) + continue; e = ahash_data(n, i, set->dsize); if (SET_WITH_TIMEOUT(set) && ip_set_timeout_expired(ext_timeout(e, set))) @@ -989,9 +1151,10 @@ mtype_list(const struct ip_set *set, if (!nested) { if (cb->args[IPSET_CB_ARG0] == first) { nla_nest_cancel(skb, atd); - return -EMSGSIZE; - } else - goto nla_put_failure; + ret = -EMSGSIZE; + goto out; + } + goto nla_put_failure; } if (mtype_data_list(skb, e)) goto nla_put_failure; @@ -1004,7 +1167,7 @@ mtype_list(const struct ip_set *set, /* Set listing finished */ cb->args[IPSET_CB_ARG0] = 0; - return 0; + goto out; nla_put_failure: nlmsg_trim(skb, incomplete); @@ -1012,20 +1175,24 @@ nla_put_failure: pr_warn("Can't list set %s: one bucket does not fit into a message. Please report it!\n", set->name); cb->args[IPSET_CB_ARG0] = 0; - return -EMSGSIZE; + ret = -EMSGSIZE; + } else { + ipset_nest_end(skb, atd); } - ipset_nest_end(skb, atd); - return 0; +out: + rcu_read_unlock(); + return ret; } static int IPSET_TOKEN(MTYPE, _kadt)(struct ip_set *set, const struct sk_buff *skb, - const struct xt_action_param *par, - enum ipset_adt adt, struct ip_set_adt_opt *opt); + const struct xt_action_param *par, + enum ipset_adt adt, struct ip_set_adt_opt *opt); static int IPSET_TOKEN(MTYPE, _uadt)(struct ip_set *set, struct nlattr *tb[], - enum ipset_adt adt, u32 *lineno, u32 flags, bool retried); + enum ipset_adt adt, u32 *lineno, u32 flags, + bool retried); static const struct ip_set_type_variant mtype_variant = { .kadt = mtype_kadt, @@ -1039,6 +1206,7 @@ static const struct ip_set_type_variant mtype_variant = { .flush = mtype_flush, .head = mtype_head, .list = mtype_list, + .uref = mtype_uref, .resize = mtype_resize, .same_set = mtype_same_set, }; @@ -1076,12 +1244,14 @@ IPSET_TOKEN(HTYPE, _create)(struct net *net, struct ip_set *set, if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_HASHSIZE) || !ip_set_optattr_netorder(tb, IPSET_ATTR_MAXELEM) || -#ifdef IP_SET_HASH_WITH_MARKMASK - !ip_set_optattr_netorder(tb, IPSET_ATTR_MARKMASK) || -#endif !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) return -IPSET_ERR_PROTOCOL; +#ifdef IP_SET_HASH_WITH_MARKMASK + /* Separated condition in order to avoid directive in argument list */ + if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_MARKMASK))) + return -IPSET_ERR_PROTOCOL; +#endif if (tb[IPSET_ATTR_HASHSIZE]) { hashsize = ip_set_get_h32(tb[IPSET_ATTR_HASHSIZE]); @@ -1104,7 +1274,7 @@ IPSET_TOKEN(HTYPE, _create)(struct net *net, struct ip_set *set, #endif #ifdef IP_SET_HASH_WITH_MARKMASK if (tb[IPSET_ATTR_MARKMASK]) { - markmask = ntohl(nla_get_u32(tb[IPSET_ATTR_MARKMASK])); + markmask = ntohl(nla_get_be32(tb[IPSET_ATTR_MARKMASK])); if (markmask == 0) return -IPSET_ERR_INVALID_MARKMASK; diff --git a/net/netfilter/ipset/ip_set_hash_ip.c b/net/netfilter/ipset/ip_set_hash_ip.c index 54df48b5c455..9d6bf19f7b78 100644 --- a/net/netfilter/ipset/ip_set_hash_ip.c +++ b/net/netfilter/ipset/ip_set_hash_ip.c @@ -108,18 +108,12 @@ hash_ip4_uadt(struct ip_set *set, struct nlattr *tb[], u32 ip = 0, ip_to = 0, hosts; int ret = 0; - if (unlikely(!tb[IPSET_ATTR_IP] || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) - return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP])) + return -IPSET_ERR_PROTOCOL; + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP], &ip); if (ret) return ret; @@ -164,8 +158,8 @@ hash_ip4_uadt(struct ip_set *set, struct nlattr *tb[], if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; } return ret; } @@ -246,20 +240,20 @@ hash_ip6_uadt(struct ip_set *set, struct nlattr *tb[], struct ip_set_ext ext = IP_SET_INIT_UEXT(set); int ret; - if (unlikely(!tb[IPSET_ATTR_IP] || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE) || - tb[IPSET_ATTR_IP_TO] || - tb[IPSET_ATTR_CIDR])) - return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP])) + return -IPSET_ERR_PROTOCOL; + if (unlikely(tb[IPSET_ATTR_IP_TO])) + return -IPSET_ERR_HASH_RANGE_UNSUPPORTED; + if (unlikely(tb[IPSET_ATTR_CIDR])) { + u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); + + if (cidr != HOST_MASK) + return -IPSET_ERR_INVALID_CIDR; + } + ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &e.ip); if (ret) return ret; @@ -321,6 +315,7 @@ hash_ip_init(void) static void __exit hash_ip_fini(void) { + rcu_barrier(); ip_set_type_unregister(&hash_ip_type); } diff --git a/net/netfilter/ipset/ip_set_hash_ipmark.c b/net/netfilter/ipset/ip_set_hash_ipmark.c index d231248eb3e2..a0695a2ab585 100644 --- a/net/netfilter/ipset/ip_set_hash_ipmark.c +++ b/net/netfilter/ipset/ip_set_hash_ipmark.c @@ -108,19 +108,13 @@ hash_ipmark4_uadt(struct ip_set *set, struct nlattr *tb[], u32 ip, ip_to = 0; int ret; - if (unlikely(!tb[IPSET_ATTR_IP] || - !ip_set_attr_netorder(tb, IPSET_ATTR_MARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) - return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || + !ip_set_attr_netorder(tb, IPSET_ATTR_MARK))) + return -IPSET_ERR_PROTOCOL; + ret = ip_set_get_ipaddr4(tb[IPSET_ATTR_IP], &e.ip); if (ret) return ret; @@ -161,8 +155,8 @@ hash_ipmark4_uadt(struct ip_set *set, struct nlattr *tb[], if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; } return ret; } @@ -212,7 +206,6 @@ hash_ipmark6_data_next(struct hash_ipmark4_elem *next, #define IP_SET_EMIT_CREATE #include "ip_set_hash_gen.h" - static int hash_ipmark6_kadt(struct ip_set *set, const struct sk_buff *skb, const struct xt_action_param *par, @@ -240,20 +233,20 @@ hash_ipmark6_uadt(struct ip_set *set, struct nlattr *tb[], struct ip_set_ext ext = IP_SET_INIT_UEXT(set); int ret; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || - !ip_set_attr_netorder(tb, IPSET_ATTR_MARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE) || - tb[IPSET_ATTR_IP_TO] || - tb[IPSET_ATTR_CIDR])) + !ip_set_attr_netorder(tb, IPSET_ATTR_MARK))) return -IPSET_ERR_PROTOCOL; + if (unlikely(tb[IPSET_ATTR_IP_TO])) + return -IPSET_ERR_HASH_RANGE_UNSUPPORTED; + if (unlikely(tb[IPSET_ATTR_CIDR])) { + u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (cidr != HOST_MASK) + return -IPSET_ERR_INVALID_CIDR; + } ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &e.ip); if (ret) @@ -274,10 +267,8 @@ hash_ipmark6_uadt(struct ip_set *set, struct nlattr *tb[], ret = adtfn(set, &e, &ext, &ext, flags); if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; - return ret; + return 0; } static struct ip_set_type hash_ipmark_type __read_mostly = { @@ -325,6 +316,7 @@ hash_ipmark_init(void) static void __exit hash_ipmark_fini(void) { + rcu_barrier(); ip_set_type_unregister(&hash_ipmark_type); } diff --git a/net/netfilter/ipset/ip_set_hash_ipport.c b/net/netfilter/ipset/ip_set_hash_ipport.c index a47c29f12090..9d84b3dff603 100644 --- a/net/netfilter/ipset/ip_set_hash_ipport.c +++ b/net/netfilter/ipset/ip_set_hash_ipport.c @@ -116,20 +116,14 @@ hash_ipport4_uadt(struct ip_set *set, struct nlattr *tb[], bool with_ports = false; int ret; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) + !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO))) return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); - ret = ip_set_get_ipaddr4(tb[IPSET_ATTR_IP], &e.ip); if (ret) return ret; @@ -146,8 +140,9 @@ hash_ipport4_uadt(struct ip_set *set, struct nlattr *tb[], if (e.proto == 0) return -IPSET_ERR_INVALID_PROTO; - } else + } else { return -IPSET_ERR_MISSING_PROTO; + } if (!(with_ports || e.proto == IPPROTO_ICMP)) e.port = 0; @@ -193,8 +188,8 @@ hash_ipport4_uadt(struct ip_set *set, struct nlattr *tb[], if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; } } return ret; @@ -279,21 +274,21 @@ hash_ipport6_uadt(struct ip_set *set, struct nlattr *tb[], bool with_ports = false; int ret; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE) || - tb[IPSET_ATTR_IP_TO] || - tb[IPSET_ATTR_CIDR])) + !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO))) return -IPSET_ERR_PROTOCOL; + if (unlikely(tb[IPSET_ATTR_IP_TO])) + return -IPSET_ERR_HASH_RANGE_UNSUPPORTED; + if (unlikely(tb[IPSET_ATTR_CIDR])) { + u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (cidr != HOST_MASK) + return -IPSET_ERR_INVALID_CIDR; + } ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &e.ip); if (ret) @@ -311,8 +306,9 @@ hash_ipport6_uadt(struct ip_set *set, struct nlattr *tb[], if (e.proto == 0) return -IPSET_ERR_INVALID_PROTO; - } else + } else { return -IPSET_ERR_MISSING_PROTO; + } if (!(with_ports || e.proto == IPPROTO_ICMPV6)) e.port = 0; @@ -335,8 +331,8 @@ hash_ipport6_uadt(struct ip_set *set, struct nlattr *tb[], if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; } return ret; } @@ -388,6 +384,7 @@ hash_ipport_init(void) static void __exit hash_ipport_fini(void) { + rcu_barrier(); ip_set_type_unregister(&hash_ipport_type); } diff --git a/net/netfilter/ipset/ip_set_hash_ipportip.c b/net/netfilter/ipset/ip_set_hash_ipportip.c index 89615f134845..215b7b942038 100644 --- a/net/netfilter/ipset/ip_set_hash_ipportip.c +++ b/net/netfilter/ipset/ip_set_hash_ipportip.c @@ -63,7 +63,7 @@ hash_ipportip4_data_equal(const struct hash_ipportip4_elem *ip1, static bool hash_ipportip4_data_list(struct sk_buff *skb, - const struct hash_ipportip4_elem *data) + const struct hash_ipportip4_elem *data) { if (nla_put_ipaddr4(skb, IPSET_ATTR_IP, data->ip) || nla_put_ipaddr4(skb, IPSET_ATTR_IP2, data->ip2) || @@ -119,20 +119,14 @@ hash_ipportip4_uadt(struct ip_set *set, struct nlattr *tb[], bool with_ports = false; int ret; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IP2] || !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) + !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO))) return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); - ret = ip_set_get_ipaddr4(tb[IPSET_ATTR_IP], &e.ip); if (ret) return ret; @@ -153,8 +147,9 @@ hash_ipportip4_uadt(struct ip_set *set, struct nlattr *tb[], if (e.proto == 0) return -IPSET_ERR_INVALID_PROTO; - } else + } else { return -IPSET_ERR_MISSING_PROTO; + } if (!(with_ports || e.proto == IPPROTO_ICMP)) e.port = 0; @@ -200,8 +195,8 @@ hash_ipportip4_uadt(struct ip_set *set, struct nlattr *tb[], if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; } } return ret; @@ -290,21 +285,21 @@ hash_ipportip6_uadt(struct ip_set *set, struct nlattr *tb[], bool with_ports = false; int ret; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IP2] || !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE) || - tb[IPSET_ATTR_IP_TO] || - tb[IPSET_ATTR_CIDR])) + !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO))) return -IPSET_ERR_PROTOCOL; + if (unlikely(tb[IPSET_ATTR_IP_TO])) + return -IPSET_ERR_HASH_RANGE_UNSUPPORTED; + if (unlikely(tb[IPSET_ATTR_CIDR])) { + u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (cidr != HOST_MASK) + return -IPSET_ERR_INVALID_CIDR; + } ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &e.ip); if (ret) @@ -326,8 +321,9 @@ hash_ipportip6_uadt(struct ip_set *set, struct nlattr *tb[], if (e.proto == 0) return -IPSET_ERR_INVALID_PROTO; - } else + } else { return -IPSET_ERR_MISSING_PROTO; + } if (!(with_ports || e.proto == IPPROTO_ICMPV6)) e.port = 0; @@ -350,8 +346,8 @@ hash_ipportip6_uadt(struct ip_set *set, struct nlattr *tb[], if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; } return ret; } @@ -403,6 +399,7 @@ hash_ipportip_init(void) static void __exit hash_ipportip_fini(void) { + rcu_barrier(); ip_set_type_unregister(&hash_ipportip_type); } diff --git a/net/netfilter/ipset/ip_set_hash_ipportnet.c b/net/netfilter/ipset/ip_set_hash_ipportnet.c index 6ba7a7e083f9..9ca719625ea3 100644 --- a/net/netfilter/ipset/ip_set_hash_ipportnet.c +++ b/net/netfilter/ipset/ip_set_hash_ipportnet.c @@ -141,7 +141,7 @@ hash_ipportnet4_kadt(struct ip_set *set, const struct sk_buff *skb, const struct hash_ipportnet *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_ipportnet4_elem e = { - .cidr = IP_SET_INIT_CIDR(h->nets[0].cidr[0], HOST_MASK) - 1, + .cidr = INIT_CIDR(h->nets[0].cidr[0], HOST_MASK), }; struct ip_set_ext ext = IP_SET_INIT_KEXT(skb, opt, set); @@ -173,21 +173,15 @@ hash_ipportnet4_uadt(struct ip_set *set, struct nlattr *tb[], u8 cidr; int ret; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IP2] || !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) + !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); - ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP], &ip); if (ret) return ret; @@ -215,14 +209,16 @@ hash_ipportnet4_uadt(struct ip_set *set, struct nlattr *tb[], if (e.proto == 0) return -IPSET_ERR_INVALID_PROTO; - } else + } else { return -IPSET_ERR_MISSING_PROTO; + } if (!(with_ports || e.proto == IPPROTO_ICMP)) e.port = 0; if (tb[IPSET_ATTR_CADT_FLAGS]) { u32 cadt_flags = ip_set_get_h32(tb[IPSET_ATTR_CADT_FLAGS]); + if (cadt_flags & IPSET_FLAG_NOMATCH) flags |= (IPSET_FLAG_NOMATCH << 16); } @@ -269,8 +265,9 @@ hash_ipportnet4_uadt(struct ip_set *set, struct nlattr *tb[], swap(ip2_from, ip2_to); if (ip2_from + UINT_MAX == ip2_to) return -IPSET_ERR_HASH_RANGE; - } else + } else { ip_set_mask_from_to(ip2_from, ip2_to, e.cidr + 1); + } if (retried) ip = ntohl(h->next.ip); @@ -293,8 +290,8 @@ hash_ipportnet4_uadt(struct ip_set *set, struct nlattr *tb[], if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; ip2 = ip2_last + 1; } } @@ -395,7 +392,7 @@ hash_ipportnet6_kadt(struct ip_set *set, const struct sk_buff *skb, const struct hash_ipportnet *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_ipportnet6_elem e = { - .cidr = IP_SET_INIT_CIDR(h->nets[0].cidr[0], HOST_MASK) - 1, + .cidr = INIT_CIDR(h->nets[0].cidr[0], HOST_MASK), }; struct ip_set_ext ext = IP_SET_INIT_KEXT(skb, opt, set); @@ -426,24 +423,22 @@ hash_ipportnet6_uadt(struct ip_set *set, struct nlattr *tb[], u8 cidr; int ret; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IP2] || !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE) || - tb[IPSET_ATTR_IP_TO] || - tb[IPSET_ATTR_CIDR])) + !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) return -IPSET_ERR_PROTOCOL; if (unlikely(tb[IPSET_ATTR_IP_TO])) return -IPSET_ERR_HASH_RANGE_UNSUPPORTED; + if (unlikely(tb[IPSET_ATTR_CIDR])) { + u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (cidr != HOST_MASK) + return -IPSET_ERR_INVALID_CIDR; + } ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &e.ip); if (ret) @@ -474,14 +469,16 @@ hash_ipportnet6_uadt(struct ip_set *set, struct nlattr *tb[], if (e.proto == 0) return -IPSET_ERR_INVALID_PROTO; - } else + } else { return -IPSET_ERR_MISSING_PROTO; + } if (!(with_ports || e.proto == IPPROTO_ICMPV6)) e.port = 0; if (tb[IPSET_ATTR_CADT_FLAGS]) { u32 cadt_flags = ip_set_get_h32(tb[IPSET_ATTR_CADT_FLAGS]); + if (cadt_flags & IPSET_FLAG_NOMATCH) flags |= (IPSET_FLAG_NOMATCH << 16); } @@ -505,8 +502,8 @@ hash_ipportnet6_uadt(struct ip_set *set, struct nlattr *tb[], if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; } return ret; } @@ -562,6 +559,7 @@ hash_ipportnet_init(void) static void __exit hash_ipportnet_fini(void) { + rcu_barrier(); ip_set_type_unregister(&hash_ipportnet_type); } diff --git a/net/netfilter/ipset/ip_set_hash_mac.c b/net/netfilter/ipset/ip_set_hash_mac.c index 1f8668d7a538..f1e7d2c0f685 100644 --- a/net/netfilter/ipset/ip_set_hash_mac.c +++ b/net/netfilter/ipset/ip_set_hash_mac.c @@ -89,10 +89,10 @@ hash_mac4_kadt(struct ip_set *set, const struct sk_buff *skb, return 0; if (skb_mac_header(skb) < skb->head || - (skb_mac_header(skb) + ETH_HLEN) > skb->data) + (skb_mac_header(skb) + ETH_HLEN) > skb->data) return -EINVAL; - memcpy(e.ether, eth_hdr(skb)->h_source, ETH_ALEN); + ether_addr_copy(e.ether, eth_hdr(skb)->h_source); if (memcmp(e.ether, invalid_ether, ETH_ALEN) == 0) return -EINVAL; return adtfn(set, &e, &ext, &opt->ext, opt->cmdflags); @@ -107,22 +107,16 @@ hash_mac4_uadt(struct ip_set *set, struct nlattr *tb[], struct ip_set_ext ext = IP_SET_INIT_UEXT(set); int ret; - if (unlikely(!tb[IPSET_ATTR_ETHER] || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) - return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_ETHER])) + return -IPSET_ERR_PROTOCOL; + ret = ip_set_get_extensions(set, tb, &ext); if (ret) return ret; - memcpy(e.ether, nla_data(tb[IPSET_ATTR_ETHER]), ETH_ALEN); + ether_addr_copy(e.ether, nla_data(tb[IPSET_ATTR_ETHER])); if (memcmp(e.ether, invalid_ether, ETH_ALEN) == 0) return -IPSET_ERR_HASH_ELEM; @@ -171,6 +165,7 @@ hash_mac_init(void) static void __exit hash_mac_fini(void) { + rcu_barrier(); ip_set_type_unregister(&hash_mac_type); } diff --git a/net/netfilter/ipset/ip_set_hash_net.c b/net/netfilter/ipset/ip_set_hash_net.c index 2e63dad8644d..3e4bffdc1cc0 100644 --- a/net/netfilter/ipset/ip_set_hash_net.c +++ b/net/netfilter/ipset/ip_set_hash_net.c @@ -120,7 +120,7 @@ hash_net4_kadt(struct ip_set *set, const struct sk_buff *skb, const struct hash_net *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_net4_elem e = { - .cidr = IP_SET_INIT_CIDR(h->nets[0].cidr[0], HOST_MASK), + .cidr = INIT_CIDR(h->nets[0].cidr[0], HOST_MASK), }; struct ip_set_ext ext = IP_SET_INIT_KEXT(skb, opt, set); @@ -146,19 +146,13 @@ hash_net4_uadt(struct ip_set *set, struct nlattr *tb[], u32 ip = 0, ip_to = 0, last; int ret; - if (unlikely(!tb[IPSET_ATTR_IP] || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) - return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || + !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) + return -IPSET_ERR_PROTOCOL; + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP], &ip); if (ret) return ret; @@ -175,6 +169,7 @@ hash_net4_uadt(struct ip_set *set, struct nlattr *tb[], if (tb[IPSET_ATTR_CADT_FLAGS]) { u32 cadt_flags = ip_set_get_h32(tb[IPSET_ATTR_CADT_FLAGS]); + if (cadt_flags & IPSET_FLAG_NOMATCH) flags |= (IPSET_FLAG_NOMATCH << 16); } @@ -182,7 +177,7 @@ hash_net4_uadt(struct ip_set *set, struct nlattr *tb[], if (adt == IPSET_TEST || !tb[IPSET_ATTR_IP_TO]) { e.ip = htonl(ip & ip_set_hostmask(e.cidr)); ret = adtfn(set, &e, &ext, &ext, flags); - return ip_set_enomatch(ret, flags, adt, set) ? -ret: + return ip_set_enomatch(ret, flags, adt, set) ? -ret : ip_set_eexist(ret, flags) ? 0 : ret; } @@ -204,8 +199,8 @@ hash_net4_uadt(struct ip_set *set, struct nlattr *tb[], ret = adtfn(set, &e, &ext, &ext, flags); if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; ip = last + 1; } return ret; @@ -294,7 +289,7 @@ hash_net6_kadt(struct ip_set *set, const struct sk_buff *skb, const struct hash_net *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_net6_elem e = { - .cidr = IP_SET_INIT_CIDR(h->nets[0].cidr[0], HOST_MASK), + .cidr = INIT_CIDR(h->nets[0].cidr[0], HOST_MASK), }; struct ip_set_ext ext = IP_SET_INIT_KEXT(skb, opt, set); @@ -318,21 +313,15 @@ hash_net6_uadt(struct ip_set *set, struct nlattr *tb[], struct ip_set_ext ext = IP_SET_INIT_UEXT(set); int ret; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) + !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) return -IPSET_ERR_PROTOCOL; if (unlikely(tb[IPSET_ATTR_IP_TO])) return -IPSET_ERR_HASH_RANGE_UNSUPPORTED; - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); - ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &e.ip); if (ret) return ret; @@ -341,16 +330,17 @@ hash_net6_uadt(struct ip_set *set, struct nlattr *tb[], if (ret) return ret; - if (tb[IPSET_ATTR_CIDR]) + if (tb[IPSET_ATTR_CIDR]) { e.cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); - - if (!e.cidr || e.cidr > HOST_MASK) - return -IPSET_ERR_INVALID_CIDR; + if (!e.cidr || e.cidr > HOST_MASK) + return -IPSET_ERR_INVALID_CIDR; + } ip6_netmask(&e.ip, e.cidr); if (tb[IPSET_ATTR_CADT_FLAGS]) { u32 cadt_flags = ip_set_get_h32(tb[IPSET_ATTR_CADT_FLAGS]); + if (cadt_flags & IPSET_FLAG_NOMATCH) flags |= (IPSET_FLAG_NOMATCH << 16); } @@ -404,6 +394,7 @@ hash_net_init(void) static void __exit hash_net_fini(void) { + rcu_barrier(); ip_set_type_unregister(&hash_net_type); } diff --git a/net/netfilter/ipset/ip_set_hash_netiface.c b/net/netfilter/ipset/ip_set_hash_netiface.c index fe481f677f56..43d8c9896fa3 100644 --- a/net/netfilter/ipset/ip_set_hash_netiface.c +++ b/net/netfilter/ipset/ip_set_hash_netiface.c @@ -13,7 +13,6 @@ #include <linux/skbuff.h> #include <linux/errno.h> #include <linux/random.h> -#include <linux/rbtree.h> #include <net/ip.h> #include <net/ipv6.h> #include <net/netlink.h> @@ -37,88 +36,13 @@ MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>"); IP_SET_MODULE_DESC("hash:net,iface", IPSET_TYPE_REV_MIN, IPSET_TYPE_REV_MAX); MODULE_ALIAS("ip_set_hash:net,iface"); -/* Interface name rbtree */ - -struct iface_node { - struct rb_node node; - char iface[IFNAMSIZ]; -}; - -#define iface_data(n) (rb_entry(n, struct iface_node, node)->iface) - -static void -rbtree_destroy(struct rb_root *root) -{ - struct iface_node *node, *next; - - rbtree_postorder_for_each_entry_safe(node, next, root, node) - kfree(node); - - *root = RB_ROOT; -} - -static int -iface_test(struct rb_root *root, const char **iface) -{ - struct rb_node *n = root->rb_node; - - while (n) { - const char *d = iface_data(n); - int res = strcmp(*iface, d); - - if (res < 0) - n = n->rb_left; - else if (res > 0) - n = n->rb_right; - else { - *iface = d; - return 1; - } - } - return 0; -} - -static int -iface_add(struct rb_root *root, const char **iface) -{ - struct rb_node **n = &(root->rb_node), *p = NULL; - struct iface_node *d; - - while (*n) { - char *ifname = iface_data(*n); - int res = strcmp(*iface, ifname); - - p = *n; - if (res < 0) - n = &((*n)->rb_left); - else if (res > 0) - n = &((*n)->rb_right); - else { - *iface = ifname; - return 0; - } - } - - d = kzalloc(sizeof(*d), GFP_ATOMIC); - if (!d) - return -ENOMEM; - strcpy(d->iface, *iface); - - rb_link_node(&d->node, p, n); - rb_insert_color(&d->node, root); - - *iface = d->iface; - return 0; -} - /* Type specific function prefix */ #define HTYPE hash_netiface #define IP_SET_HASH_WITH_NETS -#define IP_SET_HASH_WITH_RBTREE #define IP_SET_HASH_WITH_MULTI #define IP_SET_HASH_WITH_NET0 -#define STREQ(a, b) (strcmp(a, b) == 0) +#define STRLCPY(a, b) strlcpy(a, b, IFNAMSIZ) /* IPv4 variant */ @@ -137,7 +61,7 @@ struct hash_netiface4_elem { u8 cidr; u8 nomatch; u8 elem; - const char *iface; + char iface[IFNAMSIZ]; }; /* Common functions */ @@ -151,7 +75,7 @@ hash_netiface4_data_equal(const struct hash_netiface4_elem *ip1, ip1->cidr == ip2->cidr && (++*multi) && ip1->physdev == ip2->physdev && - ip1->iface == ip2->iface; + strcmp(ip1->iface, ip2->iface) == 0; } static inline int @@ -219,7 +143,7 @@ static const char *get_physindev_name(const struct sk_buff *skb) return dev ? dev->name : NULL; } -static const char *get_phyoutdev_name(const struct sk_buff *skb) +static const char *get_physoutdev_name(const struct sk_buff *skb) { struct net_device *dev = nf_bridge_get_physoutdev(skb); @@ -235,11 +159,10 @@ hash_netiface4_kadt(struct ip_set *set, const struct sk_buff *skb, struct hash_netiface *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_netiface4_elem e = { - .cidr = IP_SET_INIT_CIDR(h->nets[0].cidr[0], HOST_MASK), + .cidr = INIT_CIDR(h->nets[0].cidr[0], HOST_MASK), .elem = 1, }; struct ip_set_ext ext = IP_SET_INIT_KEXT(skb, opt, set); - int ret; if (e.cidr == 0) return -EINVAL; @@ -249,35 +172,25 @@ hash_netiface4_kadt(struct ip_set *set, const struct sk_buff *skb, ip4addrptr(skb, opt->flags & IPSET_DIM_ONE_SRC, &e.ip); e.ip &= ip_set_netmask(e.cidr); -#define IFACE(dir) (par->dir ? par->dir->name : NULL) +#define IFACE(dir) (par->dir ? par->dir->name : "") #define SRCDIR (opt->flags & IPSET_DIM_TWO_SRC) if (opt->cmdflags & IPSET_FLAG_PHYSDEV) { #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) - e.iface = SRCDIR ? get_physindev_name(skb) : - get_phyoutdev_name(skb); + const char *eiface = SRCDIR ? get_physindev_name(skb) : + get_physoutdev_name(skb); - if (!e.iface) + if (!eiface) return -EINVAL; + STRLCPY(e.iface, eiface); e.physdev = 1; -#else - e.iface = NULL; #endif - } else - e.iface = SRCDIR ? IFACE(in) : IFACE(out); + } else { + STRLCPY(e.iface, SRCDIR ? IFACE(in) : IFACE(out)); + } - if (!e.iface) + if (strlen(e.iface) == 0) return -EINVAL; - ret = iface_test(&h->rbtree, &e.iface); - if (adt == IPSET_ADD) { - if (!ret) { - ret = iface_add(&h->rbtree, &e.iface); - if (ret) - return ret; - } - } else if (!ret) - return ret; - return adtfn(set, &e, &ext, &opt->ext, opt->cmdflags); } @@ -290,23 +203,16 @@ hash_netiface4_uadt(struct ip_set *set, struct nlattr *tb[], struct hash_netiface4_elem e = { .cidr = HOST_MASK, .elem = 1 }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); u32 ip = 0, ip_to = 0, last; - char iface[IFNAMSIZ]; int ret; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IFACE] || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) + !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); - ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP], &ip); if (ret) return ret; @@ -320,21 +226,11 @@ hash_netiface4_uadt(struct ip_set *set, struct nlattr *tb[], if (e.cidr > HOST_MASK) return -IPSET_ERR_INVALID_CIDR; } - - strcpy(iface, nla_data(tb[IPSET_ATTR_IFACE])); - e.iface = iface; - ret = iface_test(&h->rbtree, &e.iface); - if (adt == IPSET_ADD) { - if (!ret) { - ret = iface_add(&h->rbtree, &e.iface); - if (ret) - return ret; - } - } else if (!ret) - return ret; + nla_strlcpy(e.iface, tb[IPSET_ATTR_IFACE], IFNAMSIZ); if (tb[IPSET_ATTR_CADT_FLAGS]) { u32 cadt_flags = ip_set_get_h32(tb[IPSET_ATTR_CADT_FLAGS]); + if (cadt_flags & IPSET_FLAG_PHYSDEV) e.physdev = 1; if (cadt_flags & IPSET_FLAG_NOMATCH) @@ -355,8 +251,9 @@ hash_netiface4_uadt(struct ip_set *set, struct nlattr *tb[], swap(ip, ip_to); if (ip + UINT_MAX == ip_to) return -IPSET_ERR_HASH_RANGE; - } else + } else { ip_set_mask_from_to(ip, ip_to, e.cidr); + } if (retried) ip = ntohl(h->next.ip); @@ -367,8 +264,8 @@ hash_netiface4_uadt(struct ip_set *set, struct nlattr *tb[], if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; ip = last + 1; } return ret; @@ -390,7 +287,7 @@ struct hash_netiface6_elem { u8 cidr; u8 nomatch; u8 elem; - const char *iface; + char iface[IFNAMSIZ]; }; /* Common functions */ @@ -404,7 +301,7 @@ hash_netiface6_data_equal(const struct hash_netiface6_elem *ip1, ip1->cidr == ip2->cidr && (++*multi) && ip1->physdev == ip2->physdev && - ip1->iface == ip2->iface; + strcmp(ip1->iface, ip2->iface) == 0; } static inline int @@ -475,11 +372,10 @@ hash_netiface6_kadt(struct ip_set *set, const struct sk_buff *skb, struct hash_netiface *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_netiface6_elem e = { - .cidr = IP_SET_INIT_CIDR(h->nets[0].cidr[0], HOST_MASK), + .cidr = INIT_CIDR(h->nets[0].cidr[0], HOST_MASK), .elem = 1, }; struct ip_set_ext ext = IP_SET_INIT_KEXT(skb, opt, set); - int ret; if (e.cidr == 0) return -EINVAL; @@ -491,60 +387,43 @@ hash_netiface6_kadt(struct ip_set *set, const struct sk_buff *skb, if (opt->cmdflags & IPSET_FLAG_PHYSDEV) { #if IS_ENABLED(CONFIG_BRIDGE_NETFILTER) - e.iface = SRCDIR ? get_physindev_name(skb) : - get_phyoutdev_name(skb); - if (!e.iface) - return -EINVAL; + const char *eiface = SRCDIR ? get_physindev_name(skb) : + get_physoutdev_name(skb); + if (!eiface) + return -EINVAL; + STRLCPY(e.iface, eiface); e.physdev = 1; -#else - e.iface = NULL; #endif - } else - e.iface = SRCDIR ? IFACE(in) : IFACE(out); + } else { + STRLCPY(e.iface, SRCDIR ? IFACE(in) : IFACE(out)); + } - if (!e.iface) + if (strlen(e.iface) == 0) return -EINVAL; - ret = iface_test(&h->rbtree, &e.iface); - if (adt == IPSET_ADD) { - if (!ret) { - ret = iface_add(&h->rbtree, &e.iface); - if (ret) - return ret; - } - } else if (!ret) - return ret; return adtfn(set, &e, &ext, &opt->ext, opt->cmdflags); } static int hash_netiface6_uadt(struct ip_set *set, struct nlattr *tb[], - enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) + enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) { - struct hash_netiface *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_netiface6_elem e = { .cidr = HOST_MASK, .elem = 1 }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); - char iface[IFNAMSIZ]; int ret; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IFACE] || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) + !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) return -IPSET_ERR_PROTOCOL; if (unlikely(tb[IPSET_ATTR_IP_TO])) return -IPSET_ERR_HASH_RANGE_UNSUPPORTED; - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); - ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &e.ip); if (ret) return ret; @@ -553,26 +432,19 @@ hash_netiface6_uadt(struct ip_set *set, struct nlattr *tb[], if (ret) return ret; - if (tb[IPSET_ATTR_CIDR]) + if (tb[IPSET_ATTR_CIDR]) { e.cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); - if (e.cidr > HOST_MASK) - return -IPSET_ERR_INVALID_CIDR; + if (e.cidr > HOST_MASK) + return -IPSET_ERR_INVALID_CIDR; + } + ip6_netmask(&e.ip, e.cidr); - strcpy(iface, nla_data(tb[IPSET_ATTR_IFACE])); - e.iface = iface; - ret = iface_test(&h->rbtree, &e.iface); - if (adt == IPSET_ADD) { - if (!ret) { - ret = iface_add(&h->rbtree, &e.iface); - if (ret) - return ret; - } - } else if (!ret) - return ret; + nla_strlcpy(e.iface, tb[IPSET_ATTR_IFACE], IFNAMSIZ); if (tb[IPSET_ATTR_CADT_FLAGS]) { u32 cadt_flags = ip_set_get_h32(tb[IPSET_ATTR_CADT_FLAGS]); + if (cadt_flags & IPSET_FLAG_PHYSDEV) e.physdev = 1; if (cadt_flags & IPSET_FLAG_NOMATCH) @@ -633,6 +505,7 @@ hash_netiface_init(void) static void __exit hash_netiface_fini(void) { + rcu_barrier(); ip_set_type_unregister(&hash_netiface_type); } diff --git a/net/netfilter/ipset/ip_set_hash_netnet.c b/net/netfilter/ipset/ip_set_hash_netnet.c index 847047483560..3c862c0a76d1 100644 --- a/net/netfilter/ipset/ip_set_hash_netnet.c +++ b/net/netfilter/ipset/ip_set_hash_netnet.c @@ -57,8 +57,8 @@ struct hash_netnet4_elem { static inline bool hash_netnet4_data_equal(const struct hash_netnet4_elem *ip1, - const struct hash_netnet4_elem *ip2, - u32 *multi) + const struct hash_netnet4_elem *ip2, + u32 *multi) { return ip1->ipcmp == ip2->ipcmp && ip1->ccmp == ip2->ccmp; @@ -84,7 +84,7 @@ hash_netnet4_data_reset_flags(struct hash_netnet4_elem *elem, u8 *flags) static inline void hash_netnet4_data_reset_elem(struct hash_netnet4_elem *elem, - struct hash_netnet4_elem *orig) + struct hash_netnet4_elem *orig) { elem->ip[1] = orig->ip[1]; } @@ -103,7 +103,7 @@ hash_netnet4_data_netmask(struct hash_netnet4_elem *elem, u8 cidr, bool inner) static bool hash_netnet4_data_list(struct sk_buff *skb, - const struct hash_netnet4_elem *data) + const struct hash_netnet4_elem *data) { u32 flags = data->nomatch ? IPSET_FLAG_NOMATCH : 0; @@ -122,7 +122,7 @@ nla_put_failure: static inline void hash_netnet4_data_next(struct hash_netnet4_elem *next, - const struct hash_netnet4_elem *d) + const struct hash_netnet4_elem *d) { next->ipcmp = d->ipcmp; } @@ -133,16 +133,16 @@ hash_netnet4_data_next(struct hash_netnet4_elem *next, static int hash_netnet4_kadt(struct ip_set *set, const struct sk_buff *skb, - const struct xt_action_param *par, - enum ipset_adt adt, struct ip_set_adt_opt *opt) + const struct xt_action_param *par, + enum ipset_adt adt, struct ip_set_adt_opt *opt) { const struct hash_netnet *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_netnet4_elem e = { }; struct ip_set_ext ext = IP_SET_INIT_KEXT(skb, opt, set); - e.cidr[0] = IP_SET_INIT_CIDR(h->nets[0].cidr[0], HOST_MASK); - e.cidr[1] = IP_SET_INIT_CIDR(h->nets[0].cidr[1], HOST_MASK); + e.cidr[0] = INIT_CIDR(h->nets[0].cidr[0], HOST_MASK); + e.cidr[1] = INIT_CIDR(h->nets[0].cidr[1], HOST_MASK); if (adt == IPSET_TEST) e.ccmp = (HOST_MASK << (sizeof(e.cidr[0]) * 8)) | HOST_MASK; @@ -156,31 +156,23 @@ hash_netnet4_kadt(struct ip_set *set, const struct sk_buff *skb, static int hash_netnet4_uadt(struct ip_set *set, struct nlattr *tb[], - enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) + enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) { const struct hash_netnet *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; - struct hash_netnet4_elem e = { }; + struct hash_netnet4_elem e = { .cidr = { HOST_MASK, HOST_MASK, }, }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); u32 ip = 0, ip_to = 0, last; u32 ip2 = 0, ip2_from = 0, ip2_to = 0, last2; - u8 cidr, cidr2; int ret; - e.cidr[0] = e.cidr[1] = HOST_MASK; - if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IP2] || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) - return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IP2] || + !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) + return -IPSET_ERR_PROTOCOL; + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP], &ip); if (ret) return ret; @@ -194,21 +186,20 @@ hash_netnet4_uadt(struct ip_set *set, struct nlattr *tb[], return ret; if (tb[IPSET_ATTR_CIDR]) { - cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); - if (!cidr || cidr > HOST_MASK) + e.cidr[0] = nla_get_u8(tb[IPSET_ATTR_CIDR]); + if (!e.cidr[0] || e.cidr[0] > HOST_MASK) return -IPSET_ERR_INVALID_CIDR; - e.cidr[0] = cidr; } if (tb[IPSET_ATTR_CIDR2]) { - cidr2 = nla_get_u8(tb[IPSET_ATTR_CIDR2]); - if (!cidr2 || cidr2 > HOST_MASK) + e.cidr[1] = nla_get_u8(tb[IPSET_ATTR_CIDR2]); + if (!e.cidr[1] || e.cidr[1] > HOST_MASK) return -IPSET_ERR_INVALID_CIDR; - e.cidr[1] = cidr2; } if (tb[IPSET_ATTR_CADT_FLAGS]) { u32 cadt_flags = ip_set_get_h32(tb[IPSET_ATTR_CADT_FLAGS]); + if (cadt_flags & IPSET_FLAG_NOMATCH) flags |= (IPSET_FLAG_NOMATCH << 16); } @@ -231,8 +222,9 @@ hash_netnet4_uadt(struct ip_set *set, struct nlattr *tb[], swap(ip, ip_to); if (unlikely(ip + UINT_MAX == ip_to)) return -IPSET_ERR_HASH_RANGE; - } else + } else { ip_set_mask_from_to(ip, ip_to, e.cidr[0]); + } ip2_to = ip2_from; if (tb[IPSET_ATTR_IP2_TO]) { @@ -243,28 +235,27 @@ hash_netnet4_uadt(struct ip_set *set, struct nlattr *tb[], swap(ip2_from, ip2_to); if (unlikely(ip2_from + UINT_MAX == ip2_to)) return -IPSET_ERR_HASH_RANGE; - } else + } else { ip_set_mask_from_to(ip2_from, ip2_to, e.cidr[1]); + } if (retried) ip = ntohl(h->next.ip[0]); while (!after(ip, ip_to)) { e.ip[0] = htonl(ip); - last = ip_set_range_to_cidr(ip, ip_to, &cidr); - e.cidr[0] = cidr; + last = ip_set_range_to_cidr(ip, ip_to, &e.cidr[0]); ip2 = (retried && ip == ntohl(h->next.ip[0])) ? ntohl(h->next.ip[1]) : ip2_from; while (!after(ip2, ip2_to)) { e.ip[1] = htonl(ip2); - last2 = ip_set_range_to_cidr(ip2, ip2_to, &cidr2); - e.cidr[1] = cidr2; + last2 = ip_set_range_to_cidr(ip2, ip2_to, &e.cidr[1]); ret = adtfn(set, &e, &ext, &ext, flags); if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; ip2 = last2 + 1; } ip = last + 1; @@ -288,8 +279,8 @@ struct hash_netnet6_elem { static inline bool hash_netnet6_data_equal(const struct hash_netnet6_elem *ip1, - const struct hash_netnet6_elem *ip2, - u32 *multi) + const struct hash_netnet6_elem *ip2, + u32 *multi) { return ipv6_addr_equal(&ip1->ip[0].in6, &ip2->ip[0].in6) && ipv6_addr_equal(&ip1->ip[1].in6, &ip2->ip[1].in6) && @@ -316,7 +307,7 @@ hash_netnet6_data_reset_flags(struct hash_netnet6_elem *elem, u8 *flags) static inline void hash_netnet6_data_reset_elem(struct hash_netnet6_elem *elem, - struct hash_netnet6_elem *orig) + struct hash_netnet6_elem *orig) { elem->ip[1] = orig->ip[1]; } @@ -335,7 +326,7 @@ hash_netnet6_data_netmask(struct hash_netnet6_elem *elem, u8 cidr, bool inner) static bool hash_netnet6_data_list(struct sk_buff *skb, - const struct hash_netnet6_elem *data) + const struct hash_netnet6_elem *data) { u32 flags = data->nomatch ? IPSET_FLAG_NOMATCH : 0; @@ -354,7 +345,7 @@ nla_put_failure: static inline void hash_netnet6_data_next(struct hash_netnet4_elem *next, - const struct hash_netnet6_elem *d) + const struct hash_netnet6_elem *d) { } @@ -368,18 +359,18 @@ hash_netnet6_data_next(struct hash_netnet4_elem *next, static int hash_netnet6_kadt(struct ip_set *set, const struct sk_buff *skb, - const struct xt_action_param *par, - enum ipset_adt adt, struct ip_set_adt_opt *opt) + const struct xt_action_param *par, + enum ipset_adt adt, struct ip_set_adt_opt *opt) { const struct hash_netnet *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_netnet6_elem e = { }; struct ip_set_ext ext = IP_SET_INIT_KEXT(skb, opt, set); - e.cidr[0] = IP_SET_INIT_CIDR(h->nets[0].cidr[0], HOST_MASK); - e.cidr[1] = IP_SET_INIT_CIDR(h->nets[0].cidr[1], HOST_MASK); + e.cidr[0] = INIT_CIDR(h->nets[0].cidr[0], HOST_MASK); + e.cidr[1] = INIT_CIDR(h->nets[0].cidr[1], HOST_MASK); if (adt == IPSET_TEST) - e.ccmp = (HOST_MASK << (sizeof(u8)*8)) | HOST_MASK; + e.ccmp = (HOST_MASK << (sizeof(u8) * 8)) | HOST_MASK; ip6addrptr(skb, opt->flags & IPSET_DIM_ONE_SRC, &e.ip[0].in6); ip6addrptr(skb, opt->flags & IPSET_DIM_TWO_SRC, &e.ip[1].in6); @@ -391,29 +382,22 @@ hash_netnet6_kadt(struct ip_set *set, const struct sk_buff *skb, static int hash_netnet6_uadt(struct ip_set *set, struct nlattr *tb[], - enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) + enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) { ipset_adtfn adtfn = set->variant->adt[adt]; - struct hash_netnet6_elem e = { }; + struct hash_netnet6_elem e = { .cidr = { HOST_MASK, HOST_MASK, }, }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); int ret; - e.cidr[0] = e.cidr[1] = HOST_MASK; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IP2] || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) + !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) return -IPSET_ERR_PROTOCOL; if (unlikely(tb[IPSET_ATTR_IP_TO] || tb[IPSET_ATTR_IP2_TO])) return -IPSET_ERR_HASH_RANGE_UNSUPPORTED; - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); - ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &e.ip[0]); if (ret) return ret; @@ -426,21 +410,24 @@ hash_netnet6_uadt(struct ip_set *set, struct nlattr *tb[], if (ret) return ret; - if (tb[IPSET_ATTR_CIDR]) + if (tb[IPSET_ATTR_CIDR]) { e.cidr[0] = nla_get_u8(tb[IPSET_ATTR_CIDR]); + if (!e.cidr[0] || e.cidr[0] > HOST_MASK) + return -IPSET_ERR_INVALID_CIDR; + } - if (tb[IPSET_ATTR_CIDR2]) + if (tb[IPSET_ATTR_CIDR2]) { e.cidr[1] = nla_get_u8(tb[IPSET_ATTR_CIDR2]); - - if (!e.cidr[0] || e.cidr[0] > HOST_MASK || !e.cidr[1] || - e.cidr[1] > HOST_MASK) - return -IPSET_ERR_INVALID_CIDR; + if (!e.cidr[1] || e.cidr[1] > HOST_MASK) + return -IPSET_ERR_INVALID_CIDR; + } ip6_netmask(&e.ip[0], e.cidr[0]); ip6_netmask(&e.ip[1], e.cidr[1]); if (tb[IPSET_ATTR_CADT_FLAGS]) { u32 cadt_flags = ip_set_get_h32(tb[IPSET_ATTR_CADT_FLAGS]); + if (cadt_flags & IPSET_FLAG_NOMATCH) flags |= (IPSET_FLAG_NOMATCH << 16); } @@ -497,6 +484,7 @@ hash_netnet_init(void) static void __exit hash_netnet_fini(void) { + rcu_barrier(); ip_set_type_unregister(&hash_netnet_type); } diff --git a/net/netfilter/ipset/ip_set_hash_netport.c b/net/netfilter/ipset/ip_set_hash_netport.c index 8273819c1a2f..731813e0f08c 100644 --- a/net/netfilter/ipset/ip_set_hash_netport.c +++ b/net/netfilter/ipset/ip_set_hash_netport.c @@ -136,7 +136,7 @@ hash_netport4_kadt(struct ip_set *set, const struct sk_buff *skb, const struct hash_netport *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_netport4_elem e = { - .cidr = IP_SET_INIT_CIDR(h->nets[0].cidr[0], HOST_MASK) - 1, + .cidr = INIT_CIDR(h->nets[0].cidr[0], HOST_MASK), }; struct ip_set_ext ext = IP_SET_INIT_KEXT(skb, opt, set); @@ -166,21 +166,15 @@ hash_netport4_uadt(struct ip_set *set, struct nlattr *tb[], u8 cidr; int ret; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) + !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); - ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP], &ip); if (ret) return ret; @@ -204,8 +198,9 @@ hash_netport4_uadt(struct ip_set *set, struct nlattr *tb[], if (e.proto == 0) return -IPSET_ERR_INVALID_PROTO; - } else + } else { return -IPSET_ERR_MISSING_PROTO; + } if (!(with_ports || e.proto == IPPROTO_ICMP)) e.port = 0; @@ -214,6 +209,7 @@ hash_netport4_uadt(struct ip_set *set, struct nlattr *tb[], if (tb[IPSET_ATTR_CADT_FLAGS]) { u32 cadt_flags = ip_set_get_h32(tb[IPSET_ATTR_CADT_FLAGS]); + if (cadt_flags & IPSET_FLAG_NOMATCH) flags |= (IPSET_FLAG_NOMATCH << 16); } @@ -239,8 +235,9 @@ hash_netport4_uadt(struct ip_set *set, struct nlattr *tb[], swap(ip, ip_to); if (ip + UINT_MAX == ip_to) return -IPSET_ERR_HASH_RANGE; - } else + } else { ip_set_mask_from_to(ip, ip_to, e.cidr + 1); + } if (retried) ip = ntohl(h->next.ip); @@ -256,8 +253,8 @@ hash_netport4_uadt(struct ip_set *set, struct nlattr *tb[], if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; } ip = last + 1; } @@ -354,7 +351,7 @@ hash_netport6_kadt(struct ip_set *set, const struct sk_buff *skb, const struct hash_netport *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_netport6_elem e = { - .cidr = IP_SET_INIT_CIDR(h->nets[0].cidr[0], HOST_MASK) - 1, + .cidr = INIT_CIDR(h->nets[0].cidr[0], HOST_MASK), }; struct ip_set_ext ext = IP_SET_INIT_KEXT(skb, opt, set); @@ -384,23 +381,17 @@ hash_netport6_uadt(struct ip_set *set, struct nlattr *tb[], u8 cidr; int ret; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) + !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) return -IPSET_ERR_PROTOCOL; if (unlikely(tb[IPSET_ATTR_IP_TO])) return -IPSET_ERR_HASH_RANGE_UNSUPPORTED; - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); - ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &e.ip); if (ret) return ret; @@ -425,14 +416,16 @@ hash_netport6_uadt(struct ip_set *set, struct nlattr *tb[], if (e.proto == 0) return -IPSET_ERR_INVALID_PROTO; - } else + } else { return -IPSET_ERR_MISSING_PROTO; + } if (!(with_ports || e.proto == IPPROTO_ICMPV6)) e.port = 0; if (tb[IPSET_ATTR_CADT_FLAGS]) { u32 cadt_flags = ip_set_get_h32(tb[IPSET_ATTR_CADT_FLAGS]); + if (cadt_flags & IPSET_FLAG_NOMATCH) flags |= (IPSET_FLAG_NOMATCH << 16); } @@ -456,8 +449,8 @@ hash_netport6_uadt(struct ip_set *set, struct nlattr *tb[], if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; } return ret; } @@ -510,6 +503,7 @@ hash_netport_init(void) static void __exit hash_netport_fini(void) { + rcu_barrier(); ip_set_type_unregister(&hash_netport_type); } diff --git a/net/netfilter/ipset/ip_set_hash_netportnet.c b/net/netfilter/ipset/ip_set_hash_netportnet.c index 1451a8ac938f..0c68734f5cc4 100644 --- a/net/netfilter/ipset/ip_set_hash_netportnet.c +++ b/net/netfilter/ipset/ip_set_hash_netportnet.c @@ -62,8 +62,8 @@ struct hash_netportnet4_elem { static inline bool hash_netportnet4_data_equal(const struct hash_netportnet4_elem *ip1, - const struct hash_netportnet4_elem *ip2, - u32 *multi) + const struct hash_netportnet4_elem *ip2, + u32 *multi) { return ip1->ipcmp == ip2->ipcmp && ip1->ccmp == ip2->ccmp && @@ -91,7 +91,7 @@ hash_netportnet4_data_reset_flags(struct hash_netportnet4_elem *elem, u8 *flags) static inline void hash_netportnet4_data_reset_elem(struct hash_netportnet4_elem *elem, - struct hash_netportnet4_elem *orig) + struct hash_netportnet4_elem *orig) { elem->ip[1] = orig->ip[1]; } @@ -111,7 +111,7 @@ hash_netportnet4_data_netmask(struct hash_netportnet4_elem *elem, static bool hash_netportnet4_data_list(struct sk_buff *skb, - const struct hash_netportnet4_elem *data) + const struct hash_netportnet4_elem *data) { u32 flags = data->nomatch ? IPSET_FLAG_NOMATCH : 0; @@ -132,7 +132,7 @@ nla_put_failure: static inline void hash_netportnet4_data_next(struct hash_netportnet4_elem *next, - const struct hash_netportnet4_elem *d) + const struct hash_netportnet4_elem *d) { next->ipcmp = d->ipcmp; next->port = d->port; @@ -144,16 +144,16 @@ hash_netportnet4_data_next(struct hash_netportnet4_elem *next, static int hash_netportnet4_kadt(struct ip_set *set, const struct sk_buff *skb, - const struct xt_action_param *par, - enum ipset_adt adt, struct ip_set_adt_opt *opt) + const struct xt_action_param *par, + enum ipset_adt adt, struct ip_set_adt_opt *opt) { const struct hash_netportnet *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_netportnet4_elem e = { }; struct ip_set_ext ext = IP_SET_INIT_KEXT(skb, opt, set); - e.cidr[0] = IP_SET_INIT_CIDR(h->nets[0].cidr[0], HOST_MASK); - e.cidr[1] = IP_SET_INIT_CIDR(h->nets[0].cidr[1], HOST_MASK); + e.cidr[0] = INIT_CIDR(h->nets[0].cidr[0], HOST_MASK); + e.cidr[1] = INIT_CIDR(h->nets[0].cidr[1], HOST_MASK); if (adt == IPSET_TEST) e.ccmp = (HOST_MASK << (sizeof(e.cidr[0]) * 8)) | HOST_MASK; @@ -171,34 +171,26 @@ hash_netportnet4_kadt(struct ip_set *set, const struct sk_buff *skb, static int hash_netportnet4_uadt(struct ip_set *set, struct nlattr *tb[], - enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) + enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) { const struct hash_netportnet *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; - struct hash_netportnet4_elem e = { }; + struct hash_netportnet4_elem e = { .cidr = { HOST_MASK, HOST_MASK, }, }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); u32 ip = 0, ip_to = 0, ip_last, p = 0, port, port_to; u32 ip2_from = 0, ip2_to = 0, ip2_last, ip2; bool with_ports = false; - u8 cidr, cidr2; int ret; - e.cidr[0] = e.cidr[1] = HOST_MASK; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IP2] || !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) + !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); - ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP], &ip); if (ret) return ret; @@ -212,17 +204,15 @@ hash_netportnet4_uadt(struct ip_set *set, struct nlattr *tb[], return ret; if (tb[IPSET_ATTR_CIDR]) { - cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); - if (!cidr || cidr > HOST_MASK) + e.cidr[0] = nla_get_u8(tb[IPSET_ATTR_CIDR]); + if (!e.cidr[0] || e.cidr[0] > HOST_MASK) return -IPSET_ERR_INVALID_CIDR; - e.cidr[0] = cidr; } if (tb[IPSET_ATTR_CIDR2]) { - cidr = nla_get_u8(tb[IPSET_ATTR_CIDR2]); - if (!cidr || cidr > HOST_MASK) + e.cidr[1] = nla_get_u8(tb[IPSET_ATTR_CIDR2]); + if (!e.cidr[1] || e.cidr[1] > HOST_MASK) return -IPSET_ERR_INVALID_CIDR; - e.cidr[1] = cidr; } e.port = nla_get_be16(tb[IPSET_ATTR_PORT]); @@ -233,14 +223,16 @@ hash_netportnet4_uadt(struct ip_set *set, struct nlattr *tb[], if (e.proto == 0) return -IPSET_ERR_INVALID_PROTO; - } else + } else { return -IPSET_ERR_MISSING_PROTO; + } if (!(with_ports || e.proto == IPPROTO_ICMP)) e.port = 0; if (tb[IPSET_ATTR_CADT_FLAGS]) { u32 cadt_flags = ip_set_get_h32(tb[IPSET_ATTR_CADT_FLAGS]); + if (cadt_flags & IPSET_FLAG_NOMATCH) flags |= (IPSET_FLAG_NOMATCH << 16); } @@ -264,8 +256,9 @@ hash_netportnet4_uadt(struct ip_set *set, struct nlattr *tb[], swap(ip, ip_to); if (unlikely(ip + UINT_MAX == ip_to)) return -IPSET_ERR_HASH_RANGE; - } else + } else { ip_set_mask_from_to(ip, ip_to, e.cidr[0]); + } port_to = port = ntohs(e.port); if (tb[IPSET_ATTR_PORT_TO]) { @@ -283,16 +276,16 @@ hash_netportnet4_uadt(struct ip_set *set, struct nlattr *tb[], swap(ip2_from, ip2_to); if (unlikely(ip2_from + UINT_MAX == ip2_to)) return -IPSET_ERR_HASH_RANGE; - } else + } else { ip_set_mask_from_to(ip2_from, ip2_to, e.cidr[1]); + } if (retried) ip = ntohl(h->next.ip[0]); while (!after(ip, ip_to)) { e.ip[0] = htonl(ip); - ip_last = ip_set_range_to_cidr(ip, ip_to, &cidr); - e.cidr[0] = cidr; + ip_last = ip_set_range_to_cidr(ip, ip_to, &e.cidr[0]); p = retried && ip == ntohl(h->next.ip[0]) ? ntohs(h->next.port) : port; for (; p <= port_to; p++) { @@ -303,13 +296,12 @@ hash_netportnet4_uadt(struct ip_set *set, struct nlattr *tb[], while (!after(ip2, ip2_to)) { e.ip[1] = htonl(ip2); ip2_last = ip_set_range_to_cidr(ip2, ip2_to, - &cidr2); - e.cidr[1] = cidr2; + &e.cidr[1]); ret = adtfn(set, &e, &ext, &ext, flags); if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; ip2 = ip2_last + 1; } } @@ -336,8 +328,8 @@ struct hash_netportnet6_elem { static inline bool hash_netportnet6_data_equal(const struct hash_netportnet6_elem *ip1, - const struct hash_netportnet6_elem *ip2, - u32 *multi) + const struct hash_netportnet6_elem *ip2, + u32 *multi) { return ipv6_addr_equal(&ip1->ip[0].in6, &ip2->ip[0].in6) && ipv6_addr_equal(&ip1->ip[1].in6, &ip2->ip[1].in6) && @@ -366,7 +358,7 @@ hash_netportnet6_data_reset_flags(struct hash_netportnet6_elem *elem, u8 *flags) static inline void hash_netportnet6_data_reset_elem(struct hash_netportnet6_elem *elem, - struct hash_netportnet6_elem *orig) + struct hash_netportnet6_elem *orig) { elem->ip[1] = orig->ip[1]; } @@ -386,7 +378,7 @@ hash_netportnet6_data_netmask(struct hash_netportnet6_elem *elem, static bool hash_netportnet6_data_list(struct sk_buff *skb, - const struct hash_netportnet6_elem *data) + const struct hash_netportnet6_elem *data) { u32 flags = data->nomatch ? IPSET_FLAG_NOMATCH : 0; @@ -407,7 +399,7 @@ nla_put_failure: static inline void hash_netportnet6_data_next(struct hash_netportnet4_elem *next, - const struct hash_netportnet6_elem *d) + const struct hash_netportnet6_elem *d) { next->port = d->port; } @@ -422,16 +414,16 @@ hash_netportnet6_data_next(struct hash_netportnet4_elem *next, static int hash_netportnet6_kadt(struct ip_set *set, const struct sk_buff *skb, - const struct xt_action_param *par, - enum ipset_adt adt, struct ip_set_adt_opt *opt) + const struct xt_action_param *par, + enum ipset_adt adt, struct ip_set_adt_opt *opt) { const struct hash_netportnet *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_netportnet6_elem e = { }; struct ip_set_ext ext = IP_SET_INIT_KEXT(skb, opt, set); - e.cidr[0] = IP_SET_INIT_CIDR(h->nets[0].cidr[0], HOST_MASK); - e.cidr[1] = IP_SET_INIT_CIDR(h->nets[0].cidr[1], HOST_MASK); + e.cidr[0] = INIT_CIDR(h->nets[0].cidr[0], HOST_MASK); + e.cidr[1] = INIT_CIDR(h->nets[0].cidr[1], HOST_MASK); if (adt == IPSET_TEST) e.ccmp = (HOST_MASK << (sizeof(u8) * 8)) | HOST_MASK; @@ -449,34 +441,27 @@ hash_netportnet6_kadt(struct ip_set *set, const struct sk_buff *skb, static int hash_netportnet6_uadt(struct ip_set *set, struct nlattr *tb[], - enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) + enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) { const struct hash_netportnet *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; - struct hash_netportnet6_elem e = { }; + struct hash_netportnet6_elem e = { .cidr = { HOST_MASK, HOST_MASK, }, }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); u32 port, port_to; bool with_ports = false; int ret; - e.cidr[0] = e.cidr[1] = HOST_MASK; + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IP2] || !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) + !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) return -IPSET_ERR_PROTOCOL; if (unlikely(tb[IPSET_ATTR_IP_TO] || tb[IPSET_ATTR_IP2_TO])) return -IPSET_ERR_HASH_RANGE_UNSUPPORTED; - if (tb[IPSET_ATTR_LINENO]) - *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); - ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &e.ip[0]); if (ret) return ret; @@ -489,15 +474,17 @@ hash_netportnet6_uadt(struct ip_set *set, struct nlattr *tb[], if (ret) return ret; - if (tb[IPSET_ATTR_CIDR]) + if (tb[IPSET_ATTR_CIDR]) { e.cidr[0] = nla_get_u8(tb[IPSET_ATTR_CIDR]); + if (!e.cidr[0] || e.cidr[0] > HOST_MASK) + return -IPSET_ERR_INVALID_CIDR; + } - if (tb[IPSET_ATTR_CIDR2]) + if (tb[IPSET_ATTR_CIDR2]) { e.cidr[1] = nla_get_u8(tb[IPSET_ATTR_CIDR2]); - - if (unlikely(!e.cidr[0] || e.cidr[0] > HOST_MASK || !e.cidr[1] || - e.cidr[1] > HOST_MASK)) - return -IPSET_ERR_INVALID_CIDR; + if (!e.cidr[1] || e.cidr[1] > HOST_MASK) + return -IPSET_ERR_INVALID_CIDR; + } ip6_netmask(&e.ip[0], e.cidr[0]); ip6_netmask(&e.ip[1], e.cidr[1]); @@ -510,14 +497,16 @@ hash_netportnet6_uadt(struct ip_set *set, struct nlattr *tb[], if (e.proto == 0) return -IPSET_ERR_INVALID_PROTO; - } else + } else { return -IPSET_ERR_MISSING_PROTO; + } if (!(with_ports || e.proto == IPPROTO_ICMPV6)) e.port = 0; if (tb[IPSET_ATTR_CADT_FLAGS]) { u32 cadt_flags = ip_set_get_h32(tb[IPSET_ATTR_CADT_FLAGS]); + if (cadt_flags & IPSET_FLAG_NOMATCH) flags |= (IPSET_FLAG_NOMATCH << 16); } @@ -541,8 +530,8 @@ hash_netportnet6_uadt(struct ip_set *set, struct nlattr *tb[], if (ret && !ip_set_eexist(ret, flags)) return ret; - else - ret = 0; + + ret = 0; } return ret; } @@ -598,6 +587,7 @@ hash_netportnet_init(void) static void __exit hash_netportnet_fini(void) { + rcu_barrier(); ip_set_type_unregister(&hash_netportnet_type); } diff --git a/net/netfilter/ipset/ip_set_list_set.c b/net/netfilter/ipset/ip_set_list_set.c index 5bd3b1eae3fa..a1fe5377a2b3 100644 --- a/net/netfilter/ipset/ip_set_list_set.c +++ b/net/netfilter/ipset/ip_set_list_set.c @@ -9,6 +9,7 @@ #include <linux/module.h> #include <linux/ip.h> +#include <linux/rculist.h> #include <linux/skbuff.h> #include <linux/errno.h> @@ -27,6 +28,8 @@ MODULE_ALIAS("ip_set_list:set"); /* Member elements */ struct set_elem { + struct rcu_head rcu; + struct list_head list; ip_set_id_t id; }; @@ -41,12 +44,9 @@ struct list_set { u32 size; /* size of set list array */ struct timer_list gc; /* garbage collection */ struct net *net; /* namespace */ - struct set_elem members[0]; /* the set members */ + struct list_head members; /* the set members */ }; -#define list_set_elem(set, map, id) \ - (struct set_elem *)((void *)(map)->members + (id) * (set)->dsize) - static int list_set_ktest(struct ip_set *set, const struct sk_buff *skb, const struct xt_action_param *par, @@ -54,17 +54,14 @@ list_set_ktest(struct ip_set *set, const struct sk_buff *skb, { struct list_set *map = set->data; struct set_elem *e; - u32 i, cmdflags = opt->cmdflags; + u32 cmdflags = opt->cmdflags; int ret; /* Don't lookup sub-counters at all */ opt->cmdflags &= ~IPSET_FLAG_MATCH_COUNTERS; if (opt->cmdflags & IPSET_FLAG_SKIP_SUBCOUNTER_UPDATE) opt->cmdflags &= ~IPSET_FLAG_SKIP_COUNTER_UPDATE; - for (i = 0; i < map->size; i++) { - e = list_set_elem(set, map, i); - if (e->id == IPSET_INVALID_ID) - return 0; + list_for_each_entry_rcu(e, &map->members, list) { if (SET_WITH_TIMEOUT(set) && ip_set_timeout_expired(ext_timeout(e, set))) continue; @@ -91,13 +88,9 @@ list_set_kadd(struct ip_set *set, const struct sk_buff *skb, { struct list_set *map = set->data; struct set_elem *e; - u32 i; int ret; - for (i = 0; i < map->size; i++) { - e = list_set_elem(set, map, i); - if (e->id == IPSET_INVALID_ID) - return 0; + list_for_each_entry(e, &map->members, list) { if (SET_WITH_TIMEOUT(set) && ip_set_timeout_expired(ext_timeout(e, set))) continue; @@ -115,13 +108,9 @@ list_set_kdel(struct ip_set *set, const struct sk_buff *skb, { struct list_set *map = set->data; struct set_elem *e; - u32 i; int ret; - for (i = 0; i < map->size; i++) { - e = list_set_elem(set, map, i); - if (e->id == IPSET_INVALID_ID) - return 0; + list_for_each_entry(e, &map->members, list) { if (SET_WITH_TIMEOUT(set) && ip_set_timeout_expired(ext_timeout(e, set))) continue; @@ -138,110 +127,65 @@ list_set_kadt(struct ip_set *set, const struct sk_buff *skb, enum ipset_adt adt, struct ip_set_adt_opt *opt) { struct ip_set_ext ext = IP_SET_INIT_KEXT(skb, opt, set); + int ret = -EINVAL; + rcu_read_lock(); switch (adt) { case IPSET_TEST: - return list_set_ktest(set, skb, par, opt, &ext); + ret = list_set_ktest(set, skb, par, opt, &ext); + break; case IPSET_ADD: - return list_set_kadd(set, skb, par, opt, &ext); + ret = list_set_kadd(set, skb, par, opt, &ext); + break; case IPSET_DEL: - return list_set_kdel(set, skb, par, opt, &ext); + ret = list_set_kdel(set, skb, par, opt, &ext); + break; default: break; } - return -EINVAL; -} - -static bool -id_eq(const struct ip_set *set, u32 i, ip_set_id_t id) -{ - const struct list_set *map = set->data; - const struct set_elem *e; - - if (i >= map->size) - return 0; + rcu_read_unlock(); - e = list_set_elem(set, map, i); - return !!(e->id == id && - !(SET_WITH_TIMEOUT(set) && - ip_set_timeout_expired(ext_timeout(e, set)))); + return ret; } -static int -list_set_add(struct ip_set *set, u32 i, struct set_adt_elem *d, - const struct ip_set_ext *ext) -{ - struct list_set *map = set->data; - struct set_elem *e = list_set_elem(set, map, i); +/* Userspace interfaces: we are protected by the nfnl mutex */ - if (e->id != IPSET_INVALID_ID) { - if (i == map->size - 1) { - /* Last element replaced: e.g. add new,before,last */ - ip_set_put_byindex(map->net, e->id); - ip_set_ext_destroy(set, e); - } else { - struct set_elem *x = list_set_elem(set, map, - map->size - 1); - - /* Last element pushed off */ - if (x->id != IPSET_INVALID_ID) { - ip_set_put_byindex(map->net, x->id); - ip_set_ext_destroy(set, x); - } - memmove(list_set_elem(set, map, i + 1), e, - set->dsize * (map->size - (i + 1))); - /* Extensions must be initialized to zero */ - memset(e, 0, set->dsize); - } - } - - e->id = d->id; - if (SET_WITH_TIMEOUT(set)) - ip_set_timeout_set(ext_timeout(e, set), ext->timeout); - if (SET_WITH_COUNTER(set)) - ip_set_init_counter(ext_counter(e, set), ext); - if (SET_WITH_COMMENT(set)) - ip_set_init_comment(ext_comment(e, set), ext); - if (SET_WITH_SKBINFO(set)) - ip_set_init_skbinfo(ext_skbinfo(e, set), ext); - return 0; -} - -static int -list_set_del(struct ip_set *set, u32 i) +static void +__list_set_del(struct ip_set *set, struct set_elem *e) { struct list_set *map = set->data; - struct set_elem *e = list_set_elem(set, map, i); ip_set_put_byindex(map->net, e->id); + /* We may call it, because we don't have a to be destroyed + * extension which is used by the kernel. + */ ip_set_ext_destroy(set, e); + kfree_rcu(e, rcu); +} - if (i < map->size - 1) - memmove(e, list_set_elem(set, map, i + 1), - set->dsize * (map->size - (i + 1))); +static inline void +list_set_del(struct ip_set *set, struct set_elem *e) +{ + list_del_rcu(&e->list); + __list_set_del(set, e); +} - /* Last element */ - e = list_set_elem(set, map, map->size - 1); - e->id = IPSET_INVALID_ID; - return 0; +static inline void +list_set_replace(struct ip_set *set, struct set_elem *e, struct set_elem *old) +{ + list_replace_rcu(&old->list, &e->list); + __list_set_del(set, old); } static void set_cleanup_entries(struct ip_set *set) { struct list_set *map = set->data; - struct set_elem *e; - u32 i = 0; + struct set_elem *e, *n; - while (i < map->size) { - e = list_set_elem(set, map, i); - if (e->id != IPSET_INVALID_ID && - ip_set_timeout_expired(ext_timeout(e, set))) - list_set_del(set, i); - /* Check element moved to position i in next loop */ - else - i++; - } + list_for_each_entry_safe(e, n, &map->members, list) + if (ip_set_timeout_expired(ext_timeout(e, set))) + list_set_del(set, e); } static int @@ -250,31 +194,46 @@ list_set_utest(struct ip_set *set, void *value, const struct ip_set_ext *ext, { struct list_set *map = set->data; struct set_adt_elem *d = value; - struct set_elem *e; - u32 i; + struct set_elem *e, *next, *prev = NULL; int ret; - for (i = 0; i < map->size; i++) { - e = list_set_elem(set, map, i); - if (e->id == IPSET_INVALID_ID) - return 0; - else if (SET_WITH_TIMEOUT(set) && - ip_set_timeout_expired(ext_timeout(e, set))) + list_for_each_entry(e, &map->members, list) { + if (SET_WITH_TIMEOUT(set) && + ip_set_timeout_expired(ext_timeout(e, set))) continue; - else if (e->id != d->id) + else if (e->id != d->id) { + prev = e; continue; + } - if (d->before == 0) - return 1; - else if (d->before > 0) - ret = id_eq(set, i + 1, d->refid); - else - ret = i > 0 && id_eq(set, i - 1, d->refid); + if (d->before == 0) { + ret = 1; + } else if (d->before > 0) { + next = list_next_entry(e, list); + ret = !list_is_last(&e->list, &map->members) && + next->id == d->refid; + } else { + ret = prev && prev->id == d->refid; + } return ret; } return 0; } +static void +list_set_init_extensions(struct ip_set *set, const struct ip_set_ext *ext, + struct set_elem *e) +{ + if (SET_WITH_COUNTER(set)) + ip_set_init_counter(ext_counter(e, set), ext); + if (SET_WITH_COMMENT(set)) + ip_set_init_comment(ext_comment(e, set), ext); + if (SET_WITH_SKBINFO(set)) + ip_set_init_skbinfo(ext_skbinfo(e, set), ext); + /* Update timeout last */ + if (SET_WITH_TIMEOUT(set)) + ip_set_timeout_set(ext_timeout(e, set), ext->timeout); +} static int list_set_uadd(struct ip_set *set, void *value, const struct ip_set_ext *ext, @@ -282,60 +241,78 @@ list_set_uadd(struct ip_set *set, void *value, const struct ip_set_ext *ext, { struct list_set *map = set->data; struct set_adt_elem *d = value; - struct set_elem *e; + struct set_elem *e, *n, *prev, *next; bool flag_exist = flags & IPSET_FLAG_EXIST; - u32 i, ret = 0; if (SET_WITH_TIMEOUT(set)) set_cleanup_entries(set); - /* Check already added element */ - for (i = 0; i < map->size; i++) { - e = list_set_elem(set, map, i); - if (e->id == IPSET_INVALID_ID) - goto insert; - else if (e->id != d->id) + /* Find where to add the new entry */ + n = prev = next = NULL; + list_for_each_entry(e, &map->members, list) { + if (SET_WITH_TIMEOUT(set) && + ip_set_timeout_expired(ext_timeout(e, set))) continue; - - if ((d->before > 1 && !id_eq(set, i + 1, d->refid)) || - (d->before < 0 && - (i == 0 || !id_eq(set, i - 1, d->refid)))) - /* Before/after doesn't match */ + else if (d->id == e->id) + n = e; + else if (d->before == 0 || e->id != d->refid) + continue; + else if (d->before > 0) + next = e; + else + prev = e; + } + /* Re-add already existing element */ + if (n) { + if ((d->before > 0 && !next) || + (d->before < 0 && !prev)) return -IPSET_ERR_REF_EXIST; if (!flag_exist) - /* Can't re-add */ return -IPSET_ERR_EXIST; /* Update extensions */ - ip_set_ext_destroy(set, e); + ip_set_ext_destroy(set, n); + list_set_init_extensions(set, ext, n); - if (SET_WITH_TIMEOUT(set)) - ip_set_timeout_set(ext_timeout(e, set), ext->timeout); - if (SET_WITH_COUNTER(set)) - ip_set_init_counter(ext_counter(e, set), ext); - if (SET_WITH_COMMENT(set)) - ip_set_init_comment(ext_comment(e, set), ext); - if (SET_WITH_SKBINFO(set)) - ip_set_init_skbinfo(ext_skbinfo(e, set), ext); /* Set is already added to the list */ ip_set_put_byindex(map->net, d->id); return 0; } -insert: - ret = -IPSET_ERR_LIST_FULL; - for (i = 0; i < map->size && ret == -IPSET_ERR_LIST_FULL; i++) { - e = list_set_elem(set, map, i); - if (e->id == IPSET_INVALID_ID) - ret = d->before != 0 ? -IPSET_ERR_REF_EXIST - : list_set_add(set, i, d, ext); - else if (e->id != d->refid) - continue; - else if (d->before > 0) - ret = list_set_add(set, i, d, ext); - else if (i + 1 < map->size) - ret = list_set_add(set, i + 1, d, ext); + /* Add new entry */ + if (d->before == 0) { + /* Append */ + n = list_empty(&map->members) ? NULL : + list_last_entry(&map->members, struct set_elem, list); + } else if (d->before > 0) { + /* Insert after next element */ + if (!list_is_last(&next->list, &map->members)) + n = list_next_entry(next, list); + } else { + /* Insert before prev element */ + if (prev->list.prev != &map->members) + n = list_prev_entry(prev, list); } + /* Can we replace a timed out entry? */ + if (n && + !(SET_WITH_TIMEOUT(set) && + ip_set_timeout_expired(ext_timeout(n, set)))) + n = NULL; + + e = kzalloc(set->dsize, GFP_KERNEL); + if (!e) + return -ENOMEM; + e->id = d->id; + INIT_LIST_HEAD(&e->list); + list_set_init_extensions(set, ext, e); + if (n) + list_set_replace(set, e, n); + else if (next) + list_add_tail_rcu(&e->list, &next->list); + else if (prev) + list_add_rcu(&e->list, &prev->list); + else + list_add_tail_rcu(&e->list, &map->members); - return ret; + return 0; } static int @@ -344,32 +321,30 @@ list_set_udel(struct ip_set *set, void *value, const struct ip_set_ext *ext, { struct list_set *map = set->data; struct set_adt_elem *d = value; - struct set_elem *e; - u32 i; - - for (i = 0; i < map->size; i++) { - e = list_set_elem(set, map, i); - if (e->id == IPSET_INVALID_ID) - return d->before != 0 ? -IPSET_ERR_REF_EXIST - : -IPSET_ERR_EXIST; - else if (SET_WITH_TIMEOUT(set) && - ip_set_timeout_expired(ext_timeout(e, set))) + struct set_elem *e, *next, *prev = NULL; + + list_for_each_entry(e, &map->members, list) { + if (SET_WITH_TIMEOUT(set) && + ip_set_timeout_expired(ext_timeout(e, set))) continue; - else if (e->id != d->id) + else if (e->id != d->id) { + prev = e; continue; + } - if (d->before == 0) - return list_set_del(set, i); - else if (d->before > 0) { - if (!id_eq(set, i + 1, d->refid)) + if (d->before > 0) { + next = list_next_entry(e, list); + if (list_is_last(&e->list, &map->members) || + next->id != d->refid) return -IPSET_ERR_REF_EXIST; - return list_set_del(set, i); - } else if (i == 0 || !id_eq(set, i - 1, d->refid)) - return -IPSET_ERR_REF_EXIST; - else - return list_set_del(set, i); + } else if (d->before < 0) { + if (!prev || prev->id != d->refid) + return -IPSET_ERR_REF_EXIST; + } + list_set_del(set, e); + return 0; } - return -IPSET_ERR_EXIST; + return d->before != 0 ? -IPSET_ERR_REF_EXIST : -IPSET_ERR_EXIST; } static int @@ -383,19 +358,13 @@ list_set_uadt(struct ip_set *set, struct nlattr *tb[], struct ip_set *s; int ret = 0; - if (unlikely(!tb[IPSET_ATTR_NAME] || - !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_PACKETS) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_BYTES) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBMARK) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBPRIO) || - !ip_set_optattr_netorder(tb, IPSET_ATTR_SKBQUEUE))) - return -IPSET_ERR_PROTOCOL; - if (tb[IPSET_ATTR_LINENO]) *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + if (unlikely(!tb[IPSET_ATTR_NAME] || + !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) + return -IPSET_ERR_PROTOCOL; + ret = ip_set_get_extensions(set, tb, &ext); if (ret) return ret; @@ -410,6 +379,7 @@ list_set_uadt(struct ip_set *set, struct nlattr *tb[], if (tb[IPSET_ATTR_CADT_FLAGS]) { u32 f = ip_set_get_h32(tb[IPSET_ATTR_CADT_FLAGS]); + e.before = f & IPSET_FLAG_BEFORE; } @@ -447,27 +417,26 @@ static void list_set_flush(struct ip_set *set) { struct list_set *map = set->data; - struct set_elem *e; - u32 i; - - for (i = 0; i < map->size; i++) { - e = list_set_elem(set, map, i); - if (e->id != IPSET_INVALID_ID) { - ip_set_put_byindex(map->net, e->id); - ip_set_ext_destroy(set, e); - e->id = IPSET_INVALID_ID; - } - } + struct set_elem *e, *n; + + list_for_each_entry_safe(e, n, &map->members, list) + list_set_del(set, e); } static void list_set_destroy(struct ip_set *set) { struct list_set *map = set->data; + struct set_elem *e, *n; if (SET_WITH_TIMEOUT(set)) del_timer_sync(&map->gc); - list_set_flush(set); + list_for_each_entry_safe(e, n, &map->members, list) { + list_del(&e->list); + ip_set_put_byindex(map->net, e->id); + ip_set_ext_destroy(set, e); + kfree(e); + } kfree(map); set->data = NULL; @@ -478,6 +447,11 @@ list_set_head(struct ip_set *set, struct sk_buff *skb) { const struct list_set *map = set->data; struct nlattr *nested; + struct set_elem *e; + u32 n = 0; + + list_for_each_entry(e, &map->members, list) + n++; nested = ipset_nest_start(skb, IPSET_ATTR_DATA); if (!nested) @@ -485,7 +459,7 @@ list_set_head(struct ip_set *set, struct sk_buff *skb) if (nla_put_net32(skb, IPSET_ATTR_SIZE, htonl(map->size)) || nla_put_net32(skb, IPSET_ATTR_REFERENCES, htonl(set->ref - 1)) || nla_put_net32(skb, IPSET_ATTR_MEMSIZE, - htonl(sizeof(*map) + map->size * set->dsize))) + htonl(sizeof(*map) + n * set->dsize))) goto nla_put_failure; if (unlikely(ip_set_put_flags(skb, set))) goto nla_put_failure; @@ -502,18 +476,22 @@ list_set_list(const struct ip_set *set, { const struct list_set *map = set->data; struct nlattr *atd, *nested; - u32 i, first = cb->args[IPSET_CB_ARG0]; - const struct set_elem *e; + u32 i = 0, first = cb->args[IPSET_CB_ARG0]; + struct set_elem *e; + int ret = 0; atd = ipset_nest_start(skb, IPSET_ATTR_ADT); if (!atd) return -EMSGSIZE; - for (; cb->args[IPSET_CB_ARG0] < map->size; - cb->args[IPSET_CB_ARG0]++) { - i = cb->args[IPSET_CB_ARG0]; - e = list_set_elem(set, map, i); - if (e->id == IPSET_INVALID_ID) - goto finish; + list_for_each_entry(e, &map->members, list) { + if (i == first) + break; + i++; + } + + rcu_read_lock(); + list_for_each_entry_from(e, &map->members, list) { + i++; if (SET_WITH_TIMEOUT(set) && ip_set_timeout_expired(ext_timeout(e, set))) continue; @@ -521,9 +499,10 @@ list_set_list(const struct ip_set *set, if (!nested) { if (i == first) { nla_nest_cancel(skb, atd); - return -EMSGSIZE; - } else - goto nla_put_failure; + ret = -EMSGSIZE; + goto out; + } + goto nla_put_failure; } if (nla_put_string(skb, IPSET_ATTR_NAME, ip_set_name_byindex(map->net, e->id))) @@ -532,20 +511,23 @@ list_set_list(const struct ip_set *set, goto nla_put_failure; ipset_nest_end(skb, nested); } -finish: + ipset_nest_end(skb, atd); /* Set listing finished */ cb->args[IPSET_CB_ARG0] = 0; - return 0; + goto out; nla_put_failure: nla_nest_cancel(skb, nested); if (unlikely(i == first)) { cb->args[IPSET_CB_ARG0] = 0; - return -EMSGSIZE; + ret = -EMSGSIZE; } + cb->args[IPSET_CB_ARG0] = i - 1; ipset_nest_end(skb, atd); - return 0; +out: + rcu_read_unlock(); + return ret; } static bool @@ -577,12 +559,12 @@ static const struct ip_set_type_variant set_variant = { static void list_set_gc(unsigned long ul_set) { - struct ip_set *set = (struct ip_set *) ul_set; + struct ip_set *set = (struct ip_set *)ul_set; struct list_set *map = set->data; - write_lock_bh(&set->lock); + spin_lock_bh(&set->lock); set_cleanup_entries(set); - write_unlock_bh(&set->lock); + spin_unlock_bh(&set->lock); map->gc.expires = jiffies + IPSET_GC_PERIOD(set->timeout) * HZ; add_timer(&map->gc); @@ -594,7 +576,7 @@ list_set_gc_init(struct ip_set *set, void (*gc)(unsigned long ul_set)) struct list_set *map = set->data; init_timer(&map->gc); - map->gc.data = (unsigned long) set; + map->gc.data = (unsigned long)set; map->gc.function = gc; map->gc.expires = jiffies + IPSET_GC_PERIOD(set->timeout) * HZ; add_timer(&map->gc); @@ -606,24 +588,16 @@ static bool init_list_set(struct net *net, struct ip_set *set, u32 size) { struct list_set *map; - struct set_elem *e; - u32 i; - map = kzalloc(sizeof(*map) + - min_t(u32, size, IP_SET_LIST_MAX_SIZE) * set->dsize, - GFP_KERNEL); + map = kzalloc(sizeof(*map), GFP_KERNEL); if (!map) return false; map->size = size; map->net = net; + INIT_LIST_HEAD(&map->members); set->data = map; - for (i = 0; i < size; i++) { - e = list_set_elem(set, map, i); - e->id = IPSET_INVALID_ID; - } - return true; } @@ -696,6 +670,7 @@ list_set_init(void) static void __exit list_set_fini(void) { + rcu_barrier(); ip_set_type_unregister(&list_set_type); } diff --git a/net/netfilter/ipset/pfxlen.c b/net/netfilter/ipset/pfxlen.c index 04d15fdc99ee..1c8a42c1056c 100644 --- a/net/netfilter/ipset/pfxlen.c +++ b/net/netfilter/ipset/pfxlen.c @@ -1,9 +1,7 @@ #include <linux/export.h> #include <linux/netfilter/ipset/pfxlen.h> -/* - * Prefixlen maps for fast conversions, by Jan Engelhardt. - */ +/* Prefixlen maps for fast conversions, by Jan Engelhardt. */ #define E(a, b, c, d) \ {.ip6 = { \ @@ -11,8 +9,7 @@ htonl(c), htonl(d), \ } } -/* - * This table works for both IPv4 and IPv6; +/* This table works for both IPv4 and IPv6; * just use prefixlen_netmask_map[prefixlength].ip. */ const union nf_inet_addr ip_set_netmask_map[] = { @@ -149,13 +146,12 @@ const union nf_inet_addr ip_set_netmask_map[] = { EXPORT_SYMBOL_GPL(ip_set_netmask_map); #undef E -#define E(a, b, c, d) \ - {.ip6 = { (__force __be32) a, (__force __be32) b, \ - (__force __be32) c, (__force __be32) d, \ +#define E(a, b, c, d) \ + {.ip6 = { (__force __be32)a, (__force __be32)b, \ + (__force __be32)c, (__force __be32)d, \ } } -/* - * This table works for both IPv4 and IPv6; +/* This table works for both IPv4 and IPv6; * just use prefixlen_hostmask_map[prefixlength].ip. */ const union nf_inet_addr ip_set_hostmask_map[] = { diff --git a/net/netfilter/nf_conntrack_proto_generic.c b/net/netfilter/nf_conntrack_proto_generic.c index 60865f110309..2281be419a74 100644 --- a/net/netfilter/nf_conntrack_proto_generic.c +++ b/net/netfilter/nf_conntrack_proto_generic.c @@ -90,7 +90,13 @@ static int generic_packet(struct nf_conn *ct, static bool generic_new(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, unsigned int *timeouts) { - return nf_generic_should_process(nf_ct_protonum(ct)); + bool ret; + + ret = nf_generic_should_process(nf_ct_protonum(ct)); + if (!ret) + pr_warn_once("conntrack: generic helper won't handle protocol %d. Please consider loading the specific helper module.\n", + nf_ct_protonum(ct)); + return ret; } #if IS_ENABLED(CONFIG_NF_CT_NETLINK_TIMEOUT) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 4528f122bcd2..cfe636808541 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -127,13 +127,46 @@ static void nft_trans_destroy(struct nft_trans *trans) kfree(trans); } +int nft_register_basechain(struct nft_base_chain *basechain, + unsigned int hook_nops) +{ + if (basechain->flags & NFT_BASECHAIN_DISABLED) + return 0; + + return nf_register_hooks(basechain->ops, hook_nops); +} +EXPORT_SYMBOL_GPL(nft_register_basechain); + +void nft_unregister_basechain(struct nft_base_chain *basechain, + unsigned int hook_nops) +{ + if (basechain->flags & NFT_BASECHAIN_DISABLED) + return; + + nf_unregister_hooks(basechain->ops, hook_nops); +} +EXPORT_SYMBOL_GPL(nft_unregister_basechain); + +static int nf_tables_register_hooks(const struct nft_table *table, + struct nft_chain *chain, + unsigned int hook_nops) +{ + if (table->flags & NFT_TABLE_F_DORMANT || + !(chain->flags & NFT_BASE_CHAIN)) + return 0; + + return nft_register_basechain(nft_base_chain(chain), hook_nops); +} + static void nf_tables_unregister_hooks(const struct nft_table *table, - const struct nft_chain *chain, + struct nft_chain *chain, unsigned int hook_nops) { - if (!(table->flags & NFT_TABLE_F_DORMANT) && - chain->flags & NFT_BASE_CHAIN) - nf_unregister_hooks(nft_base_chain(chain)->ops, hook_nops); + if (table->flags & NFT_TABLE_F_DORMANT || + !(chain->flags & NFT_BASE_CHAIN)) + return; + + nft_unregister_basechain(nft_base_chain(chain), hook_nops); } /* Internal table flags */ @@ -399,8 +432,6 @@ static const struct nla_policy nft_table_policy[NFTA_TABLE_MAX + 1] = { [NFTA_TABLE_NAME] = { .type = NLA_STRING, .len = NFT_TABLE_MAXNAMELEN - 1 }, [NFTA_TABLE_FLAGS] = { .type = NLA_U32 }, - [NFTA_TABLE_DEV] = { .type = NLA_STRING, - .len = IFNAMSIZ - 1 }, }; static int nf_tables_fill_table_info(struct sk_buff *skb, struct net *net, @@ -425,10 +456,6 @@ static int nf_tables_fill_table_info(struct sk_buff *skb, struct net *net, nla_put_be32(skb, NFTA_TABLE_USE, htonl(table->use))) goto nla_put_failure; - if (table->dev && - nla_put_string(skb, NFTA_TABLE_DEV, table->dev->name)) - goto nla_put_failure; - nlmsg_end(skb, nlh); return 0; @@ -566,7 +593,7 @@ static int nf_tables_table_enable(const struct nft_af_info *afi, if (!(chain->flags & NFT_BASE_CHAIN)) continue; - err = nf_register_hooks(nft_base_chain(chain)->ops, afi->nops); + err = nft_register_basechain(nft_base_chain(chain), afi->nops); if (err < 0) goto err; @@ -581,20 +608,20 @@ err: if (i-- <= 0) break; - nf_unregister_hooks(nft_base_chain(chain)->ops, afi->nops); + nft_unregister_basechain(nft_base_chain(chain), afi->nops); } return err; } static void nf_tables_table_disable(const struct nft_af_info *afi, - struct nft_table *table) + struct nft_table *table) { struct nft_chain *chain; list_for_each_entry(chain, &table->chains, list) { if (chain->flags & NFT_BASE_CHAIN) - nf_unregister_hooks(nft_base_chain(chain)->ops, - afi->nops); + nft_unregister_basechain(nft_base_chain(chain), + afi->nops); } } @@ -614,11 +641,6 @@ static int nf_tables_updtable(struct nft_ctx *ctx) if (flags == ctx->table->flags) return 0; - if ((ctx->afi->flags & NFT_AF_NEEDS_DEV) && - ctx->nla[NFTA_TABLE_DEV] && - nla_strcmp(ctx->nla[NFTA_TABLE_DEV], ctx->table->dev->name)) - return -EOPNOTSUPP; - trans = nft_trans_alloc(ctx, NFT_MSG_NEWTABLE, sizeof(struct nft_trans_table)); if (trans == NULL) @@ -656,7 +678,6 @@ static int nf_tables_newtable(struct sock *nlsk, struct sk_buff *skb, struct nft_table *table; struct net *net = sock_net(skb->sk); int family = nfmsg->nfgen_family; - struct net_device *dev = NULL; u32 flags = 0; struct nft_ctx ctx; int err; @@ -691,20 +712,6 @@ static int nf_tables_newtable(struct sock *nlsk, struct sk_buff *skb, return -EINVAL; } - if (afi->flags & NFT_AF_NEEDS_DEV) { - char ifname[IFNAMSIZ]; - - if (!nla[NFTA_TABLE_DEV]) - return -EOPNOTSUPP; - - nla_strlcpy(ifname, nla[NFTA_TABLE_DEV], IFNAMSIZ); - dev = dev_get_by_name(net, ifname); - if (!dev) - return -ENOENT; - } else if (nla[NFTA_TABLE_DEV]) { - return -EOPNOTSUPP; - } - err = -EAFNOSUPPORT; if (!try_module_get(afi->owner)) goto err1; @@ -718,7 +725,6 @@ static int nf_tables_newtable(struct sock *nlsk, struct sk_buff *skb, INIT_LIST_HEAD(&table->chains); INIT_LIST_HEAD(&table->sets); table->flags = flags; - table->dev = dev; nft_ctx_init(&ctx, skb, nlh, afi, table, NULL, nla); err = nft_trans_table_add(&ctx, NFT_MSG_NEWTABLE); @@ -732,9 +738,6 @@ err3: err2: module_put(afi->owner); err1: - if (dev != NULL) - dev_put(dev); - return err; } @@ -838,9 +841,6 @@ static void nf_tables_table_destroy(struct nft_ctx *ctx) { BUG_ON(ctx->table->use > 0); - if (ctx->table->dev) - dev_put(ctx->table->dev); - kfree(ctx->table); module_put(ctx->afi->owner); } @@ -916,6 +916,8 @@ static const struct nla_policy nft_chain_policy[NFTA_CHAIN_MAX + 1] = { static const struct nla_policy nft_hook_policy[NFTA_HOOK_MAX + 1] = { [NFTA_HOOK_HOOKNUM] = { .type = NLA_U32 }, [NFTA_HOOK_PRIORITY] = { .type = NLA_U32 }, + [NFTA_HOOK_DEV] = { .type = NLA_STRING, + .len = IFNAMSIZ - 1 }, }; static int nft_dump_stats(struct sk_buff *skb, struct nft_stats __percpu *stats) @@ -989,6 +991,9 @@ static int nf_tables_fill_chain_info(struct sk_buff *skb, struct net *net, goto nla_put_failure; if (nla_put_be32(skb, NFTA_HOOK_PRIORITY, htonl(ops->priority))) goto nla_put_failure; + if (basechain->dev_name[0] && + nla_put_string(skb, NFTA_HOOK_DEV, basechain->dev_name)) + goto nla_put_failure; nla_nest_end(skb, nest); if (nla_put_be32(skb, NFTA_CHAIN_POLICY, @@ -1200,9 +1205,13 @@ static void nf_tables_chain_destroy(struct nft_chain *chain) BUG_ON(chain->use > 0); if (chain->flags & NFT_BASE_CHAIN) { - module_put(nft_base_chain(chain)->type->owner); - free_percpu(nft_base_chain(chain)->stats); - kfree(nft_base_chain(chain)); + struct nft_base_chain *basechain = nft_base_chain(chain); + + module_put(basechain->type->owner); + free_percpu(basechain->stats); + if (basechain->ops[0].dev != NULL) + dev_put(basechain->ops[0].dev); + kfree(basechain); } else { kfree(chain); } @@ -1221,6 +1230,7 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, struct nlattr *ha[NFTA_HOOK_MAX + 1]; struct net *net = sock_net(skb->sk); int family = nfmsg->nfgen_family; + struct net_device *dev = NULL; u8 policy = NF_ACCEPT; u64 handle = 0; unsigned int i; @@ -1360,17 +1370,43 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, return -ENOENT; hookfn = type->hooks[hooknum]; + if (afi->flags & NFT_AF_NEEDS_DEV) { + char ifname[IFNAMSIZ]; + + if (!ha[NFTA_HOOK_DEV]) { + module_put(type->owner); + return -EOPNOTSUPP; + } + + nla_strlcpy(ifname, ha[NFTA_HOOK_DEV], IFNAMSIZ); + dev = dev_get_by_name(net, ifname); + if (!dev) { + module_put(type->owner); + return -ENOENT; + } + } else if (ha[NFTA_HOOK_DEV]) { + module_put(type->owner); + return -EOPNOTSUPP; + } + basechain = kzalloc(sizeof(*basechain), GFP_KERNEL); if (basechain == NULL) { module_put(type->owner); + if (dev != NULL) + dev_put(dev); return -ENOMEM; } + if (dev != NULL) + strncpy(basechain->dev_name, dev->name, IFNAMSIZ); + if (nla[NFTA_CHAIN_COUNTERS]) { stats = nft_stats_alloc(nla[NFTA_CHAIN_COUNTERS]); if (IS_ERR(stats)) { module_put(type->owner); kfree(basechain); + if (dev != NULL) + dev_put(dev); return PTR_ERR(stats); } basechain->stats = stats; @@ -1379,6 +1415,8 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, if (stats == NULL) { module_put(type->owner); kfree(basechain); + if (dev != NULL) + dev_put(dev); return -ENOMEM; } rcu_assign_pointer(basechain->stats, stats); @@ -1396,7 +1434,7 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, ops->priority = priority; ops->priv = chain; ops->hook = afi->hooks[ops->hooknum]; - ops->dev = table->dev; + ops->dev = dev; if (hookfn) ops->hook = hookfn; if (afi->hook_ops_init) @@ -1416,12 +1454,9 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, chain->table = table; nla_strlcpy(chain->name, name, NFT_CHAIN_MAXNAMELEN); - if (!(table->flags & NFT_TABLE_F_DORMANT) && - chain->flags & NFT_BASE_CHAIN) { - err = nf_register_hooks(nft_base_chain(chain)->ops, afi->nops); - if (err < 0) - goto err1; - } + err = nf_tables_register_hooks(table, chain, afi->nops); + if (err < 0) + goto err1; nft_ctx_init(&ctx, skb, nlh, afi, table, chain, nla); err = nft_trans_chain_add(&ctx, NFT_MSG_NEWCHAIN); diff --git a/net/netfilter/nf_tables_netdev.c b/net/netfilter/nf_tables_netdev.c index 04cb17057f46..2cae4d4a03b7 100644 --- a/net/netfilter/nf_tables_netdev.c +++ b/net/netfilter/nf_tables_netdev.c @@ -8,6 +8,7 @@ #include <linux/init.h> #include <linux/module.h> +#include <linux/netdevice.h> #include <net/netfilter/nf_tables.h> #include <linux/ip.h> #include <linux/ipv6.h> @@ -157,6 +158,77 @@ static const struct nf_chain_type nft_filter_chain_netdev = { .hook_mask = (1 << NF_NETDEV_INGRESS), }; +static void nft_netdev_event(unsigned long event, struct nft_af_info *afi, + struct net_device *dev, struct nft_table *table, + struct nft_base_chain *basechain) +{ + switch (event) { + case NETDEV_REGISTER: + if (strcmp(basechain->dev_name, dev->name) != 0) + return; + + BUG_ON(!(basechain->flags & NFT_BASECHAIN_DISABLED)); + + dev_hold(dev); + basechain->ops[0].dev = dev; + basechain->flags &= ~NFT_BASECHAIN_DISABLED; + if (!(table->flags & NFT_TABLE_F_DORMANT)) + nft_register_basechain(basechain, afi->nops); + break; + case NETDEV_UNREGISTER: + if (strcmp(basechain->dev_name, dev->name) != 0) + return; + + BUG_ON(basechain->flags & NFT_BASECHAIN_DISABLED); + + if (!(table->flags & NFT_TABLE_F_DORMANT)) + nft_unregister_basechain(basechain, afi->nops); + + dev_put(basechain->ops[0].dev); + basechain->ops[0].dev = NULL; + basechain->flags |= NFT_BASECHAIN_DISABLED; + break; + case NETDEV_CHANGENAME: + if (dev->ifindex != basechain->ops[0].dev->ifindex) + return; + + strncpy(basechain->dev_name, dev->name, IFNAMSIZ); + break; + } +} + +static int nf_tables_netdev_event(struct notifier_block *this, + unsigned long event, void *ptr) +{ + struct net_device *dev = netdev_notifier_info_to_dev(ptr); + struct nft_af_info *afi; + struct nft_table *table; + struct nft_chain *chain; + + nfnl_lock(NFNL_SUBSYS_NFTABLES); + list_for_each_entry(afi, &dev_net(dev)->nft.af_info, list) { + if (afi->family != NFPROTO_NETDEV) + continue; + + list_for_each_entry(table, &afi->tables, list) { + list_for_each_entry(chain, &table->chains, list) { + if (!(chain->flags & NFT_BASE_CHAIN)) + continue; + + nft_netdev_event(event, afi, dev, table, + nft_base_chain(chain)); + } + } + } + nfnl_unlock(NFNL_SUBSYS_NFTABLES); + + return NOTIFY_DONE; +} + +static struct notifier_block nf_tables_netdev_notifier = { + .notifier_call = nf_tables_netdev_event, +}; + static int __init nf_tables_netdev_init(void) { int ret; @@ -166,11 +238,14 @@ static int __init nf_tables_netdev_init(void) if (ret < 0) nft_unregister_chain_type(&nft_filter_chain_netdev); + register_netdevice_notifier(&nf_tables_netdev_notifier); + return ret; } static void __exit nf_tables_netdev_exit(void) { + unregister_netdevice_notifier(&nf_tables_netdev_notifier); unregister_pernet_subsys(&nf_tables_netdev_net_ops); nft_unregister_chain_type(&nft_filter_chain_netdev); } diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 83032464a4bd..d324fe71260c 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -658,35 +658,23 @@ EXPORT_SYMBOL_GPL(xt_compat_target_to_user); struct xt_table_info *xt_alloc_table_info(unsigned int size) { - struct xt_table_info *newinfo; - int cpu; + struct xt_table_info *info = NULL; + size_t sz = sizeof(*info) + size; /* Pedantry: prevent them from hitting BUG() in vmalloc.c --RR */ if ((SMP_ALIGN(size) >> PAGE_SHIFT) + 2 > totalram_pages) return NULL; - newinfo = kzalloc(XT_TABLE_INFO_SZ, GFP_KERNEL); - if (!newinfo) - return NULL; - - newinfo->size = size; - - for_each_possible_cpu(cpu) { - if (size <= PAGE_SIZE) - newinfo->entries[cpu] = kmalloc_node(size, - GFP_KERNEL, - cpu_to_node(cpu)); - else - newinfo->entries[cpu] = vmalloc_node(size, - cpu_to_node(cpu)); - - if (newinfo->entries[cpu] == NULL) { - xt_free_table_info(newinfo); + if (sz <= (PAGE_SIZE << PAGE_ALLOC_COSTLY_ORDER)) + info = kmalloc(sz, GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY); + if (!info) { + info = vmalloc(sz); + if (!info) return NULL; - } } - - return newinfo; + memset(info, 0, sizeof(*info)); + info->size = size; + return info; } EXPORT_SYMBOL(xt_alloc_table_info); @@ -694,9 +682,6 @@ void xt_free_table_info(struct xt_table_info *info) { int cpu; - for_each_possible_cpu(cpu) - kvfree(info->entries[cpu]); - if (info->jumpstack != NULL) { for_each_possible_cpu(cpu) kvfree(info->jumpstack[cpu]); @@ -705,7 +690,7 @@ void xt_free_table_info(struct xt_table_info *info) free_percpu(info->stackptr); - kfree(info); + kvfree(info); } EXPORT_SYMBOL(xt_free_table_info); diff --git a/net/netfilter/xt_set.c b/net/netfilter/xt_set.c index b103e9627716..5669e5b453f4 100644 --- a/net/netfilter/xt_set.c +++ b/net/netfilter/xt_set.c @@ -9,7 +9,8 @@ */ /* Kernel module which implements the set match and SET target - * for netfilter/iptables. */ + * for netfilter/iptables. + */ #include <linux/module.h> #include <linux/skbuff.h> @@ -53,6 +54,7 @@ static bool set_match_v0(const struct sk_buff *skb, struct xt_action_param *par) { const struct xt_set_info_match_v0 *info = par->matchinfo; + ADT_OPT(opt, par->family, info->match_set.u.compat.dim, info->match_set.u.compat.flags, 0, UINT_MAX); @@ -69,10 +71,10 @@ compat_flags(struct xt_set_info_v0 *info) info->u.compat.dim = IPSET_DIM_ZERO; if (info->u.flags[0] & IPSET_MATCH_INV) info->u.compat.flags |= IPSET_INV_MATCH; - for (i = 0; i < IPSET_DIM_MAX-1 && info->u.flags[i]; i++) { + for (i = 0; i < IPSET_DIM_MAX - 1 && info->u.flags[i]; i++) { info->u.compat.dim++; if (info->u.flags[i] & IPSET_SRC) - info->u.compat.flags |= (1<<info->u.compat.dim); + info->u.compat.flags |= (1 << info->u.compat.dim); } } @@ -89,7 +91,7 @@ set_match_v0_checkentry(const struct xt_mtchk_param *par) info->match_set.index); return -ENOENT; } - if (info->match_set.u.flags[IPSET_DIM_MAX-1] != 0) { + if (info->match_set.u.flags[IPSET_DIM_MAX - 1] != 0) { pr_warn("Protocol error: set match dimension is over the limit!\n"); ip_set_nfnl_put(par->net, info->match_set.index); return -ERANGE; @@ -115,6 +117,7 @@ static bool set_match_v1(const struct sk_buff *skb, struct xt_action_param *par) { const struct xt_set_info_match_v1 *info = par->matchinfo; + ADT_OPT(opt, par->family, info->match_set.dim, info->match_set.flags, 0, UINT_MAX); @@ -179,9 +182,10 @@ static bool set_match_v3(const struct sk_buff *skb, struct xt_action_param *par) { const struct xt_set_info_match_v3 *info = par->matchinfo; + int ret; + ADT_OPT(opt, par->family, info->match_set.dim, info->match_set.flags, info->flags, UINT_MAX); - int ret; if (info->packets.op != IPSET_COUNTER_NONE || info->bytes.op != IPSET_COUNTER_NONE) @@ -225,9 +229,10 @@ static bool set_match_v4(const struct sk_buff *skb, struct xt_action_param *par) { const struct xt_set_info_match_v4 *info = par->matchinfo; + int ret; + ADT_OPT(opt, par->family, info->match_set.dim, info->match_set.flags, info->flags, UINT_MAX); - int ret; if (info->packets.op != IPSET_COUNTER_NONE || info->bytes.op != IPSET_COUNTER_NONE) @@ -253,6 +258,7 @@ static unsigned int set_target_v0(struct sk_buff *skb, const struct xt_action_param *par) { const struct xt_set_info_target_v0 *info = par->targinfo; + ADT_OPT(add_opt, par->family, info->add_set.u.compat.dim, info->add_set.u.compat.flags, 0, UINT_MAX); ADT_OPT(del_opt, par->family, info->del_set.u.compat.dim, @@ -291,8 +297,8 @@ set_target_v0_checkentry(const struct xt_tgchk_param *par) return -ENOENT; } } - if (info->add_set.u.flags[IPSET_DIM_MAX-1] != 0 || - info->del_set.u.flags[IPSET_DIM_MAX-1] != 0) { + if (info->add_set.u.flags[IPSET_DIM_MAX - 1] != 0 || + info->del_set.u.flags[IPSET_DIM_MAX - 1] != 0) { pr_warn("Protocol error: SET target dimension is over the limit!\n"); if (info->add_set.index != IPSET_INVALID_ID) ip_set_nfnl_put(par->net, info->add_set.index); @@ -325,6 +331,7 @@ static unsigned int set_target_v1(struct sk_buff *skb, const struct xt_action_param *par) { const struct xt_set_info_target_v1 *info = par->targinfo; + ADT_OPT(add_opt, par->family, info->add_set.dim, info->add_set.flags, 0, UINT_MAX); ADT_OPT(del_opt, par->family, info->del_set.dim, @@ -393,6 +400,7 @@ static unsigned int set_target_v2(struct sk_buff *skb, const struct xt_action_param *par) { const struct xt_set_info_target_v2 *info = par->targinfo; + ADT_OPT(add_opt, par->family, info->add_set.dim, info->add_set.flags, info->flags, info->timeout); ADT_OPT(del_opt, par->family, info->del_set.dim, @@ -400,8 +408,8 @@ set_target_v2(struct sk_buff *skb, const struct xt_action_param *par) /* Normalize to fit into jiffies */ if (add_opt.ext.timeout != IPSET_NO_TIMEOUT && - add_opt.ext.timeout > UINT_MAX/MSEC_PER_SEC) - add_opt.ext.timeout = UINT_MAX/MSEC_PER_SEC; + add_opt.ext.timeout > UINT_MAX / MSEC_PER_SEC) + add_opt.ext.timeout = UINT_MAX / MSEC_PER_SEC; if (info->add_set.index != IPSET_INVALID_ID) ip_set_add(info->add_set.index, skb, par, &add_opt); if (info->del_set.index != IPSET_INVALID_ID) @@ -419,6 +427,8 @@ static unsigned int set_target_v3(struct sk_buff *skb, const struct xt_action_param *par) { const struct xt_set_info_target_v3 *info = par->targinfo; + int ret; + ADT_OPT(add_opt, par->family, info->add_set.dim, info->add_set.flags, info->flags, info->timeout); ADT_OPT(del_opt, par->family, info->del_set.dim, @@ -426,12 +436,10 @@ set_target_v3(struct sk_buff *skb, const struct xt_action_param *par) ADT_OPT(map_opt, par->family, info->map_set.dim, info->map_set.flags, 0, UINT_MAX); - int ret; - /* Normalize to fit into jiffies */ if (add_opt.ext.timeout != IPSET_NO_TIMEOUT && - add_opt.ext.timeout > UINT_MAX/MSEC_PER_SEC) - add_opt.ext.timeout = UINT_MAX/MSEC_PER_SEC; + add_opt.ext.timeout > UINT_MAX / MSEC_PER_SEC) + add_opt.ext.timeout = UINT_MAX / MSEC_PER_SEC; if (info->add_set.index != IPSET_INVALID_ID) ip_set_add(info->add_set.index, skb, par, &add_opt); if (info->del_set.index != IPSET_INVALID_ID) @@ -457,7 +465,6 @@ set_target_v3(struct sk_buff *skb, const struct xt_action_param *par) return XT_CONTINUE; } - static int set_target_v3_checkentry(const struct xt_tgchk_param *par) { @@ -497,8 +504,7 @@ set_target_v3_checkentry(const struct xt_tgchk_param *par) !(par->hook_mask & (1 << NF_INET_FORWARD | 1 << NF_INET_LOCAL_OUT | 1 << NF_INET_POST_ROUTING))) { - pr_warn("mapping of prio or/and queue is allowed only" - "from OUTPUT/FORWARD/POSTROUTING chains\n"); + pr_warn("mapping of prio or/and queue is allowed only from OUTPUT/FORWARD/POSTROUTING chains\n"); return -EINVAL; } index = ip_set_nfnl_get_byindex(par->net, @@ -519,8 +525,7 @@ set_target_v3_checkentry(const struct xt_tgchk_param *par) if (info->add_set.dim > IPSET_DIM_MAX || info->del_set.dim > IPSET_DIM_MAX || info->map_set.dim > IPSET_DIM_MAX) { - pr_warn("Protocol error: SET target dimension " - "is over the limit!\n"); + pr_warn("Protocol error: SET target dimension is over the limit!\n"); if (info->add_set.index != IPSET_INVALID_ID) ip_set_nfnl_put(par->net, info->add_set.index); if (info->del_set.index != IPSET_INVALID_ID) @@ -546,7 +551,6 @@ set_target_v3_destroy(const struct xt_tgdtor_param *par) ip_set_nfnl_put(par->net, info->map_set.index); } - static struct xt_match set_matches[] __read_mostly = { { .name = "set", |