diff options
Diffstat (limited to 'net/netfilter')
-rw-r--r-- | net/netfilter/ipset/ip_set_core.c | 5 | ||||
-rw-r--r-- | net/netfilter/ipvs/ip_vs_core.c | 2 | ||||
-rw-r--r-- | net/netfilter/ipvs/ip_vs_xmit.c | 22 | ||||
-rw-r--r-- | net/netfilter/nf_nat_core.c | 24 | ||||
-rw-r--r-- | net/netfilter/nf_tables_api.c | 1269 | ||||
-rw-r--r-- | net/netfilter/nfnetlink.c | 10 | ||||
-rw-r--r-- | net/netfilter/nfnetlink_acct.c | 86 | ||||
-rw-r--r-- | net/netfilter/nft_ct.c | 96 | ||||
-rw-r--r-- | net/netfilter/nft_hash.c | 59 | ||||
-rw-r--r-- | net/netfilter/nft_lookup.c | 10 | ||||
-rw-r--r-- | net/netfilter/nft_meta.c | 103 | ||||
-rw-r--r-- | net/netfilter/nft_rbtree.c | 43 | ||||
-rw-r--r-- | net/netfilter/xt_bpf.c | 5 | ||||
-rw-r--r-- | net/netfilter/xt_nfacct.c | 5 | ||||
-rw-r--r-- | net/netfilter/xt_recent.c | 5 |
15 files changed, 1237 insertions, 507 deletions
diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c index 117208321f16..ec8114fae50b 100644 --- a/net/netfilter/ipset/ip_set_core.c +++ b/net/netfilter/ipset/ip_set_core.c @@ -271,10 +271,7 @@ ip_set_free(void *members) { pr_debug("%p: free with %s\n", members, is_vmalloc_addr(members) ? "vfree" : "kfree"); - if (is_vmalloc_addr(members)) - vfree(members); - else - kfree(members); + kvfree(members); } EXPORT_SYMBOL_GPL(ip_set_free); diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 3d2d2c8108ca..e6836755c45d 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -97,7 +97,7 @@ const char *ip_vs_proto_name(unsigned int proto) return "ICMPv6"; #endif default: - sprintf(buf, "IP_%d", proto); + sprintf(buf, "IP_%u", proto); return buf; } } diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index c47444e4cf8c..73ba1cc7a88d 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -562,7 +562,7 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ip_send_check(iph); /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; ip_vs_send_or_cont(NFPROTO_IPV4, skb, cp, 0); rcu_read_unlock(); @@ -590,7 +590,7 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, goto tx_error; /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; ip_vs_send_or_cont(NFPROTO_IPV6, skb, cp, 0); rcu_read_unlock(); @@ -684,7 +684,7 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, MTU problem. */ /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; rc = ip_vs_nat_send_or_cont(NFPROTO_IPV4, skb, cp, local); rcu_read_unlock(); @@ -774,7 +774,7 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, MTU problem. */ /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; rc = ip_vs_nat_send_or_cont(NFPROTO_IPV6, skb, cp, local); rcu_read_unlock(); @@ -883,10 +883,10 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, iph->daddr = cp->daddr.ip; iph->saddr = saddr; iph->ttl = old_iph->ttl; - ip_select_ident(skb, &rt->dst, NULL); + ip_select_ident(skb, NULL); /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; ret = ip_vs_tunnel_xmit_prepare(skb, cp); if (ret == NF_ACCEPT) @@ -974,7 +974,7 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, iph->hop_limit = old_iph->hop_limit; /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; ret = ip_vs_tunnel_xmit_prepare(skb, cp); if (ret == NF_ACCEPT) @@ -1023,7 +1023,7 @@ ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ip_send_check(ip_hdr(skb)); /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; ip_vs_send_or_cont(NFPROTO_IPV4, skb, cp, 0); rcu_read_unlock(); @@ -1060,7 +1060,7 @@ ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, } /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; ip_vs_send_or_cont(NFPROTO_IPV6, skb, cp, 0); rcu_read_unlock(); @@ -1157,7 +1157,7 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, ip_vs_nat_icmp(skb, pp, cp, 0); /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; rc = ip_vs_nat_send_or_cont(NFPROTO_IPV4, skb, cp, local); rcu_read_unlock(); @@ -1249,7 +1249,7 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, ip_vs_nat_icmp_v6(skb, pp, cp, 0); /* Another hack: avoid icmp_send in ip_fragment */ - skb->local_df = 1; + skb->ignore_df = 1; rc = ip_vs_nat_send_or_cont(NFPROTO_IPV6, skb, cp, local); rcu_read_unlock(); diff --git a/net/netfilter/nf_nat_core.c b/net/netfilter/nf_nat_core.c index 52ca952b802c..09096a670c45 100644 --- a/net/netfilter/nf_nat_core.c +++ b/net/netfilter/nf_nat_core.c @@ -358,6 +358,19 @@ out: rcu_read_unlock(); } +struct nf_conn_nat *nf_ct_nat_ext_add(struct nf_conn *ct) +{ + struct nf_conn_nat *nat = nfct_nat(ct); + if (nat) + return nat; + + if (!nf_ct_is_confirmed(ct)) + nat = nf_ct_ext_add(ct, NF_CT_EXT_NAT, GFP_ATOMIC); + + return nat; +} +EXPORT_SYMBOL_GPL(nf_ct_nat_ext_add); + unsigned int nf_nat_setup_info(struct nf_conn *ct, const struct nf_nat_range *range, @@ -368,14 +381,9 @@ nf_nat_setup_info(struct nf_conn *ct, struct nf_conn_nat *nat; /* nat helper or nfctnetlink also setup binding */ - nat = nfct_nat(ct); - if (!nat) { - nat = nf_ct_ext_add(ct, NF_CT_EXT_NAT, GFP_ATOMIC); - if (nat == NULL) { - pr_debug("failed to add NAT extension\n"); - return NF_ACCEPT; - } - } + nat = nf_ct_nat_ext_add(ct); + if (nat == NULL) + return NF_ACCEPT; NF_CT_ASSERT(maniptype == NF_NAT_MANIP_SRC || maniptype == NF_NAT_MANIP_DST); diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 3fd159db9f06..624e083125b9 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -88,6 +88,45 @@ nf_tables_afinfo_lookup(struct net *net, int family, bool autoload) return ERR_PTR(-EAFNOSUPPORT); } +static void nft_ctx_init(struct nft_ctx *ctx, + const struct sk_buff *skb, + const struct nlmsghdr *nlh, + struct nft_af_info *afi, + struct nft_table *table, + struct nft_chain *chain, + const struct nlattr * const *nla) +{ + ctx->net = sock_net(skb->sk); + ctx->afi = afi; + ctx->table = table; + ctx->chain = chain; + ctx->nla = nla; + ctx->portid = NETLINK_CB(skb).portid; + ctx->report = nlmsg_report(nlh); + ctx->seq = nlh->nlmsg_seq; +} + +static struct nft_trans *nft_trans_alloc(struct nft_ctx *ctx, int msg_type, + u32 size) +{ + struct nft_trans *trans; + + trans = kzalloc(sizeof(struct nft_trans) + size, GFP_KERNEL); + if (trans == NULL) + return NULL; + + trans->msg_type = msg_type; + trans->ctx = *ctx; + + return trans; +} + +static void nft_trans_destroy(struct nft_trans *trans) +{ + list_del(&trans->list); + kfree(trans); +} + /* * Tables */ @@ -197,20 +236,13 @@ nla_put_failure: return -1; } -static int nf_tables_table_notify(const struct sk_buff *oskb, - const struct nlmsghdr *nlh, - const struct nft_table *table, - int event, int family) +static int nf_tables_table_notify(const struct nft_ctx *ctx, int event) { struct sk_buff *skb; - u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; - u32 seq = nlh ? nlh->nlmsg_seq : 0; - struct net *net = oskb ? sock_net(oskb->sk) : &init_net; - bool report; int err; - report = nlh ? nlmsg_report(nlh) : false; - if (!report && !nfnetlink_has_listeners(net, NFNLGRP_NFTABLES)) + if (!ctx->report && + !nfnetlink_has_listeners(ctx->net, NFNLGRP_NFTABLES)) return 0; err = -ENOBUFS; @@ -218,18 +250,20 @@ static int nf_tables_table_notify(const struct sk_buff *oskb, if (skb == NULL) goto err; - err = nf_tables_fill_table_info(skb, portid, seq, event, 0, - family, table); + err = nf_tables_fill_table_info(skb, ctx->portid, ctx->seq, event, 0, + ctx->afi->family, ctx->table); if (err < 0) { kfree_skb(skb); goto err; } - err = nfnetlink_send(skb, net, portid, NFNLGRP_NFTABLES, report, - GFP_KERNEL); + err = nfnetlink_send(skb, ctx->net, ctx->portid, NFNLGRP_NFTABLES, + ctx->report, GFP_KERNEL); err: - if (err < 0) - nfnetlink_set_err(net, portid, NFNLGRP_NFTABLES, err); + if (err < 0) { + nfnetlink_set_err(ctx->net, ctx->portid, NFNLGRP_NFTABLES, + err); + } return err; } @@ -269,6 +303,9 @@ done: return skb->len; } +/* Internal table flags */ +#define NFT_TABLE_INACTIVE (1 << 15) + static int nf_tables_gettable(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) @@ -295,6 +332,8 @@ static int nf_tables_gettable(struct sock *nlsk, struct sk_buff *skb, table = nf_tables_table_lookup(afi, nla[NFTA_TABLE_NAME]); if (IS_ERR(table)) return PTR_ERR(table); + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb2) @@ -343,7 +382,7 @@ err: return err; } -static int nf_tables_table_disable(const struct nft_af_info *afi, +static void nf_tables_table_disable(const struct nft_af_info *afi, struct nft_table *table) { struct nft_chain *chain; @@ -353,45 +392,63 @@ static int nf_tables_table_disable(const struct nft_af_info *afi, nf_unregister_hooks(nft_base_chain(chain)->ops, afi->nops); } - - return 0; } -static int nf_tables_updtable(struct sock *nlsk, struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nlattr * const nla[], - struct nft_af_info *afi, struct nft_table *table) +static int nf_tables_updtable(struct nft_ctx *ctx) { - const struct nfgenmsg *nfmsg = nlmsg_data(nlh); - int family = nfmsg->nfgen_family, ret = 0; + struct nft_trans *trans; + u32 flags; + int ret = 0; - if (nla[NFTA_TABLE_FLAGS]) { - u32 flags; + if (!ctx->nla[NFTA_TABLE_FLAGS]) + return 0; - flags = ntohl(nla_get_be32(nla[NFTA_TABLE_FLAGS])); - if (flags & ~NFT_TABLE_F_DORMANT) - return -EINVAL; + flags = ntohl(nla_get_be32(ctx->nla[NFTA_TABLE_FLAGS])); + if (flags & ~NFT_TABLE_F_DORMANT) + return -EINVAL; + + trans = nft_trans_alloc(ctx, NFT_MSG_NEWTABLE, + sizeof(struct nft_trans_table)); + if (trans == NULL) + return -ENOMEM; - if ((flags & NFT_TABLE_F_DORMANT) && - !(table->flags & NFT_TABLE_F_DORMANT)) { - ret = nf_tables_table_disable(afi, table); - if (ret >= 0) - table->flags |= NFT_TABLE_F_DORMANT; - } else if (!(flags & NFT_TABLE_F_DORMANT) && - table->flags & NFT_TABLE_F_DORMANT) { - ret = nf_tables_table_enable(afi, table); - if (ret >= 0) - table->flags &= ~NFT_TABLE_F_DORMANT; + if ((flags & NFT_TABLE_F_DORMANT) && + !(ctx->table->flags & NFT_TABLE_F_DORMANT)) { + nft_trans_table_enable(trans) = false; + } else if (!(flags & NFT_TABLE_F_DORMANT) && + ctx->table->flags & NFT_TABLE_F_DORMANT) { + ret = nf_tables_table_enable(ctx->afi, ctx->table); + if (ret >= 0) { + ctx->table->flags &= ~NFT_TABLE_F_DORMANT; + nft_trans_table_enable(trans) = true; } - if (ret < 0) - goto err; } + if (ret < 0) + goto err; - nf_tables_table_notify(skb, nlh, table, NFT_MSG_NEWTABLE, family); + nft_trans_table_update(trans) = true; + list_add_tail(&trans->list, &ctx->net->nft.commit_list); + return 0; err: + nft_trans_destroy(trans); return ret; } +static int nft_trans_table_add(struct nft_ctx *ctx, int msg_type) +{ + struct nft_trans *trans; + + trans = nft_trans_alloc(ctx, msg_type, sizeof(struct nft_trans_table)); + if (trans == NULL) + return -ENOMEM; + + if (msg_type == NFT_MSG_NEWTABLE) + ctx->table->flags |= NFT_TABLE_INACTIVE; + + list_add_tail(&trans->list, &ctx->net->nft.commit_list); + return 0; +} + static int nf_tables_newtable(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) @@ -403,6 +460,8 @@ static int nf_tables_newtable(struct sock *nlsk, struct sk_buff *skb, struct net *net = sock_net(skb->sk); int family = nfmsg->nfgen_family; u32 flags = 0; + struct nft_ctx ctx; + int err; afi = nf_tables_afinfo_lookup(net, family, true); if (IS_ERR(afi)) @@ -417,11 +476,15 @@ static int nf_tables_newtable(struct sock *nlsk, struct sk_buff *skb, } if (table != NULL) { + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; if (nlh->nlmsg_flags & NLM_F_EXCL) return -EEXIST; if (nlh->nlmsg_flags & NLM_F_REPLACE) return -EOPNOTSUPP; - return nf_tables_updtable(nlsk, skb, nlh, nla, afi, table); + + nft_ctx_init(&ctx, skb, nlh, afi, table, NULL, nla); + return nf_tables_updtable(&ctx); } if (nla[NFTA_TABLE_FLAGS]) { @@ -444,8 +507,14 @@ static int nf_tables_newtable(struct sock *nlsk, struct sk_buff *skb, INIT_LIST_HEAD(&table->sets); table->flags = flags; + nft_ctx_init(&ctx, skb, nlh, afi, table, NULL, nla); + err = nft_trans_table_add(&ctx, NFT_MSG_NEWTABLE); + if (err < 0) { + kfree(table); + module_put(afi->owner); + return err; + } list_add_tail(&table->list, &afi->tables); - nf_tables_table_notify(skb, nlh, table, NFT_MSG_NEWTABLE, family); return 0; } @@ -457,7 +526,8 @@ static int nf_tables_deltable(struct sock *nlsk, struct sk_buff *skb, struct nft_af_info *afi; struct nft_table *table; struct net *net = sock_net(skb->sk); - int family = nfmsg->nfgen_family; + int family = nfmsg->nfgen_family, err; + struct nft_ctx ctx; afi = nf_tables_afinfo_lookup(net, family, false); if (IS_ERR(afi)) @@ -466,17 +536,28 @@ static int nf_tables_deltable(struct sock *nlsk, struct sk_buff *skb, table = nf_tables_table_lookup(afi, nla[NFTA_TABLE_NAME]); if (IS_ERR(table)) return PTR_ERR(table); - - if (!list_empty(&table->chains) || !list_empty(&table->sets)) + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; + if (table->use > 0) return -EBUSY; + nft_ctx_init(&ctx, skb, nlh, afi, table, NULL, nla); + err = nft_trans_table_add(&ctx, NFT_MSG_DELTABLE); + if (err < 0) + return err; + list_del(&table->list); - nf_tables_table_notify(skb, nlh, table, NFT_MSG_DELTABLE, family); - kfree(table); - module_put(afi->owner); return 0; } +static void nf_tables_table_destroy(struct nft_ctx *ctx) +{ + BUG_ON(ctx->table->use > 0); + + kfree(ctx->table); + module_put(ctx->afi->owner); +} + int nft_register_chain_type(const struct nf_chain_type *ctype) { int err = 0; @@ -541,7 +622,7 @@ static const struct nla_policy nft_chain_policy[NFTA_CHAIN_MAX + 1] = { .len = NFT_CHAIN_MAXNAMELEN - 1 }, [NFTA_CHAIN_HOOK] = { .type = NLA_NESTED }, [NFTA_CHAIN_POLICY] = { .type = NLA_U32 }, - [NFTA_CHAIN_TYPE] = { .type = NLA_NUL_STRING }, + [NFTA_CHAIN_TYPE] = { .type = NLA_STRING }, [NFTA_CHAIN_COUNTERS] = { .type = NLA_NESTED }, }; @@ -637,21 +718,13 @@ nla_put_failure: return -1; } -static int nf_tables_chain_notify(const struct sk_buff *oskb, - const struct nlmsghdr *nlh, - const struct nft_table *table, - const struct nft_chain *chain, - int event, int family) +static int nf_tables_chain_notify(const struct nft_ctx *ctx, int event) { struct sk_buff *skb; - u32 portid = oskb ? NETLINK_CB(oskb).portid : 0; - struct net *net = oskb ? sock_net(oskb->sk) : &init_net; - u32 seq = nlh ? nlh->nlmsg_seq : 0; - bool report; int err; - report = nlh ? nlmsg_report(nlh) : false; - if (!report && !nfnetlink_has_listeners(net, NFNLGRP_NFTABLES)) + if (!ctx->report && + !nfnetlink_has_listeners(ctx->net, NFNLGRP_NFTABLES)) return 0; err = -ENOBUFS; @@ -659,18 +732,21 @@ static int nf_tables_chain_notify(const struct sk_buff *oskb, if (skb == NULL) goto err; - err = nf_tables_fill_chain_info(skb, portid, seq, event, 0, family, - table, chain); + err = nf_tables_fill_chain_info(skb, ctx->portid, ctx->seq, event, 0, + ctx->afi->family, ctx->table, + ctx->chain); if (err < 0) { kfree_skb(skb); goto err; } - err = nfnetlink_send(skb, net, portid, NFNLGRP_NFTABLES, report, - GFP_KERNEL); + err = nfnetlink_send(skb, ctx->net, ctx->portid, NFNLGRP_NFTABLES, + ctx->report, GFP_KERNEL); err: - if (err < 0) - nfnetlink_set_err(net, portid, NFNLGRP_NFTABLES, err); + if (err < 0) { + nfnetlink_set_err(ctx->net, ctx->portid, NFNLGRP_NFTABLES, + err); + } return err; } @@ -740,10 +816,14 @@ static int nf_tables_getchain(struct sock *nlsk, struct sk_buff *skb, table = nf_tables_table_lookup(afi, nla[NFTA_CHAIN_TABLE]); if (IS_ERR(table)) return PTR_ERR(table); + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; chain = nf_tables_chain_lookup(table, nla[NFTA_CHAIN_NAME]); if (IS_ERR(chain)) return PTR_ERR(chain); + if (chain->flags & NFT_CHAIN_INACTIVE) + return -ENOENT; skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb2) @@ -767,8 +847,7 @@ static const struct nla_policy nft_counter_policy[NFTA_COUNTER_MAX + 1] = { [NFTA_COUNTER_BYTES] = { .type = NLA_U64 }, }; -static int -nf_tables_counters(struct nft_base_chain *chain, const struct nlattr *attr) +static struct nft_stats __percpu *nft_stats_alloc(const struct nlattr *attr) { struct nlattr *tb[NFTA_COUNTER_MAX+1]; struct nft_stats __percpu *newstats; @@ -777,14 +856,14 @@ nf_tables_counters(struct nft_base_chain *chain, const struct nlattr *attr) err = nla_parse_nested(tb, NFTA_COUNTER_MAX, attr, nft_counter_policy); if (err < 0) - return err; + return ERR_PTR(err); if (!tb[NFTA_COUNTER_BYTES] || !tb[NFTA_COUNTER_PACKETS]) - return -EINVAL; + return ERR_PTR(-EINVAL); newstats = alloc_percpu(struct nft_stats); if (newstats == NULL) - return -ENOMEM; + return ERR_PTR(-ENOMEM); /* Restore old counters on this cpu, no problem. Per-cpu statistics * are not exposed to userspace. @@ -793,6 +872,12 @@ nf_tables_counters(struct nft_base_chain *chain, const struct nlattr *attr) stats->bytes = be64_to_cpu(nla_get_be64(tb[NFTA_COUNTER_BYTES])); stats->pkts = be64_to_cpu(nla_get_be64(tb[NFTA_COUNTER_PACKETS])); + return newstats; +} + +static void nft_chain_stats_replace(struct nft_base_chain *chain, + struct nft_stats __percpu *newstats) +{ if (chain->stats) { struct nft_stats __percpu *oldstats = nft_dereference(chain->stats); @@ -802,17 +887,43 @@ nf_tables_counters(struct nft_base_chain *chain, const struct nlattr *attr) free_percpu(oldstats); } else rcu_assign_pointer(chain->stats, newstats); +} + +static int nft_trans_chain_add(struct nft_ctx *ctx, int msg_type) +{ + struct nft_trans *trans; + trans = nft_trans_alloc(ctx, msg_type, sizeof(struct nft_trans_chain)); + if (trans == NULL) + return -ENOMEM; + + if (msg_type == NFT_MSG_NEWCHAIN) + ctx->chain->flags |= NFT_CHAIN_INACTIVE; + + list_add_tail(&trans->list, &ctx->net->nft.commit_list); return 0; } +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)); + } else { + kfree(chain); + } +} + static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); const struct nlattr * uninitialized_var(name); - const struct nft_af_info *afi; + struct nft_af_info *afi; struct nft_table *table; struct nft_chain *chain; struct nft_base_chain *basechain = NULL; @@ -822,8 +933,10 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, u8 policy = NF_ACCEPT; u64 handle = 0; unsigned int i; + struct nft_stats __percpu *stats; int err; bool create; + struct nft_ctx ctx; create = nlh->nlmsg_flags & NLM_F_CREATE ? true : false; @@ -869,6 +982,11 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, } if (chain != NULL) { + struct nft_stats *stats = NULL; + struct nft_trans *trans; + + if (chain->flags & NFT_CHAIN_INACTIVE) + return -ENOENT; if (nlh->nlmsg_flags & NLM_F_EXCL) return -EEXIST; if (nlh->nlmsg_flags & NLM_F_REPLACE) @@ -882,19 +1000,31 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, if (!(chain->flags & NFT_BASE_CHAIN)) return -EOPNOTSUPP; - err = nf_tables_counters(nft_base_chain(chain), - nla[NFTA_CHAIN_COUNTERS]); - if (err < 0) - return err; + stats = nft_stats_alloc(nla[NFTA_CHAIN_COUNTERS]); + if (IS_ERR(stats)) + return PTR_ERR(stats); } - if (nla[NFTA_CHAIN_POLICY]) - nft_base_chain(chain)->policy = policy; + nft_ctx_init(&ctx, skb, nlh, afi, table, chain, nla); + trans = nft_trans_alloc(&ctx, NFT_MSG_NEWCHAIN, + sizeof(struct nft_trans_chain)); + if (trans == NULL) + return -ENOMEM; + + nft_trans_chain_stats(trans) = stats; + nft_trans_chain_update(trans) = true; - if (nla[NFTA_CHAIN_HANDLE] && name) - nla_strlcpy(chain->name, name, NFT_CHAIN_MAXNAMELEN); + if (nla[NFTA_CHAIN_POLICY]) + nft_trans_chain_policy(trans) = policy; + else + nft_trans_chain_policy(trans) = -1; - goto notify; + if (nla[NFTA_CHAIN_HANDLE] && name) { + nla_strlcpy(nft_trans_chain_name(trans), name, + NFT_CHAIN_MAXNAMELEN); + } + list_add_tail(&trans->list, &net->nft.commit_list); + return 0; } if (table->use == UINT_MAX) @@ -939,23 +1069,21 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, return -ENOMEM; if (nla[NFTA_CHAIN_COUNTERS]) { - err = nf_tables_counters(basechain, - nla[NFTA_CHAIN_COUNTERS]); - if (err < 0) { + stats = nft_stats_alloc(nla[NFTA_CHAIN_COUNTERS]); + if (IS_ERR(stats)) { module_put(type->owner); kfree(basechain); - return err; + return PTR_ERR(stats); } + basechain->stats = stats; } else { - struct nft_stats __percpu *newstats; - - newstats = alloc_percpu(struct nft_stats); - if (newstats == NULL) { + stats = alloc_percpu(struct nft_stats); + if (IS_ERR(stats)) { module_put(type->owner); kfree(basechain); - return -ENOMEM; + return PTR_ERR(stats); } - rcu_assign_pointer(basechain->stats, newstats); + rcu_assign_pointer(basechain->stats, stats); } basechain->type = type; @@ -992,31 +1120,27 @@ static int nf_tables_newchain(struct sock *nlsk, struct sk_buff *skb, 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) { - module_put(basechain->type->owner); - free_percpu(basechain->stats); - kfree(basechain); - return err; - } + if (err < 0) + goto err1; } - list_add_tail(&chain->list, &table->chains); - table->use++; -notify: - nf_tables_chain_notify(skb, nlh, table, chain, NFT_MSG_NEWCHAIN, - family); - return 0; -} -static void nf_tables_chain_destroy(struct nft_chain *chain) -{ - BUG_ON(chain->use > 0); + nft_ctx_init(&ctx, skb, nlh, afi, table, chain, nla); + err = nft_trans_chain_add(&ctx, NFT_MSG_NEWCHAIN); + if (err < 0) + goto err2; - 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)); - } else - kfree(chain); + table->use++; + list_add_tail(&chain->list, &table->chains); + return 0; +err2: + if (!(table->flags & NFT_TABLE_F_DORMANT) && + chain->flags & NFT_BASE_CHAIN) { + nf_unregister_hooks(nft_base_chain(chain)->ops, + afi->nops); + } +err1: + nf_tables_chain_destroy(chain); + return err; } static int nf_tables_delchain(struct sock *nlsk, struct sk_buff *skb, @@ -1024,11 +1148,13 @@ static int nf_tables_delchain(struct sock *nlsk, struct sk_buff *skb, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); - const struct nft_af_info *afi; + struct nft_af_info *afi; struct nft_table *table; struct nft_chain *chain; struct net *net = sock_net(skb->sk); int family = nfmsg->nfgen_family; + struct nft_ctx ctx; + int err; afi = nf_tables_afinfo_lookup(net, family, false); if (IS_ERR(afi)) @@ -1037,48 +1163,27 @@ static int nf_tables_delchain(struct sock *nlsk, struct sk_buff *skb, table = nf_tables_table_lookup(afi, nla[NFTA_CHAIN_TABLE]); if (IS_ERR(table)) return PTR_ERR(table); + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; chain = nf_tables_chain_lookup(table, nla[NFTA_CHAIN_NAME]); if (IS_ERR(chain)) return PTR_ERR(chain); - - if (!list_empty(&chain->rules) || chain->use > 0) + if (chain->flags & NFT_CHAIN_INACTIVE) + return -ENOENT; + if (chain->use > 0) return -EBUSY; - list_del(&chain->list); - table->use--; - - if (!(table->flags & NFT_TABLE_F_DORMANT) && - chain->flags & NFT_BASE_CHAIN) - nf_unregister_hooks(nft_base_chain(chain)->ops, afi->nops); - - nf_tables_chain_notify(skb, nlh, table, chain, NFT_MSG_DELCHAIN, - family); - - /* Make sure all rule references are gone before this is released */ - synchronize_rcu(); + nft_ctx_init(&ctx, skb, nlh, afi, table, chain, nla); + err = nft_trans_chain_add(&ctx, NFT_MSG_DELCHAIN); + if (err < 0) + return err; - nf_tables_chain_destroy(chain); + table->use--; + list_del(&chain->list); return 0; } -static void nft_ctx_init(struct nft_ctx *ctx, - const struct sk_buff *skb, - const struct nlmsghdr *nlh, - const struct nft_af_info *afi, - const struct nft_table *table, - const struct nft_chain *chain, - const struct nlattr * const *nla) -{ - ctx->net = sock_net(skb->sk); - ctx->skb = skb; - ctx->nlh = nlh; - ctx->afi = afi; - ctx->table = table; - ctx->chain = chain; - ctx->nla = nla; -} - /* * Expressions */ @@ -1093,7 +1198,10 @@ static void nft_ctx_init(struct nft_ctx *ctx, int nft_register_expr(struct nft_expr_type *type) { nfnl_lock(NFNL_SUBSYS_NFTABLES); - list_add_tail(&type->list, &nf_tables_expressions); + if (type->family == NFPROTO_UNSPEC) + list_add_tail(&type->list, &nf_tables_expressions); + else + list_add(&type->list, &nf_tables_expressions); nfnl_unlock(NFNL_SUBSYS_NFTABLES); return 0; } @@ -1361,22 +1469,15 @@ nla_put_failure: return -1; } -static int nf_tables_rule_notify(const struct sk_buff *oskb, - const struct nlmsghdr *nlh, - const struct nft_table *table, - const struct nft_chain *chain, +static int nf_tables_rule_notify(const struct nft_ctx *ctx, const struct nft_rule *rule, - int event, u32 flags, int family) + int event) { struct sk_buff *skb; - u32 portid = NETLINK_CB(oskb).portid; - struct net *net = oskb ? sock_net(oskb->sk) : &init_net; - u32 seq = nlh->nlmsg_seq; - bool report; int err; - report = nlmsg_report(nlh); - if (!report && !nfnetlink_has_listeners(net, NFNLGRP_NFTABLES)) + if (!ctx->report && + !nfnetlink_has_listeners(ctx->net, NFNLGRP_NFTABLES)) return 0; err = -ENOBUFS; @@ -1384,18 +1485,21 @@ static int nf_tables_rule_notify(const struct sk_buff *oskb, if (skb == NULL) goto err; - err = nf_tables_fill_rule_info(skb, portid, seq, event, flags, - family, table, chain, rule); + err = nf_tables_fill_rule_info(skb, ctx->portid, ctx->seq, event, 0, + ctx->afi->family, ctx->table, + ctx->chain, rule); if (err < 0) { kfree_skb(skb); goto err; } - err = nfnetlink_send(skb, net, portid, NFNLGRP_NFTABLES, report, - GFP_KERNEL); + err = nfnetlink_send(skb, ctx->net, ctx->portid, NFNLGRP_NFTABLES, + ctx->report, GFP_KERNEL); err: - if (err < 0) - nfnetlink_set_err(net, portid, NFNLGRP_NFTABLES, err); + if (err < 0) { + nfnetlink_set_err(ctx->net, ctx->portid, NFNLGRP_NFTABLES, + err); + } return err; } @@ -1511,10 +1615,14 @@ static int nf_tables_getrule(struct sock *nlsk, struct sk_buff *skb, table = nf_tables_table_lookup(afi, nla[NFTA_RULE_TABLE]); if (IS_ERR(table)) return PTR_ERR(table); + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; chain = nf_tables_chain_lookup(table, nla[NFTA_RULE_CHAIN]); if (IS_ERR(chain)) return PTR_ERR(chain); + if (chain->flags & NFT_CHAIN_INACTIVE) + return -ENOENT; rule = nf_tables_rule_lookup(chain, nla[NFTA_RULE_HANDLE]); if (IS_ERR(rule)) @@ -1554,37 +1662,36 @@ static void nf_tables_rule_destroy(const struct nft_ctx *ctx, kfree(rule); } -#define NFT_RULE_MAXEXPRS 128 - -static struct nft_expr_info *info; - -static struct nft_rule_trans * -nf_tables_trans_add(struct nft_ctx *ctx, struct nft_rule *rule) +static struct nft_trans *nft_trans_rule_add(struct nft_ctx *ctx, int msg_type, + struct nft_rule *rule) { - struct nft_rule_trans *rupd; + struct nft_trans *trans; - rupd = kmalloc(sizeof(struct nft_rule_trans), GFP_KERNEL); - if (rupd == NULL) - return NULL; + trans = nft_trans_alloc(ctx, msg_type, sizeof(struct nft_trans_rule)); + if (trans == NULL) + return NULL; - rupd->ctx = *ctx; - rupd->rule = rule; - list_add_tail(&rupd->list, &ctx->net->nft.commit_list); + nft_trans_rule(trans) = rule; + list_add_tail(&trans->list, &ctx->net->nft.commit_list); - return rupd; + return trans; } +#define NFT_RULE_MAXEXPRS 128 + +static struct nft_expr_info *info; + static int nf_tables_newrule(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); - const struct nft_af_info *afi; + struct nft_af_info *afi; struct net *net = sock_net(skb->sk); struct nft_table *table; struct nft_chain *chain; struct nft_rule *rule, *old_rule = NULL; - struct nft_rule_trans *repl = NULL; + struct nft_trans *trans = NULL; struct nft_expr *expr; struct nft_ctx ctx; struct nlattr *tmp; @@ -1682,8 +1789,9 @@ static int nf_tables_newrule(struct sock *nlsk, struct sk_buff *skb, if (nlh->nlmsg_flags & NLM_F_REPLACE) { if (nft_rule_is_active_next(net, old_rule)) { - repl = nf_tables_trans_add(&ctx, old_rule); - if (repl == NULL) { + trans = nft_trans_rule_add(&ctx, NFT_MSG_NEWRULE, + old_rule); + if (trans == NULL) { err = -ENOMEM; goto err2; } @@ -1705,19 +1813,19 @@ static int nf_tables_newrule(struct sock *nlsk, struct sk_buff *skb, list_add_rcu(&rule->list, &chain->rules); } - if (nf_tables_trans_add(&ctx, rule) == NULL) { + if (nft_trans_rule_add(&ctx, NFT_MSG_NEWRULE, rule) == NULL) { err = -ENOMEM; goto err3; } + chain->use++; return 0; err3: list_del_rcu(&rule->list); - if (repl) { - list_del_rcu(&repl->rule->list); - list_del(&repl->list); - nft_rule_clear(net, repl->rule); - kfree(repl); + if (trans) { + list_del_rcu(&nft_trans_rule(trans)->list); + nft_rule_clear(net, nft_trans_rule(trans)); + nft_trans_destroy(trans); } err2: nf_tables_rule_destroy(&ctx, rule); @@ -1734,9 +1842,10 @@ nf_tables_delrule_one(struct nft_ctx *ctx, struct nft_rule *rule) { /* You cannot delete the same rule twice */ if (nft_rule_is_active_next(ctx->net, rule)) { - if (nf_tables_trans_add(ctx, rule) == NULL) + if (nft_trans_rule_add(ctx, NFT_MSG_DELRULE, rule) == NULL) return -ENOMEM; nft_rule_disactivate_next(ctx->net, rule); + ctx->chain->use--; return 0; } return -ENOENT; @@ -1760,9 +1869,9 @@ static int nf_tables_delrule(struct sock *nlsk, struct sk_buff *skb, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); - const struct nft_af_info *afi; + struct nft_af_info *afi; struct net *net = sock_net(skb->sk); - const struct nft_table *table; + struct nft_table *table; struct nft_chain *chain = NULL; struct nft_rule *rule; int family = nfmsg->nfgen_family, err = 0; @@ -1775,6 +1884,8 @@ static int nf_tables_delrule(struct sock *nlsk, struct sk_buff *skb, table = nf_tables_table_lookup(afi, nla[NFTA_RULE_TABLE]); if (IS_ERR(table)) return PTR_ERR(table); + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; if (nla[NFTA_RULE_CHAIN]) { chain = nf_tables_chain_lookup(table, nla[NFTA_RULE_CHAIN]); @@ -1807,88 +1918,6 @@ static int nf_tables_delrule(struct sock *nlsk, struct sk_buff *skb, return err; } -static int nf_tables_commit(struct sk_buff *skb) -{ - struct net *net = sock_net(skb->sk); - struct nft_rule_trans *rupd, *tmp; - - /* Bump generation counter, invalidate any dump in progress */ - net->nft.genctr++; - - /* A new generation has just started */ - net->nft.gencursor = gencursor_next(net); - - /* Make sure all packets have left the previous generation before - * purging old rules. - */ - synchronize_rcu(); - - list_for_each_entry_safe(rupd, tmp, &net->nft.commit_list, list) { - /* This rule was inactive in the past and just became active. - * Clear the next bit of the genmask since its meaning has - * changed, now it is the future. - */ - if (nft_rule_is_active(net, rupd->rule)) { - nft_rule_clear(net, rupd->rule); - nf_tables_rule_notify(skb, rupd->ctx.nlh, - rupd->ctx.table, rupd->ctx.chain, - rupd->rule, NFT_MSG_NEWRULE, 0, - rupd->ctx.afi->family); - list_del(&rupd->list); - kfree(rupd); - continue; - } - - /* This rule is in the past, get rid of it */ - list_del_rcu(&rupd->rule->list); - nf_tables_rule_notify(skb, rupd->ctx.nlh, - rupd->ctx.table, rupd->ctx.chain, - rupd->rule, NFT_MSG_DELRULE, 0, - rupd->ctx.afi->family); - } - - /* Make sure we don't see any packet traversing old rules */ - synchronize_rcu(); - - /* Now we can safely release unused old rules */ - list_for_each_entry_safe(rupd, tmp, &net->nft.commit_list, list) { - nf_tables_rule_destroy(&rupd->ctx, rupd->rule); - list_del(&rupd->list); - kfree(rupd); - } - - return 0; -} - -static int nf_tables_abort(struct sk_buff *skb) -{ - struct net *net = sock_net(skb->sk); - struct nft_rule_trans *rupd, *tmp; - - list_for_each_entry_safe(rupd, tmp, &net->nft.commit_list, list) { - if (!nft_rule_is_active_next(net, rupd->rule)) { - nft_rule_clear(net, rupd->rule); - list_del(&rupd->list); - kfree(rupd); - continue; - } - - /* This rule is inactive, get rid of it */ - list_del_rcu(&rupd->rule->list); - } - - /* Make sure we don't see any packet accessing aborted rules */ - synchronize_rcu(); - - list_for_each_entry_safe(rupd, tmp, &net->nft.commit_list, list) { - nf_tables_rule_destroy(&rupd->ctx, rupd->rule); - list_del(&rupd->list); - kfree(rupd); - } - - return 0; -} - /* * Sets */ @@ -1912,9 +1941,18 @@ void nft_unregister_set(struct nft_set_ops *ops) } EXPORT_SYMBOL_GPL(nft_unregister_set); -static const struct nft_set_ops *nft_select_set_ops(const struct nlattr * const nla[]) +/* + * Select a set implementation based on the data characteristics and the + * given policy. The total memory use might not be known if no size is + * given, in that case the amount of memory per element is used. + */ +static const struct nft_set_ops * +nft_select_set_ops(const struct nlattr * const nla[], + const struct nft_set_desc *desc, + enum nft_set_policies policy) { - const struct nft_set_ops *ops; + const struct nft_set_ops *ops, *bops; + struct nft_set_estimate est, best; u32 features; #ifdef CONFIG_MODULES @@ -1932,15 +1970,45 @@ static const struct nft_set_ops *nft_select_set_ops(const struct nlattr * const features &= NFT_SET_INTERVAL | NFT_SET_MAP; } - // FIXME: implement selection properly + bops = NULL; + best.size = ~0; + best.class = ~0; + list_for_each_entry(ops, &nf_tables_set_ops, list) { if ((ops->features & features) != features) continue; + if (!ops->estimate(desc, features, &est)) + continue; + + switch (policy) { + case NFT_SET_POL_PERFORMANCE: + if (est.class < best.class) + break; + if (est.class == best.class && est.size < best.size) + break; + continue; + case NFT_SET_POL_MEMORY: + if (est.size < best.size) + break; + if (est.size == best.size && est.class < best.class) + break; + continue; + default: + break; + } + if (!try_module_get(ops->owner)) continue; - return ops; + if (bops != NULL) + module_put(bops->owner); + + bops = ops; + best = est; } + if (bops != NULL) + return bops; + return ERR_PTR(-EOPNOTSUPP); } @@ -1953,6 +2021,13 @@ static const struct nla_policy nft_set_policy[NFTA_SET_MAX + 1] = { [NFTA_SET_KEY_LEN] = { .type = NLA_U32 }, [NFTA_SET_DATA_TYPE] = { .type = NLA_U32 }, [NFTA_SET_DATA_LEN] = { .type = NLA_U32 }, + [NFTA_SET_POLICY] = { .type = NLA_U32 }, + [NFTA_SET_DESC] = { .type = NLA_NESTED }, + [NFTA_SET_ID] = { .type = NLA_U32 }, +}; + +static const struct nla_policy nft_set_desc_policy[NFTA_SET_DESC_MAX + 1] = { + [NFTA_SET_DESC_SIZE] = { .type = NLA_U32 }, }; static int nft_ctx_init_from_setattr(struct nft_ctx *ctx, @@ -1962,8 +2037,8 @@ static int nft_ctx_init_from_setattr(struct nft_ctx *ctx, { struct net *net = sock_net(skb->sk); const struct nfgenmsg *nfmsg = nlmsg_data(nlh); - const struct nft_af_info *afi = NULL; - const struct nft_table *table = NULL; + struct nft_af_info *afi = NULL; + struct nft_table *table = NULL; if (nfmsg->nfgen_family != NFPROTO_UNSPEC) { afi = nf_tables_afinfo_lookup(net, nfmsg->nfgen_family, false); @@ -1978,6 +2053,8 @@ static int nft_ctx_init_from_setattr(struct nft_ctx *ctx, table = nf_tables_table_lookup(afi, nla[NFTA_SET_TABLE]); if (IS_ERR(table)) return PTR_ERR(table); + if (table->flags & NFT_TABLE_INACTIVE) + return -ENOENT; } nft_ctx_init(ctx, skb, nlh, afi, table, NULL, nla); @@ -1999,13 +2076,27 @@ struct nft_set *nf_tables_set_lookup(const struct nft_table *table, return ERR_PTR(-ENOENT); } +struct nft_set *nf_tables_set_lookup_byid(const struct net *net, + const struct nlattr *nla) +{ + struct nft_trans *trans; + u32 id = ntohl(nla_get_be32(nla)); + + list_for_each_entry(trans, &net->nft.commit_list, list) { + if (trans->msg_type == NFT_MSG_NEWSET && + id == nft_trans_set_id(trans)) + return nft_trans_set(trans); + } + return ERR_PTR(-ENOENT); +} + static int nf_tables_set_alloc_name(struct nft_ctx *ctx, struct nft_set *set, const char *name) { const struct nft_set *i; const char *p; unsigned long *inuse; - unsigned int n = 0; + unsigned int n = 0, min = 0; p = strnchr(name, IFNAMSIZ, '%'); if (p != NULL) { @@ -2015,23 +2106,28 @@ static int nf_tables_set_alloc_name(struct nft_ctx *ctx, struct nft_set *set, inuse = (unsigned long *)get_zeroed_page(GFP_KERNEL); if (inuse == NULL) return -ENOMEM; - +cont: list_for_each_entry(i, &ctx->table->sets, list) { int tmp; if (!sscanf(i->name, name, &tmp)) continue; - if (tmp < 0 || tmp >= BITS_PER_BYTE * PAGE_SIZE) + if (tmp < min || tmp >= min + BITS_PER_BYTE * PAGE_SIZE) continue; - set_bit(tmp, inuse); + set_bit(tmp - min, inuse); } n = find_first_zero_bit(inuse, BITS_PER_BYTE * PAGE_SIZE); + if (n >= BITS_PER_BYTE * PAGE_SIZE) { + min += BITS_PER_BYTE * PAGE_SIZE; + memset(inuse, 0, PAGE_SIZE); + goto cont; + } free_page((unsigned long)inuse); } - snprintf(set->name, sizeof(set->name), name, n); + snprintf(set->name, sizeof(set->name), name, min + n); list_for_each_entry(i, &ctx->table->sets, list) { if (!strcmp(set->name, i->name)) return -ENFILE; @@ -2044,8 +2140,9 @@ static int nf_tables_fill_set(struct sk_buff *skb, const struct nft_ctx *ctx, { struct nfgenmsg *nfmsg; struct nlmsghdr *nlh; - u32 portid = NETLINK_CB(ctx->skb).portid; - u32 seq = ctx->nlh->nlmsg_seq; + struct nlattr *desc; + u32 portid = ctx->portid; + u32 seq = ctx->seq; event |= NFNL_SUBSYS_NFTABLES << 8; nlh = nlmsg_put(skb, portid, seq, event, sizeof(struct nfgenmsg), @@ -2077,6 +2174,14 @@ static int nf_tables_fill_set(struct sk_buff *skb, const struct nft_ctx *ctx, goto nla_put_failure; } + desc = nla_nest_start(skb, NFTA_SET_DESC); + if (desc == NULL) + goto nla_put_failure; + if (set->size && + nla_put_be32(skb, NFTA_SET_DESC_SIZE, htonl(set->size))) + goto nla_put_failure; + nla_nest_end(skb, desc); + return nlmsg_end(skb, nlh); nla_put_failure: @@ -2086,19 +2191,18 @@ nla_put_failure: static int nf_tables_set_notify(const struct nft_ctx *ctx, const struct nft_set *set, - int event) + int event, gfp_t gfp_flags) { struct sk_buff *skb; - u32 portid = NETLINK_CB(ctx->skb).portid; - bool report; + u32 portid = ctx->portid; int err; - report = nlmsg_report(ctx->nlh); - if (!report && !nfnetlink_has_listeners(ctx->net, NFNLGRP_NFTABLES)) + if (!ctx->report && + !nfnetlink_has_listeners(ctx->net, NFNLGRP_NFTABLES)) return 0; err = -ENOBUFS; - skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + skb = nlmsg_new(NLMSG_GOODSIZE, gfp_flags); if (skb == NULL) goto err; @@ -2108,8 +2212,8 @@ static int nf_tables_set_notify(const struct nft_ctx *ctx, goto err; } - err = nfnetlink_send(skb, ctx->net, portid, NFNLGRP_NFTABLES, report, - GFP_KERNEL); + err = nfnetlink_send(skb, ctx->net, portid, NFNLGRP_NFTABLES, + ctx->report, gfp_flags); err: if (err < 0) nfnetlink_set_err(ctx->net, portid, NFNLGRP_NFTABLES, err); @@ -2183,7 +2287,7 @@ static int nf_tables_dump_sets_all(struct nft_ctx *ctx, struct sk_buff *skb, { const struct nft_set *set; unsigned int idx, s_idx = cb->args[0]; - const struct nft_af_info *afi; + struct nft_af_info *afi; struct nft_table *table, *cur_table = (struct nft_table *)cb->args[2]; struct net *net = sock_net(skb->sk); int cur_family = cb->args[3]; @@ -2260,6 +2364,8 @@ static int nf_tables_dump_sets(struct sk_buff *skb, struct netlink_callback *cb) return ret; } +#define NFT_SET_INACTIVE (1 << 15) /* Internal set flag */ + static int nf_tables_getset(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) @@ -2289,6 +2395,8 @@ static int nf_tables_getset(struct sock *nlsk, struct sk_buff *skb, set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_NAME]); if (IS_ERR(set)) return PTR_ERR(set); + if (set->flags & NFT_SET_INACTIVE) + return -ENOENT; skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (skb2 == NULL) @@ -2305,13 +2413,50 @@ err: return err; } +static int nf_tables_set_desc_parse(const struct nft_ctx *ctx, + struct nft_set_desc *desc, + const struct nlattr *nla) +{ + struct nlattr *da[NFTA_SET_DESC_MAX + 1]; + int err; + + err = nla_parse_nested(da, NFTA_SET_DESC_MAX, nla, nft_set_desc_policy); + if (err < 0) + return err; + + if (da[NFTA_SET_DESC_SIZE] != NULL) + desc->size = ntohl(nla_get_be32(da[NFTA_SET_DESC_SIZE])); + + return 0; +} + +static int nft_trans_set_add(struct nft_ctx *ctx, int msg_type, + struct nft_set *set) +{ + struct nft_trans *trans; + + trans = nft_trans_alloc(ctx, msg_type, sizeof(struct nft_trans_set)); + if (trans == NULL) + return -ENOMEM; + + if (msg_type == NFT_MSG_NEWSET && ctx->nla[NFTA_SET_ID] != NULL) { + nft_trans_set_id(trans) = + ntohl(nla_get_be32(ctx->nla[NFTA_SET_ID])); + set->flags |= NFT_SET_INACTIVE; + } + nft_trans_set(trans) = set; + list_add_tail(&trans->list, &ctx->net->nft.commit_list); + + return 0; +} + static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); const struct nft_set_ops *ops; - const struct nft_af_info *afi; + struct nft_af_info *afi; struct net *net = sock_net(skb->sk); struct nft_table *table; struct nft_set *set; @@ -2319,14 +2464,18 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, char name[IFNAMSIZ]; unsigned int size; bool create; - u32 ktype, klen, dlen, dtype, flags; + u32 ktype, dtype, flags, policy; + struct nft_set_desc desc; int err; if (nla[NFTA_SET_TABLE] == NULL || nla[NFTA_SET_NAME] == NULL || - nla[NFTA_SET_KEY_LEN] == NULL) + nla[NFTA_SET_KEY_LEN] == NULL || + nla[NFTA_SET_ID] == NULL) return -EINVAL; + memset(&desc, 0, sizeof(desc)); + ktype = NFT_DATA_VALUE; if (nla[NFTA_SET_KEY_TYPE] != NULL) { ktype = ntohl(nla_get_be32(nla[NFTA_SET_KEY_TYPE])); @@ -2334,8 +2483,8 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, return -EINVAL; } - klen = ntohl(nla_get_be32(nla[NFTA_SET_KEY_LEN])); - if (klen == 0 || klen > FIELD_SIZEOF(struct nft_data, data)) + desc.klen = ntohl(nla_get_be32(nla[NFTA_SET_KEY_LEN])); + if (desc.klen == 0 || desc.klen > FIELD_SIZEOF(struct nft_data, data)) return -EINVAL; flags = 0; @@ -2347,7 +2496,6 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, } dtype = 0; - dlen = 0; if (nla[NFTA_SET_DATA_TYPE] != NULL) { if (!(flags & NFT_SET_MAP)) return -EINVAL; @@ -2360,15 +2508,25 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, if (dtype != NFT_DATA_VERDICT) { if (nla[NFTA_SET_DATA_LEN] == NULL) return -EINVAL; - dlen = ntohl(nla_get_be32(nla[NFTA_SET_DATA_LEN])); - if (dlen == 0 || - dlen > FIELD_SIZEOF(struct nft_data, data)) + desc.dlen = ntohl(nla_get_be32(nla[NFTA_SET_DATA_LEN])); + if (desc.dlen == 0 || + desc.dlen > FIELD_SIZEOF(struct nft_data, data)) return -EINVAL; } else - dlen = sizeof(struct nft_data); + desc.dlen = sizeof(struct nft_data); } else if (flags & NFT_SET_MAP) return -EINVAL; + policy = NFT_SET_POL_PERFORMANCE; + if (nla[NFTA_SET_POLICY] != NULL) + policy = ntohl(nla_get_be32(nla[NFTA_SET_POLICY])); + + if (nla[NFTA_SET_DESC] != NULL) { + err = nf_tables_set_desc_parse(&ctx, &desc, nla[NFTA_SET_DESC]); + if (err < 0) + return err; + } + create = nlh->nlmsg_flags & NLM_F_CREATE ? true : false; afi = nf_tables_afinfo_lookup(net, nfmsg->nfgen_family, create); @@ -2399,7 +2557,7 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, if (!(nlh->nlmsg_flags & NLM_F_CREATE)) return -ENOENT; - ops = nft_select_set_ops(nla); + ops = nft_select_set_ops(nla, &desc, policy); if (IS_ERR(ops)) return PTR_ERR(ops); @@ -2420,17 +2578,22 @@ static int nf_tables_newset(struct sock *nlsk, struct sk_buff *skb, INIT_LIST_HEAD(&set->bindings); set->ops = ops; set->ktype = ktype; - set->klen = klen; + set->klen = desc.klen; set->dtype = dtype; - set->dlen = dlen; + set->dlen = desc.dlen; set->flags = flags; + set->size = desc.size; + + err = ops->init(set, &desc, nla); + if (err < 0) + goto err2; - err = ops->init(set, nla); + err = nft_trans_set_add(&ctx, NFT_MSG_NEWSET, set); if (err < 0) goto err2; list_add_tail(&set->list, &table->sets); - nf_tables_set_notify(&ctx, set, NFT_MSG_NEWSET); + table->use++; return 0; err2: @@ -2440,16 +2603,20 @@ err1: return err; } -static void nf_tables_set_destroy(const struct nft_ctx *ctx, struct nft_set *set) +static void nft_set_destroy(struct nft_set *set) { - list_del(&set->list); - nf_tables_set_notify(ctx, set, NFT_MSG_DELSET); - set->ops->destroy(set); module_put(set->ops->owner); kfree(set); } +static void nf_tables_set_destroy(const struct nft_ctx *ctx, struct nft_set *set) +{ + list_del(&set->list); + nf_tables_set_notify(ctx, set, NFT_MSG_DELSET, GFP_ATOMIC); + nft_set_destroy(set); +} + static int nf_tables_delset(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) @@ -2471,10 +2638,17 @@ static int nf_tables_delset(struct sock *nlsk, struct sk_buff *skb, set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_NAME]); if (IS_ERR(set)) return PTR_ERR(set); + if (set->flags & NFT_SET_INACTIVE) + return -ENOENT; if (!list_empty(&set->bindings)) return -EBUSY; - nf_tables_set_destroy(&ctx, set); + err = nft_trans_set_add(&ctx, NFT_MSG_DELSET, set); + if (err < 0) + return err; + + list_del(&set->list); + ctx.table->use--; return 0; } @@ -2534,7 +2708,8 @@ void nf_tables_unbind_set(const struct nft_ctx *ctx, struct nft_set *set, { list_del(&binding->list); - if (list_empty(&set->bindings) && set->flags & NFT_SET_ANONYMOUS) + if (list_empty(&set->bindings) && set->flags & NFT_SET_ANONYMOUS && + !(set->flags & NFT_SET_INACTIVE)) nf_tables_set_destroy(ctx, set); } @@ -2552,16 +2727,18 @@ static const struct nla_policy nft_set_elem_list_policy[NFTA_SET_ELEM_LIST_MAX + [NFTA_SET_ELEM_LIST_TABLE] = { .type = NLA_STRING }, [NFTA_SET_ELEM_LIST_SET] = { .type = NLA_STRING }, [NFTA_SET_ELEM_LIST_ELEMENTS] = { .type = NLA_NESTED }, + [NFTA_SET_ELEM_LIST_SET_ID] = { .type = NLA_U32 }, }; static int nft_ctx_init_from_elemattr(struct nft_ctx *ctx, const struct sk_buff *skb, const struct nlmsghdr *nlh, - const struct nlattr * const nla[]) + const struct nlattr * const nla[], + bool trans) { const struct nfgenmsg *nfmsg = nlmsg_data(nlh); - const struct nft_af_info *afi; - const struct nft_table *table; + struct nft_af_info *afi; + struct nft_table *table; struct net *net = sock_net(skb->sk); afi = nf_tables_afinfo_lookup(net, nfmsg->nfgen_family, false); @@ -2571,6 +2748,8 @@ static int nft_ctx_init_from_elemattr(struct nft_ctx *ctx, table = nf_tables_table_lookup(afi, nla[NFTA_SET_ELEM_LIST_TABLE]); if (IS_ERR(table)) return PTR_ERR(table); + if (!trans && (table->flags & NFT_TABLE_INACTIVE)) + return -ENOENT; nft_ctx_init(ctx, skb, nlh, afi, table, NULL, nla); return 0; @@ -2644,13 +2823,16 @@ static int nf_tables_dump_set(struct sk_buff *skb, struct netlink_callback *cb) if (err < 0) return err; - err = nft_ctx_init_from_elemattr(&ctx, cb->skb, cb->nlh, (void *)nla); + err = nft_ctx_init_from_elemattr(&ctx, cb->skb, cb->nlh, (void *)nla, + false); if (err < 0) return err; set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); if (IS_ERR(set)) return PTR_ERR(set); + if (set->flags & NFT_SET_INACTIVE) + return -ENOENT; event = NFT_MSG_NEWSETELEM; event |= NFNL_SUBSYS_NFTABLES << 8; @@ -2707,13 +2889,15 @@ static int nf_tables_getsetelem(struct sock *nlsk, struct sk_buff *skb, struct nft_ctx ctx; int err; - err = nft_ctx_init_from_elemattr(&ctx, skb, nlh, nla); + err = nft_ctx_init_from_elemattr(&ctx, skb, nlh, nla, false); if (err < 0) return err; set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); if (IS_ERR(set)) return PTR_ERR(set); + if (set->flags & NFT_SET_INACTIVE) + return -ENOENT; if (nlh->nlmsg_flags & NLM_F_DUMP) { struct netlink_dump_control c = { @@ -2724,7 +2908,98 @@ static int nf_tables_getsetelem(struct sock *nlsk, struct sk_buff *skb, return -EOPNOTSUPP; } -static int nft_add_set_elem(const struct nft_ctx *ctx, struct nft_set *set, +static int nf_tables_fill_setelem_info(struct sk_buff *skb, + const struct nft_ctx *ctx, u32 seq, + u32 portid, int event, u16 flags, + const struct nft_set *set, + const struct nft_set_elem *elem) +{ + struct nfgenmsg *nfmsg; + struct nlmsghdr *nlh; + struct nlattr *nest; + int err; + + event |= NFNL_SUBSYS_NFTABLES << 8; + nlh = nlmsg_put(skb, portid, seq, event, sizeof(struct nfgenmsg), + flags); + if (nlh == NULL) + goto nla_put_failure; + + nfmsg = nlmsg_data(nlh); + nfmsg->nfgen_family = ctx->afi->family; + nfmsg->version = NFNETLINK_V0; + nfmsg->res_id = 0; + + if (nla_put_string(skb, NFTA_SET_TABLE, ctx->table->name)) + goto nla_put_failure; + if (nla_put_string(skb, NFTA_SET_NAME, set->name)) + goto nla_put_failure; + + nest = nla_nest_start(skb, NFTA_SET_ELEM_LIST_ELEMENTS); + if (nest == NULL) + goto nla_put_failure; + + err = nf_tables_fill_setelem(skb, set, elem); + if (err < 0) + goto nla_put_failure; + + nla_nest_end(skb, nest); + + return nlmsg_end(skb, nlh); + +nla_put_failure: + nlmsg_trim(skb, nlh); + return -1; +} + +static int nf_tables_setelem_notify(const struct nft_ctx *ctx, + const struct nft_set *set, + const struct nft_set_elem *elem, + int event, u16 flags) +{ + struct net *net = ctx->net; + u32 portid = ctx->portid; + struct sk_buff *skb; + int err; + + if (!ctx->report && !nfnetlink_has_listeners(net, NFNLGRP_NFTABLES)) + return 0; + + err = -ENOBUFS; + skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + if (skb == NULL) + goto err; + + err = nf_tables_fill_setelem_info(skb, ctx, 0, portid, event, flags, + set, elem); + if (err < 0) { + kfree_skb(skb); + goto err; + } + + err = nfnetlink_send(skb, net, portid, NFNLGRP_NFTABLES, ctx->report, + GFP_KERNEL); +err: + if (err < 0) + nfnetlink_set_err(net, portid, NFNLGRP_NFTABLES, err); + return err; +} + +static struct nft_trans *nft_trans_elem_alloc(struct nft_ctx *ctx, + int msg_type, + struct nft_set *set) +{ + struct nft_trans *trans; + + trans = nft_trans_alloc(ctx, msg_type, sizeof(struct nft_trans_elem)); + if (trans == NULL) + return NULL; + + nft_trans_elem_set(trans) = set; + return trans; +} + +static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set, const struct nlattr *attr) { struct nlattr *nla[NFTA_SET_ELEM_MAX + 1]; @@ -2732,8 +3007,12 @@ static int nft_add_set_elem(const struct nft_ctx *ctx, struct nft_set *set, struct nft_set_elem elem; struct nft_set_binding *binding; enum nft_registers dreg; + struct nft_trans *trans; int err; + if (set->size && set->nelems == set->size) + return -ENFILE; + err = nla_parse_nested(nla, NFTA_SET_ELEM_MAX, attr, nft_set_elem_policy); if (err < 0) @@ -2786,7 +3065,7 @@ static int nft_add_set_elem(const struct nft_ctx *ctx, struct nft_set *set, struct nft_ctx bind_ctx = { .afi = ctx->afi, .table = ctx->table, - .chain = binding->chain, + .chain = (struct nft_chain *)binding->chain, }; err = nft_validate_data_load(&bind_ctx, dreg, @@ -2796,12 +3075,20 @@ static int nft_add_set_elem(const struct nft_ctx *ctx, struct nft_set *set, } } + trans = nft_trans_elem_alloc(ctx, NFT_MSG_NEWSETELEM, set); + if (trans == NULL) + goto err3; + err = set->ops->insert(set, &elem); if (err < 0) - goto err3; + goto err4; + nft_trans_elem(trans) = elem; + list_add_tail(&trans->list, &ctx->net->nft.commit_list); return 0; +err4: + kfree(trans); err3: if (nla[NFTA_SET_ELEM_DATA] != NULL) nft_data_uninit(&elem.data, d2.type); @@ -2815,35 +3102,46 @@ static int nf_tables_newsetelem(struct sock *nlsk, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const nla[]) { + struct net *net = sock_net(skb->sk); const struct nlattr *attr; struct nft_set *set; struct nft_ctx ctx; - int rem, err; + int rem, err = 0; - err = nft_ctx_init_from_elemattr(&ctx, skb, nlh, nla); + err = nft_ctx_init_from_elemattr(&ctx, skb, nlh, nla, true); if (err < 0) return err; set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET]); - if (IS_ERR(set)) - return PTR_ERR(set); + if (IS_ERR(set)) { + if (nla[NFTA_SET_ELEM_LIST_SET_ID]) { + set = nf_tables_set_lookup_byid(net, + nla[NFTA_SET_ELEM_LIST_SET_ID]); + } + if (IS_ERR(set)) + return PTR_ERR(set); + } + if (!list_empty(&set->bindings) && set->flags & NFT_SET_CONSTANT) return -EBUSY; nla_for_each_nested(attr, nla[NFTA_SET_ELEM_LIST_ELEMENTS], rem) { err = nft_add_set_elem(&ctx, set, attr); if (err < 0) - return err; + break; + + set->nelems++; } - return 0; + return err; } -static int nft_del_setelem(const struct nft_ctx *ctx, struct nft_set *set, +static int nft_del_setelem(struct nft_ctx *ctx, struct nft_set *set, const struct nlattr *attr) { struct nlattr *nla[NFTA_SET_ELEM_MAX + 1]; struct nft_data_desc desc; struct nft_set_elem elem; + struct nft_trans *trans; int err; err = nla_parse_nested(nla, NFTA_SET_ELEM_MAX, attr, @@ -2867,7 +3165,12 @@ static int nft_del_setelem(const struct nft_ctx *ctx, struct nft_set *set, if (err < 0) goto err2; - set->ops->remove(set, &elem); + trans = nft_trans_elem_alloc(ctx, NFT_MSG_DELSETELEM, set); + if (trans == NULL) + goto err2; + + nft_trans_elem(trans) = elem; + list_add_tail(&trans->list, &ctx->net->nft.commit_list); nft_data_uninit(&elem.key, NFT_DATA_VALUE); if (set->flags & NFT_SET_MAP) @@ -2886,9 +3189,9 @@ static int nf_tables_delsetelem(struct sock *nlsk, struct sk_buff *skb, const struct nlattr *attr; struct nft_set *set; struct nft_ctx ctx; - int rem, err; + int rem, err = 0; - err = nft_ctx_init_from_elemattr(&ctx, skb, nlh, nla); + err = nft_ctx_init_from_elemattr(&ctx, skb, nlh, nla, false); if (err < 0) return err; @@ -2901,14 +3204,16 @@ static int nf_tables_delsetelem(struct sock *nlsk, struct sk_buff *skb, nla_for_each_nested(attr, nla[NFTA_SET_ELEM_LIST_ELEMENTS], rem) { err = nft_del_setelem(&ctx, set, attr); if (err < 0) - return err; + break; + + set->nelems--; } - return 0; + return err; } static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { [NFT_MSG_NEWTABLE] = { - .call = nf_tables_newtable, + .call_batch = nf_tables_newtable, .attr_count = NFTA_TABLE_MAX, .policy = nft_table_policy, }, @@ -2918,12 +3223,12 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { .policy = nft_table_policy, }, [NFT_MSG_DELTABLE] = { - .call = nf_tables_deltable, + .call_batch = nf_tables_deltable, .attr_count = NFTA_TABLE_MAX, .policy = nft_table_policy, }, [NFT_MSG_NEWCHAIN] = { - .call = nf_tables_newchain, + .call_batch = nf_tables_newchain, .attr_count = NFTA_CHAIN_MAX, .policy = nft_chain_policy, }, @@ -2933,7 +3238,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { .policy = nft_chain_policy, }, [NFT_MSG_DELCHAIN] = { - .call = nf_tables_delchain, + .call_batch = nf_tables_delchain, .attr_count = NFTA_CHAIN_MAX, .policy = nft_chain_policy, }, @@ -2953,7 +3258,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { .policy = nft_rule_policy, }, [NFT_MSG_NEWSET] = { - .call = nf_tables_newset, + .call_batch = nf_tables_newset, .attr_count = NFTA_SET_MAX, .policy = nft_set_policy, }, @@ -2963,12 +3268,12 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { .policy = nft_set_policy, }, [NFT_MSG_DELSET] = { - .call = nf_tables_delset, + .call_batch = nf_tables_delset, .attr_count = NFTA_SET_MAX, .policy = nft_set_policy, }, [NFT_MSG_NEWSETELEM] = { - .call = nf_tables_newsetelem, + .call_batch = nf_tables_newsetelem, .attr_count = NFTA_SET_ELEM_LIST_MAX, .policy = nft_set_elem_list_policy, }, @@ -2978,12 +3283,282 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = { .policy = nft_set_elem_list_policy, }, [NFT_MSG_DELSETELEM] = { - .call = nf_tables_delsetelem, + .call_batch = nf_tables_delsetelem, .attr_count = NFTA_SET_ELEM_LIST_MAX, .policy = nft_set_elem_list_policy, }, }; +static void nft_chain_commit_update(struct nft_trans *trans) +{ + struct nft_base_chain *basechain; + + if (nft_trans_chain_name(trans)[0]) + strcpy(trans->ctx.chain->name, nft_trans_chain_name(trans)); + + if (!(trans->ctx.chain->flags & NFT_BASE_CHAIN)) + return; + + basechain = nft_base_chain(trans->ctx.chain); + nft_chain_stats_replace(basechain, nft_trans_chain_stats(trans)); + + switch (nft_trans_chain_policy(trans)) { + case NF_DROP: + case NF_ACCEPT: + basechain->policy = nft_trans_chain_policy(trans); + break; + } +} + +/* Schedule objects for release via rcu to make sure no packets are accesing + * removed rules. + */ +static void nf_tables_commit_release_rcu(struct rcu_head *rt) +{ + struct nft_trans *trans = container_of(rt, struct nft_trans, rcu_head); + + switch (trans->msg_type) { + case NFT_MSG_DELTABLE: + nf_tables_table_destroy(&trans->ctx); + break; + case NFT_MSG_DELCHAIN: + nf_tables_chain_destroy(trans->ctx.chain); + break; + case NFT_MSG_DELRULE: + nf_tables_rule_destroy(&trans->ctx, nft_trans_rule(trans)); + break; + case NFT_MSG_DELSET: + nft_set_destroy(nft_trans_set(trans)); + break; + } + kfree(trans); +} + +static int nf_tables_commit(struct sk_buff *skb) +{ + struct net *net = sock_net(skb->sk); + struct nft_trans *trans, *next; + struct nft_set *set; + + /* Bump generation counter, invalidate any dump in progress */ + net->nft.genctr++; + + /* A new generation has just started */ + net->nft.gencursor = gencursor_next(net); + + /* Make sure all packets have left the previous generation before + * purging old rules. + */ + synchronize_rcu(); + + list_for_each_entry_safe(trans, next, &net->nft.commit_list, list) { + switch (trans->msg_type) { + case NFT_MSG_NEWTABLE: + if (nft_trans_table_update(trans)) { + if (!nft_trans_table_enable(trans)) { + nf_tables_table_disable(trans->ctx.afi, + trans->ctx.table); + trans->ctx.table->flags |= NFT_TABLE_F_DORMANT; + } + } else { + trans->ctx.table->flags &= ~NFT_TABLE_INACTIVE; + } + nf_tables_table_notify(&trans->ctx, NFT_MSG_NEWTABLE); + nft_trans_destroy(trans); + break; + case NFT_MSG_DELTABLE: + nf_tables_table_notify(&trans->ctx, NFT_MSG_DELTABLE); + break; + case NFT_MSG_NEWCHAIN: + if (nft_trans_chain_update(trans)) + nft_chain_commit_update(trans); + else + trans->ctx.chain->flags &= ~NFT_CHAIN_INACTIVE; + + nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN); + nft_trans_destroy(trans); + break; + case NFT_MSG_DELCHAIN: + nf_tables_chain_notify(&trans->ctx, NFT_MSG_DELCHAIN); + if (!(trans->ctx.table->flags & NFT_TABLE_F_DORMANT) && + trans->ctx.chain->flags & NFT_BASE_CHAIN) { + nf_unregister_hooks(nft_base_chain(trans->ctx.chain)->ops, + trans->ctx.afi->nops); + } + break; + case NFT_MSG_NEWRULE: + nft_rule_clear(trans->ctx.net, nft_trans_rule(trans)); + nf_tables_rule_notify(&trans->ctx, + nft_trans_rule(trans), + NFT_MSG_NEWRULE); + nft_trans_destroy(trans); + break; + case NFT_MSG_DELRULE: + list_del_rcu(&nft_trans_rule(trans)->list); + nf_tables_rule_notify(&trans->ctx, + nft_trans_rule(trans), + NFT_MSG_DELRULE); + break; + case NFT_MSG_NEWSET: + nft_trans_set(trans)->flags &= ~NFT_SET_INACTIVE; + /* This avoids hitting -EBUSY when deleting the table + * from the transaction. + */ + if (nft_trans_set(trans)->flags & NFT_SET_ANONYMOUS && + !list_empty(&nft_trans_set(trans)->bindings)) + trans->ctx.table->use--; + + nf_tables_set_notify(&trans->ctx, nft_trans_set(trans), + NFT_MSG_NEWSET, GFP_KERNEL); + nft_trans_destroy(trans); + break; + case NFT_MSG_DELSET: + nf_tables_set_notify(&trans->ctx, nft_trans_set(trans), + NFT_MSG_DELSET, GFP_KERNEL); + break; + case NFT_MSG_NEWSETELEM: + nf_tables_setelem_notify(&trans->ctx, + nft_trans_elem_set(trans), + &nft_trans_elem(trans), + NFT_MSG_NEWSETELEM, 0); + nft_trans_destroy(trans); + break; + case NFT_MSG_DELSETELEM: + nf_tables_setelem_notify(&trans->ctx, + nft_trans_elem_set(trans), + &nft_trans_elem(trans), + NFT_MSG_DELSETELEM, 0); + set = nft_trans_elem_set(trans); + set->ops->get(set, &nft_trans_elem(trans)); + set->ops->remove(set, &nft_trans_elem(trans)); + nft_trans_destroy(trans); + break; + } + } + + list_for_each_entry_safe(trans, next, &net->nft.commit_list, list) { + list_del(&trans->list); + trans->ctx.nla = NULL; + call_rcu(&trans->rcu_head, nf_tables_commit_release_rcu); + } + + return 0; +} + +/* Schedule objects for release via rcu to make sure no packets are accesing + * aborted rules. + */ +static void nf_tables_abort_release_rcu(struct rcu_head *rt) +{ + struct nft_trans *trans = container_of(rt, struct nft_trans, rcu_head); + + switch (trans->msg_type) { + case NFT_MSG_NEWTABLE: + nf_tables_table_destroy(&trans->ctx); + break; + case NFT_MSG_NEWCHAIN: + nf_tables_chain_destroy(trans->ctx.chain); + break; + case NFT_MSG_NEWRULE: + nf_tables_rule_destroy(&trans->ctx, nft_trans_rule(trans)); + break; + case NFT_MSG_NEWSET: + nft_set_destroy(nft_trans_set(trans)); + break; + } + kfree(trans); +} + +static int nf_tables_abort(struct sk_buff *skb) +{ + struct net *net = sock_net(skb->sk); + struct nft_trans *trans, *next; + struct nft_set *set; + + list_for_each_entry_safe(trans, next, &net->nft.commit_list, list) { + switch (trans->msg_type) { + case NFT_MSG_NEWTABLE: + if (nft_trans_table_update(trans)) { + if (nft_trans_table_enable(trans)) { + nf_tables_table_disable(trans->ctx.afi, + trans->ctx.table); + trans->ctx.table->flags |= NFT_TABLE_F_DORMANT; + } + nft_trans_destroy(trans); + } else { + list_del(&trans->ctx.table->list); + } + break; + case NFT_MSG_DELTABLE: + list_add_tail(&trans->ctx.table->list, + &trans->ctx.afi->tables); + nft_trans_destroy(trans); + break; + case NFT_MSG_NEWCHAIN: + if (nft_trans_chain_update(trans)) { + if (nft_trans_chain_stats(trans)) + free_percpu(nft_trans_chain_stats(trans)); + + nft_trans_destroy(trans); + } else { + trans->ctx.table->use--; + list_del(&trans->ctx.chain->list); + if (!(trans->ctx.table->flags & NFT_TABLE_F_DORMANT) && + trans->ctx.chain->flags & NFT_BASE_CHAIN) { + nf_unregister_hooks(nft_base_chain(trans->ctx.chain)->ops, + trans->ctx.afi->nops); + } + } + break; + case NFT_MSG_DELCHAIN: + trans->ctx.table->use++; + list_add_tail(&trans->ctx.chain->list, + &trans->ctx.table->chains); + nft_trans_destroy(trans); + break; + case NFT_MSG_NEWRULE: + trans->ctx.chain->use--; + list_del_rcu(&nft_trans_rule(trans)->list); + break; + case NFT_MSG_DELRULE: + trans->ctx.chain->use++; + nft_rule_clear(trans->ctx.net, nft_trans_rule(trans)); + nft_trans_destroy(trans); + break; + case NFT_MSG_NEWSET: + trans->ctx.table->use--; + list_del(&nft_trans_set(trans)->list); + break; + case NFT_MSG_DELSET: + trans->ctx.table->use++; + list_add_tail(&nft_trans_set(trans)->list, + &trans->ctx.table->sets); + nft_trans_destroy(trans); + break; + case NFT_MSG_NEWSETELEM: + nft_trans_elem_set(trans)->nelems--; + set = nft_trans_elem_set(trans); + set->ops->get(set, &nft_trans_elem(trans)); + set->ops->remove(set, &nft_trans_elem(trans)); + nft_trans_destroy(trans); + break; + case NFT_MSG_DELSETELEM: + nft_trans_elem_set(trans)->nelems++; + nft_trans_destroy(trans); + break; + } + } + + list_for_each_entry_safe_reverse(trans, next, + &net->nft.commit_list, list) { + list_del(&trans->list); + trans->ctx.nla = NULL; + call_rcu(&trans->rcu_head, nf_tables_abort_release_rcu); + } + + return 0; +} + static const struct nfnetlink_subsystem nf_tables_subsys = { .name = "nf_tables", .subsys_id = NFNL_SUBSYS_NFTABLES, diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index 23ef77c60fff..c138b8fbe280 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -399,19 +399,17 @@ static void nfnetlink_rcv(struct sk_buff *skb) } #ifdef CONFIG_MODULES -static void nfnetlink_bind(int group) +static int nfnetlink_bind(int group) { const struct nfnetlink_subsystem *ss; int type = nfnl_group2type[group]; rcu_read_lock(); ss = nfnetlink_get_subsys(type); - if (!ss) { - rcu_read_unlock(); - request_module("nfnetlink-subsys-%d", type); - return; - } rcu_read_unlock(); + if (!ss) + request_module("nfnetlink-subsys-%d", type); + return 0; } #endif diff --git a/net/netfilter/nfnetlink_acct.c b/net/netfilter/nfnetlink_acct.c index c7b6d466a662..2baa125c2e8d 100644 --- a/net/netfilter/nfnetlink_acct.c +++ b/net/netfilter/nfnetlink_acct.c @@ -32,18 +32,24 @@ static LIST_HEAD(nfnl_acct_list); struct nf_acct { atomic64_t pkts; atomic64_t bytes; + unsigned long flags; struct list_head head; atomic_t refcnt; char name[NFACCT_NAME_MAX]; struct rcu_head rcu_head; + char data[0]; }; +#define NFACCT_F_QUOTA (NFACCT_F_QUOTA_PKTS | NFACCT_F_QUOTA_BYTES) + static int nfnl_acct_new(struct sock *nfnl, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const tb[]) { struct nf_acct *nfacct, *matching = NULL; char *acct_name; + unsigned int size = 0; + u32 flags = 0; if (!tb[NFACCT_NAME]) return -EINVAL; @@ -68,15 +74,38 @@ nfnl_acct_new(struct sock *nfnl, struct sk_buff *skb, /* reset counters if you request a replacement. */ atomic64_set(&matching->pkts, 0); atomic64_set(&matching->bytes, 0); + smp_mb__before_atomic(); + /* reset overquota flag if quota is enabled. */ + if ((matching->flags & NFACCT_F_QUOTA)) + clear_bit(NFACCT_F_OVERQUOTA, &matching->flags); return 0; } return -EBUSY; } - nfacct = kzalloc(sizeof(struct nf_acct), GFP_KERNEL); + if (tb[NFACCT_FLAGS]) { + flags = ntohl(nla_get_be32(tb[NFACCT_FLAGS])); + if (flags & ~NFACCT_F_QUOTA) + return -EOPNOTSUPP; + if ((flags & NFACCT_F_QUOTA) == NFACCT_F_QUOTA) + return -EINVAL; + if (flags & NFACCT_F_OVERQUOTA) + return -EINVAL; + + size += sizeof(u64); + } + + nfacct = kzalloc(sizeof(struct nf_acct) + size, GFP_KERNEL); if (nfacct == NULL) return -ENOMEM; + if (flags & NFACCT_F_QUOTA) { + u64 *quota = (u64 *)nfacct->data; + + *quota = be64_to_cpu(nla_get_be64(tb[NFACCT_QUOTA])); + nfacct->flags = flags; + } + strncpy(nfacct->name, nla_data(tb[NFACCT_NAME]), NFACCT_NAME_MAX); if (tb[NFACCT_BYTES]) { @@ -117,6 +146,9 @@ nfnl_acct_fill_info(struct sk_buff *skb, u32 portid, u32 seq, u32 type, if (type == NFNL_MSG_ACCT_GET_CTRZERO) { pkts = atomic64_xchg(&acct->pkts, 0); bytes = atomic64_xchg(&acct->bytes, 0); + smp_mb__before_atomic(); + if (acct->flags & NFACCT_F_QUOTA) + clear_bit(NFACCT_F_OVERQUOTA, &acct->flags); } else { pkts = atomic64_read(&acct->pkts); bytes = atomic64_read(&acct->bytes); @@ -125,7 +157,13 @@ nfnl_acct_fill_info(struct sk_buff *skb, u32 portid, u32 seq, u32 type, nla_put_be64(skb, NFACCT_BYTES, cpu_to_be64(bytes)) || nla_put_be32(skb, NFACCT_USE, htonl(atomic_read(&acct->refcnt)))) goto nla_put_failure; + if (acct->flags & NFACCT_F_QUOTA) { + u64 *quota = (u64 *)acct->data; + if (nla_put_be32(skb, NFACCT_FLAGS, htonl(acct->flags)) || + nla_put_be64(skb, NFACCT_QUOTA, cpu_to_be64(*quota))) + goto nla_put_failure; + } nlmsg_end(skb, nlh); return skb->len; @@ -270,6 +308,8 @@ static const struct nla_policy nfnl_acct_policy[NFACCT_MAX+1] = { [NFACCT_NAME] = { .type = NLA_NUL_STRING, .len = NFACCT_NAME_MAX-1 }, [NFACCT_BYTES] = { .type = NLA_U64 }, [NFACCT_PKTS] = { .type = NLA_U64 }, + [NFACCT_FLAGS] = { .type = NLA_U32 }, + [NFACCT_QUOTA] = { .type = NLA_U64 }, }; static const struct nfnl_callback nfnl_acct_cb[NFNL_MSG_ACCT_MAX] = { @@ -336,6 +376,50 @@ void nfnl_acct_update(const struct sk_buff *skb, struct nf_acct *nfacct) } EXPORT_SYMBOL_GPL(nfnl_acct_update); +static void nfnl_overquota_report(struct nf_acct *nfacct) +{ + int ret; + struct sk_buff *skb; + + skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC); + if (skb == NULL) + return; + + ret = nfnl_acct_fill_info(skb, 0, 0, NFNL_MSG_ACCT_OVERQUOTA, 0, + nfacct); + if (ret <= 0) { + kfree_skb(skb); + return; + } + netlink_broadcast(init_net.nfnl, skb, 0, NFNLGRP_ACCT_QUOTA, + GFP_ATOMIC); +} + +int nfnl_acct_overquota(const struct sk_buff *skb, struct nf_acct *nfacct) +{ + u64 now; + u64 *quota; + int ret = NFACCT_UNDERQUOTA; + + /* no place here if we don't have a quota */ + if (!(nfacct->flags & NFACCT_F_QUOTA)) + return NFACCT_NO_QUOTA; + + quota = (u64 *)nfacct->data; + now = (nfacct->flags & NFACCT_F_QUOTA_PKTS) ? + atomic64_read(&nfacct->pkts) : atomic64_read(&nfacct->bytes); + + ret = now > *quota; + + if (now >= *quota && + !test_and_set_bit(NFACCT_F_OVERQUOTA, &nfacct->flags)) { + nfnl_overquota_report(nfacct); + } + + return ret; +} +EXPORT_SYMBOL_GPL(nfnl_acct_overquota); + static int __init nfnl_acct_init(void) { int ret; diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c index bd0d41e69341..cc5603016242 100644 --- a/net/netfilter/nft_ct.c +++ b/net/netfilter/nft_ct.c @@ -215,22 +215,14 @@ static void nft_ct_l3proto_module_put(uint8_t family) nf_ct_l3proto_module_put(family); } -static int nft_ct_init_validate_get(const struct nft_expr *expr, - const struct nlattr * const tb[]) +static int nft_ct_get_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) { struct nft_ct *priv = nft_expr_priv(expr); + int err; - if (tb[NFTA_CT_DIRECTION] != NULL) { - priv->dir = nla_get_u8(tb[NFTA_CT_DIRECTION]); - switch (priv->dir) { - case IP_CT_DIR_ORIGINAL: - case IP_CT_DIR_REPLY: - break; - default: - return -EINVAL; - } - } - + priv->key = ntohl(nla_get_be32(tb[NFTA_CT_KEY])); switch (priv->key) { case NFT_CT_STATE: case NFT_CT_DIRECTION: @@ -262,55 +254,55 @@ static int nft_ct_init_validate_get(const struct nft_expr *expr, return -EOPNOTSUPP; } - return 0; -} - -static int nft_ct_init_validate_set(uint32_t key) -{ - switch (key) { - case NFT_CT_MARK: - break; - default: - return -EOPNOTSUPP; + if (tb[NFTA_CT_DIRECTION] != NULL) { + priv->dir = nla_get_u8(tb[NFTA_CT_DIRECTION]); + switch (priv->dir) { + case IP_CT_DIR_ORIGINAL: + case IP_CT_DIR_REPLY: + break; + default: + return -EINVAL; + } } + priv->dreg = ntohl(nla_get_be32(tb[NFTA_CT_DREG])); + err = nft_validate_output_register(priv->dreg); + if (err < 0) + return err; + + err = nft_validate_data_load(ctx, priv->dreg, NULL, NFT_DATA_VALUE); + if (err < 0) + return err; + + err = nft_ct_l3proto_try_module_get(ctx->afi->family); + if (err < 0) + return err; + return 0; } -static int nft_ct_init(const struct nft_ctx *ctx, - const struct nft_expr *expr, - const struct nlattr * const tb[]) +static int nft_ct_set_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) { struct nft_ct *priv = nft_expr_priv(expr); int err; priv->key = ntohl(nla_get_be32(tb[NFTA_CT_KEY])); - - if (tb[NFTA_CT_DREG]) { - err = nft_ct_init_validate_get(expr, tb); - if (err < 0) - return err; - - priv->dreg = ntohl(nla_get_be32(tb[NFTA_CT_DREG])); - err = nft_validate_output_register(priv->dreg); - if (err < 0) - return err; - - err = nft_validate_data_load(ctx, priv->dreg, NULL, - NFT_DATA_VALUE); - if (err < 0) - return err; - } else { - err = nft_ct_init_validate_set(priv->key); - if (err < 0) - return err; - - priv->sreg = ntohl(nla_get_be32(tb[NFTA_CT_SREG])); - err = nft_validate_input_register(priv->sreg); - if (err < 0) - return err; + switch (priv->key) { +#ifdef CONFIG_NF_CONNTRACK_MARK + case NFT_CT_MARK: + break; +#endif + default: + return -EOPNOTSUPP; } + priv->sreg = ntohl(nla_get_be32(tb[NFTA_CT_SREG])); + err = nft_validate_input_register(priv->sreg); + if (err < 0) + return err; + err = nft_ct_l3proto_try_module_get(ctx->afi->family); if (err < 0) return err; @@ -370,7 +362,7 @@ static const struct nft_expr_ops nft_ct_get_ops = { .type = &nft_ct_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_ct)), .eval = nft_ct_get_eval, - .init = nft_ct_init, + .init = nft_ct_get_init, .destroy = nft_ct_destroy, .dump = nft_ct_get_dump, }; @@ -379,7 +371,7 @@ static const struct nft_expr_ops nft_ct_set_ops = { .type = &nft_ct_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_ct)), .eval = nft_ct_set_eval, - .init = nft_ct_init, + .init = nft_ct_set_init, .destroy = nft_ct_destroy, .dump = nft_ct_set_dump, }; diff --git a/net/netfilter/nft_hash.c b/net/netfilter/nft_hash.c index 3b1ad876d6b0..4080ed6a072b 100644 --- a/net/netfilter/nft_hash.c +++ b/net/netfilter/nft_hash.c @@ -12,6 +12,7 @@ #include <linux/init.h> #include <linux/module.h> #include <linux/list.h> +#include <linux/log2.h> #include <linux/jhash.h> #include <linux/netlink.h> #include <linux/vmalloc.h> @@ -19,7 +20,7 @@ #include <linux/netfilter/nf_tables.h> #include <net/netfilter/nf_tables.h> -#define NFT_HASH_MIN_SIZE 4 +#define NFT_HASH_MIN_SIZE 4UL struct nft_hash { struct nft_hash_table __rcu *tbl; @@ -27,7 +28,6 @@ struct nft_hash { struct nft_hash_table { unsigned int size; - unsigned int elements; struct nft_hash_elem __rcu *buckets[]; }; @@ -76,10 +76,12 @@ static bool nft_hash_lookup(const struct nft_set *set, static void nft_hash_tbl_free(const struct nft_hash_table *tbl) { - if (is_vmalloc_addr(tbl)) - vfree(tbl); - else - kfree(tbl); + kvfree(tbl); +} + +static unsigned int nft_hash_tbl_size(unsigned int nelem) +{ + return max(roundup_pow_of_two(nelem * 4 / 3), NFT_HASH_MIN_SIZE); } static struct nft_hash_table *nft_hash_tbl_alloc(unsigned int nbuckets) @@ -161,7 +163,6 @@ static int nft_hash_tbl_expand(const struct nft_set *set, struct nft_hash *priv) break; } } - ntbl->elements = tbl->elements; /* Publish new table */ rcu_assign_pointer(priv->tbl, ntbl); @@ -201,7 +202,6 @@ static int nft_hash_tbl_shrink(const struct nft_set *set, struct nft_hash *priv) ; RCU_INIT_POINTER(*pprev, tbl->buckets[i + ntbl->size]); } - ntbl->elements = tbl->elements; /* Publish new table */ rcu_assign_pointer(priv->tbl, ntbl); @@ -237,10 +237,9 @@ static int nft_hash_insert(const struct nft_set *set, h = nft_hash_data(&he->key, tbl->size, set->klen); RCU_INIT_POINTER(he->next, tbl->buckets[h]); rcu_assign_pointer(tbl->buckets[h], he); - tbl->elements++; /* Expand table when exceeding 75% load */ - if (tbl->elements > tbl->size / 4 * 3) + if (set->nelems + 1 > tbl->size / 4 * 3) nft_hash_tbl_expand(set, priv); return 0; @@ -268,10 +267,9 @@ static void nft_hash_remove(const struct nft_set *set, RCU_INIT_POINTER(*pprev, he->next); synchronize_rcu(); kfree(he); - tbl->elements--; /* Shrink table beneath 30% load */ - if (tbl->elements < tbl->size * 3 / 10 && + if (set->nelems - 1 < tbl->size * 3 / 10 && tbl->size > NFT_HASH_MIN_SIZE) nft_hash_tbl_shrink(set, priv); } @@ -335,17 +333,23 @@ static unsigned int nft_hash_privsize(const struct nlattr * const nla[]) } static int nft_hash_init(const struct nft_set *set, + const struct nft_set_desc *desc, const struct nlattr * const tb[]) { struct nft_hash *priv = nft_set_priv(set); struct nft_hash_table *tbl; + unsigned int size; if (unlikely(!nft_hash_rnd_initted)) { get_random_bytes(&nft_hash_rnd, 4); nft_hash_rnd_initted = true; } - tbl = nft_hash_tbl_alloc(NFT_HASH_MIN_SIZE); + size = NFT_HASH_MIN_SIZE; + if (desc->size) + size = nft_hash_tbl_size(desc->size); + + tbl = nft_hash_tbl_alloc(size); if (tbl == NULL) return -ENOMEM; RCU_INIT_POINTER(priv->tbl, tbl); @@ -369,8 +373,37 @@ static void nft_hash_destroy(const struct nft_set *set) kfree(tbl); } +static bool nft_hash_estimate(const struct nft_set_desc *desc, u32 features, + struct nft_set_estimate *est) +{ + unsigned int esize; + + esize = sizeof(struct nft_hash_elem); + if (features & NFT_SET_MAP) + esize += FIELD_SIZEOF(struct nft_hash_elem, data[0]); + + if (desc->size) { + est->size = sizeof(struct nft_hash) + + nft_hash_tbl_size(desc->size) * + sizeof(struct nft_hash_elem *) + + desc->size * esize; + } else { + /* Resizing happens when the load drops below 30% or goes + * above 75%. The average of 52.5% load (approximated by 50%) + * is used for the size estimation of the hash buckets, + * meaning we calculate two buckets per element. + */ + est->size = esize + 2 * sizeof(struct nft_hash_elem *); + } + + est->class = NFT_SET_CLASS_O_1; + + return true; +} + static struct nft_set_ops nft_hash_ops __read_mostly = { .privsize = nft_hash_privsize, + .estimate = nft_hash_estimate, .init = nft_hash_init, .destroy = nft_hash_destroy, .get = nft_hash_get, diff --git a/net/netfilter/nft_lookup.c b/net/netfilter/nft_lookup.c index 7fd2bea8aa23..6404a726d17b 100644 --- a/net/netfilter/nft_lookup.c +++ b/net/netfilter/nft_lookup.c @@ -56,8 +56,14 @@ static int nft_lookup_init(const struct nft_ctx *ctx, return -EINVAL; set = nf_tables_set_lookup(ctx->table, tb[NFTA_LOOKUP_SET]); - if (IS_ERR(set)) - return PTR_ERR(set); + if (IS_ERR(set)) { + if (tb[NFTA_LOOKUP_SET_ID]) { + set = nf_tables_set_lookup_byid(ctx->net, + tb[NFTA_LOOKUP_SET_ID]); + } + if (IS_ERR(set)) + return PTR_ERR(set); + } priv->sreg = ntohl(nla_get_be32(tb[NFTA_LOOKUP_SREG])); err = nft_validate_input_register(priv->sreg); diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c index 425cf39af890..852b178c6ae7 100644 --- a/net/netfilter/nft_meta.c +++ b/net/netfilter/nft_meta.c @@ -18,18 +18,11 @@ #include <net/sock.h> #include <net/tcp_states.h> /* for TCP_TIME_WAIT */ #include <net/netfilter/nf_tables.h> +#include <net/netfilter/nft_meta.h> -struct nft_meta { - enum nft_meta_keys key:8; - union { - enum nft_registers dreg:8; - enum nft_registers sreg:8; - }; -}; - -static void nft_meta_get_eval(const struct nft_expr *expr, - struct nft_data data[NFT_REG_MAX + 1], - const struct nft_pktinfo *pkt) +void nft_meta_get_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt) { const struct nft_meta *priv = nft_expr_priv(expr); const struct sk_buff *skb = pkt->skb; @@ -140,10 +133,11 @@ static void nft_meta_get_eval(const struct nft_expr *expr, err: data[NFT_REG_VERDICT].verdict = NFT_BREAK; } +EXPORT_SYMBOL_GPL(nft_meta_get_eval); -static void nft_meta_set_eval(const struct nft_expr *expr, - struct nft_data data[NFT_REG_MAX + 1], - const struct nft_pktinfo *pkt) +void nft_meta_set_eval(const struct nft_expr *expr, + struct nft_data data[NFT_REG_MAX + 1], + const struct nft_pktinfo *pkt) { const struct nft_meta *meta = nft_expr_priv(expr); struct sk_buff *skb = pkt->skb; @@ -163,28 +157,24 @@ static void nft_meta_set_eval(const struct nft_expr *expr, WARN_ON(1); } } +EXPORT_SYMBOL_GPL(nft_meta_set_eval); -static const struct nla_policy nft_meta_policy[NFTA_META_MAX + 1] = { +const struct nla_policy nft_meta_policy[NFTA_META_MAX + 1] = { [NFTA_META_DREG] = { .type = NLA_U32 }, [NFTA_META_KEY] = { .type = NLA_U32 }, [NFTA_META_SREG] = { .type = NLA_U32 }, }; +EXPORT_SYMBOL_GPL(nft_meta_policy); -static int nft_meta_init_validate_set(uint32_t key) +int nft_meta_get_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) { - switch (key) { - case NFT_META_MARK: - case NFT_META_PRIORITY: - case NFT_META_NFTRACE: - return 0; - default: - return -EOPNOTSUPP; - } -} + struct nft_meta *priv = nft_expr_priv(expr); + int err; -static int nft_meta_init_validate_get(uint32_t key) -{ - switch (key) { + priv->key = ntohl(nla_get_be32(tb[NFTA_META_KEY])); + switch (priv->key) { case NFT_META_LEN: case NFT_META_PROTOCOL: case NFT_META_NFPROTO: @@ -205,39 +195,41 @@ static int nft_meta_init_validate_get(uint32_t key) #ifdef CONFIG_NETWORK_SECMARK case NFT_META_SECMARK: #endif - return 0; + break; default: return -EOPNOTSUPP; } + priv->dreg = ntohl(nla_get_be32(tb[NFTA_META_DREG])); + err = nft_validate_output_register(priv->dreg); + if (err < 0) + return err; + + err = nft_validate_data_load(ctx, priv->dreg, NULL, NFT_DATA_VALUE); + if (err < 0) + return err; + + return 0; } +EXPORT_SYMBOL_GPL(nft_meta_get_init); -static int nft_meta_init(const struct nft_ctx *ctx, const struct nft_expr *expr, - const struct nlattr * const tb[]) +int nft_meta_set_init(const struct nft_ctx *ctx, + const struct nft_expr *expr, + const struct nlattr * const tb[]) { struct nft_meta *priv = nft_expr_priv(expr); int err; priv->key = ntohl(nla_get_be32(tb[NFTA_META_KEY])); - - if (tb[NFTA_META_DREG]) { - err = nft_meta_init_validate_get(priv->key); - if (err < 0) - return err; - - priv->dreg = ntohl(nla_get_be32(tb[NFTA_META_DREG])); - err = nft_validate_output_register(priv->dreg); - if (err < 0) - return err; - - return nft_validate_data_load(ctx, priv->dreg, NULL, - NFT_DATA_VALUE); + switch (priv->key) { + case NFT_META_MARK: + case NFT_META_PRIORITY: + case NFT_META_NFTRACE: + break; + default: + return -EOPNOTSUPP; } - err = nft_meta_init_validate_set(priv->key); - if (err < 0) - return err; - priv->sreg = ntohl(nla_get_be32(tb[NFTA_META_SREG])); err = nft_validate_input_register(priv->sreg); if (err < 0) @@ -245,9 +237,10 @@ static int nft_meta_init(const struct nft_ctx *ctx, const struct nft_expr *expr, return 0; } +EXPORT_SYMBOL_GPL(nft_meta_set_init); -static int nft_meta_get_dump(struct sk_buff *skb, - const struct nft_expr *expr) +int nft_meta_get_dump(struct sk_buff *skb, + const struct nft_expr *expr) { const struct nft_meta *priv = nft_expr_priv(expr); @@ -260,9 +253,10 @@ static int nft_meta_get_dump(struct sk_buff *skb, nla_put_failure: return -1; } +EXPORT_SYMBOL_GPL(nft_meta_get_dump); -static int nft_meta_set_dump(struct sk_buff *skb, - const struct nft_expr *expr) +int nft_meta_set_dump(struct sk_buff *skb, + const struct nft_expr *expr) { const struct nft_meta *priv = nft_expr_priv(expr); @@ -276,13 +270,14 @@ static int nft_meta_set_dump(struct sk_buff *skb, nla_put_failure: return -1; } +EXPORT_SYMBOL_GPL(nft_meta_set_dump); static struct nft_expr_type nft_meta_type; static const struct nft_expr_ops nft_meta_get_ops = { .type = &nft_meta_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_meta)), .eval = nft_meta_get_eval, - .init = nft_meta_init, + .init = nft_meta_get_init, .dump = nft_meta_get_dump, }; @@ -290,7 +285,7 @@ static const struct nft_expr_ops nft_meta_set_ops = { .type = &nft_meta_type, .size = NFT_EXPR_SIZE(sizeof(struct nft_meta)), .eval = nft_meta_set_eval, - .init = nft_meta_init, + .init = nft_meta_set_init, .dump = nft_meta_set_dump, }; diff --git a/net/netfilter/nft_rbtree.c b/net/netfilter/nft_rbtree.c index e21d69d13506..e1836ff88199 100644 --- a/net/netfilter/nft_rbtree.c +++ b/net/netfilter/nft_rbtree.c @@ -18,6 +18,8 @@ #include <linux/netfilter/nf_tables.h> #include <net/netfilter/nf_tables.h> +static DEFINE_SPINLOCK(nft_rbtree_lock); + struct nft_rbtree { struct rb_root root; }; @@ -38,6 +40,7 @@ static bool nft_rbtree_lookup(const struct nft_set *set, const struct rb_node *parent = priv->root.rb_node; int d; + spin_lock_bh(&nft_rbtree_lock); while (parent != NULL) { rbe = rb_entry(parent, struct nft_rbtree_elem, node); @@ -53,6 +56,8 @@ found: goto out; if (set->flags & NFT_SET_MAP) nft_data_copy(data, rbe->data); + + spin_unlock_bh(&nft_rbtree_lock); return true; } } @@ -62,6 +67,7 @@ found: goto found; } out: + spin_unlock_bh(&nft_rbtree_lock); return false; } @@ -124,9 +130,12 @@ static int nft_rbtree_insert(const struct nft_set *set, !(rbe->flags & NFT_SET_ELEM_INTERVAL_END)) nft_data_copy(rbe->data, &elem->data); + spin_lock_bh(&nft_rbtree_lock); err = __nft_rbtree_insert(set, rbe); if (err < 0) kfree(rbe); + + spin_unlock_bh(&nft_rbtree_lock); return err; } @@ -136,7 +145,9 @@ static void nft_rbtree_remove(const struct nft_set *set, struct nft_rbtree *priv = nft_set_priv(set); struct nft_rbtree_elem *rbe = elem->cookie; + spin_lock_bh(&nft_rbtree_lock); rb_erase(&rbe->node, &priv->root); + spin_unlock_bh(&nft_rbtree_lock); kfree(rbe); } @@ -147,6 +158,7 @@ static int nft_rbtree_get(const struct nft_set *set, struct nft_set_elem *elem) struct nft_rbtree_elem *rbe; int d; + spin_lock_bh(&nft_rbtree_lock); while (parent != NULL) { rbe = rb_entry(parent, struct nft_rbtree_elem, node); @@ -161,9 +173,11 @@ static int nft_rbtree_get(const struct nft_set *set, struct nft_set_elem *elem) !(rbe->flags & NFT_SET_ELEM_INTERVAL_END)) nft_data_copy(&elem->data, rbe->data); elem->flags = rbe->flags; + spin_unlock_bh(&nft_rbtree_lock); return 0; } } + spin_unlock_bh(&nft_rbtree_lock); return -ENOENT; } @@ -176,6 +190,7 @@ static void nft_rbtree_walk(const struct nft_ctx *ctx, struct nft_set_elem elem; struct rb_node *node; + spin_lock_bh(&nft_rbtree_lock); for (node = rb_first(&priv->root); node != NULL; node = rb_next(node)) { if (iter->count < iter->skip) goto cont; @@ -188,11 +203,14 @@ static void nft_rbtree_walk(const struct nft_ctx *ctx, elem.flags = rbe->flags; iter->err = iter->fn(ctx, set, iter, &elem); - if (iter->err < 0) + if (iter->err < 0) { + spin_unlock_bh(&nft_rbtree_lock); return; + } cont: iter->count++; } + spin_unlock_bh(&nft_rbtree_lock); } static unsigned int nft_rbtree_privsize(const struct nlattr * const nla[]) @@ -201,6 +219,7 @@ static unsigned int nft_rbtree_privsize(const struct nlattr * const nla[]) } static int nft_rbtree_init(const struct nft_set *set, + const struct nft_set_desc *desc, const struct nlattr * const nla[]) { struct nft_rbtree *priv = nft_set_priv(set); @@ -215,15 +234,37 @@ static void nft_rbtree_destroy(const struct nft_set *set) struct nft_rbtree_elem *rbe; struct rb_node *node; + spin_lock_bh(&nft_rbtree_lock); while ((node = priv->root.rb_node) != NULL) { rb_erase(node, &priv->root); rbe = rb_entry(node, struct nft_rbtree_elem, node); nft_rbtree_elem_destroy(set, rbe); } + spin_unlock_bh(&nft_rbtree_lock); +} + +static bool nft_rbtree_estimate(const struct nft_set_desc *desc, u32 features, + struct nft_set_estimate *est) +{ + unsigned int nsize; + + nsize = sizeof(struct nft_rbtree_elem); + if (features & NFT_SET_MAP) + nsize += FIELD_SIZEOF(struct nft_rbtree_elem, data[0]); + + if (desc->size) + est->size = sizeof(struct nft_rbtree) + desc->size * nsize; + else + est->size = nsize; + + est->class = NFT_SET_CLASS_O_LOG_N; + + return true; } static struct nft_set_ops nft_rbtree_ops __read_mostly = { .privsize = nft_rbtree_privsize, + .estimate = nft_rbtree_estimate, .init = nft_rbtree_init, .destroy = nft_rbtree_destroy, .insert = nft_rbtree_insert, diff --git a/net/netfilter/xt_bpf.c b/net/netfilter/xt_bpf.c index 12d4da8e6c77..bbffdbdaf603 100644 --- a/net/netfilter/xt_bpf.c +++ b/net/netfilter/xt_bpf.c @@ -23,10 +23,11 @@ MODULE_ALIAS("ip6t_bpf"); static int bpf_mt_check(const struct xt_mtchk_param *par) { struct xt_bpf_info *info = par->matchinfo; - struct sock_fprog program; + struct sock_fprog_kern program; program.len = info->bpf_program_num_elem; - program.filter = (struct sock_filter __user *) info->bpf_program; + program.filter = info->bpf_program; + if (sk_unattached_filter_create(&info->filter, &program)) { pr_info("bpf: check failed: parse error\n"); return -EINVAL; diff --git a/net/netfilter/xt_nfacct.c b/net/netfilter/xt_nfacct.c index b3be0ef21f19..8c646ed9c921 100644 --- a/net/netfilter/xt_nfacct.c +++ b/net/netfilter/xt_nfacct.c @@ -21,11 +21,14 @@ MODULE_ALIAS("ip6t_nfacct"); static bool nfacct_mt(const struct sk_buff *skb, struct xt_action_param *par) { + int overquota; const struct xt_nfacct_match_info *info = par->targinfo; nfnl_acct_update(skb, info->nfacct); - return true; + overquota = nfnl_acct_overquota(skb, info->nfacct); + + return overquota == NFACCT_UNDERQUOTA ? false : true; } static int diff --git a/net/netfilter/xt_recent.c b/net/netfilter/xt_recent.c index 1e657cf715c4..a9faae89f955 100644 --- a/net/netfilter/xt_recent.c +++ b/net/netfilter/xt_recent.c @@ -313,10 +313,7 @@ out: static void recent_table_free(void *addr) { - if (is_vmalloc_addr(addr)) - vfree(addr); - else - kfree(addr); + kvfree(addr); } static int recent_mt_check(const struct xt_mtchk_param *par, |