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
*/
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;
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;
}
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[])
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)
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;
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[])
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))
}
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]) {
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;
}
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))
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;
if (!list_empty(&table->chains) || !list_empty(&table->sets))
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)
+{
+ kfree(ctx->table);
+ module_put(ctx->afi->owner);
+}
+
int nft_register_chain_type(const struct nf_chain_type *ctype)
{
int err = 0;
.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 },
};
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;
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;
}
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)
[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;
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.
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);
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;
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;
}
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)
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)
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;
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);
+ 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,
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))
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;
if (!list_empty(&chain->rules) || 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);
+ 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
*/
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;
}
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;
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;
}
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))
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;
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;
}
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;
}
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);
{
/* 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);
return 0;
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;
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]);
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
*/
}
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
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);
}
[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,
{
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);
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);
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) {
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;
{
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),
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:
int event)
{
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;
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_KERNEL);
err:
if (err < 0)
nfnetlink_set_err(ctx->net, portid, NFNLGRP_NFTABLES, err);
{
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];
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[])
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)
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;
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]));
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;
}
dtype = 0;
- dlen = 0;
if (nla[NFTA_SET_DATA_TYPE] != NULL) {
if (!(flags & NFT_SET_MAP))
return -EINVAL;
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);
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);
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, nla);
+ err = ops->init(set, &desc, nla);
+ if (err < 0)
+ goto err2;
+
+ 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);
return 0;
err2:
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);
+ 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[])
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);
return 0;
}
{
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);
}
[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);
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;
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;
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 = {
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];
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)
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,
}
}
+ 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(&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);
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;
}
- 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,
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(&trans->list, &ctx->net->nft.commit_list);
nft_data_uninit(&elem.key, NFT_DATA_VALUE);
if (set->flags & NFT_SET_MAP)
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;
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;
}
- 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,
},
.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,
},
.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,
},
.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,
},
.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,
},
.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;
+ trans->ctx.table->use++;
+ }
+ nf_tables_chain_notify(&trans->ctx, NFT_MSG_NEWCHAIN);
+ nft_trans_destroy(trans);
+ break;
+ case NFT_MSG_DELCHAIN:
+ trans->ctx.table->use--;
+ 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;
+ nf_tables_set_notify(&trans->ctx, nft_trans_set(trans),
+ NFT_MSG_NEWSET);
+ nft_trans_destroy(trans);
+ break;
+ case NFT_MSG_DELSET:
+ nf_tables_set_notify(&trans->ctx, nft_trans_set(trans),
+ NFT_MSG_DELSET);
+ break;
+ case NFT_MSG_NEWSETELEM:
+ nft_trans_elem_set(trans)->nelems++;
+ 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:
+ nft_trans_elem_set(trans)->nelems--;
+ 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 {
+ 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:
+ list_add_tail(&trans->ctx.chain->list,
+ &trans->ctx.table->chains);
+ nft_trans_destroy(trans);
+ break;
+ case NFT_MSG_NEWRULE:
+ list_del_rcu(&nft_trans_rule(trans)->list);
+ break;
+ case NFT_MSG_DELRULE:
+ nft_rule_clear(trans->ctx.net, nft_trans_rule(trans));
+ nft_trans_destroy(trans);
+ break;
+ case NFT_MSG_NEWSET:
+ list_del(&nft_trans_set(trans)->list);
+ break;
+ case NFT_MSG_DELSET:
+ list_add_tail(&nft_trans_set(trans)->list,
+ &trans->ctx.table->sets);
+ nft_trans_destroy(trans);
+ break;
+ case NFT_MSG_NEWSETELEM:
+ 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_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_abort_release_rcu);
+ }
+
+ return 0;
+}
+
static const struct nfnetlink_subsystem nf_tables_subsys = {
.name = "nf_tables",
.subsys_id = NFNL_SUBSYS_NFTABLES,
git.openpandora.org / pandora-kernel.git / blobdiff