pkt_sched: act_xt support new Xtables interface

Message ID	50D1AB7E.5060000@mojatatu.com
State	Not Applicable
Headers	show Return-Path: <netfilter-devel-owner@vger.kernel.org> Message-ID: <50D1AB7E.5060000@mojatatu.com> Date: Wed, 19 Dec 2012 06:56:46 -0500 From: Jamal Hadi Salim <jhs@mojatatu.com> User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/17.0 Thunderbird/17.0 MIME-Version: 1.0 To: Hasan Chowdhury <shemonc@gmail.com> CC: Stephen Hemminger <shemminger@vyatta.com>, Jan Engelhardt <jengelh@inai.de>, Yury Stankevich <urykhy@gmail.com>, "netdev@vger.kernel.org" <netdev@vger.kernel.org>, pablo@netfilter.org, netfilter-devel@vger.kernel.org Subject: [PATCH] pkt_sched: act_xt support new Xtables interface References: <50C4821D.5090206@gmail.com> <50C9B4BB.9060609@mojatatu.com> <50CCE961.5050204@mojatatu.com> <alpine.LNX.2.01.1212160002330.4901@nerf07.vanv.qr> <CAASe=fQT2pVOK0uctdaKL+aOrF8nYeTMfoF15kmd-rC02+7Vnw@mail.gmail.com> <50CDFB6A.3090806@mojatatu.com> <50CE1A04.1000405@mojatatu.com> <alpine.LNX.2.01.1212162003340.27614@nerf07.vanv.qr> <50CE3203.9080007@mojatatu.com> <50CF1071.1050405@mojatatu.com> <CAASe=fRuJdtisEvp7uo=PHwN3nKHqsYDW4Om1gk2MK-vyNvBrA@mail.gmail.com> <50D06177.2090905@mojatatu.com> <CAASe=fR6Hm2dxp=1wDchtrzqnaH6qacHpg2wrsqLfmGpPbQ9Fg@mail.gmail.com> <50D1A8A7.1090002@mojatatu.com> In-Reply-To: <50D1A8A7.1090002@mojatatu.com> Content-Type: multipart/mixed; boundary="------------050503020000070306090906" Sender: netfilter-devel-owner@vger.kernel.org Precedence: bulk

diff --git a/net/sched/Kconfig b/net/sched/Kconfig index 235e01a..1693973 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -578,12 +578,25 @@ config NET_ACT_MIRRED config NET_ACT_IPT tristate "IPtables targets" depends on NET_CLS_ACT && NETFILTER && IP_NF_IPTABLES + select NET_ACT_XT ---help--- Say Y here to be able to invoke iptables targets after successful - classification. + classification. Better yet choose NET_ACT_XT since this version + will eventually be obsoleted. To compile this code as a module, choose M here: the module will be called act_ipt. +config NET_ACT_XT + tristate "New IPtables targets" + depends on NET_CLS_ACT && NETFILTER && IP_NF_IPTABLES + ---help--- + Say Y here to be able to invoke iptables targets after successful + classification using the new xtables mechanism. This mechanism + will eventually replace NET_ACT_IPT + + To compile this code as a module, choose M here: the + module will be called act_xt. + config NET_ACT_NAT tristate "Stateless NAT" diff --git a/net/sched/Makefile b/net/sched/Makefile index 978cbf0..10a1136 100644 --- a/net/sched/Makefile +++ b/net/sched/Makefile @@ -11,6 +11,7 @@ obj-$(CONFIG_NET_ACT_POLICE) += act_police.o obj-$(CONFIG_NET_ACT_GACT) += act_gact.o obj-$(CONFIG_NET_ACT_MIRRED) += act_mirred.o obj-$(CONFIG_NET_ACT_IPT) += act_ipt.o +obj-$(CONFIG_NET_ACT_XT) += act_xt.o obj-$(CONFIG_NET_ACT_NAT) += act_nat.o obj-$(CONFIG_NET_ACT_PEDIT) += act_pedit.o obj-$(CONFIG_NET_ACT_SIMP) += act_simple.o diff --git a/net/sched/act_xt.c b/net/sched/act_xt.c new file mode 100644 index 0000000..589cfe6 --- /dev/null +++ b/net/sched/act_xt.c @@ -0,0 +1,324 @@ +/* + * net/sched/act_xt.c iptables target interface + * + *TODO: Add other tables. For now we only support the ipv4 table targets + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Copyright: Jamal Hadi Salim (2002-12) + */ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/errno.h> +#include <linux/skbuff.h> +#include <linux/rtnetlink.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/slab.h> +#include <net/netlink.h> +#include <net/pkt_sched.h> +#include <linux/tc_act/tc_ipt.h> +#include <net/tc_act/tc_ipt.h> + +#include <linux/netfilter_ipv4/ip_tables.h> + +#define IPT_TAB_MASK 15 +static struct tcf_common *tcf_ipt_ht[IPT_TAB_MASK + 1]; +static u32 ipt_idx_gen; +static DEFINE_RWLOCK(ipt_lock); + +static struct tcf_hashinfo ipt_hash_info = { + .htab = tcf_ipt_ht, + .hmask = IPT_TAB_MASK, + .lock = &ipt_lock, +}; + +static int ipt_init_target(struct xt_entry_target *t, char *table, + unsigned int hook) +{ + struct xt_tgchk_param par; + struct xt_target *target; + int ret = 0; + + target = xt_request_find_target(AF_INET, t->u.user.name, + t->u.user.revision); + if (IS_ERR(target)) + return PTR_ERR(target); + + t->u.kernel.target = target; + par.table = table; + par.entryinfo = NULL; + par.target = target; + par.targinfo = t->data; + par.hook_mask = hook; + par.family = NFPROTO_IPV4; + + ret = xt_check_target(&par, t->u.target_size - sizeof(*t), 0, false); + if (ret < 0) { + module_put(t->u.kernel.target->me); + return ret; + } + return 0; +} + +static void ipt_destroy_target(struct xt_entry_target *t) +{ + struct xt_tgdtor_param par = { + .target = t->u.kernel.target, + .targinfo = t->data, + }; + if (par.target->destroy != NULL) + par.target->destroy(&par); + module_put(par.target->me); +} + +static int tcf_ipt_release(struct tcf_ipt *ipt, int bind) +{ + int ret = 0; + if (ipt) { + if (bind) + ipt->tcf_bindcnt--; + ipt->tcf_refcnt--; + if (ipt->tcf_bindcnt <= 0 && ipt->tcf_refcnt <= 0) { + ipt_destroy_target(ipt->tcfi_t); + kfree(ipt->tcfi_tname); + kfree(ipt->tcfi_t); + tcf_hash_destroy(&ipt->common, &ipt_hash_info); + ret = ACT_P_DELETED; + } + } + return ret; +} + +static const struct nla_policy ipt_policy[TCA_IPT_MAX + 1] = { + [TCA_IPT_TABLE] = {.type = NLA_STRING,.len = IFNAMSIZ}, + [TCA_IPT_HOOK] = {.type = NLA_U32}, + [TCA_IPT_INDEX] = {.type = NLA_U32}, + [TCA_IPT_TARG] = {.len = sizeof(struct xt_entry_target)}, +}; + +static int tcf_ipt_init(struct nlattr *nla, struct nlattr *est, + struct tc_action *a, int ovr, int bind) +{ + struct nlattr *tb[TCA_IPT_MAX + 1]; + struct tcf_ipt *ipt; + struct tcf_common *pc; + struct xt_entry_target *td, *t; + char *tname; + int ret = 0, err; + u32 hook = 0; + u32 index = 0; + + if (nla == NULL) + return -EINVAL; + + err = nla_parse_nested(tb, TCA_IPT_MAX, nla, ipt_policy); + if (err < 0) + return err; + + if (tb[TCA_IPT_HOOK] == NULL) + return -EINVAL; + if (tb[TCA_IPT_TARG] == NULL) + return -EINVAL; + + td = (struct xt_entry_target *)nla_data(tb[TCA_IPT_TARG]); + if (nla_len(tb[TCA_IPT_TARG]) < td->u.target_size) + return -EINVAL; + + if (tb[TCA_IPT_INDEX] != NULL) + index = nla_get_u32(tb[TCA_IPT_INDEX]); + + pc = tcf_hash_check(index, a, bind, &ipt_hash_info); + if (!pc) { + pc = tcf_hash_create(index, est, a, sizeof(*ipt), bind, + &ipt_idx_gen, &ipt_hash_info); + if (IS_ERR(pc)) + return PTR_ERR(pc); + ret = ACT_P_CREATED; + } else { + if (!ovr) { + tcf_ipt_release(to_ipt(pc), bind); + return -EEXIST; + } + } + ipt = to_ipt(pc); + + hook = nla_get_u32(tb[TCA_IPT_HOOK]); + + err = -ENOMEM; + tname = kmalloc(IFNAMSIZ, GFP_KERNEL); + if (unlikely(!tname)) + goto err1; + if (tb[TCA_IPT_TABLE] == NULL || + nla_strlcpy(tname, tb[TCA_IPT_TABLE], IFNAMSIZ) >= IFNAMSIZ) + strcpy(tname, "mangle"); + + t = kmemdup(td, td->u.target_size, GFP_KERNEL); + if (unlikely(!t)) + goto err2; + + err = ipt_init_target(t, tname, hook); + if (err < 0) + goto err3; + + spin_lock_bh(&ipt->tcf_lock); + if (ret != ACT_P_CREATED) { + ipt_destroy_target(ipt->tcfi_t); + kfree(ipt->tcfi_tname); + kfree(ipt->tcfi_t); + } + ipt->tcfi_tname = tname; + ipt->tcfi_t = t; + ipt->tcfi_hook = hook; + spin_unlock_bh(&ipt->tcf_lock); + if (ret == ACT_P_CREATED) + tcf_hash_insert(pc, &ipt_hash_info); + return ret; + +err3: + kfree(t); +err2: + kfree(tname); +err1: + if (ret == ACT_P_CREATED) { + if (est) + gen_kill_estimator(&pc->tcfc_bstats, + &pc->tcfc_rate_est); + kfree_rcu(pc, tcfc_rcu); + } + return err; +} + +static int tcf_ipt_cleanup(struct tc_action *a, int bind) +{ + struct tcf_ipt *ipt = a->priv; + return tcf_ipt_release(ipt, bind); +} + +static int tcf_ipt(struct sk_buff *skb, const struct tc_action *a, + struct tcf_result *res) +{ + int ret = 0, result = 0; + struct tcf_ipt *ipt = a->priv; + struct xt_action_param par; + + if (skb_cloned(skb)) { + if (pskb_expand_head(skb, 0, 0, GFP_ATOMIC)) + return TC_ACT_UNSPEC; + } + + spin_lock(&ipt->tcf_lock); + + ipt->tcf_tm.lastuse = jiffies; + bstats_update(&ipt->tcf_bstats, skb); + + /* yes, we have to worry about both in and out dev + * worry later - danger - this API seems to have changed + * from earlier kernels + */ + par.in = skb->dev; + par.out = NULL; + par.hooknum = ipt->tcfi_hook; + par.target = ipt->tcfi_t->u.kernel.target; + par.targinfo = ipt->tcfi_t->data; + ret = par.target->target(skb, &par); + + switch (ret) { + case NF_ACCEPT: + result = TC_ACT_OK; + break; + case NF_DROP: + result = TC_ACT_SHOT; + ipt->tcf_qstats.drops++; + break; + case XT_CONTINUE: + result = TC_ACT_PIPE; + break; + default: + net_notice_ratelimited + ("tc filter: Bogus netfilter code %d assume ACCEPT\n", ret); + result = TC_POLICE_OK; + break; + } + spin_unlock(&ipt->tcf_lock); + return result; + +} + +static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, + int ref) +{ + unsigned char *b = skb_tail_pointer(skb); + struct tcf_ipt *ipt = a->priv; + struct xt_entry_target *t; + struct tcf_t tm; + struct tc_cnt c; + + /* for simple targets kernel size == user size + * user name = target name + * for foolproof you need to not assume this + */ + + t = kmemdup(ipt->tcfi_t, ipt->tcfi_t->u.user.target_size, GFP_ATOMIC); + if (unlikely(!t)) + goto nla_put_failure; + + c.bindcnt = ipt->tcf_bindcnt - bind; + c.refcnt = ipt->tcf_refcnt - ref; + strcpy(t->u.user.name, ipt->tcfi_t->u.kernel.target->name); + + if (nla_put(skb, TCA_IPT_TARG, ipt->tcfi_t->u.user.target_size, t) || + nla_put_u32(skb, TCA_IPT_INDEX, ipt->tcf_index) || + nla_put_u32(skb, TCA_IPT_HOOK, ipt->tcfi_hook) || + nla_put(skb, TCA_IPT_CNT, sizeof(struct tc_cnt), &c) || + nla_put_string(skb, TCA_IPT_TABLE, ipt->tcfi_tname)) + goto nla_put_failure; + tm.install = jiffies_to_clock_t(jiffies - ipt->tcf_tm.install); + tm.lastuse = jiffies_to_clock_t(jiffies - ipt->tcf_tm.lastuse); + tm.expires = jiffies_to_clock_t(ipt->tcf_tm.expires); + if (nla_put(skb, TCA_IPT_TM, sizeof(tm), &tm)) + goto nla_put_failure; + kfree(t); + return skb->len; + +nla_put_failure: + nlmsg_trim(skb, b); + kfree(t); + return -1; +} + +static struct tc_action_ops act_ipt_ops = { + .kind = "xt", + .hinfo = &ipt_hash_info, + .type = TCA_ACT_IPT, + .capab = TCA_CAP_NONE, + .owner = THIS_MODULE, + .act = tcf_ipt, + .dump = tcf_ipt_dump, + .cleanup = tcf_ipt_cleanup, + .lookup = tcf_hash_search, + .init = tcf_ipt_init, + .walk = tcf_generic_walker +}; + +MODULE_AUTHOR("Jamal Hadi Salim(2002-12)"); +MODULE_DESCRIPTION("New Iptables target actions"); +MODULE_LICENSE("GPL"); + +static int __init ipt_init_module(void) +{ + return tcf_register_action(&act_ipt_ops); +} + +static void __exit ipt_cleanup_module(void) +{ + tcf_unregister_action(&act_ipt_ops); +} + +module_init(ipt_init_module); +module_exit(ipt_cleanup_module);

pkt_sched: act_xt support new Xtables interface

Commit Message

Comments

Patch