From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pablo Neira Ayuso Subject: [RFC][PATCH] ctnetlink port to nf_conntrack take #3 Date: Mon, 05 Dec 2005 11:42:47 +0100 Message-ID: <439419A7.7040100@eurodev.net> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="------------050000040707000002080806" Cc: Patrick McHardy , Yasuyuki Kozakai Return-path: To: Netfilter Development Mailinglist List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: netfilter-devel-bounces@lists.netfilter.org Errors-To: netfilter-devel-bounces@lists.netfilter.org List-Id: netfilter-devel.vger.kernel.org This is a multi-part message in MIME format. --------------050000040707000002080806 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi! Attached the 3rd incarnation of the ctnetlink port to nf_conntrack, it includes all recent Patrick's bugfixes and Yasuyuki's comments. nf_conntrack_netlink.c is available at: http://people.netfilter.org/pablo/nf_conntrack_netlink.c -- Pablo --------------050000040707000002080806 Content-Type: text/plain; name="x" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="x" diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index cc48256..64b82b7 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -94,6 +94,9 @@ struct nf_conn /* Current number of expected connections */ unsigned int expecting; + /* Unique ID that identifies this conntrack*/ + unsigned int id; + /* Helper. if any */ struct nf_conntrack_helper *helper; @@ -140,6 +143,9 @@ struct nf_conntrack_expect /* Usage count. */ atomic_t use; + /* Unique ID */ + unsigned int id; + /* Flags */ unsigned int flags; @@ -190,6 +196,31 @@ static inline void nf_ct_put(struct nf_c nf_conntrack_put(&ct->ct_general); } +extern struct nf_conntrack_tuple_hash * +__nf_conntrack_find(const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack); + +extern void nf_conntrack_hash_insert(struct nf_conn *ct); + +extern struct nf_conntrack_expect * +__nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple); + +extern struct nf_conntrack_expect * +nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple); + +extern void nf_ct_unlink_expect(struct nf_conntrack_expect *exp); + +extern void nf_ct_remove_expectations(struct nf_conn *ct); + +extern void nf_conntrack_flush(void); + +extern struct nf_conntrack_helper * +nf_ct_helper_find_get( const struct nf_conntrack_tuple *tuple); +extern void nf_ct_helper_put(struct nf_conntrack_helper *helper); + +extern struct nf_conntrack_helper * +__nf_conntrack_helper_find_byname(const char *name); + /* call to create an explicit dependency on nf_conntrack. */ extern void need_nf_conntrack(void); diff --git a/include/net/netfilter/nf_conntrack_helper.h b/include/net/netfilter/nf_conntrack_helper.h index 5a66b2a..86ec817 100644 --- a/include/net/netfilter/nf_conntrack_helper.h +++ b/include/net/netfilter/nf_conntrack_helper.h @@ -33,6 +33,8 @@ struct nf_conntrack_helper unsigned int protoff, struct nf_conn *ct, enum ip_conntrack_info conntrackinfo); + + int (*to_nfattr)(struct sk_buff *skb, const struct nf_conn *ct); }; extern int nf_conntrack_helper_register(struct nf_conntrack_helper *); diff --git a/include/net/netfilter/nf_conntrack_l3proto.h b/include/net/netfilter/nf_conntrack_l3proto.h index 01663e5..67856eb 100644 --- a/include/net/netfilter/nf_conntrack_l3proto.h +++ b/include/net/netfilter/nf_conntrack_l3proto.h @@ -14,6 +14,8 @@ #include #include +struct nfattr; + struct nf_conntrack_l3proto { /* Next pointer. */ @@ -70,6 +72,12 @@ struct nf_conntrack_l3proto u_int32_t (*get_features)(const struct nf_conntrack_tuple *tuple); + int (*tuple_to_nfattr)(struct sk_buff *skb, + const struct nf_conntrack_tuple *t); + + int (*nfattr_to_tuple)(struct nfattr *tb[], + struct nf_conntrack_tuple *t); + /* Module (if any) which this is connected to. */ struct module *me; }; @@ -81,11 +89,16 @@ extern int nf_conntrack_l3proto_register extern void nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto); static inline struct nf_conntrack_l3proto * -nf_ct_find_l3proto(u_int16_t l3proto) +__nf_ct_l3proto_find(u_int16_t l3proto) { return nf_ct_l3protos[l3proto]; } +extern struct nf_conntrack_l3proto * +nf_ct_l3proto_find_get(u_int16_t l3proto); + +extern void nf_ct_l3proto_put(struct nf_conntrack_l3proto *p); + /* Existing built-in protocols */ extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4; extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6; diff --git a/include/net/netfilter/nf_conntrack_protocol.h b/include/net/netfilter/nf_conntrack_protocol.h index b3afda3..1f33737 100644 --- a/include/net/netfilter/nf_conntrack_protocol.h +++ b/include/net/netfilter/nf_conntrack_protocol.h @@ -12,6 +12,7 @@ #include struct seq_file; +struct nfattr; struct nf_conntrack_protocol { @@ -66,6 +67,18 @@ struct nf_conntrack_protocol enum ip_conntrack_info *ctinfo, int pf, unsigned int hooknum); + /* convert protoinfo to nfnetink attributes */ + int (*to_nfattr)(struct sk_buff *skb, struct nfattr *nfa, + const struct nf_conn *ct); + + /* convert nfnetlink attributes to protoinfo */ + int (*from_nfattr)(struct nfattr *tb[], struct nf_conn *ct); + + int (*tuple_to_nfattr)(struct sk_buff *skb, + const struct nf_conntrack_tuple *t); + int (*nfattr_to_tuple)(struct nfattr *tb[], + struct nf_conntrack_tuple *t); + /* Module (if any) which this is connected to. */ struct module *me; }; @@ -80,12 +93,23 @@ extern struct nf_conntrack_protocol nf_c extern struct nf_conntrack_protocol **nf_ct_protos[PF_MAX]; extern struct nf_conntrack_protocol * -nf_ct_find_proto(u_int16_t l3proto, u_int8_t protocol); +__nf_ct_proto_find(u_int16_t l3proto, u_int8_t protocol); + +extern struct nf_conntrack_protocol * +nf_ct_proto_find_get(u_int16_t l3proto, u_int8_t protocol); + +extern void nf_ct_proto_put(struct nf_conntrack_protocol *p); /* Protocol registration. */ extern int nf_conntrack_protocol_register(struct nf_conntrack_protocol *proto); extern void nf_conntrack_protocol_unregister(struct nf_conntrack_protocol *proto); +/* Generic netlink helpers */ +extern int nf_ct_port_tuple_to_nfattr(struct sk_buff *skb, + const struct nf_conntrack_tuple *tuple); +extern int nf_ct_port_nfattr_to_tuple(struct nfattr *tb[], + struct nf_conntrack_tuple *t); + /* Log invalid packets */ extern unsigned int nf_ct_log_invalid; diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h diff --git a/net/ipv4/netfilter/ip_conntrack_netlink.c b/net/ipv4/netfilter/ip_conntrack_netlink.c diff --git a/net/ipv4/netfilter/ip_conntrack_proto_icmp.c b/net/ipv4/netfilter/ip_conntrack_proto_icmp.c diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index 8202c1c..385867e 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -392,6 +392,48 @@ getorigdst(struct sock *sk, int optval, return -ENOENT; } +#if defined(CONFIG_NF_CT_NETLINK) || \ + defined(CONFIG_NF_CT_NETLINK_MODULE) + +#include +#include + +static int ipv4_tuple_to_nfattr(struct sk_buff *skb, + const struct nf_conntrack_tuple *tuple) +{ + NFA_PUT(skb, CTA_IP_V4_SRC, sizeof(u_int32_t), + &tuple->src.u3.ip); + NFA_PUT(skb, CTA_IP_V4_DST, sizeof(u_int32_t), + &tuple->dst.u3.ip); + return 0; + +nfattr_failure: + return -1; +} + +static const size_t cta_min_ip[CTA_IP_MAX] = { + [CTA_IP_V4_SRC-1] = sizeof(u_int32_t), + [CTA_IP_V4_DST-1] = sizeof(u_int32_t), +}; + +static int ipv4_nfattr_to_tuple(struct nfattr *tb[], + struct nf_conntrack_tuple *t) +{ + if (!tb[CTA_IP_V4_SRC-1] || !tb[CTA_IP_V4_DST-1]) + return -EINVAL; + + if (nfattr_bad_size(tb, CTA_IP_MAX, cta_min_ip)) + return -EINVAL; + + t->src.u3.ip = + *(u_int32_t *)NFA_DATA(tb[CTA_IP_V4_SRC-1]); + t->dst.u3.ip = + *(u_int32_t *)NFA_DATA(tb[CTA_IP_V4_DST-1]); + + return 0; +} +#endif + static struct nf_sockopt_ops so_getorigdst = { .pf = PF_INET, .get_optmin = SO_ORIGINAL_DST, @@ -408,6 +450,11 @@ struct nf_conntrack_l3proto nf_conntrack .print_conntrack = ipv4_print_conntrack, .prepare = ipv4_prepare, .get_features = ipv4_get_features, +#if defined(CONFIG_NF_CT_NETLINK) || \ + defined(CONFIG_NF_CT_NETLINK_MODULE) + .tuple_to_nfattr = ipv4_tuple_to_nfattr, + .nfattr_to_tuple = ipv4_nfattr_to_tuple, +#endif .me = THIS_MODULE, }; diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c index 7ddb5c0..973e653 100644 --- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -116,16 +116,16 @@ static int icmp_packet(struct nf_conn *c return NF_ACCEPT; } +static u_int8_t valid_new[] + = { [ICMP_ECHO] = 1, + [ICMP_TIMESTAMP] = 1, + [ICMP_INFO_REQUEST] = 1, + [ICMP_ADDRESS] = 1 }; + /* Called when a new connection for this protocol found. */ static int icmp_new(struct nf_conn *conntrack, const struct sk_buff *skb, unsigned int dataoff) { - static u_int8_t valid_new[] - = { [ICMP_ECHO] = 1, - [ICMP_TIMESTAMP] = 1, - [ICMP_INFO_REQUEST] = 1, - [ICMP_ADDRESS] = 1 }; - if (conntrack->tuplehash[0].tuple.dst.u.icmp.type >= sizeof(valid_new) || !valid_new[conntrack->tuplehash[0].tuple.dst.u.icmp.type]) { /* Can't create a new ICMP `conn' with this. */ @@ -168,7 +168,7 @@ icmp_error_message(struct sk_buff *skb, return -NF_ACCEPT; } - innerproto = nf_ct_find_proto(PF_INET, inside->ip.protocol); + innerproto = __nf_ct_proto_find(PF_INET, inside->ip.protocol); dataoff = skb->nh.iph->ihl*4 + sizeof(inside->icmp); /* Are they talking about one of our connections? */ if (!nf_ct_get_tuple(skb, dataoff, dataoff + inside->ip.ihl*4, PF_INET, @@ -281,6 +281,60 @@ checksum_skipped: return icmp_error_message(skb, ctinfo, hooknum); } +#if defined(CONFIG_NF_CT_NETLINK) || \ + defined(CONFIG_NF_CT_NETLINK_MODULE) + +#include +#include + +static int icmp_tuple_to_nfattr(struct sk_buff *skb, + const struct nf_conntrack_tuple *t) +{ + NFA_PUT(skb, CTA_PROTO_ICMP_ID, sizeof(u_int16_t), + &t->src.u.icmp.id); + NFA_PUT(skb, CTA_PROTO_ICMP_TYPE, sizeof(u_int8_t), + &t->dst.u.icmp.type); + NFA_PUT(skb, CTA_PROTO_ICMP_CODE, sizeof(u_int8_t), + &t->dst.u.icmp.code); + + return 0; + +nfattr_failure: + return -1; +} + +static const size_t cta_min_proto[CTA_PROTO_MAX] = { + [CTA_PROTO_ICMP_TYPE-1] = sizeof(u_int8_t), + [CTA_PROTO_ICMP_CODE-1] = sizeof(u_int8_t), + [CTA_PROTO_ICMP_ID-1] = sizeof(u_int16_t) +}; + +static int icmp_nfattr_to_tuple(struct nfattr *tb[], + struct nf_conntrack_tuple *tuple) +{ + if (!tb[CTA_PROTO_ICMP_TYPE-1] + || !tb[CTA_PROTO_ICMP_CODE-1] + || !tb[CTA_PROTO_ICMP_ID-1]) + return -EINVAL; + + if (nfattr_bad_size(tb, CTA_PROTO_MAX, cta_min_proto)) + return -EINVAL; + + tuple->dst.u.icmp.type = + *(u_int8_t *)NFA_DATA(tb[CTA_PROTO_ICMP_TYPE-1]); + tuple->dst.u.icmp.code = + *(u_int8_t *)NFA_DATA(tb[CTA_PROTO_ICMP_CODE-1]); + tuple->src.u.icmp.id = + *(u_int16_t *)NFA_DATA(tb[CTA_PROTO_ICMP_ID-1]); + + if (tuple->dst.u.icmp.type >= sizeof(valid_new) + || !valid_new[tuple->dst.u.icmp.type]) + return -EINVAL; + + return 0; +} +#endif + struct nf_conntrack_protocol nf_conntrack_protocol_icmp = { .list = { NULL, NULL }, @@ -295,7 +349,12 @@ struct nf_conntrack_protocol nf_conntrac .new = icmp_new, .error = icmp_error, .destroy = NULL, - .me = NULL + .me = NULL, +#if defined(CONFIG_NF_CT_NETLINK) || \ + defined(CONFIG_NF_CT_NETLINK_MODULE) + .tuple_to_nfattr = icmp_tuple_to_nfattr, + .nfattr_to_tuple = icmp_nfattr_to_tuple, +#endif }; EXPORT_SYMBOL(nf_conntrack_protocol_icmp); diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c index 753a3ae..5b08699 100644 --- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c @@ -401,6 +401,48 @@ static ctl_table nf_ct_net_table[] = { }; #endif +#if defined(CONFIG_NF_CT_NETLINK) || \ + defined(CONFIG_NF_CT_NETLINK_MODULE) + +#include +#include + +static int ipv6_tuple_to_nfattr(struct sk_buff *skb, + const struct nf_conntrack_tuple *tuple) +{ + NFA_PUT(skb, CTA_IP_V6_SRC, sizeof(u_int32_t) * 4, + &tuple->src.u3.ip6); + NFA_PUT(skb, CTA_IP_V6_DST, sizeof(u_int32_t) * 4, + &tuple->dst.u3.ip6); + return 0; + +nfattr_failure: + return -1; +} + +static const size_t cta_min_ip[CTA_IP_MAX] = { + [CTA_IP_V6_SRC-1] = sizeof(u_int32_t)*4, + [CTA_IP_V6_DST-1] = sizeof(u_int32_t)*4, +}; + +static int ipv6_nfattr_to_tuple(struct nfattr *tb[], + struct nf_conntrack_tuple *t) +{ + if (!tb[CTA_IP_V6_SRC-1] || !tb[CTA_IP_V6_DST-1]) + return -EINVAL; + + if (nfattr_bad_size(tb, CTA_IP_MAX, cta_min_ip)) + return -EINVAL; + + memcpy(&t->src.u3.ip6, NFA_DATA(tb[CTA_IP_V6_SRC-1]), + sizeof(u_int32_t) * 4); + memcpy(&t->dst.u3.ip6, NFA_DATA(tb[CTA_IP_V6_DST-1]), + sizeof(u_int32_t) * 4); + + return 0; +} +#endif + struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6 = { .l3proto = PF_INET6, .name = "ipv6", @@ -410,6 +452,11 @@ struct nf_conntrack_l3proto nf_conntrack .print_conntrack = ipv6_print_conntrack, .prepare = ipv6_prepare, .get_features = ipv6_get_features, +#if defined(CONFIG_NF_CT_NETLINK) || \ + defined(CONFIG_NF_CT_NETLINK_MODULE) + .tuple_to_nfattr = ipv6_tuple_to_nfattr, + .nfattr_to_tuple = ipv6_nfattr_to_tuple, +#endif .me = THIS_MODULE, }; diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c index c0f1da5..00574f7 100644 --- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -185,7 +185,7 @@ icmpv6_error_message(struct sk_buff *skb return -NF_ACCEPT; } - inproto = nf_ct_find_proto(PF_INET6, inprotonum); + inproto = __nf_ct_proto_find(PF_INET6, inprotonum); /* Are they talking about one of our connections? */ if (!nf_ct_get_tuple(skb, inip6off, inprotoff, PF_INET6, inprotonum, diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index a84f922..d6a0b96 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -95,4 +95,11 @@ config NF_CONNTRACK_FTP To compile it as a module, choose M here. If unsure, say N. +config NF_CT_NETLINK + tristate 'Connection tracking netlink interface (EXPERIMENTAL)' + depends on EXPERIMENTAL && NF_CONNTRACK && NETFILTER_NETLINK + depends on NF_CONNTRACK!=y || NETFILTER_NETLINK!=m + help + This option enables support for a netlink-based userspace interface + endmenu diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 55f019a..cb21831 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -13,3 +13,6 @@ obj-$(CONFIG_NF_CONNTRACK_FTP) += nf_con # SCTP protocol connection tracking obj-$(CONFIG_NF_CT_PROTO_SCTP) += nf_conntrack_proto_sctp.o + +# netlink interface for nf_conntrack +obj-$(CONFIG_NF_CT_NETLINK) += nf_conntrack_netlink.o diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 1da6783..f7cbeee 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -82,6 +82,8 @@ unsigned int nf_ct_log_invalid; static LIST_HEAD(unconfirmed); static int nf_conntrack_vmalloc; +static unsigned int nf_conntrack_next_id = 1; +static unsigned int nf_conntrack_expect_next_id = 1; #ifdef CONFIG_NF_CONNTRACK_EVENTS struct notifier_block *nf_conntrack_chain; struct notifier_block *nf_conntrack_expect_chain; @@ -184,7 +186,7 @@ DECLARE_MUTEX(nf_ct_cache_mutex); extern struct nf_conntrack_protocol nf_conntrack_generic_protocol; struct nf_conntrack_protocol * -nf_ct_find_proto(u_int16_t l3proto, u_int8_t protocol) +__nf_ct_proto_find(u_int16_t l3proto, u_int8_t protocol) { if (unlikely(nf_ct_protos[l3proto] == NULL)) return &nf_conntrack_generic_protocol; @@ -192,6 +194,50 @@ nf_ct_find_proto(u_int16_t l3proto, u_in return nf_ct_protos[l3proto][protocol]; } +/* this is guaranteed to always return a valid protocol helper, since + * it falls back to generic_protocol */ +struct nf_conntrack_protocol * +nf_ct_proto_find_get(u_int16_t l3proto, u_int8_t protocol) +{ + struct nf_conntrack_protocol *p; + + preempt_disable(); + p = __nf_ct_proto_find(l3proto, protocol); + if (p) { + if (!try_module_get(p->me)) + p = &nf_conntrack_generic_protocol; + } + preempt_enable(); + + return p; +} + +void nf_ct_proto_put(struct nf_conntrack_protocol *p) +{ + module_put(p->me); +} + +struct nf_conntrack_l3proto * +nf_ct_l3proto_find_get(u_int16_t l3proto) +{ + struct nf_conntrack_l3proto *p; + + preempt_disable(); + p = __nf_ct_l3proto_find(l3proto); + if (p) { + if (!try_module_get(p->me)) + p = &nf_conntrack_generic_l3proto; + } + preempt_enable(); + + return p; +} + +void nf_ct_l3proto_put(struct nf_conntrack_l3proto *p) +{ + module_put(p->me); +} + static int nf_conntrack_hash_rnd_initted; static unsigned int nf_conntrack_hash_rnd; @@ -384,7 +430,7 @@ nf_ct_invert_tuple(struct nf_conntrack_t } /* nf_conntrack_expect helper functions */ -static void nf_ct_unlink_expect(struct nf_conntrack_expect *exp) +void nf_ct_unlink_expect(struct nf_conntrack_expect *exp) { ASSERT_WRITE_LOCK(&nf_conntrack_lock); NF_CT_ASSERT(!timer_pending(&exp->timeout)); @@ -404,6 +450,33 @@ static void expectation_timed_out(unsign nf_conntrack_expect_put(exp); } +struct nf_conntrack_expect * +__nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple) +{ + struct nf_conntrack_expect *i; + + list_for_each_entry(i, &nf_conntrack_expect_list, list) { + if (nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask)) { + atomic_inc(&i->use); + return i; + } + } + return NULL; +} + +/* Just find a expectation corresponding to a tuple. */ +struct nf_conntrack_expect * +nf_conntrack_expect_find(const struct nf_conntrack_tuple *tuple) +{ + struct nf_conntrack_expect *i; + + read_lock_bh(&nf_conntrack_lock); + i = __nf_conntrack_expect_find(tuple); + read_unlock_bh(&nf_conntrack_lock); + + return i; +} + /* If an expectation for this connection is found, it gets delete from * global list then returned. */ static struct nf_conntrack_expect * @@ -432,7 +505,7 @@ find_expectation(const struct nf_conntra } /* delete all expectations for this conntrack */ -static void remove_expectations(struct nf_conn *ct) +void nf_ct_remove_expectations(struct nf_conn *ct) { struct nf_conntrack_expect *i, *tmp; @@ -462,7 +535,7 @@ clean_from_lists(struct nf_conn *ct) LIST_DELETE(&nf_conntrack_hash[hr], &ct->tuplehash[IP_CT_DIR_REPLY]); /* Destroy all pending expectations */ - remove_expectations(ct); + nf_ct_remove_expectations(ct); } static void @@ -482,12 +555,11 @@ destroy_conntrack(struct nf_conntrack *n /* To make sure we don't get any weird locking issues here: * destroy_conntrack() MUST NOT be called with a write lock * to nf_conntrack_lock!!! -HW */ - l3proto = nf_ct_find_l3proto(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num); + l3proto = __nf_ct_l3proto_find(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num); if (l3proto && l3proto->destroy) l3proto->destroy(ct); - proto = nf_ct_find_proto(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num, - ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.protonum); + proto = __nf_ct_proto_find(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num, ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.protonum); if (proto && proto->destroy) proto->destroy(ct); @@ -499,7 +571,7 @@ destroy_conntrack(struct nf_conntrack *n * except TFTP can create an expectation on the first packet, * before connection is in the list, so we need to clean here, * too. */ - remove_expectations(ct); + nf_ct_remove_expectations(ct); /* We overload first tuple to link into unconfirmed list. */ if (!nf_ct_is_confirmed(ct)) { @@ -540,7 +612,7 @@ conntrack_tuple_cmp(const struct nf_conn && nf_ct_tuple_equal(tuple, &i->tuple); } -static struct nf_conntrack_tuple_hash * +struct nf_conntrack_tuple_hash * __nf_conntrack_find(const struct nf_conntrack_tuple *tuple, const struct nf_conn *ignored_conntrack) { @@ -575,6 +647,29 @@ nf_conntrack_find_get(const struct nf_co return h; } +static void __nf_conntrack_hash_insert(struct nf_conn *ct, + unsigned int hash, + unsigned int repl_hash) +{ + ct->id = ++nf_conntrack_next_id; + list_prepend(&nf_conntrack_hash[hash], + &ct->tuplehash[IP_CT_DIR_ORIGINAL].list); + list_prepend(&nf_conntrack_hash[repl_hash], + &ct->tuplehash[IP_CT_DIR_REPLY].list); +} + +void nf_conntrack_hash_insert(struct nf_conn *ct) +{ + unsigned int hash, repl_hash; + + hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + repl_hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); + + write_lock_bh(&nf_conntrack_lock); + __nf_conntrack_hash_insert(ct, hash, repl_hash); + write_unlock_bh(&nf_conntrack_lock); +} + /* Confirm a connection given skb; places it in hash table */ int __nf_conntrack_confirm(struct sk_buff **pskb) @@ -621,10 +716,7 @@ __nf_conntrack_confirm(struct sk_buff ** /* Remove from unconfirmed list */ list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list); - list_prepend(&nf_conntrack_hash[hash], - &ct->tuplehash[IP_CT_DIR_ORIGINAL]); - list_prepend(&nf_conntrack_hash[repl_hash], - &ct->tuplehash[IP_CT_DIR_REPLY]); + __nf_conntrack_hash_insert(ct, hash, repl_hash); /* Timer relative to confirmation time, not original setting time, otherwise we'd get timer wrap in weird delay cases. */ @@ -708,13 +800,41 @@ static inline int helper_cmp(const struc } static struct nf_conntrack_helper * -nf_ct_find_helper(const struct nf_conntrack_tuple *tuple) +__nf_ct_helper_find(const struct nf_conntrack_tuple *tuple) { return LIST_FIND(&helpers, helper_cmp, struct nf_conntrack_helper *, tuple); } +struct nf_conntrack_helper * +nf_ct_helper_find_get( const struct nf_conntrack_tuple *tuple) +{ + struct nf_conntrack_helper *helper; + + /* need nf_conntrack_lock to assure that helper exists until + * try_module_get() is called */ + read_lock_bh(&nf_conntrack_lock); + + helper = __nf_ct_helper_find(tuple); + if (helper) { + /* need to increase module usage count to assure helper will + * not go away while the caller is e.g. busy putting a + * conntrack in the hash that uses the helper */ + if (!try_module_get(helper->me)) + helper = NULL; + } + + read_unlock_bh(&nf_conntrack_lock); + + return helper; +} + +void nf_ct_helper_put(struct nf_conntrack_helper *helper) +{ + module_put(helper->me); +} + static struct nf_conn * __nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, const struct nf_conntrack_tuple *repl, @@ -744,7 +864,7 @@ __nf_conntrack_alloc(const struct nf_con /* find features needed by this conntrack. */ features = l3proto->get_features(orig); read_lock_bh(&nf_conntrack_lock); - if (nf_ct_find_helper(repl) != NULL) + if (__nf_ct_helper_find(repl) != NULL) features |= NF_CT_F_HELP; read_unlock_bh(&nf_conntrack_lock); @@ -794,7 +914,7 @@ struct nf_conn *nf_conntrack_alloc(const { struct nf_conntrack_l3proto *l3proto; - l3proto = nf_ct_find_l3proto(orig->src.l3num); + l3proto = __nf_ct_l3proto_find(orig->src.l3num); return __nf_conntrack_alloc(orig, repl, l3proto); } @@ -853,7 +973,7 @@ init_conntrack(const struct nf_conntrack nf_conntrack_get(&conntrack->master->ct_general); NF_CT_STAT_INC(expect_new); } else { - conntrack->helper = nf_ct_find_helper(&repl_tuple); + conntrack->helper = __nf_ct_helper_find(&repl_tuple); NF_CT_STAT_INC(new); } @@ -947,13 +1067,13 @@ nf_conntrack_in(int pf, unsigned int hoo return NF_ACCEPT; } - l3proto = nf_ct_find_l3proto((u_int16_t)pf); + l3proto = __nf_ct_l3proto_find((u_int16_t)pf); if ((ret = l3proto->prepare(pskb, hooknum, &dataoff, &protonum)) <= 0) { DEBUGP("not prepared to track yet or error occured\n"); return -ret; } - proto = nf_ct_find_proto((u_int16_t)pf, protonum); + proto = __nf_ct_proto_find((u_int16_t)pf, protonum); /* It may be an special packet, error, unclean... * inverse of the return code tells to the netfilter @@ -1002,9 +1122,9 @@ int nf_ct_invert_tuplepr(struct nf_connt const struct nf_conntrack_tuple *orig) { return nf_ct_invert_tuple(inverse, orig, - nf_ct_find_l3proto(orig->src.l3num), - nf_ct_find_proto(orig->src.l3num, - orig->dst.protonum)); + __nf_ct_l3proto_find(orig->src.l3num), + __nf_ct_proto_find(orig->src.l3num, + orig->dst.protonum)); } /* Would two expected things clash? */ @@ -1096,6 +1216,7 @@ static void nf_conntrack_expect_insert(s exp->timeout.expires = jiffies + exp->master->helper->timeout * HZ; add_timer(&exp->timeout); + exp->id = ++nf_conntrack_expect_next_id; atomic_inc(&exp->use); NF_CT_STAT_INC(expect_create); } @@ -1175,7 +1296,7 @@ void nf_conntrack_alter_reply(struct nf_ conntrack->tuplehash[IP_CT_DIR_REPLY].tuple = *newreply; if (!conntrack->master && conntrack->expecting == 0) - conntrack->helper = nf_ct_find_helper(newreply); + conntrack->helper = __nf_ct_helper_find(newreply); write_unlock_bh(&nf_conntrack_lock); } @@ -1200,6 +1321,19 @@ int nf_conntrack_helper_register(struct return 0; } +struct nf_conntrack_helper * +__nf_conntrack_helper_find_byname(const char *name) +{ + struct nf_conntrack_helper *h; + + list_for_each_entry(h, &helpers, list) { + if (!strcmp(h->name, name)) + return h; + } + + return NULL; +} + static inline int unhelp(struct nf_conntrack_tuple_hash *i, const struct nf_conntrack_helper *me) { @@ -1283,6 +1417,51 @@ void __nf_ct_refresh_acct(struct nf_conn nf_conntrack_event_cache(event, skb); } +#if defined(CONFIG_NF_CT_NETLINK) || \ + defined(CONFIG_NF_CT_NETLINK_MODULE) + +#include +#include + +/* Generic function for tcp/udp/sctp/dccp and alike. This needs to be + * in ip_conntrack_core, since we don't want the protocols to autoload + * or depend on ctnetlink */ +int nf_ct_port_tuple_to_nfattr(struct sk_buff *skb, + const struct nf_conntrack_tuple *tuple) +{ + NFA_PUT(skb, CTA_PROTO_SRC_PORT, sizeof(u_int16_t), + &tuple->src.u.tcp.port); + NFA_PUT(skb, CTA_PROTO_DST_PORT, sizeof(u_int16_t), + &tuple->dst.u.tcp.port); + return 0; + +nfattr_failure: + return -1; +} + +static const size_t cta_min_proto[CTA_PROTO_MAX] = { + [CTA_PROTO_SRC_PORT-1] = sizeof(u_int16_t), + [CTA_PROTO_DST_PORT-1] = sizeof(u_int16_t) +}; + +int nf_ct_port_nfattr_to_tuple(struct nfattr *tb[], + struct nf_conntrack_tuple *t) +{ + if (!tb[CTA_PROTO_SRC_PORT-1] || !tb[CTA_PROTO_DST_PORT-1]) + return -EINVAL; + + if (nfattr_bad_size(tb, CTA_PROTO_MAX, cta_min_proto)) + return -EINVAL; + + t->src.u.tcp.port = + *(u_int16_t *)NFA_DATA(tb[CTA_PROTO_SRC_PORT-1]); + t->dst.u.tcp.port = + *(u_int16_t *)NFA_DATA(tb[CTA_PROTO_DST_PORT-1]); + + return 0; +} +#endif + /* Used by ipt_REJECT and ip6t_REJECT. */ void __nf_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb) { @@ -1365,6 +1544,11 @@ static void free_conntrack_hash(struct l get_order(sizeof(struct list_head) * size)); } +void nf_conntrack_flush() +{ + nf_ct_iterate_cleanup(kill_all, NULL); +} + /* Mishearing the voices in his head, our hero wonders how he's supposed to kill the mall. */ void nf_conntrack_cleanup(void) @@ -1378,7 +1562,7 @@ void nf_conntrack_cleanup(void) nf_ct_event_cache_flush(); i_see_dead_people: - nf_ct_iterate_cleanup(kill_all, NULL); + nf_conntrack_flush(); if (atomic_read(&nf_conntrack_count) != 0) { schedule(); goto i_see_dead_people; diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 5a6fcf3..cb32d22 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -1136,6 +1136,63 @@ static int tcp_new(struct nf_conn *connt receiver->td_scale); return 1; } + +#if defined(CONFIG_NF_CT_NETLINK) || \ + defined(CONFIG_NF_CT_NETLINK_MODULE) + +#include +#include + +static int tcp_to_nfattr(struct sk_buff *skb, struct nfattr *nfa, + const struct nf_conn *ct) +{ + struct nfattr *nest_parms; + + read_lock_bh(&tcp_lock); + nest_parms = NFA_NEST(skb, CTA_PROTOINFO_TCP); + NFA_PUT(skb, CTA_PROTOINFO_TCP_STATE, sizeof(u_int8_t), + &ct->proto.tcp.state); + read_unlock_bh(&tcp_lock); + + NFA_NEST_END(skb, nest_parms); + + return 0; + +nfattr_failure: + read_unlock_bh(&tcp_lock); + return -1; +} + +static const size_t cta_min_tcp[CTA_PROTOINFO_TCP_MAX] = { + [CTA_PROTOINFO_TCP_STATE-1] = sizeof(u_int8_t), +}; + +static int nfattr_to_tcp(struct nfattr *cda[], struct nf_conn *ct) +{ + struct nfattr *attr = cda[CTA_PROTOINFO_TCP-1]; + struct nfattr *tb[CTA_PROTOINFO_TCP_MAX]; + + /* updates could not contain anything about the private + * protocol info, in that case skip the parsing */ + if (!attr) + return 0; + + nfattr_parse_nested(tb, CTA_PROTOINFO_TCP_MAX, attr); + + if (nfattr_bad_size(tb, CTA_PROTOINFO_TCP_MAX, cta_min_tcp)) + return -EINVAL; + + if (!tb[CTA_PROTOINFO_TCP_STATE-1]) + return -EINVAL; + + write_lock_bh(&tcp_lock); + ct->proto.tcp.state = + *(u_int8_t *)NFA_DATA(tb[CTA_PROTOINFO_TCP_STATE-1]); + write_unlock_bh(&tcp_lock); + + return 0; +} +#endif struct nf_conntrack_protocol nf_conntrack_protocol_tcp4 = { @@ -1149,6 +1206,13 @@ struct nf_conntrack_protocol nf_conntrac .packet = tcp_packet, .new = tcp_new, .error = tcp_error4, +#if defined(CONFIG_NF_CT_NETLINK) || \ + defined(CONFIG_NF_CT_NETLINK_MODULE) + .to_nfattr = tcp_to_nfattr, + .from_nfattr = nfattr_to_tcp, + .tuple_to_nfattr = nf_ct_port_tuple_to_nfattr, + .nfattr_to_tuple = nf_ct_port_nfattr_to_tuple, +#endif }; struct nf_conntrack_protocol nf_conntrack_protocol_tcp6 = @@ -1163,6 +1227,13 @@ struct nf_conntrack_protocol nf_conntrac .packet = tcp_packet, .new = tcp_new, .error = tcp_error6, +#if defined(CONFIG_NF_CT_NETLINK) || \ + defined(CONFIG_NF_CT_NETLINK_MODULE) + .to_nfattr = tcp_to_nfattr, + .from_nfattr = nfattr_to_tcp, + .tuple_to_nfattr = nf_ct_port_tuple_to_nfattr, + .nfattr_to_tuple = nf_ct_port_nfattr_to_tuple, +#endif }; EXPORT_SYMBOL(nf_conntrack_protocol_tcp4); diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c index 3cae7ce..1a592a5 100644 --- a/net/netfilter/nf_conntrack_proto_udp.c +++ b/net/netfilter/nf_conntrack_proto_udp.c @@ -196,6 +196,11 @@ struct nf_conntrack_protocol nf_conntrac .packet = udp_packet, .new = udp_new, .error = udp_error4, +#if defined(CONFIG_NF_CT_NETLINK) || \ + defined(CONFIG_NF_CT_NETLINK_MODULE) + .tuple_to_nfattr = nf_ct_port_tuple_to_nfattr, + .nfattr_to_tuple = nf_ct_port_nfattr_to_tuple, +#endif }; struct nf_conntrack_protocol nf_conntrack_protocol_udp6 = @@ -210,6 +215,11 @@ struct nf_conntrack_protocol nf_conntrac .packet = udp_packet, .new = udp_new, .error = udp_error6, +#if defined(CONFIG_NF_CT_NETLINK) || \ + defined(CONFIG_NF_CT_NETLINK_MODULE) + .tuple_to_nfattr = nf_ct_port_tuple_to_nfattr, + .nfattr_to_tuple = nf_ct_port_nfattr_to_tuple, +#endif }; EXPORT_SYMBOL(nf_conntrack_protocol_udp4); diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 5af381f..4c82f37 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -161,14 +161,14 @@ static int ct_seq_show(struct seq_file * if (NF_CT_DIRECTION(hash)) return 0; - l3proto = nf_ct_find_l3proto(conntrack->tuplehash[IP_CT_DIR_ORIGINAL] - .tuple.src.l3num); + l3proto = __nf_ct_l3proto_find(conntrack->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.src.l3num); NF_CT_ASSERT(l3proto); - proto = nf_ct_find_proto(conntrack->tuplehash[IP_CT_DIR_ORIGINAL] - .tuple.src.l3num, - conntrack->tuplehash[IP_CT_DIR_ORIGINAL] - .tuple.dst.protonum); + proto = __nf_ct_proto_find(conntrack->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.src.l3num, + conntrack->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.dst.protonum); NF_CT_ASSERT(proto); if (seq_printf(s, "%-8s %u %-8s %u %ld ", @@ -307,9 +307,9 @@ static int exp_seq_show(struct seq_file expect->tuple.src.l3num, expect->tuple.dst.protonum); print_tuple(s, &expect->tuple, - nf_ct_find_l3proto(expect->tuple.src.l3num), - nf_ct_find_proto(expect->tuple.src.l3num, - expect->tuple.dst.protonum)); + __nf_ct_l3proto_find(expect->tuple.src.l3num), + __nf_ct_proto_find(expect->tuple.src.l3num, + expect->tuple.dst.protonum)); return seq_putc(s, '\n'); } @@ -847,7 +847,11 @@ EXPORT_SYMBOL(nf_conntrack_helper_unregi EXPORT_SYMBOL(nf_ct_iterate_cleanup); EXPORT_SYMBOL(__nf_ct_refresh_acct); EXPORT_SYMBOL(nf_ct_protos); -EXPORT_SYMBOL(nf_ct_find_proto); +EXPORT_SYMBOL(__nf_ct_proto_find); +EXPORT_SYMBOL(nf_ct_proto_find_get); +EXPORT_SYMBOL(nf_ct_proto_put); +EXPORT_SYMBOL(nf_ct_l3proto_find_get); +EXPORT_SYMBOL(nf_ct_l3proto_put); EXPORT_SYMBOL(nf_ct_l3protos); EXPORT_SYMBOL(nf_conntrack_expect_alloc); EXPORT_SYMBOL(nf_conntrack_expect_put); @@ -867,3 +871,18 @@ EXPORT_SYMBOL(nf_ct_get_tuple); EXPORT_SYMBOL(nf_ct_invert_tuple); EXPORT_SYMBOL(nf_conntrack_in); EXPORT_SYMBOL(__nf_conntrack_attach); +EXPORT_SYMBOL(nf_conntrack_alloc); +EXPORT_SYMBOL(nf_conntrack_free); +EXPORT_SYMBOL(nf_conntrack_flush); +EXPORT_SYMBOL(nf_ct_remove_expectations); +EXPORT_SYMBOL(nf_ct_helper_find_get); +EXPORT_SYMBOL(nf_ct_helper_put); +EXPORT_SYMBOL(__nf_conntrack_helper_find_byname); +EXPORT_SYMBOL(__nf_conntrack_find); +EXPORT_SYMBOL(nf_ct_unlink_expect); +EXPORT_SYMBOL(nf_conntrack_hash_insert); +EXPORT_SYMBOL(__nf_conntrack_expect_find); +EXPORT_SYMBOL(nf_conntrack_expect_find); +EXPORT_SYMBOL(nf_conntrack_expect_list); +EXPORT_SYMBOL(nf_ct_port_tuple_to_nfattr); +EXPORT_SYMBOL(nf_ct_port_nfattr_to_tuple); --------------050000040707000002080806--