From 9fb9cbb1082d6b31fb45aa1a14432449a0df6cf1 Mon Sep 17 00:00:00 2001 From: Yasuyuki Kozakai Date: Wed, 9 Nov 2005 16:38:16 -0800 Subject: [NETFILTER]: Add nf_conntrack subsystem. The existing connection tracking subsystem in netfilter can only handle ipv4. There were basically two choices present to add connection tracking support for ipv6. We could either duplicate all of the ipv4 connection tracking code into an ipv6 counterpart, or (the choice taken by these patches) we could design a generic layer that could handle both ipv4 and ipv6 and thus requiring only one sub-protocol (TCP, UDP, etc.) connection tracking helper module to be written. In fact nf_conntrack is capable of working with any layer 3 protocol. The existing ipv4 specific conntrack code could also not deal with the pecularities of doing connection tracking on ipv6, which is also cured here. For example, these issues include: 1) ICMPv6 handling, which is used for neighbour discovery in ipv6 thus some messages such as these should not participate in connection tracking since effectively they are like ARP messages 2) fragmentation must be handled differently in ipv6, because the simplistic "defrag, connection track and NAT, refrag" (which the existing ipv4 connection tracking does) approach simply isn't feasible in ipv6 3) ipv6 extension header parsing must occur at the correct spots before and after connection tracking decisions, and there were no provisions for this in the existing connection tracking design 4) ipv6 has no need for stateful NAT The ipv4 specific conntrack layer is kept around, until all of the ipv4 specific conntrack helpers are ported over to nf_conntrack and it is feature complete. Once that occurs, the old conntrack stuff will get placed into the feature-removal-schedule and we will fully kill it off 6 months later. Signed-off-by: Yasuyuki Kozakai Signed-off-by: Harald Welte Signed-off-by: Arnaldo Carvalho de Melo --- net/core/skbuff.c | 15 + net/ipv4/netfilter/Kconfig | 41 +- net/ipv4/netfilter/Makefile | 6 + net/ipv4/netfilter/ip_conntrack_netlink.c | 2 +- net/ipv4/netfilter/ipt_CLUSTERIP.c | 12 +- net/ipv4/netfilter/ipt_CONNMARK.c | 22 +- net/ipv4/netfilter/ipt_NOTRACK.c | 4 +- net/ipv4/netfilter/ipt_connbytes.c | 39 +- net/ipv4/netfilter/ipt_connmark.c | 10 +- net/ipv4/netfilter/ipt_conntrack.c | 96 ++ net/ipv4/netfilter/ipt_helper.c | 54 + net/ipv4/netfilter/ipt_state.c | 6 +- net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 571 +++++++++ net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 301 +++++ net/ipv6/ip6_input.c | 5 + net/ipv6/ip6_output.c | 6 + net/ipv6/netfilter/Kconfig | 14 + net/ipv6/netfilter/Makefile | 6 + net/ipv6/netfilter/ip6t_MARK.c | 6 +- net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 556 +++++++++ net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 272 +++++ net/ipv6/netfilter/nf_conntrack_reasm.c | 885 ++++++++++++++ net/ipv6/raw.c | 4 +- net/netfilter/Kconfig | 74 ++ net/netfilter/Makefile | 8 + net/netfilter/nf_conntrack_core.c | 1538 ++++++++++++++++++++++++ net/netfilter/nf_conntrack_ftp.c | 698 +++++++++++ net/netfilter/nf_conntrack_l3proto_generic.c | 98 ++ net/netfilter/nf_conntrack_proto_generic.c | 85 ++ net/netfilter/nf_conntrack_proto_sctp.c | 670 +++++++++++ net/netfilter/nf_conntrack_proto_tcp.c | 1162 ++++++++++++++++++ net/netfilter/nf_conntrack_proto_udp.c | 216 ++++ net/netfilter/nf_conntrack_standalone.c | 869 +++++++++++++ 33 files changed, 8289 insertions(+), 62 deletions(-) create mode 100644 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c create mode 100644 net/ipv4/netfilter/nf_conntrack_proto_icmp.c create mode 100644 net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c create mode 100644 net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c create mode 100644 net/ipv6/netfilter/nf_conntrack_reasm.c create mode 100644 net/netfilter/nf_conntrack_core.c create mode 100644 net/netfilter/nf_conntrack_ftp.c create mode 100644 net/netfilter/nf_conntrack_l3proto_generic.c create mode 100644 net/netfilter/nf_conntrack_proto_generic.c create mode 100644 net/netfilter/nf_conntrack_proto_sctp.c create mode 100644 net/netfilter/nf_conntrack_proto_tcp.c create mode 100644 net/netfilter/nf_conntrack_proto_udp.c create mode 100644 net/netfilter/nf_conntrack_standalone.c (limited to 'net') diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 95501e40100e..b7d13a4fff48 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -336,6 +336,9 @@ void __kfree_skb(struct sk_buff *skb) } #ifdef CONFIG_NETFILTER nf_conntrack_put(skb->nfct); +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + nf_conntrack_put_reasm(skb->nfct_reasm); +#endif #ifdef CONFIG_BRIDGE_NETFILTER nf_bridge_put(skb->nf_bridge); #endif @@ -414,9 +417,17 @@ struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t gfp_mask) C(nfct); nf_conntrack_get(skb->nfct); C(nfctinfo); +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + C(nfct_reasm); + nf_conntrack_get_reasm(skb->nfct_reasm); +#endif #if defined(CONFIG_IP_VS) || defined(CONFIG_IP_VS_MODULE) C(ipvs_property); #endif +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + C(nfct_reasm); + nf_conntrack_get_reasm(skb->nfct_reasm); +#endif #ifdef CONFIG_BRIDGE_NETFILTER C(nf_bridge); nf_bridge_get(skb->nf_bridge); @@ -474,6 +485,10 @@ static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old) new->nfct = old->nfct; nf_conntrack_get(old->nfct); new->nfctinfo = old->nfctinfo; +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + new->nfct_reasm = old->nfct_reasm; + nf_conntrack_get_reasm(old->nfct_reasm); +#endif #if defined(CONFIG_IP_VS) || defined(CONFIG_IP_VS_MODULE) new->ipvs_property = old->ipvs_property; #endif diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index 7d917e4ce1d9..9d3c8b5f327e 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -5,6 +5,20 @@ menu "IP: Netfilter Configuration" depends on INET && NETFILTER +config NF_CONNTRACK_IPV4 + tristate "IPv4 support for new connection tracking (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + ---help--- + Connection tracking keeps a record of what packets have passed + through your machine, in order to figure out how they are related + into connections. + + This is IPv4 support on Layer 3 independent connection tracking. + Layer 3 independent connection tracking is experimental scheme + which generalize ip_conntrack to support other layer 3 protocols. + + To compile it as a module, choose M here. If unsure, say N. + # connection tracking, helpers and protocols config IP_NF_CONNTRACK tristate "Connection tracking (required for masq/NAT)" @@ -209,8 +223,8 @@ config IP_NF_MATCH_PKTTYPE tristate "Packet type match support" depends on IP_NF_IPTABLES help - Packet type matching allows you to match a packet by - its "class", eg. BROADCAST, MULTICAST, ... + Packet type matching allows you to match a packet by + its "class", eg. BROADCAST, MULTICAST, ... Typical usage: iptables -A INPUT -m pkttype --pkt-type broadcast -j LOG @@ -317,7 +331,8 @@ config IP_NF_MATCH_TCPMSS config IP_NF_MATCH_HELPER tristate "Helper match support" - depends on IP_NF_CONNTRACK && IP_NF_IPTABLES + depends on IP_NF_IPTABLES + depends on IP_NF_CONNTRACK || NF_CONNTRACK_IPV4 help Helper matching allows you to match packets in dynamic connections tracked by a conntrack-helper, ie. ip_conntrack_ftp @@ -326,7 +341,8 @@ config IP_NF_MATCH_HELPER config IP_NF_MATCH_STATE tristate "Connection state match support" - depends on IP_NF_CONNTRACK && IP_NF_IPTABLES + depends on IP_NF_IPTABLES + depends on IP_NF_CONNTRACK || NF_CONNTRACK_IPV4 help Connection state matching allows you to match packets based on their relationship to a tracked connection (ie. previous packets). This @@ -336,7 +352,8 @@ config IP_NF_MATCH_STATE config IP_NF_MATCH_CONNTRACK tristate "Connection tracking match support" - depends on IP_NF_CONNTRACK && IP_NF_IPTABLES + depends on IP_NF_IPTABLES + depends on IP_NF_CONNTRACK || NF_CONNTRACK_IPV4 help This is a general conntrack match module, a superset of the state match. @@ -422,7 +439,8 @@ config IP_NF_MATCH_COMMENT config IP_NF_MATCH_CONNMARK tristate 'Connection mark match support' - depends on IP_NF_CONNTRACK_MARK && IP_NF_IPTABLES + depends on IP_NF_IPTABLES + depends on IP_NF_CONNTRACK_MARK || (NF_CONNTRACK_MARK && NF_CONNTRACK_IPV4) help This option adds a `connmark' match, which allows you to match the connection mark value previously set for the session by `CONNMARK'. @@ -433,7 +451,8 @@ config IP_NF_MATCH_CONNMARK config IP_NF_MATCH_CONNBYTES tristate 'Connection byte/packet counter match support' - depends on IP_NF_CT_ACCT && IP_NF_IPTABLES + depends on IP_NF_IPTABLES + depends on IP_NF_CT_ACCT || (NF_CT_ACCT && NF_CONNTRACK_IPV4) help This option adds a `connbytes' match, which allows you to match the number of bytes and/or packets for each direction within a connection. @@ -747,7 +766,8 @@ config IP_NF_TARGET_TTL config IP_NF_TARGET_CONNMARK tristate 'CONNMARK target support' - depends on IP_NF_CONNTRACK_MARK && IP_NF_MANGLE + depends on IP_NF_MANGLE + depends on IP_NF_CONNTRACK_MARK || (NF_CONNTRACK_MARK && NF_CONNTRACK_IPV4) help This option adds a `CONNMARK' target, which allows one to manipulate the connection mark value. Similar to the MARK target, but @@ -759,7 +779,8 @@ config IP_NF_TARGET_CONNMARK config IP_NF_TARGET_CLUSTERIP tristate "CLUSTERIP target support (EXPERIMENTAL)" - depends on IP_NF_CONNTRACK_MARK && IP_NF_IPTABLES && EXPERIMENTAL + depends on IP_NF_IPTABLES && EXPERIMENTAL + depends on IP_NF_CONNTRACK_MARK || (NF_CONNTRACK_MARK && NF_CONNTRACK_IPV4) help The CLUSTERIP target allows you to build load-balancing clusters of network servers without having a dedicated load-balancing @@ -782,7 +803,7 @@ config IP_NF_RAW config IP_NF_TARGET_NOTRACK tristate 'NOTRACK target support' depends on IP_NF_RAW - depends on IP_NF_CONNTRACK + depends on IP_NF_CONNTRACK || NF_CONNTRACK_IPV4 help The NOTRACK target allows a select rule to specify which packets *not* to enter the conntrack/NAT diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index dab4b58dd31e..058c48e258fc 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -103,3 +103,9 @@ obj-$(CONFIG_IP_NF_ARP_MANGLE) += arpt_mangle.o obj-$(CONFIG_IP_NF_ARPFILTER) += arptable_filter.o obj-$(CONFIG_IP_NF_QUEUE) += ip_queue.o + +# objects for l3 independent conntrack +nf_conntrack_ipv4-objs := nf_conntrack_l3proto_ipv4.o nf_conntrack_proto_icmp.o + +# l3 independent conntrack +obj-$(CONFIG_NF_CONNTRACK_IPV4) += nf_conntrack_ipv4.o diff --git a/net/ipv4/netfilter/ip_conntrack_netlink.c b/net/ipv4/netfilter/ip_conntrack_netlink.c index 5c1c0a3d1c4b..d2a4fec22862 100644 --- a/net/ipv4/netfilter/ip_conntrack_netlink.c +++ b/net/ipv4/netfilter/ip_conntrack_netlink.c @@ -1376,7 +1376,7 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, ip_conntrack_expect_put(exp); } } - write_unlock(&ip_conntrack_lock); + write_unlock_bh(&ip_conntrack_lock); } else { /* This basically means we have to flush everything*/ write_lock_bh(&ip_conntrack_lock); diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index 9bcb398fbc1f..45c52d8f4d99 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -29,7 +29,7 @@ #include #include -#include +#include #define CLUSTERIP_VERSION "0.8" @@ -316,14 +316,14 @@ target(struct sk_buff **pskb, { const struct ipt_clusterip_tgt_info *cipinfo = targinfo; enum ip_conntrack_info ctinfo; - struct ip_conntrack *ct = ip_conntrack_get((*pskb), &ctinfo); - u_int32_t hash; + u_int32_t *mark, hash; /* don't need to clusterip_config_get() here, since refcount * is only decremented by destroy() - and ip_tables guarantees * that the ->target() function isn't called after ->destroy() */ - if (!ct) { + mark = nf_ct_get_mark((*pskb), &ctinfo); + if (mark == NULL) { printk(KERN_ERR "CLUSTERIP: no conntrack!\n"); /* FIXME: need to drop invalid ones, since replies * to outgoing connections of other nodes will be @@ -346,7 +346,7 @@ target(struct sk_buff **pskb, switch (ctinfo) { case IP_CT_NEW: - ct->mark = hash; + *mark = hash; break; case IP_CT_RELATED: case IP_CT_RELATED+IP_CT_IS_REPLY: @@ -363,7 +363,7 @@ target(struct sk_buff **pskb, #ifdef DEBUG_CLUSTERP DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); #endif - DEBUGP("hash=%u ct_hash=%u ", hash, ct->mark); + DEBUGP("hash=%u ct_hash=%u ", hash, *mark); if (!clusterip_responsible(cipinfo->config, hash)) { DEBUGP("not responsible\n"); return NF_DROP; diff --git a/net/ipv4/netfilter/ipt_CONNMARK.c b/net/ipv4/netfilter/ipt_CONNMARK.c index 05d66ab59424..8acac5a40a92 100644 --- a/net/ipv4/netfilter/ipt_CONNMARK.c +++ b/net/ipv4/netfilter/ipt_CONNMARK.c @@ -29,7 +29,7 @@ MODULE_LICENSE("GPL"); #include #include -#include +#include static unsigned int target(struct sk_buff **pskb, @@ -43,24 +43,24 @@ target(struct sk_buff **pskb, u_int32_t diff; u_int32_t nfmark; u_int32_t newmark; + u_int32_t ctinfo; + u_int32_t *ctmark = nf_ct_get_mark(*pskb, &ctinfo); - enum ip_conntrack_info ctinfo; - struct ip_conntrack *ct = ip_conntrack_get((*pskb), &ctinfo); - if (ct) { + if (ctmark) { switch(markinfo->mode) { case IPT_CONNMARK_SET: - newmark = (ct->mark & ~markinfo->mask) | markinfo->mark; - if (newmark != ct->mark) - ct->mark = newmark; + newmark = (*ctmark & ~markinfo->mask) | markinfo->mark; + if (newmark != *ctmark) + *ctmark = newmark; break; case IPT_CONNMARK_SAVE: - newmark = (ct->mark & ~markinfo->mask) | ((*pskb)->nfmark & markinfo->mask); - if (ct->mark != newmark) - ct->mark = newmark; + newmark = (*ctmark & ~markinfo->mask) | ((*pskb)->nfmark & markinfo->mask); + if (*ctmark != newmark) + *ctmark = newmark; break; case IPT_CONNMARK_RESTORE: nfmark = (*pskb)->nfmark; - diff = (ct->mark ^ nfmark) & markinfo->mask; + diff = (*ctmark ^ nfmark) & markinfo->mask; if (diff != 0) (*pskb)->nfmark = nfmark ^ diff; break; diff --git a/net/ipv4/netfilter/ipt_NOTRACK.c b/net/ipv4/netfilter/ipt_NOTRACK.c index a4bb9b3bc292..e3c69d072c6e 100644 --- a/net/ipv4/netfilter/ipt_NOTRACK.c +++ b/net/ipv4/netfilter/ipt_NOTRACK.c @@ -5,7 +5,7 @@ #include #include -#include +#include static unsigned int target(struct sk_buff **pskb, @@ -23,7 +23,7 @@ target(struct sk_buff **pskb, If there is a real ct entry correspondig to this packet, it'll hang aroun till timing out. We don't deal with it for performance reasons. JK */ - (*pskb)->nfct = &ip_conntrack_untracked.ct_general; + nf_ct_untrack(*pskb); (*pskb)->nfctinfo = IP_CT_NEW; nf_conntrack_get((*pskb)->nfct); diff --git a/net/ipv4/netfilter/ipt_connbytes.c b/net/ipv4/netfilter/ipt_connbytes.c index df4a42c6da22..d68a048b7176 100644 --- a/net/ipv4/netfilter/ipt_connbytes.c +++ b/net/ipv4/netfilter/ipt_connbytes.c @@ -10,7 +10,7 @@ */ #include #include -#include +#include #include #include @@ -46,60 +46,59 @@ match(const struct sk_buff *skb, int *hotdrop) { const struct ipt_connbytes_info *sinfo = matchinfo; - enum ip_conntrack_info ctinfo; - struct ip_conntrack *ct; u_int64_t what = 0; /* initialize to make gcc happy */ + const struct ip_conntrack_counter *counters; - if (!(ct = ip_conntrack_get((struct sk_buff *)skb, &ctinfo))) + if (!(counters = nf_ct_get_counters(skb))) return 0; /* no match */ switch (sinfo->what) { case IPT_CONNBYTES_PKTS: switch (sinfo->direction) { case IPT_CONNBYTES_DIR_ORIGINAL: - what = ct->counters[IP_CT_DIR_ORIGINAL].packets; + what = counters[IP_CT_DIR_ORIGINAL].packets; break; case IPT_CONNBYTES_DIR_REPLY: - what = ct->counters[IP_CT_DIR_REPLY].packets; + what = counters[IP_CT_DIR_REPLY].packets; break; case IPT_CONNBYTES_DIR_BOTH: - what = ct->counters[IP_CT_DIR_ORIGINAL].packets; - what += ct->counters[IP_CT_DIR_REPLY].packets; + what = counters[IP_CT_DIR_ORIGINAL].packets; + what += counters[IP_CT_DIR_REPLY].packets; break; } break; case IPT_CONNBYTES_BYTES: switch (sinfo->direction) { case IPT_CONNBYTES_DIR_ORIGINAL: - what = ct->counters[IP_CT_DIR_ORIGINAL].bytes; + what = counters[IP_CT_DIR_ORIGINAL].bytes; break; case IPT_CONNBYTES_DIR_REPLY: - what = ct->counters[IP_CT_DIR_REPLY].bytes; + what = counters[IP_CT_DIR_REPLY].bytes; break; case IPT_CONNBYTES_DIR_BOTH: - what = ct->counters[IP_CT_DIR_ORIGINAL].bytes; - what += ct->counters[IP_CT_DIR_REPLY].bytes; + what = counters[IP_CT_DIR_ORIGINAL].bytes; + what += counters[IP_CT_DIR_REPLY].bytes; break; } break; case IPT_CONNBYTES_AVGPKT: switch (sinfo->direction) { case IPT_CONNBYTES_DIR_ORIGINAL: - what = div64_64(ct->counters[IP_CT_DIR_ORIGINAL].bytes, - ct->counters[IP_CT_DIR_ORIGINAL].packets); + what = div64_64(counters[IP_CT_DIR_ORIGINAL].bytes, + counters[IP_CT_DIR_ORIGINAL].packets); break; case IPT_CONNBYTES_DIR_REPLY: - what = div64_64(ct->counters[IP_CT_DIR_REPLY].bytes, - ct->counters[IP_CT_DIR_REPLY].packets); + what = div64_64(counters[IP_CT_DIR_REPLY].bytes, + counters[IP_CT_DIR_REPLY].packets); break; case IPT_CONNBYTES_DIR_BOTH: { u_int64_t bytes; u_int64_t pkts; - bytes = ct->counters[IP_CT_DIR_ORIGINAL].bytes + - ct->counters[IP_CT_DIR_REPLY].bytes; - pkts = ct->counters[IP_CT_DIR_ORIGINAL].packets+ - ct->counters[IP_CT_DIR_REPLY].packets; + bytes = counters[IP_CT_DIR_ORIGINAL].bytes + + counters[IP_CT_DIR_REPLY].bytes; + pkts = counters[IP_CT_DIR_ORIGINAL].packets+ + counters[IP_CT_DIR_REPLY].packets; /* FIXME_THEORETICAL: what to do if sum * overflows ? */ diff --git a/net/ipv4/netfilter/ipt_connmark.c b/net/ipv4/netfilter/ipt_connmark.c index bf8de47ce004..5306ef293b92 100644 --- a/net/ipv4/netfilter/ipt_connmark.c +++ b/net/ipv4/netfilter/ipt_connmark.c @@ -28,7 +28,7 @@ MODULE_LICENSE("GPL"); #include #include -#include +#include static int match(const struct sk_buff *skb, @@ -39,12 +39,12 @@ match(const struct sk_buff *skb, int *hotdrop) { const struct ipt_connmark_info *info = matchinfo; - enum ip_conntrack_info ctinfo; - struct ip_conntrack *ct = ip_conntrack_get((struct sk_buff *)skb, &ctinfo); - if (!ct) + u_int32_t ctinfo; + const u_int32_t *ctmark = nf_ct_get_mark(skb, &ctinfo); + if (!ctmark) return 0; - return ((ct->mark & info->mask) == info->mark) ^ info->invert; + return (((*ctmark) & info->mask) == info->mark) ^ info->invert; } static int diff --git a/net/ipv4/netfilter/ipt_conntrack.c b/net/ipv4/netfilter/ipt_conntrack.c index c1d22801b7cf..c8d18705469b 100644 --- a/net/ipv4/netfilter/ipt_conntrack.c +++ b/net/ipv4/netfilter/ipt_conntrack.c @@ -10,7 +10,14 @@ #include #include + +#if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) #include +#include +#else +#include +#endif + #include #include @@ -18,6 +25,8 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Marc Boucher "); MODULE_DESCRIPTION("iptables connection tracking match module"); +#if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) + static int match(const struct sk_buff *skb, const struct net_device *in, @@ -102,6 +111,93 @@ match(const struct sk_buff *skb, return 1; } +#else /* CONFIG_IP_NF_CONNTRACK */ +static int +match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + int *hotdrop) +{ + const struct ipt_conntrack_info *sinfo = matchinfo; + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + unsigned int statebit; + + ct = nf_ct_get((struct sk_buff *)skb, &ctinfo); + +#define FWINV(bool,invflg) ((bool) ^ !!(sinfo->invflags & invflg)) + + if (ct == &nf_conntrack_untracked) + statebit = IPT_CONNTRACK_STATE_UNTRACKED; + else if (ct) + statebit = IPT_CONNTRACK_STATE_BIT(ctinfo); + else + statebit = IPT_CONNTRACK_STATE_INVALID; + + if(sinfo->flags & IPT_CONNTRACK_STATE) { + if (ct) { + if(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip != + ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip) + statebit |= IPT_CONNTRACK_STATE_SNAT; + + if(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.u3.ip != + ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip) + statebit |= IPT_CONNTRACK_STATE_DNAT; + } + + if (FWINV((statebit & sinfo->statemask) == 0, IPT_CONNTRACK_STATE)) + return 0; + } + + if(sinfo->flags & IPT_CONNTRACK_PROTO) { + if (!ct || FWINV(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum != sinfo->tuple[IP_CT_DIR_ORIGINAL].dst.protonum, IPT_CONNTRACK_PROTO)) + return 0; + } + + if(sinfo->flags & IPT_CONNTRACK_ORIGSRC) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip&sinfo->sipmsk[IP_CT_DIR_ORIGINAL].s_addr) != sinfo->tuple[IP_CT_DIR_ORIGINAL].src.ip, IPT_CONNTRACK_ORIGSRC)) + return 0; + } + + if(sinfo->flags & IPT_CONNTRACK_ORIGDST) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.u3.ip&sinfo->dipmsk[IP_CT_DIR_ORIGINAL].s_addr) != sinfo->tuple[IP_CT_DIR_ORIGINAL].dst.ip, IPT_CONNTRACK_ORIGDST)) + return 0; + } + + if(sinfo->flags & IPT_CONNTRACK_REPLSRC) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip&sinfo->sipmsk[IP_CT_DIR_REPLY].s_addr) != sinfo->tuple[IP_CT_DIR_REPLY].src.ip, IPT_CONNTRACK_REPLSRC)) + return 0; + } + + if(sinfo->flags & IPT_CONNTRACK_REPLDST) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip&sinfo->dipmsk[IP_CT_DIR_REPLY].s_addr) != sinfo->tuple[IP_CT_DIR_REPLY].dst.ip, IPT_CONNTRACK_REPLDST)) + return 0; + } + + if(sinfo->flags & IPT_CONNTRACK_STATUS) { + if (!ct || FWINV((ct->status & sinfo->statusmask) == 0, IPT_CONNTRACK_STATUS)) + return 0; + } + + if(sinfo->flags & IPT_CONNTRACK_EXPIRES) { + unsigned long expires; + + if(!ct) + return 0; + + expires = timer_pending(&ct->timeout) ? (ct->timeout.expires - jiffies)/HZ : 0; + + if (FWINV(!(expires >= sinfo->expires_min && expires <= sinfo->expires_max), IPT_CONNTRACK_EXPIRES)) + return 0; + } + + return 1; +} + +#endif /* CONFIG_NF_IP_CONNTRACK */ + static int check(const char *tablename, const struct ipt_ip *ip, void *matchinfo, diff --git a/net/ipv4/netfilter/ipt_helper.c b/net/ipv4/netfilter/ipt_helper.c index 3e7dd014de43..bf14e1c7798a 100644 --- a/net/ipv4/netfilter/ipt_helper.c +++ b/net/ipv4/netfilter/ipt_helper.c @@ -13,9 +13,15 @@ #include #include #include +#if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) #include #include #include +#else +#include +#include +#include +#endif #include #include @@ -29,6 +35,7 @@ MODULE_DESCRIPTION("iptables helper match module"); #define DEBUGP(format, args...) #endif +#if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) static int match(const struct sk_buff *skb, const struct net_device *in, @@ -73,6 +80,53 @@ out_unlock: return ret; } +#else /* CONFIG_IP_NF_CONNTRACK */ + +static int +match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + int *hotdrop) +{ + const struct ipt_helper_info *info = matchinfo; + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + int ret = info->invert; + + ct = nf_ct_get((struct sk_buff *)skb, &ctinfo); + if (!ct) { + DEBUGP("ipt_helper: Eek! invalid conntrack?\n"); + return ret; + } + + if (!ct->master) { + DEBUGP("ipt_helper: conntrack %p has no master\n", ct); + return ret; + } + + read_lock_bh(&nf_conntrack_lock); + if (!ct->master->helper) { + DEBUGP("ipt_helper: master ct %p has no helper\n", + exp->expectant); + goto out_unlock; + } + + DEBUGP("master's name = %s , info->name = %s\n", + ct->master->helper->name, info->name); + + if (info->name[0] == '\0') + ret ^= 1; + else + ret ^= !strncmp(ct->master->helper->name, info->name, + strlen(ct->master->helper->name)); +out_unlock: + read_unlock_bh(&nf_conntrack_lock); + return ret; +} +#endif + static int check(const char *tablename, const struct ipt_ip *ip, void *matchinfo, diff --git a/net/ipv4/netfilter/ipt_state.c b/net/ipv4/netfilter/ipt_state.c index b1511b97ea5f..4d7f16b70cec 100644 --- a/net/ipv4/netfilter/ipt_state.c +++ b/net/ipv4/netfilter/ipt_state.c @@ -10,7 +10,7 @@ #include #include -#include +#include #include #include @@ -30,9 +30,9 @@ match(const struct sk_buff *skb, enum ip_conntrack_info ctinfo; unsigned int statebit; - if (skb->nfct == &ip_conntrack_untracked.ct_general) + if (nf_ct_is_untracked(skb)) statebit = IPT_STATE_UNTRACKED; - else if (!ip_conntrack_get(skb, &ctinfo)) + else if (!nf_ct_get_ctinfo(skb, &ctinfo)) statebit = IPT_STATE_INVALID; else statebit = IPT_STATE_BIT(ctinfo); diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c new file mode 100644 index 000000000000..8202c1c0afad --- /dev/null +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -0,0 +1,571 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI + * - move L3 protocol dependent part to this file. + * 23 Mar 2004: Yasuyuki Kozakai @USAGI + * - add get_features() to support various size of conntrack + * structures. + * + * Derived from net/ipv4/netfilter/ip_conntrack_standalone.c + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +DECLARE_PER_CPU(struct nf_conntrack_stat, nf_conntrack_stat); + +static int ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, + struct nf_conntrack_tuple *tuple) +{ + u_int32_t _addrs[2], *ap; + ap = skb_header_pointer(skb, nhoff + offsetof(struct iphdr, saddr), + sizeof(u_int32_t) * 2, _addrs); + if (ap == NULL) + return 0; + + tuple->src.u3.ip = ap[0]; + tuple->dst.u3.ip = ap[1]; + + return 1; +} + +static int ipv4_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + tuple->src.u3.ip = orig->dst.u3.ip; + tuple->dst.u3.ip = orig->src.u3.ip; + + return 1; +} + +static int ipv4_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "src=%u.%u.%u.%u dst=%u.%u.%u.%u ", + NIPQUAD(tuple->src.u3.ip), + NIPQUAD(tuple->dst.u3.ip)); +} + +static int ipv4_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + return 0; +} + +/* Returns new sk_buff, or NULL */ +static struct sk_buff * +nf_ct_ipv4_gather_frags(struct sk_buff *skb, u_int32_t user) +{ + skb_orphan(skb); + + local_bh_disable(); + skb = ip_defrag(skb, user); + local_bh_enable(); + + if (skb) + ip_send_check(skb->nh.iph); + + return skb; +} + +static int +ipv4_prepare(struct sk_buff **pskb, unsigned int hooknum, unsigned int *dataoff, + u_int8_t *protonum) +{ + /* Never happen */ + if ((*pskb)->nh.iph->frag_off & htons(IP_OFFSET)) { + if (net_ratelimit()) { + printk(KERN_ERR "ipv4_prepare: Frag of proto %u (hook=%u)\n", + (*pskb)->nh.iph->protocol, hooknum); + } + return -NF_DROP; + } + + *dataoff = (*pskb)->nh.raw - (*pskb)->data + (*pskb)->nh.iph->ihl*4; + *protonum = (*pskb)->nh.iph->protocol; + + return NF_ACCEPT; +} + +int nat_module_is_loaded = 0; +static u_int32_t ipv4_get_features(const struct nf_conntrack_tuple *tuple) +{ + if (nat_module_is_loaded) + return NF_CT_F_NAT; + + return NF_CT_F_BASIC; +} + +static unsigned int ipv4_confirm(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + /* We've seen it coming out the other side: confirm it */ + return nf_conntrack_confirm(pskb); +} + +static unsigned int ipv4_conntrack_help(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + + /* This is where we call the helper: as the packet goes out. */ + ct = nf_ct_get(*pskb, &ctinfo); + if (ct && ct->helper) { + unsigned int ret; + ret = ct->helper->help(pskb, + (*pskb)->nh.raw - (*pskb)->data + + (*pskb)->nh.iph->ihl*4, + ct, ctinfo); + if (ret != NF_ACCEPT) + return ret; + } + return NF_ACCEPT; +} + +static unsigned int ipv4_conntrack_defrag(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ +#if !defined(CONFIG_IP_NF_NAT) && !defined(CONFIG_IP_NF_NAT_MODULE) + /* Previously seen (loopback)? Ignore. Do this before + fragment check. */ + if ((*pskb)->nfct) + return NF_ACCEPT; +#endif + + /* Gather fragments. */ + if ((*pskb)->nh.iph->frag_off & htons(IP_MF|IP_OFFSET)) { + *pskb = nf_ct_ipv4_gather_frags(*pskb, + hooknum == NF_IP_PRE_ROUTING ? + IP_DEFRAG_CONNTRACK_IN : + IP_DEFRAG_CONNTRACK_OUT); + if (!*pskb) + return NF_STOLEN; + } + return NF_ACCEPT; +} + +static unsigned int ipv4_refrag(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct rtable *rt = (struct rtable *)(*pskb)->dst; + + /* We've seen it coming out the other side: confirm */ + if (ipv4_confirm(hooknum, pskb, in, out, okfn) != NF_ACCEPT) + return NF_DROP; + + /* Local packets are never produced too large for their + interface. We degfragment them at LOCAL_OUT, however, + so we have to refragment them here. */ + if ((*pskb)->len > dst_mtu(&rt->u.dst) && + !skb_shinfo(*pskb)->tso_size) { + /* No hook can be after us, so this should be OK. */ + ip_fragment(*pskb, okfn); + return NF_STOLEN; + } + return NF_ACCEPT; +} + +static unsigned int ipv4_conntrack_in(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + return nf_conntrack_in(PF_INET, hooknum, pskb); +} + +static unsigned int ipv4_conntrack_local(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + /* root is playing with raw sockets. */ + if ((*pskb)->len < sizeof(struct iphdr) + || (*pskb)->nh.iph->ihl * 4 < sizeof(struct iphdr)) { + if (net_ratelimit()) + printk("ipt_hook: happy cracking.\n"); + return NF_ACCEPT; + } + return nf_conntrack_in(PF_INET, hooknum, pskb); +} + +/* Connection tracking may drop packets, but never alters them, so + make it the first hook. */ +static struct nf_hook_ops ipv4_conntrack_defrag_ops = { + .hook = ipv4_conntrack_defrag, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_PRE_ROUTING, + .priority = NF_IP_PRI_CONNTRACK_DEFRAG, +}; + +static struct nf_hook_ops ipv4_conntrack_in_ops = { + .hook = ipv4_conntrack_in, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_PRE_ROUTING, + .priority = NF_IP_PRI_CONNTRACK, +}; + +static struct nf_hook_ops ipv4_conntrack_defrag_local_out_ops = { + .hook = ipv4_conntrack_defrag, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_OUT, + .priority = NF_IP_PRI_CONNTRACK_DEFRAG, +}; + +static struct nf_hook_ops ipv4_conntrack_local_out_ops = { + .hook = ipv4_conntrack_local, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_OUT, + .priority = NF_IP_PRI_CONNTRACK, +}; + +/* helpers */ +static struct nf_hook_ops ipv4_conntrack_helper_out_ops = { + .hook = ipv4_conntrack_help, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_POST_ROUTING, + .priority = NF_IP_PRI_CONNTRACK_HELPER, +}; + +static struct nf_hook_ops ipv4_conntrack_helper_in_ops = { + .hook = ipv4_conntrack_help, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_IN, + .priority = NF_IP_PRI_CONNTRACK_HELPER, +}; + + +/* Refragmenter; last chance. */ +static struct nf_hook_ops ipv4_conntrack_out_ops = { + .hook = ipv4_refrag, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_POST_ROUTING, + .priority = NF_IP_PRI_CONNTRACK_CONFIRM, +}; + +static struct nf_hook_ops ipv4_conntrack_local_in_ops = { + .hook = ipv4_confirm, + .owner = THIS_MODULE, + .pf = PF_INET, + .hooknum = NF_IP_LOCAL_IN, + .priority = NF_IP_PRI_CONNTRACK_CONFIRM, +}; + +#ifdef CONFIG_SYSCTL +/* From nf_conntrack_proto_icmp.c */ +extern unsigned long nf_ct_icmp_timeout; +static struct ctl_table_header *nf_ct_ipv4_sysctl_header; + +static ctl_table nf_ct_sysctl_table[] = { + { + .ctl_name = NET_NF_CONNTRACK_ICMP_TIMEOUT, + .procname = "nf_conntrack_icmp_timeout", + .data = &nf_ct_icmp_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_netfilter_table[] = { + { + .ctl_name = NET_NETFILTER, + .procname = "netfilter", + .mode = 0555, + .child = nf_ct_sysctl_table, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_net_table[] = { + { + .ctl_name = CTL_NET, + .procname = "net", + .mode = 0555, + .child = nf_ct_netfilter_table, + }, + { .ctl_name = 0 } +}; +#endif + +/* Fast function for those who don't want to parse /proc (and I don't + blame them). */ +/* Reversing the socket's dst/src point of view gives us the reply + mapping. */ +static int +getorigdst(struct sock *sk, int optval, void __user *user, int *len) +{ + struct inet_sock *inet = inet_sk(sk); + struct nf_conntrack_tuple_hash *h; + struct nf_conntrack_tuple tuple; + + NF_CT_TUPLE_U_BLANK(&tuple); + tuple.src.u3.ip = inet->rcv_saddr; + tuple.src.u.tcp.port = inet->sport; + tuple.dst.u3.ip = inet->daddr; + tuple.dst.u.tcp.port = inet->dport; + tuple.src.l3num = PF_INET; + tuple.dst.protonum = IPPROTO_TCP; + + /* We only do TCP at the moment: is there a better way? */ + if (strcmp(sk->sk_prot->name, "TCP")) { + DEBUGP("SO_ORIGINAL_DST: Not a TCP socket\n"); + return -ENOPROTOOPT; + } + + if ((unsigned int) *len < sizeof(struct sockaddr_in)) { + DEBUGP("SO_ORIGINAL_DST: len %u not %u\n", + *len, sizeof(struct sockaddr_in)); + return -EINVAL; + } + + h = nf_conntrack_find_get(&tuple, NULL); + if (h) { + struct sockaddr_in sin; + struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h); + + sin.sin_family = AF_INET; + sin.sin_port = ct->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.dst.u.tcp.port; + sin.sin_addr.s_addr = ct->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.dst.u3.ip; + + DEBUGP("SO_ORIGINAL_DST: %u.%u.%u.%u %u\n", + NIPQUAD(sin.sin_addr.s_addr), ntohs(sin.sin_port)); + nf_ct_put(ct); + if (copy_to_user(user, &sin, sizeof(sin)) != 0) + return -EFAULT; + else + return 0; + } + DEBUGP("SO_ORIGINAL_DST: Can't find %u.%u.%u.%u/%u-%u.%u.%u.%u/%u.\n", + NIPQUAD(tuple.src.u3.ip), ntohs(tuple.src.u.tcp.port), + NIPQUAD(tuple.dst.u3.ip), ntohs(tuple.dst.u.tcp.port)); + return -ENOENT; +} + +static struct nf_sockopt_ops so_getorigdst = { + .pf = PF_INET, + .get_optmin = SO_ORIGINAL_DST, + .get_optmax = SO_ORIGINAL_DST+1, + .get = &getorigdst, +}; + +struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 = { + .l3proto = PF_INET, + .name = "ipv4", + .pkt_to_tuple = ipv4_pkt_to_tuple, + .invert_tuple = ipv4_invert_tuple, + .print_tuple = ipv4_print_tuple, + .print_conntrack = ipv4_print_conntrack, + .prepare = ipv4_prepare, + .get_features = ipv4_get_features, + .me = THIS_MODULE, +}; + +extern struct nf_conntrack_protocol nf_conntrack_protocol_tcp4; +extern struct nf_conntrack_protocol nf_conntrack_protocol_udp4; +extern struct nf_conntrack_protocol nf_conntrack_protocol_icmp; +static int init_or_cleanup(int init) +{ + int ret = 0; + + if (!init) goto cleanup; + + ret = nf_register_sockopt(&so_getorigdst); + if (ret < 0) { + printk(KERN_ERR "Unable to register netfilter socket option\n"); + goto cleanup_nothing; + } + + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_tcp4); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register tcp.\n"); + goto cleanup_sockopt; + } + + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_udp4); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register udp.\n"); + goto cleanup_tcp; + } + + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_icmp); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register icmp.\n"); + goto cleanup_udp; + } + + ret = nf_conntrack_l3proto_register(&nf_conntrack_l3proto_ipv4); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register ipv4\n"); + goto cleanup_icmp; + } + + ret = nf_register_hook(&ipv4_conntrack_defrag_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register pre-routing defrag hook.\n"); + goto cleanup_ipv4; + } + ret = nf_register_hook(&ipv4_conntrack_defrag_local_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register local_out defrag hook.\n"); + goto cleanup_defragops; + } + + ret = nf_register_hook(&ipv4_conntrack_in_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register pre-routing hook.\n"); + goto cleanup_defraglocalops; + } + + ret = nf_register_hook(&ipv4_conntrack_local_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register local out hook.\n"); + goto cleanup_inops; + } + + ret = nf_register_hook(&ipv4_conntrack_helper_in_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register local helper hook.\n"); + goto cleanup_inandlocalops; + } + + ret = nf_register_hook(&ipv4_conntrack_helper_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register postrouting helper hook.\n"); + goto cleanup_helperinops; + } + + ret = nf_register_hook(&ipv4_conntrack_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register post-routing hook.\n"); + goto cleanup_helperoutops; + } + + ret = nf_register_hook(&ipv4_conntrack_local_in_ops); + if (ret < 0) { + printk("nf_conntrack_ipv4: can't register local in hook.\n"); + goto cleanup_inoutandlocalops; + } + +#ifdef CONFIG_SYSCTL + nf_ct_ipv4_sysctl_header = register_sysctl_table(nf_ct_net_table, 0); + if (nf_ct_ipv4_sysctl_header == NULL) { + printk("nf_conntrack: can't register to sysctl.\n"); + ret = -ENOMEM; + goto cleanup_localinops; + } +#endif + + /* For use by REJECT target */ + ip_ct_attach = __nf_conntrack_attach; + + return ret; + + cleanup: + synchronize_net(); + ip_ct_attach = NULL; +#ifdef CONFIG_SYSCTL + unregister_sysctl_table(nf_ct_ipv4_sysctl_header); + cleanup_localinops: +#endif + nf_unregister_hook(&ipv4_conntrack_local_in_ops); + cleanup_inoutandlocalops: + nf_unregister_hook(&ipv4_conntrack_out_ops); + cleanup_helperoutops: + nf_unregister_hook(&ipv4_conntrack_helper_out_ops); + cleanup_helperinops: + nf_unregister_hook(&ipv4_conntrack_helper_in_ops); + cleanup_inandlocalops: + nf_unregister_hook(&ipv4_conntrack_local_out_ops); + cleanup_inops: + nf_unregister_hook(&ipv4_conntrack_in_ops); + cleanup_defraglocalops: + nf_unregister_hook(&ipv4_conntrack_defrag_local_out_ops); + cleanup_defragops: + nf_unregister_hook(&ipv4_conntrack_defrag_ops); + cleanup_ipv4: + nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4); + cleanup_icmp: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_icmp); + cleanup_udp: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_udp4); + cleanup_tcp: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_tcp4); + cleanup_sockopt: + nf_unregister_sockopt(&so_getorigdst); + cleanup_nothing: + return ret; +} + +MODULE_LICENSE("GPL"); + +static int __init init(void) +{ + need_nf_conntrack(); + return init_or_cleanup(1); +} + +static void __exit fini(void) +{ + init_or_cleanup(0); +} + +module_init(init); +module_exit(fini); + +void need_ip_conntrack(void) +{ +} + +EXPORT_SYMBOL(need_ip_conntrack); +EXPORT_SYMBOL(nf_ct_ipv4_gather_frags); diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c new file mode 100644 index 000000000000..7ddb5c08f7b8 --- /dev/null +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -0,0 +1,301 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI + * - enable working with Layer 3 protocol independent connection tracking. + * + * Derived from net/ipv4/netfilter/ip_conntrack_proto_icmp.c + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +unsigned long nf_ct_icmp_timeout = 30*HZ; + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +static int icmp_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + struct icmphdr _hdr, *hp; + + hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); + if (hp == NULL) + return 0; + + tuple->dst.u.icmp.type = hp->type; + tuple->src.u.icmp.id = hp->un.echo.id; + tuple->dst.u.icmp.code = hp->code; + + return 1; +} + +static int icmp_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + /* Add 1; spaces filled with 0. */ + static u_int8_t invmap[] + = { [ICMP_ECHO] = ICMP_ECHOREPLY + 1, + [ICMP_ECHOREPLY] = ICMP_ECHO + 1, + [ICMP_TIMESTAMP] = ICMP_TIMESTAMPREPLY + 1, + [ICMP_TIMESTAMPREPLY] = ICMP_TIMESTAMP + 1, + [ICMP_INFO_REQUEST] = ICMP_INFO_REPLY + 1, + [ICMP_INFO_REPLY] = ICMP_INFO_REQUEST + 1, + [ICMP_ADDRESS] = ICMP_ADDRESSREPLY + 1, + [ICMP_ADDRESSREPLY] = ICMP_ADDRESS + 1}; + + if (orig->dst.u.icmp.type >= sizeof(invmap) + || !invmap[orig->dst.u.icmp.type]) + return 0; + + tuple->src.u.icmp.id = orig->src.u.icmp.id; + tuple->dst.u.icmp.type = invmap[orig->dst.u.icmp.type] - 1; + tuple->dst.u.icmp.code = orig->dst.u.icmp.code; + return 1; +} + +/* Print out the per-protocol part of the tuple. */ +static int icmp_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "type=%u code=%u id=%u ", + tuple->dst.u.icmp.type, + tuple->dst.u.icmp.code, + ntohs(tuple->src.u.icmp.id)); +} + +/* Print out the private part of the conntrack. */ +static int icmp_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + return 0; +} + +/* Returns verdict for packet, or -1 for invalid. */ +static int icmp_packet(struct nf_conn *ct, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + /* Try to delete connection immediately after all replies: + won't actually vanish as we still have skb, and del_timer + means this will only run once even if count hits zero twice + (theoretically possible with SMP) */ + if (CTINFO2DIR(ctinfo) == IP_CT_DIR_REPLY) { + if (atomic_dec_and_test(&ct->proto.icmp.count) + && del_timer(&ct->timeout)) + ct->timeout.function((unsigned long)ct); + } else { + atomic_inc(&ct->proto.icmp.count); + nf_conntrack_event_cache(IPCT_PROTOINFO_VOLATILE, skb); + nf_ct_refresh_acct(ct, ctinfo, skb, nf_ct_icmp_timeout); + } + + return NF_ACCEPT; +} + +/* 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. */ + DEBUGP("icmp: can't create new conn with type %u\n", + conntrack->tuplehash[0].tuple.dst.u.icmp.type); + NF_CT_DUMP_TUPLE(&conntrack->tuplehash[0].tuple); + return 0; + } + atomic_set(&conntrack->proto.icmp.count, 0); + return 1; +} + +extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4; +/* Returns conntrack if it dealt with ICMP, and filled in skb fields */ +static int +icmp_error_message(struct sk_buff *skb, + enum ip_conntrack_info *ctinfo, + unsigned int hooknum) +{ + struct nf_conntrack_tuple innertuple, origtuple; + struct { + struct icmphdr icmp; + struct iphdr ip; + } _in, *inside; + struct nf_conntrack_protocol *innerproto; + struct nf_conntrack_tuple_hash *h; + int dataoff; + + NF_CT_ASSERT(skb->nfct == NULL); + + /* Not enough header? */ + inside = skb_header_pointer(skb, skb->nh.iph->ihl*4, sizeof(_in), &_in); + if (inside == NULL) + return -NF_ACCEPT; + + /* Ignore ICMP's containing fragments (shouldn't happen) */ + if (inside->ip.frag_off & htons(IP_OFFSET)) { + DEBUGP("icmp_error_message: fragment of proto %u\n", + inside->ip.protocol); + return -NF_ACCEPT; + } + + innerproto = nf_ct_find_proto(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, + inside->ip.protocol, &origtuple, + &nf_conntrack_l3proto_ipv4, innerproto)) { + DEBUGP("icmp_error_message: ! get_tuple p=%u", + inside->ip.protocol); + return -NF_ACCEPT; + } + + /* Ordinarily, we'd expect the inverted tupleproto, but it's + been preserved inside the ICMP. */ + if (!nf_ct_invert_tuple(&innertuple, &origtuple, + &nf_conntrack_l3proto_ipv4, innerproto)) { + DEBUGP("icmp_error_message: no match\n"); + return -NF_ACCEPT; + } + + *ctinfo = IP_CT_RELATED; + + h = nf_conntrack_find_get(&innertuple, NULL); + if (!h) { + /* Locally generated ICMPs will match inverted if they + haven't been SNAT'ed yet */ + /* FIXME: NAT code has to handle half-done double NAT --RR */ + if (hooknum == NF_IP_LOCAL_OUT) + h = nf_conntrack_find_get(&origtuple, NULL); + + if (!h) { + DEBUGP("icmp_error_message: no match\n"); + return -NF_ACCEPT; + } + + /* Reverse direction from that found */ + if (NF_CT_DIRECTION(h) == IP_CT_DIR_REPLY) + *ctinfo += IP_CT_IS_REPLY; + } else { + if (NF_CT_DIRECTION(h) == IP_CT_DIR_REPLY) + *ctinfo += IP_CT_IS_REPLY; + } + + /* Update skb to refer to this connection */ + skb->nfct = &nf_ct_tuplehash_to_ctrack(h)->ct_general; + skb->nfctinfo = *ctinfo; + return -NF_ACCEPT; +} + +/* Small and modified version of icmp_rcv */ +static int +icmp_error(struct sk_buff *skb, unsigned int dataoff, + enum ip_conntrack_info *ctinfo, int pf, unsigned int hooknum) +{ + struct icmphdr _ih, *icmph; + + /* Not enough header? */ + icmph = skb_header_pointer(skb, skb->nh.iph->ihl*4, sizeof(_ih), &_ih); + if (icmph == NULL) { + if (LOG_INVALID(IPPROTO_ICMP)) + nf_log_packet(PF_INET, 0, skb, NULL, NULL, NULL, + "nf_ct_icmp: short packet "); + return -NF_ACCEPT; + } + + /* See ip_conntrack_proto_tcp.c */ + if (hooknum != NF_IP_PRE_ROUTING) + goto checksum_skipped; + + switch (skb->ip_summed) { + case CHECKSUM_HW: + if (!(u16)csum_fold(skb->csum)) + break; + if (LOG_INVALID(IPPROTO_ICMP)) + nf_log_packet(PF_INET, 0, skb, NULL, NULL, NULL, + "nf_ct_icmp: bad HW ICMP checksum "); + return -NF_ACCEPT; + case CHECKSUM_NONE: + if ((u16)csum_fold(skb_checksum(skb, 0, skb->len, 0))) { + if (LOG_INVALID(IPPROTO_ICMP)) + nf_log_packet(PF_INET, 0, skb, NULL, NULL, + NULL, + "nf_ct_icmp: bad ICMP checksum "); + return -NF_ACCEPT; + } + default: + break; + } + +checksum_skipped: + /* + * 18 is the highest 'known' ICMP type. Anything else is a mystery + * + * RFC 1122: 3.2.2 Unknown ICMP messages types MUST be silently + * discarded. + */ + if (icmph->type > NR_ICMP_TYPES) { + if (LOG_INVALID(IPPROTO_ICMP)) + nf_log_packet(PF_INET, 0, skb, NULL, NULL, NULL, + "nf_ct_icmp: invalid ICMP type "); + return -NF_ACCEPT; + } + + /* Need to track icmp error message? */ + if (icmph->type != ICMP_DEST_UNREACH + && icmph->type != ICMP_SOURCE_QUENCH + && icmph->type != ICMP_TIME_EXCEEDED + && icmph->type != ICMP_PARAMETERPROB + && icmph->type != ICMP_REDIRECT) + return NF_ACCEPT; + + return icmp_error_message(skb, ctinfo, hooknum); +} + +struct nf_conntrack_protocol nf_conntrack_protocol_icmp = +{ + .list = { NULL, NULL }, + .l3proto = PF_INET, + .proto = IPPROTO_ICMP, + .name = "icmp", + .pkt_to_tuple = icmp_pkt_to_tuple, + .invert_tuple = icmp_invert_tuple, + .print_tuple = icmp_print_tuple, + .print_conntrack = icmp_print_conntrack, + .packet = icmp_packet, + .new = icmp_new, + .error = icmp_error, + .destroy = NULL, + .me = NULL +}; + +EXPORT_SYMBOL(nf_conntrack_protocol_icmp); diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index 6e3480426939..a6026d2787d2 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -176,6 +176,11 @@ resubmit: if (ipprot->flags & INET6_PROTO_FINAL) { struct ipv6hdr *hdr; + /* Free reference early: we don't need it any more, + and it may hold ip_conntrack module loaded + indefinitely. */ + nf_reset(skb); + skb_postpull_rcsum(skb, skb->nh.raw, skb->h.raw - skb->nh.raw); hdr = skb->nh.ipv6h; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index dbd9767b32e4..c1fa693511a1 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -441,9 +441,15 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from) #ifdef CONFIG_NETFILTER to->nfmark = from->nfmark; /* Connection association is same as pre-frag packet */ + nf_conntrack_put(to->nfct); to->nfct = from->nfct; nf_conntrack_get(to->nfct); to->nfctinfo = from->nfctinfo; +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + nf_conntrack_put_reasm(to->nfct_reasm); + to->nfct_reasm = from->nfct_reasm; + nf_conntrack_get_reasm(to->nfct_reasm); +#endif #ifdef CONFIG_BRIDGE_NETFILTER nf_bridge_put(to->nf_bridge); to->nf_bridge = from->nf_bridge; diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig index bb7ccfe33f23..971ba60bf6e9 100644 --- a/net/ipv6/netfilter/Kconfig +++ b/net/ipv6/netfilter/Kconfig @@ -278,5 +278,19 @@ config IP6_NF_RAW If you want to compile it as a module, say M here and read . If unsure, say `N'. +config NF_CONNTRACK_IPV6 + tristate "IPv6 support for new connection tracking (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + ---help--- + Connection tracking keeps a record of what packets have passed + through your machine, in order to figure out how they are related + into connections. + + This is IPv6 support on Layer 3 independent connection tracking. + Layer 3 independent connection tracking is experimental scheme + which generalize ip_conntrack to support other layer 3 protocols. + + To compile it as a module, choose M here. If unsure, say N. + endmenu diff --git a/net/ipv6/netfilter/Makefile b/net/ipv6/netfilter/Makefile index 2b2c370e8b1c..9ab5b2ca1f59 100644 --- a/net/ipv6/netfilter/Makefile +++ b/net/ipv6/netfilter/Makefile @@ -27,3 +27,9 @@ obj-$(CONFIG_IP6_NF_TARGET_LOG) += ip6t_LOG.o obj-$(CONFIG_IP6_NF_RAW) += ip6table_raw.o obj-$(CONFIG_IP6_NF_MATCH_HL) += ip6t_hl.o obj-$(CONFIG_IP6_NF_TARGET_REJECT) += ip6t_REJECT.o + +# objects for l3 independent conntrack +nf_conntrack_ipv6-objs := nf_conntrack_l3proto_ipv6.o nf_conntrack_proto_icmpv6.o nf_conntrack_reasm.o + +# l3 independent conntrack +obj-$(CONFIG_NF_CONNTRACK_IPV6) += nf_conntrack_ipv6.o diff --git a/net/ipv6/netfilter/ip6t_MARK.c b/net/ipv6/netfilter/ip6t_MARK.c index 0c7584f92172..eab8fb864ee0 100644 --- a/net/ipv6/netfilter/ip6t_MARK.c +++ b/net/ipv6/netfilter/ip6t_MARK.c @@ -56,9 +56,9 @@ checkentry(const char *tablename, return 1; } -static struct ip6t_target ip6t_mark_reg = { - .name = "MARK", - .target = target, +static struct ip6t_target ip6t_mark_reg = { + .name = "MARK", + .target = target, .checkentry = checkentry, .me = THIS_MODULE }; diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c new file mode 100644 index 000000000000..e2c90b3a8074 --- /dev/null +++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c @@ -0,0 +1,556 @@ +/* + * Copyright (C)2004 USAGI/WIDE Project + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Author: + * Yasuyuki Kozakai @USAGI + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI + * - support Layer 3 protocol independent connection tracking. + * Based on the original ip_conntrack code which had the following + * copyright information: + * (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team + * + * 23 Mar 2004: Yasuyuki Kozakai @USAGI + * - add get_features() to support various size of conntrack + * structures. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +DECLARE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); + +static int ipv6_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, + struct nf_conntrack_tuple *tuple) +{ + u_int32_t _addrs[8], *ap; + + ap = skb_header_pointer(skb, nhoff + offsetof(struct ipv6hdr, saddr), + sizeof(_addrs), _addrs); + if (ap == NULL) + return 0; + + memcpy(tuple->src.u3.ip6, ap, sizeof(tuple->src.u3.ip6)); + memcpy(tuple->dst.u3.ip6, ap + 4, sizeof(tuple->dst.u3.ip6)); + + return 1; +} + +static int ipv6_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + memcpy(tuple->src.u3.ip6, orig->dst.u3.ip6, sizeof(tuple->src.u3.ip6)); + memcpy(tuple->dst.u3.ip6, orig->src.u3.ip6, sizeof(tuple->dst.u3.ip6)); + + return 1; +} + +static int ipv6_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "src=%x:%x:%x:%x:%x:%x:%x:%x dst=%x:%x:%x:%x:%x:%x:%x:%x ", + NIP6(*((struct in6_addr *)tuple->src.u3.ip6)), + NIP6(*((struct in6_addr *)tuple->dst.u3.ip6))); +} + +static int ipv6_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + return 0; +} + +/* + * Based on ipv6_skip_exthdr() in net/ipv6/exthdr.c + * + * This function parses (probably truncated) exthdr set "hdr" + * of length "len". "nexthdrp" initially points to some place, + * where type of the first header can be found. + * + * It skips all well-known exthdrs, and returns pointer to the start + * of unparsable area i.e. the first header with unknown type. + * if success, *nexthdr is updated by type/protocol of this header. + * + * NOTES: - it may return pointer pointing beyond end of packet, + * if the last recognized header is truncated in the middle. + * - if packet is truncated, so that all parsed headers are skipped, + * it returns -1. + * - if packet is fragmented, return pointer of the fragment header. + * - ESP is unparsable for now and considered like + * normal payload protocol. + * - Note also special handling of AUTH header. Thanks to IPsec wizards. + */ + +int nf_ct_ipv6_skip_exthdr(struct sk_buff *skb, int start, u8 *nexthdrp, + int len) +{ + u8 nexthdr = *nexthdrp; + + while (ipv6_ext_hdr(nexthdr)) { + struct ipv6_opt_hdr hdr; + int hdrlen; + + if (len < (int)sizeof(struct ipv6_opt_hdr)) + return -1; + if (nexthdr == NEXTHDR_NONE) + break; + if (nexthdr == NEXTHDR_FRAGMENT) + break; + if (skb_copy_bits(skb, start, &hdr, sizeof(hdr))) + BUG(); + if (nexthdr == NEXTHDR_AUTH) + hdrlen = (hdr.hdrlen+2)<<2; + else + hdrlen = ipv6_optlen(&hdr); + + nexthdr = hdr.nexthdr; + len -= hdrlen; + start += hdrlen; + } + + *nexthdrp = nexthdr; + return start; +} + +static int +ipv6_prepare(struct sk_buff **pskb, unsigned int hooknum, unsigned int *dataoff, + u_int8_t *protonum) +{ + unsigned int extoff; + unsigned char pnum; + int protoff; + + extoff = (u8*)((*pskb)->nh.ipv6h + 1) - (*pskb)->data; + pnum = (*pskb)->nh.ipv6h->nexthdr; + + protoff = nf_ct_ipv6_skip_exthdr(*pskb, extoff, &pnum, + (*pskb)->len - extoff); + + /* + * (protoff == (*pskb)->len) mean that the packet doesn't have no data + * except of IPv6 & ext headers. but it's tracked anyway. - YK + */ + if ((protoff < 0) || (protoff > (*pskb)->len)) { + DEBUGP("ip6_conntrack_core: can't find proto in pkt\n"); + NF_CT_STAT_INC(error); + NF_CT_STAT_INC(invalid); + return -NF_ACCEPT; + } + + *dataoff = protoff; + *protonum = pnum; + return NF_ACCEPT; +} + +static u_int32_t ipv6_get_features(const struct nf_conntrack_tuple *tuple) +{ + return NF_CT_F_BASIC; +} + +static unsigned int ipv6_confirm(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + + /* This is where we call the helper: as the packet goes out. */ + ct = nf_ct_get(*pskb, &ctinfo); + if (ct && ct->helper) { + unsigned int ret, protoff; + unsigned int extoff = (u8*)((*pskb)->nh.ipv6h + 1) + - (*pskb)->data; + unsigned char pnum = (*pskb)->nh.ipv6h->nexthdr; + + protoff = nf_ct_ipv6_skip_exthdr(*pskb, extoff, &pnum, + (*pskb)->len - extoff); + if (protoff < 0 || protoff > (*pskb)->len || + pnum == NEXTHDR_FRAGMENT) { + DEBUGP("proto header not found\n"); + return NF_ACCEPT; + } + + ret = ct->helper->help(pskb, protoff, ct, ctinfo); + if (ret != NF_ACCEPT) + return ret; + } + + /* We've seen it coming out the other side: confirm it */ + + return nf_conntrack_confirm(pskb); +} + +extern struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb); +extern void nf_ct_frag6_output(unsigned int hooknum, struct sk_buff *skb, + struct net_device *in, + struct net_device *out, + int (*okfn)(struct sk_buff *)); +static unsigned int ipv6_defrag(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct sk_buff *reasm; + + /* Previously seen (loopback)? */ + if ((*pskb)->nfct) + return NF_ACCEPT; + + reasm = nf_ct_frag6_gather(*pskb); + + /* queued */ + if (reasm == NULL) + return NF_STOLEN; + + /* error occured or not fragmented */ + if (reasm == *pskb) + return NF_ACCEPT; + + nf_ct_frag6_output(hooknum, reasm, (struct net_device *)in, + (struct net_device *)out, okfn); + + return NF_STOLEN; +} + +static unsigned int ipv6_conntrack_in(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct sk_buff *reasm = (*pskb)->nfct_reasm; + + /* This packet is fragmented and has reassembled packet. */ + if (reasm) { + /* Reassembled packet isn't parsed yet ? */ + if (!reasm->nfct) { + unsigned int ret; + + ret = nf_conntrack_in(PF_INET6, hooknum, &reasm); + if (ret != NF_ACCEPT) + return ret; + } + nf_conntrack_get(reasm->nfct); + (*pskb)->nfct = reasm->nfct; + return NF_ACCEPT; + } + + return nf_conntrack_in(PF_INET6, hooknum, pskb); +} + +static unsigned int ipv6_conntrack_local(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + /* root is playing with raw sockets. */ + if ((*pskb)->len < sizeof(struct ipv6hdr)) { + if (net_ratelimit()) + printk("ipv6_conntrack_local: packet too short\n"); + return NF_ACCEPT; + } + return ipv6_conntrack_in(hooknum, pskb, in, out, okfn); +} + +/* Connection tracking may drop packets, but never alters them, so + make it the first hook. */ +static struct nf_hook_ops ipv6_conntrack_defrag_ops = { + .hook = ipv6_defrag, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_PRE_ROUTING, + .priority = NF_IP6_PRI_CONNTRACK_DEFRAG, +}; + +static struct nf_hook_ops ipv6_conntrack_in_ops = { + .hook = ipv6_conntrack_in, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_PRE_ROUTING, + .priority = NF_IP6_PRI_CONNTRACK, +}; + +static struct nf_hook_ops ipv6_conntrack_local_out_ops = { + .hook = ipv6_conntrack_local, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_LOCAL_OUT, + .priority = NF_IP6_PRI_CONNTRACK, +}; + +static struct nf_hook_ops ipv6_conntrack_defrag_local_out_ops = { + .hook = ipv6_defrag, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_LOCAL_OUT, + .priority = NF_IP6_PRI_CONNTRACK_DEFRAG, +}; + +/* Refragmenter; last chance. */ +static struct nf_hook_ops ipv6_conntrack_out_ops = { + .hook = ipv6_confirm, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_POST_ROUTING, + .priority = NF_IP6_PRI_LAST, +}; + +static struct nf_hook_ops ipv6_conntrack_local_in_ops = { + .hook = ipv6_confirm, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_LOCAL_IN, + .priority = NF_IP6_PRI_LAST-1, +}; + +#ifdef CONFIG_SYSCTL + +/* From nf_conntrack_proto_icmpv6.c */ +extern unsigned long nf_ct_icmpv6_timeout; + +/* From nf_conntrack_frag6.c */ +extern unsigned long nf_ct_frag6_timeout; +extern unsigned long nf_ct_frag6_low_thresh; +extern unsigned long nf_ct_frag6_high_thresh; + +static struct ctl_table_header *nf_ct_ipv6_sysctl_header; + +static ctl_table nf_ct_sysctl_table[] = { + { + .ctl_name = NET_NF_CONNTRACK_ICMPV6_TIMEOUT, + .procname = "nf_conntrack_icmpv6_timeout", + .data = &nf_ct_icmpv6_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_FRAG6_TIMEOUT, + .procname = "nf_conntrack_frag6_timeout", + .data = &nf_ct_frag6_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_FRAG6_LOW_THRESH, + .procname = "nf_conntrack_frag6_low_thresh", + .data = &nf_ct_frag6_low_thresh, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_FRAG6_HIGH_THRESH, + .procname = "nf_conntrack_frag6_high_thresh", + .data = &nf_ct_frag6_high_thresh, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_netfilter_table[] = { + { + .ctl_name = NET_NETFILTER, + .procname = "netfilter", + .mode = 0555, + .child = nf_ct_sysctl_table, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_net_table[] = { + { + .ctl_name = CTL_NET, + .procname = "net", + .mode = 0555, + .child = nf_ct_netfilter_table, + }, + { .ctl_name = 0 } +}; +#endif + +struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6 = { + .l3proto = PF_INET6, + .name = "ipv6", + .pkt_to_tuple = ipv6_pkt_to_tuple, + .invert_tuple = ipv6_invert_tuple, + .print_tuple = ipv6_print_tuple, + .print_conntrack = ipv6_print_conntrack, + .prepare = ipv6_prepare, + .get_features = ipv6_get_features, + .me = THIS_MODULE, +}; + +extern struct nf_conntrack_protocol nf_conntrack_protocol_tcp6; +extern struct nf_conntrack_protocol nf_conntrack_protocol_udp6; +extern struct nf_conntrack_protocol nf_conntrack_protocol_icmpv6; +extern int nf_ct_frag6_init(void); +extern void nf_ct_frag6_cleanup(void); +static int init_or_cleanup(int init) +{ + int ret = 0; + + if (!init) goto cleanup; + + ret = nf_ct_frag6_init(); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't initialize frag6.\n"); + goto cleanup_nothing; + } + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_tcp6); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register tcp.\n"); + goto cleanup_frag6; + } + + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_udp6); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register udp.\n"); + goto cleanup_tcp; + } + + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_icmpv6); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register icmpv6.\n"); + goto cleanup_udp; + } + + ret = nf_conntrack_l3proto_register(&nf_conntrack_l3proto_ipv6); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register ipv6\n"); + goto cleanup_icmpv6; + } + + ret = nf_register_hook(&ipv6_conntrack_defrag_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register pre-routing defrag " + "hook.\n"); + goto cleanup_ipv6; + } + + ret = nf_register_hook(&ipv6_conntrack_defrag_local_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register local_out defrag " + "hook.\n"); + goto cleanup_defragops; + } + + ret = nf_register_hook(&ipv6_conntrack_in_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register pre-routing hook.\n"); + goto cleanup_defraglocalops; + } + + ret = nf_register_hook(&ipv6_conntrack_local_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register local out hook.\n"); + goto cleanup_inops; + } + + ret = nf_register_hook(&ipv6_conntrack_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register post-routing hook.\n"); + goto cleanup_inandlocalops; + } + + ret = nf_register_hook(&ipv6_conntrack_local_in_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register local in hook.\n"); + goto cleanup_inoutandlocalops; + } + +#ifdef CONFIG_SYSCTL + nf_ct_ipv6_sysctl_header = register_sysctl_table(nf_ct_net_table, 0); + if (nf_ct_ipv6_sysctl_header == NULL) { + printk("nf_conntrack: can't register to sysctl.\n"); + ret = -ENOMEM; + goto cleanup_localinops; + } +#endif + return ret; + + cleanup: + synchronize_net(); +#ifdef CONFIG_SYSCTL + unregister_sysctl_table(nf_ct_ipv6_sysctl_header); + cleanup_localinops: +#endif + nf_unregister_hook(&ipv6_conntrack_local_in_ops); + cleanup_inoutandlocalops: + nf_unregister_hook(&ipv6_conntrack_out_ops); + cleanup_inandlocalops: + nf_unregister_hook(&ipv6_conntrack_local_out_ops); + cleanup_inops: + nf_unregister_hook(&ipv6_conntrack_in_ops); + cleanup_defraglocalops: + nf_unregister_hook(&ipv6_conntrack_defrag_local_out_ops); + cleanup_defragops: + nf_unregister_hook(&ipv6_conntrack_defrag_ops); + cleanup_ipv6: + nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv6); + cleanup_icmpv6: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_icmpv6); + cleanup_udp: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_udp6); + cleanup_tcp: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_tcp6); + cleanup_frag6: + nf_ct_frag6_cleanup(); + cleanup_nothing: + return ret; +} + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Yasuyuki KOZAKAI @USAGI "); + +static int __init init(void) +{ + need_nf_conntrack(); + return init_or_cleanup(1); +} + +static void __exit fini(void) +{ + init_or_cleanup(0); +} + +module_init(init); +module_exit(fini); + +void need_ip6_conntrack(void) +{ +} + +EXPORT_SYMBOL(need_ip6_conntrack); diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c new file mode 100644 index 000000000000..c0f1da5497a9 --- /dev/null +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -0,0 +1,272 @@ +/* + * Copyright (C)2003,2004 USAGI/WIDE Project + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Author: + * Yasuyuki Kozakai @USAGI + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI + * - ICMPv6 tracking support. Derived from the original ip_conntrack code + * net/ipv4/netfilter/ip_conntrack_proto_icmp.c which had the following + * copyright information: + * (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +unsigned long nf_ct_icmpv6_timeout = 30*HZ; + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +static int icmpv6_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + struct icmp6hdr _hdr, *hp; + + hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); + if (hp == NULL) + return 0; + tuple->dst.u.icmp.type = hp->icmp6_type; + tuple->src.u.icmp.id = hp->icmp6_identifier; + tuple->dst.u.icmp.code = hp->icmp6_code; + + return 1; +} + +static int icmpv6_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + /* Add 1; spaces filled with 0. */ + static u_int8_t invmap[] = { + [ICMPV6_ECHO_REQUEST - 128] = ICMPV6_ECHO_REPLY + 1, + [ICMPV6_ECHO_REPLY - 128] = ICMPV6_ECHO_REQUEST + 1, + [ICMPV6_NI_QUERY - 128] = ICMPV6_NI_QUERY + 1, + [ICMPV6_NI_REPLY - 128] = ICMPV6_NI_REPLY +1 + }; + + __u8 type = orig->dst.u.icmp.type - 128; + if (type >= sizeof(invmap) || !invmap[type]) + return 0; + + tuple->src.u.icmp.id = orig->src.u.icmp.id; + tuple->dst.u.icmp.type = invmap[type] - 1; + tuple->dst.u.icmp.code = orig->dst.u.icmp.code; + return 1; +} + +/* Print out the per-protocol part of the tuple. */ +static int icmpv6_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "type=%u code=%u id=%u ", + tuple->dst.u.icmp.type, + tuple->dst.u.icmp.code, + ntohs(tuple->src.u.icmp.id)); +} + +/* Print out the private part of the conntrack. */ +static int icmpv6_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + return 0; +} + +/* Returns verdict for packet, or -1 for invalid. */ +static int icmpv6_packet(struct nf_conn *ct, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + /* Try to delete connection immediately after all replies: + won't actually vanish as we still have skb, and del_timer + means this will only run once even if count hits zero twice + (theoretically possible with SMP) */ + if (CTINFO2DIR(ctinfo) == IP_CT_DIR_REPLY) { + if (atomic_dec_and_test(&ct->proto.icmp.count) + && del_timer(&ct->timeout)) + ct->timeout.function((unsigned long)ct); + } else { + atomic_inc(&ct->proto.icmp.count); + nf_conntrack_event_cache(IPCT_PROTOINFO_VOLATILE, skb); + nf_ct_refresh_acct(ct, ctinfo, skb, nf_ct_icmpv6_timeout); + } + + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int icmpv6_new(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff) +{ + static u_int8_t valid_new[] = { + [ICMPV6_ECHO_REQUEST - 128] = 1, + [ICMPV6_NI_QUERY - 128] = 1 + }; + + if (conntrack->tuplehash[0].tuple.dst.u.icmp.type - 128 >= sizeof(valid_new) + || !valid_new[conntrack->tuplehash[0].tuple.dst.u.icmp.type - 128]) { + /* Can't create a new ICMPv6 `conn' with this. */ + DEBUGP("icmp: can't create new conn with type %u\n", + conntrack->tuplehash[0].tuple.dst.u.icmp.type); + NF_CT_DUMP_TUPLE(&conntrack->tuplehash[0].tuple); + return 0; + } + atomic_set(&conntrack->proto.icmp.count, 0); + return 1; +} + +extern int +nf_ct_ipv6_skip_exthdr(struct sk_buff *skb, int start, u8 *nexthdrp, int len); +extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6; +static int +icmpv6_error_message(struct sk_buff *skb, + unsigned int icmp6off, + enum ip_conntrack_info *ctinfo, + unsigned int hooknum) +{ + struct nf_conntrack_tuple intuple, origtuple; + struct nf_conntrack_tuple_hash *h; + struct icmp6hdr _hdr, *hp; + unsigned int inip6off; + struct nf_conntrack_protocol *inproto; + u_int8_t inprotonum; + unsigned int inprotoff; + + NF_CT_ASSERT(skb->nfct == NULL); + + hp = skb_header_pointer(skb, icmp6off, sizeof(_hdr), &_hdr); + if (hp == NULL) { + DEBUGP("icmpv6_error: Can't get ICMPv6 hdr.\n"); + return -NF_ACCEPT; + } + + inip6off = icmp6off + sizeof(_hdr); + if (skb_copy_bits(skb, inip6off+offsetof(struct ipv6hdr, nexthdr), + &inprotonum, sizeof(inprotonum)) != 0) { + DEBUGP("icmpv6_error: Can't get nexthdr in inner IPv6 header.\n"); + return -NF_ACCEPT; + } + inprotoff = nf_ct_ipv6_skip_exthdr(skb, + inip6off + sizeof(struct ipv6hdr), + &inprotonum, + skb->len - inip6off + - sizeof(struct ipv6hdr)); + + if ((inprotoff < 0) || (inprotoff > skb->len) || + (inprotonum == NEXTHDR_FRAGMENT)) { + DEBUGP("icmpv6_error: Can't get protocol header in ICMPv6 payload.\n"); + return -NF_ACCEPT; + } + + inproto = nf_ct_find_proto(PF_INET6, inprotonum); + + /* Are they talking about one of our connections? */ + if (!nf_ct_get_tuple(skb, inip6off, inprotoff, PF_INET6, inprotonum, + &origtuple, &nf_conntrack_l3proto_ipv6, inproto)) { + DEBUGP("icmpv6_error: Can't get tuple\n"); + return -NF_ACCEPT; + } + + /* Ordinarily, we'd expect the inverted tupleproto, but it's + been preserved inside the ICMP. */ + if (!nf_ct_invert_tuple(&intuple, &origtuple, + &nf_conntrack_l3proto_ipv6, inproto)) { + DEBUGP("icmpv6_error: Can't invert tuple\n"); + return -NF_ACCEPT; + } + + *ctinfo = IP_CT_RELATED; + + h = nf_conntrack_find_get(&intuple, NULL); + if (!h) { + DEBUGP("icmpv6_error: no match\n"); + return -NF_ACCEPT; + } else { + if (NF_CT_DIRECTION(h) == IP_CT_DIR_REPLY) + *ctinfo += IP_CT_IS_REPLY; + } + + /* Update skb to refer to this connection */ + skb->nfct = &nf_ct_tuplehash_to_ctrack(h)->ct_general; + skb->nfctinfo = *ctinfo; + return -NF_ACCEPT; +} + +static int +icmpv6_error(struct sk_buff *skb, unsigned int dataoff, + enum ip_conntrack_info *ctinfo, int pf, unsigned int hooknum) +{ + struct icmp6hdr _ih, *icmp6h; + + icmp6h = skb_header_pointer(skb, dataoff, sizeof(_ih), &_ih); + if (icmp6h == NULL) { + if (LOG_INVALID(IPPROTO_ICMPV6)) + nf_log_packet(PF_INET6, 0, skb, NULL, NULL, NULL, + "nf_ct_icmpv6: short packet "); + return -NF_ACCEPT; + } + + if (hooknum != NF_IP6_PRE_ROUTING) + goto skipped; + + /* Ignore it if the checksum's bogus. */ + if (csum_ipv6_magic(&skb->nh.ipv6h->saddr, &skb->nh.ipv6h->daddr, + skb->len - dataoff, IPPROTO_ICMPV6, + skb_checksum(skb, dataoff, + skb->len - dataoff, 0))) { + nf_log_packet(PF_INET6, 0, skb, NULL, NULL, NULL, + "nf_ct_icmpv6: ICMPv6 checksum failed\n"); + return -NF_ACCEPT; + } + +skipped: + + /* is not error message ? */ + if (icmp6h->icmp6_type >= 128) + return NF_ACCEPT; + + return icmpv6_error_message(skb, dataoff, ctinfo, hooknum); +} + +struct nf_conntrack_protocol nf_conntrack_protocol_icmpv6 = +{ + .l3proto = PF_INET6, + .proto = IPPROTO_ICMPV6, + .name = "icmpv6", + .pkt_to_tuple = icmpv6_pkt_to_tuple, + .invert_tuple = icmpv6_invert_tuple, + .print_tuple = icmpv6_print_tuple, + .print_conntrack = icmpv6_print_conntrack, + .packet = icmpv6_packet, + .new = icmpv6_new, + .error = icmpv6_error, +}; + +EXPORT_SYMBOL(nf_conntrack_protocol_icmpv6); diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c new file mode 100644 index 000000000000..7640b9bb7694 --- /dev/null +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -0,0 +1,885 @@ +/* + * IPv6 fragment reassembly for connection tracking + * + * Copyright (C)2004 USAGI/WIDE Project + * + * Author: + * Yasuyuki Kozakai @USAGI + * + * Based on: net/ipv6/reassembly.c + * + * 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. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +#define NF_CT_FRAG6_HIGH_THRESH 262144 /* == 256*1024 */ +#define NF_CT_FRAG6_LOW_THRESH 196608 /* == 192*1024 */ +#define NF_CT_FRAG6_TIMEOUT IPV6_FRAG_TIMEOUT + +int nf_ct_frag6_high_thresh = 256*1024; +int nf_ct_frag6_low_thresh = 192*1024; +int nf_ct_frag6_timeout = IPV6_FRAG_TIMEOUT; + +struct nf_ct_frag6_skb_cb +{ + struct inet6_skb_parm h; + int offset; + struct sk_buff *orig; +}; + +#define NFCT_FRAG6_CB(skb) ((struct nf_ct_frag6_skb_cb*)((skb)->cb)) + +struct nf_ct_frag6_queue +{ + struct nf_ct_frag6_queue *next; + struct list_head lru_list; /* lru list member */ + + __u32 id; /* fragment id */ + struct in6_addr saddr; + struct in6_addr daddr; + + spinlock_t lock; + atomic_t refcnt; + struct timer_list timer; /* expire timer */ + struct sk_buff *fragments; + int len; + int meat; + struct timeval stamp; + unsigned int csum; + __u8 last_in; /* has first/last segment arrived? */ +#define COMPLETE 4 +#define FIRST_IN 2 +#define LAST_IN 1 + __u16 nhoffset; + struct nf_ct_frag6_queue **pprev; +}; + +/* Hash table. */ + +#define FRAG6Q_HASHSZ 64 + +static struct nf_ct_frag6_queue *nf_ct_frag6_hash[FRAG6Q_HASHSZ]; +static rwlock_t nf_ct_frag6_lock = RW_LOCK_UNLOCKED; +static u32 nf_ct_frag6_hash_rnd; +static LIST_HEAD(nf_ct_frag6_lru_list); +int nf_ct_frag6_nqueues = 0; + +static __inline__ void __fq_unlink(struct nf_ct_frag6_queue *fq) +{ + if (fq->next) + fq->next->pprev = fq->pprev; + *fq->pprev = fq->next; + list_del(&fq->lru_list); + nf_ct_frag6_nqueues--; +} + +static __inline__ void fq_unlink(struct nf_ct_frag6_queue *fq) +{ + write_lock(&nf_ct_frag6_lock); + __fq_unlink(fq); + write_unlock(&nf_ct_frag6_lock); +} + +static unsigned int ip6qhashfn(u32 id, struct in6_addr *saddr, + struct in6_addr *daddr) +{ + u32 a, b, c; + + a = saddr->s6_addr32[0]; + b = saddr->s6_addr32[1]; + c = saddr->s6_addr32[2]; + + a += JHASH_GOLDEN_RATIO; + b += JHASH_GOLDEN_RATIO; + c += nf_ct_frag6_hash_rnd; + __jhash_mix(a, b, c); + + a += saddr->s6_addr32[3]; + b += daddr->s6_addr32[0]; + c += daddr->s6_addr32[1]; + __jhash_mix(a, b, c); + + a += daddr->s6_addr32[2]; + b += daddr->s6_addr32[3]; + c += id; + __jhash_mix(a, b, c); + + return c & (FRAG6Q_HASHSZ - 1); +} + +static struct timer_list nf_ct_frag6_secret_timer; +int nf_ct_frag6_secret_interval = 10 * 60 * HZ; + +static void nf_ct_frag6_secret_rebuild(unsigned long dummy) +{ + unsigned long now = jiffies; + int i; + + write_lock(&nf_ct_frag6_lock); + get_random_bytes(&nf_ct_frag6_hash_rnd, sizeof(u32)); + for (i = 0; i < FRAG6Q_HASHSZ; i++) { + struct nf_ct_frag6_queue *q; + + q = nf_ct_frag6_hash[i]; + while (q) { + struct nf_ct_frag6_queue *next = q->next; + unsigned int hval = ip6qhashfn(q->id, + &q->saddr, + &q->daddr); + + if (hval != i) { + /* Unlink. */ + if (q->next) + q->next->pprev = q->pprev; + *q->pprev = q->next; + + /* Relink to new hash chain. */ + if ((q->next = nf_ct_frag6_hash[hval]) != NULL) + q->next->pprev = &q->next; + nf_ct_frag6_hash[hval] = q; + q->pprev = &nf_ct_frag6_hash[hval]; + } + + q = next; + } + } + write_unlock(&nf_ct_frag6_lock); + + mod_timer(&nf_ct_frag6_secret_timer, now + nf_ct_frag6_secret_interval); +} + +atomic_t nf_ct_frag6_mem = ATOMIC_INIT(0); + +/* Memory Tracking Functions. */ +static inline void frag_kfree_skb(struct sk_buff *skb) +{ + atomic_sub(skb->truesize, &nf_ct_frag6_mem); + if (NFCT_FRAG6_CB(skb)->orig) + kfree_skb(NFCT_FRAG6_CB(skb)->orig); + + kfree_skb(skb); +} + +static inline void frag_free_queue(struct nf_ct_frag6_queue *fq) +{ + atomic_sub(sizeof(struct nf_ct_frag6_queue), &nf_ct_frag6_mem); + kfree(fq); +} + +static inline struct nf_ct_frag6_queue *frag_alloc_queue(void) +{ + struct nf_ct_frag6_queue *fq = kmalloc(sizeof(struct nf_ct_frag6_queue), GFP_ATOMIC); + + if (!fq) + return NULL; + atomic_add(sizeof(struct nf_ct_frag6_queue), &nf_ct_frag6_mem); + return fq; +} + +/* Destruction primitives. */ + +/* Complete destruction of fq. */ +static void nf_ct_frag6_destroy(struct nf_ct_frag6_queue *fq) +{ + struct sk_buff *fp; + + BUG_TRAP(fq->last_in&COMPLETE); + BUG_TRAP(del_timer(&fq->timer) == 0); + + /* Release all fragment data. */ + fp = fq->fragments; + while (fp) { + struct sk_buff *xp = fp->next; + + frag_kfree_skb(fp); + fp = xp; + } + + frag_free_queue(fq); +} + +static __inline__ void fq_put(struct nf_ct_frag6_queue *fq) +{ + if (atomic_dec_and_test(&fq->refcnt)) + nf_ct_frag6_destroy(fq); +} + +/* Kill fq entry. It is not destroyed immediately, + * because caller (and someone more) holds reference count. + */ +static __inline__ void fq_kill(struct nf_ct_frag6_queue *fq) +{ + if (del_timer(&fq->timer)) + atomic_dec(&fq->refcnt); + + if (!(fq->last_in & COMPLETE)) { + fq_unlink(fq); + atomic_dec(&fq->refcnt); + fq->last_in |= COMPLETE; + } +} + +static void nf_ct_frag6_evictor(void) +{ + struct nf_ct_frag6_queue *fq; + struct list_head *tmp; + + for (;;) { + if (atomic_read(&nf_ct_frag6_mem) <= nf_ct_frag6_low_thresh) + return; + read_lock(&nf_ct_frag6_lock); + if (list_empty(&nf_ct_frag6_lru_list)) { + read_unlock(&nf_ct_frag6_lock); + return; + } + tmp = nf_ct_frag6_lru_list.next; + fq = list_entry(tmp, struct nf_ct_frag6_queue, lru_list); + atomic_inc(&fq->refcnt); + read_unlock(&nf_ct_frag6_lock); + + spin_lock(&fq->lock); + if (!(fq->last_in&COMPLETE)) + fq_kill(fq); + spin_unlock(&fq->lock); + + fq_put(fq); + } +} + +static void nf_ct_frag6_expire(unsigned long data) +{ + struct nf_ct_frag6_queue *fq = (struct nf_ct_frag6_queue *) data; + + spin_lock(&fq->lock); + + if (fq->last_in & COMPLETE) + goto out; + + fq_kill(fq); + +out: + spin_unlock(&fq->lock); + fq_put(fq); +} + +/* Creation primitives. */ + + +static struct nf_ct_frag6_queue *nf_ct_frag6_intern(unsigned int hash, + struct nf_ct_frag6_queue *fq_in) +{ + struct nf_ct_frag6_queue *fq; + + write_lock(&nf_ct_frag6_lock); +#ifdef CONFIG_SMP + for (fq = nf_ct_frag6_hash[hash]; fq; fq = fq->next) { + if (fq->id == fq_in->id && + !ipv6_addr_cmp(&fq_in->saddr, &fq->saddr) && + !ipv6_addr_cmp(&fq_in->daddr, &fq->daddr)) { + atomic_inc(&fq->refcnt); + write_unlock(&nf_ct_frag6_lock); + fq_in->last_in |= COMPLETE; + fq_put(fq_in); + return fq; + } + } +#endif + fq = fq_in; + + if (!mod_timer(&fq->timer, jiffies + nf_ct_frag6_timeout)) + atomic_inc(&fq->refcnt); + + atomic_inc(&fq->refcnt); + if ((fq->next = nf_ct_frag6_hash[hash]) != NULL) + fq->next->pprev = &fq->next; + nf_ct_frag6_hash[hash] = fq; + fq->pprev = &nf_ct_frag6_hash[hash]; + INIT_LIST_HEAD(&fq->lru_list); + list_add_tail(&fq->lru_list, &nf_ct_frag6_lru_list); + nf_ct_frag6_nqueues++; + write_unlock(&nf_ct_frag6_lock); + return fq; +} + + +static struct nf_ct_frag6_queue * +nf_ct_frag6_create(unsigned int hash, u32 id, struct in6_addr *src, struct in6_addr *dst) +{ + struct nf_ct_frag6_queue *fq; + + if ((fq = frag_alloc_queue()) == NULL) { + DEBUGP("Can't alloc new queue\n"); + goto oom; + } + + memset(fq, 0, sizeof(struct nf_ct_frag6_queue)); + + fq->id = id; + ipv6_addr_copy(&fq->saddr, src); + ipv6_addr_copy(&fq->daddr, dst); + + init_timer(&fq->timer); + fq->timer.function = nf_ct_frag6_expire; + fq->timer.data = (long) fq; + fq->lock = SPIN_LOCK_UNLOCKED; + atomic_set(&fq->refcnt, 1); + + return nf_ct_frag6_intern(hash, fq); + +oom: + return NULL; +} + +static __inline__ struct nf_ct_frag6_queue * +fq_find(u32 id, struct in6_addr *src, struct in6_addr *dst) +{ + struct nf_ct_frag6_queue *fq; + unsigned int hash = ip6qhashfn(id, src, dst); + + read_lock(&nf_ct_frag6_lock); + for (fq = nf_ct_frag6_hash[hash]; fq; fq = fq->next) { + if (fq->id == id && + !ipv6_addr_cmp(src, &fq->saddr) && + !ipv6_addr_cmp(dst, &fq->daddr)) { + atomic_inc(&fq->refcnt); + read_unlock(&nf_ct_frag6_lock); + return fq; + } + } + read_unlock(&nf_ct_frag6_lock); + + return nf_ct_frag6_create(hash, id, src, dst); +} + + +static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb, + struct frag_hdr *fhdr, int nhoff) +{ + struct sk_buff *prev, *next; + int offset, end; + + if (fq->last_in & COMPLETE) { + DEBUGP("Allready completed\n"); + goto err; + } + + offset = ntohs(fhdr->frag_off) & ~0x7; + end = offset + (ntohs(skb->nh.ipv6h->payload_len) - + ((u8 *) (fhdr + 1) - (u8 *) (skb->nh.ipv6h + 1))); + + if ((unsigned int)end > IPV6_MAXPLEN) { + DEBUGP("offset is too large.\n"); + return -1; + } + + if (skb->ip_summed == CHECKSUM_HW) + skb->csum = csum_sub(skb->csum, + csum_partial(skb->nh.raw, + (u8*)(fhdr + 1) - skb->nh.raw, + 0)); + + /* Is this the final fragment? */ + if (!(fhdr->frag_off & htons(IP6_MF))) { + /* If we already have some bits beyond end + * or have different end, the segment is corrupted. + */ + if (end < fq->len || + ((fq->last_in & LAST_IN) && end != fq->len)) { + DEBUGP("already received last fragment\n"); + goto err; + } + fq->last_in |= LAST_IN; + fq->len = end; + } else { + /* Check if the fragment is rounded to 8 bytes. + * Required by the RFC. + */ + if (end & 0x7) { + /* RFC2460 says always send parameter problem in + * this case. -DaveM + */ + DEBUGP("the end of this fragment is not rounded to 8 bytes.\n"); + return -1; + } + if (end > fq->len) { + /* Some bits beyond end -> corruption. */ + if (fq->last_in & LAST_IN) { + DEBUGP("last packet already reached.\n"); + goto err; + } + fq->len = end; + } + } + + if (end == offset) + goto err; + + /* Point into the IP datagram 'data' part. */ + if (!pskb_pull(skb, (u8 *) (fhdr + 1) - skb->data)) { + DEBUGP("queue: message is too short.\n"); + goto err; + } + if (end-offset < skb->len) { + if (pskb_trim(skb, end - offset)) { + DEBUGP("Can't trim\n"); + goto err; + } + if (skb->ip_summed != CHECKSUM_UNNECESSARY) + skb->ip_summed = CHECKSUM_NONE; + } + + /* Find out which fragments are in front and at the back of us + * in the chain of fragments so far. We must know where to put + * this fragment, right? + */ + prev = NULL; + for (next = fq->fragments; next != NULL; next = next->next) { + if (NFCT_FRAG6_CB(next)->offset >= offset) + break; /* bingo! */ + prev = next; + } + + /* We found where to put this one. Check for overlap with + * preceding fragment, and, if needed, align things so that + * any overlaps are eliminated. + */ + if (prev) { + int i = (NFCT_FRAG6_CB(prev)->offset + prev->len) - offset; + + if (i > 0) { + offset += i; + if (end <= offset) { + DEBUGP("overlap\n"); + goto err; + } + if (!pskb_pull(skb, i)) { + DEBUGP("Can't pull\n"); + goto err; + } + if (skb->ip_summed != CHECKSUM_UNNECESSARY) + skb->ip_summed = CHECKSUM_NONE; + } + } + + /* Look for overlap with succeeding segments. + * If we can merge fragments, do it. + */ + while (next && NFCT_FRAG6_CB(next)->offset < end) { + /* overlap is 'i' bytes */ + int i = end - NFCT_FRAG6_CB(next)->offset; + + if (i < next->len) { + /* Eat head of the next overlapped fragment + * and leave the loop. The next ones cannot overlap. + */ + DEBUGP("Eat head of the overlapped parts.: %d", i); + if (!pskb_pull(next, i)) + goto err; + + /* next fragment */ + NFCT_FRAG6_CB(next)->offset += i; + fq->meat -= i; + if (next->ip_summed != CHECKSUM_UNNECESSARY) + next->ip_summed = CHECKSUM_NONE; + break; + } else { + struct sk_buff *free_it = next; + + /* Old fragmnet is completely overridden with + * new one drop it. + */ + next = next->next; + + if (prev) + prev->next = next; + else + fq->fragments = next; + + fq->meat -= free_it->len; + frag_kfree_skb(free_it); + } + } + + NFCT_FRAG6_CB(skb)->offset = offset; + + /* Insert this fragment in the chain of fragments. */ + skb->next = next; + if (prev) + prev->next = skb; + else + fq->fragments = skb; + + skb->dev = NULL; + skb_get_timestamp(skb, &fq->stamp); + fq->meat += skb->len; + atomic_add(skb->truesize, &nf_ct_frag6_mem); + + /* The first fragment. + * nhoffset is obtained from the first fragment, of course. + */ + if (offset == 0) { + fq->nhoffset = nhoff; + fq->last_in |= FIRST_IN; + } + write_lock(&nf_ct_frag6_lock); + list_move_tail(&fq->lru_list, &nf_ct_frag6_lru_list); + write_unlock(&nf_ct_frag6_lock); + return 0; + +err: + return -1; +} + +/* + * Check if this packet is complete. + * Returns NULL on failure by any reason, and pointer + * to current nexthdr field in reassembled frame. + * + * It is called with locked fq, and caller must check that + * queue is eligible for reassembly i.e. it is not COMPLETE, + * the last and the first frames arrived and all the bits are here. + */ +static struct sk_buff * +nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev) +{ + struct sk_buff *fp, *op, *head = fq->fragments; + int payload_len; + + fq_kill(fq); + + BUG_TRAP(head != NULL); + BUG_TRAP(NFCT_FRAG6_CB(head)->offset == 0); + + /* Unfragmented part is taken from the first segment. */ + payload_len = (head->data - head->nh.raw) - sizeof(struct ipv6hdr) + fq->len - sizeof(struct frag_hdr); + if (payload_len > IPV6_MAXPLEN) { + DEBUGP("payload len is too large.\n"); + goto out_oversize; + } + + /* Head of list must not be cloned. */ + if (skb_cloned(head) && pskb_expand_head(head, 0, 0, GFP_ATOMIC)) { + DEBUGP("skb is cloned but can't expand head"); + goto out_oom; + } + + /* If the first fragment is fragmented itself, we split + * it to two chunks: the first with data and paged part + * and the second, holding only fragments. */ + if (skb_shinfo(head)->frag_list) { + struct sk_buff *clone; + int i, plen = 0; + + if ((clone = alloc_skb(0, GFP_ATOMIC)) == NULL) { + DEBUGP("Can't alloc skb\n"); + goto out_oom; + } + clone->next = head->next; + head->next = clone; + skb_shinfo(clone)->frag_list = skb_shinfo(head)->frag_list; + skb_shinfo(head)->frag_list = NULL; + for (i=0; inr_frags; i++) + plen += skb_shinfo(head)->frags[i].size; + clone->len = clone->data_len = head->data_len - plen; + head->data_len -= clone->len; + head->len -= clone->len; + clone->csum = 0; + clone->ip_summed = head->ip_summed; + + NFCT_FRAG6_CB(clone)->orig = NULL; + atomic_add(clone->truesize, &nf_ct_frag6_mem); + } + + /* We have to remove fragment header from datagram and to relocate + * header in order to calculate ICV correctly. */ + head->nh.raw[fq->nhoffset] = head->h.raw[0]; + memmove(head->head + sizeof(struct frag_hdr), head->head, + (head->data - head->head) - sizeof(struct frag_hdr)); + head->mac.raw += sizeof(struct frag_hdr); + head->nh.raw += sizeof(struct frag_hdr); + + skb_shinfo(head)->frag_list = head->next; + head->h.raw = head->data; + skb_push(head, head->data - head->nh.raw); + atomic_sub(head->truesize, &nf_ct_frag6_mem); + + for (fp=head->next; fp; fp = fp->next) { + head->data_len += fp->len; + head->len += fp->len; + if (head->ip_summed != fp->ip_summed) + head->ip_summed = CHECKSUM_NONE; + else if (head->ip_summed == CHECKSUM_HW) + head->csum = csum_add(head->csum, fp->csum); + head->truesize += fp->truesize; + atomic_sub(fp->truesize, &nf_ct_frag6_mem); + } + + head->next = NULL; + head->dev = dev; + skb_set_timestamp(head, &fq->stamp); + head->nh.ipv6h->payload_len = htons(payload_len); + + /* Yes, and fold redundant checksum back. 8) */ + if (head->ip_summed == CHECKSUM_HW) + head->csum = csum_partial(head->nh.raw, head->h.raw-head->nh.raw, head->csum); + + fq->fragments = NULL; + + /* all original skbs are linked into the NFCT_FRAG6_CB(head).orig */ + fp = skb_shinfo(head)->frag_list; + if (NFCT_FRAG6_CB(fp)->orig == NULL) + /* at above code, head skb is divided into two skbs. */ + fp = fp->next; + + op = NFCT_FRAG6_CB(head)->orig; + for (; fp; fp = fp->next) { + struct sk_buff *orig = NFCT_FRAG6_CB(fp)->orig; + + op->next = orig; + op = orig; + NFCT_FRAG6_CB(fp)->orig = NULL; + } + + return head; + +out_oversize: + if (net_ratelimit()) + printk(KERN_DEBUG "nf_ct_frag6_reasm: payload len = %d\n", payload_len); + goto out_fail; +out_oom: + if (net_ratelimit()) + printk(KERN_DEBUG "nf_ct_frag6_reasm: no memory for reassembly\n"); +out_fail: + return NULL; +} + +/* + * find the header just before Fragment Header. + * + * if success return 0 and set ... + * (*prevhdrp): the value of "Next Header Field" in the header + * just before Fragment Header. + * (*prevhoff): the offset of "Next Header Field" in the header + * just before Fragment Header. + * (*fhoff) : the offset of Fragment Header. + * + * Based on ipv6_skip_hdr() in net/ipv6/exthdr.c + * + */ +static int +find_prev_fhdr(struct sk_buff *skb, u8 *prevhdrp, int *prevhoff, int *fhoff) +{ + u8 nexthdr = skb->nh.ipv6h->nexthdr; + u8 prev_nhoff = (u8 *)&skb->nh.ipv6h->nexthdr - skb->data; + int start = (u8 *)(skb->nh.ipv6h+1) - skb->data; + int len = skb->len - start; + u8 prevhdr = NEXTHDR_IPV6; + + while (nexthdr != NEXTHDR_FRAGMENT) { + struct ipv6_opt_hdr hdr; + int hdrlen; + + if (!ipv6_ext_hdr(nexthdr)) { + return -1; + } + if (len < (int)sizeof(struct ipv6_opt_hdr)) { + DEBUGP("too short\n"); + return -1; + } + if (nexthdr == NEXTHDR_NONE) { + DEBUGP("next header is none\n"); + return -1; + } + if (skb_copy_bits(skb, start, &hdr, sizeof(hdr))) + BUG(); + if (nexthdr == NEXTHDR_AUTH) + hdrlen = (hdr.hdrlen+2)<<2; + else + hdrlen = ipv6_optlen(&hdr); + + prevhdr = nexthdr; + prev_nhoff = start; + + nexthdr = hdr.nexthdr; + len -= hdrlen; + start += hdrlen; + } + + if (len < 0) + return -1; + + *prevhdrp = prevhdr; + *prevhoff = prev_nhoff; + *fhoff = start; + + return 0; +} + +struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb) +{ + struct sk_buff *clone; + struct net_device *dev = skb->dev; + struct frag_hdr *fhdr; + struct nf_ct_frag6_queue *fq; + struct ipv6hdr *hdr; + int fhoff, nhoff; + u8 prevhdr; + struct sk_buff *ret_skb = NULL; + + /* Jumbo payload inhibits frag. header */ + if (skb->nh.ipv6h->payload_len == 0) { + DEBUGP("payload len = 0\n"); + return skb; + } + + if (find_prev_fhdr(skb, &prevhdr, &nhoff, &fhoff) < 0) + return skb; + + clone = skb_clone(skb, GFP_ATOMIC); + if (clone == NULL) { + DEBUGP("Can't clone skb\n"); + return skb; + } + + NFCT_FRAG6_CB(clone)->orig = skb; + + if (!pskb_may_pull(clone, fhoff + sizeof(*fhdr))) { + DEBUGP("message is too short.\n"); + goto ret_orig; + } + + clone->h.raw = clone->data + fhoff; + hdr = clone->nh.ipv6h; + fhdr = (struct frag_hdr *)clone->h.raw; + + if (!(fhdr->frag_off & htons(0xFFF9))) { + DEBUGP("Invalid fragment offset\n"); + /* It is not a fragmented frame */ + goto ret_orig; + } + + if (atomic_read(&nf_ct_frag6_mem) > nf_ct_frag6_high_thresh) + nf_ct_frag6_evictor(); + + fq = fq_find(fhdr->identification, &hdr->saddr, &hdr->daddr); + if (fq == NULL) { + DEBUGP("Can't find and can't create new queue\n"); + goto ret_orig; + } + + spin_lock(&fq->lock); + + if (nf_ct_frag6_queue(fq, clone, fhdr, nhoff) < 0) { + spin_unlock(&fq->lock); + DEBUGP("Can't insert skb to queue\n"); + fq_put(fq); + goto ret_orig; + } + + if (fq->last_in == (FIRST_IN|LAST_IN) && fq->meat == fq->len) { + ret_skb = nf_ct_frag6_reasm(fq, dev); + if (ret_skb == NULL) + DEBUGP("Can't reassemble fragmented packets\n"); + } + spin_unlock(&fq->lock); + + fq_put(fq); + return ret_skb; + +ret_orig: + kfree_skb(clone); + return skb; +} + +void nf_ct_frag6_output(unsigned int hooknum, struct sk_buff *skb, + struct net_device *in, struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct sk_buff *s, *s2; + + for (s = NFCT_FRAG6_CB(skb)->orig; s;) { + nf_conntrack_put_reasm(s->nfct_reasm); + nf_conntrack_get_reasm(skb); + s->nfct_reasm = skb; + + s2 = s->next; + NF_HOOK_THRESH(PF_INET6, hooknum, s, in, out, okfn, + NF_IP6_PRI_CONNTRACK_DEFRAG + 1); + s = s2; + } + nf_conntrack_put_reasm(skb); +} + +int nf_ct_frag6_kfree_frags(struct sk_buff *skb) +{ + struct sk_buff *s, *s2; + + for (s = NFCT_FRAG6_CB(skb)->orig; s; s = s2) { + + s2 = s->next; + kfree_skb(s); + } + + kfree_skb(skb); + + return 0; +} + +int nf_ct_frag6_init(void) +{ + nf_ct_frag6_hash_rnd = (u32) ((num_physpages ^ (num_physpages>>7)) ^ + (jiffies ^ (jiffies >> 6))); + + init_timer(&nf_ct_frag6_secret_timer); + nf_ct_frag6_secret_timer.function = nf_ct_frag6_secret_rebuild; + nf_ct_frag6_secret_timer.expires = jiffies + + nf_ct_frag6_secret_interval; + add_timer(&nf_ct_frag6_secret_timer); + + return 0; +} + +void nf_ct_frag6_cleanup(void) +{ + del_timer(&nf_ct_frag6_secret_timer); + nf_ct_frag6_evictor(); +} diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index a1265a320b11..651c79b41eeb 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -174,8 +174,10 @@ int ipv6_raw_deliver(struct sk_buff *skb, int nexthdr) struct sk_buff *clone = skb_clone(skb, GFP_ATOMIC); /* Not releasing hash table! */ - if (clone) + if (clone) { + nf_reset(clone); rawv6_rcv(sk, clone); + } } sk = __raw_v6_lookup(sk_next(sk), nexthdr, daddr, saddr, IP6CB(skb)->iif); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 8296b38bf270..a84f9221e5f0 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -1,3 +1,6 @@ +menu "Core Netfilter Configuration" + depends on NET && NETFILTER + config NETFILTER_NETLINK tristate "Netfilter netlink interface" help @@ -22,3 +25,74 @@ config NETFILTER_NETLINK_LOG and is also scheduled to replace the old syslog-based ipt_LOG and ip6t_LOG modules. +config NF_CONNTRACK + tristate "Layer 3 Independent Connection tracking (EXPERIMENTAL)" + depends on EXPERIMENTAL && IP_NF_CONNTRACK=n + default n + ---help--- + Connection tracking keeps a record of what packets have passed + through your machine, in order to figure out how they are related + into connections. + + Layer 3 independent connection tracking is experimental scheme + which generalize ip_conntrack to support other layer 3 protocols. + + To compile it as a module, choose M here. If unsure, say N. + +config NF_CT_ACCT + bool "Connection tracking flow accounting" + depends on NF_CONNTRACK + help + If this option is enabled, the connection tracking code will + keep per-flow packet and byte counters. + + Those counters can be used for flow-based accounting or the + `connbytes' match. + + If unsure, say `N'. + +config NF_CONNTRACK_MARK + bool 'Connection mark tracking support' + depends on NF_CONNTRACK + help + This option enables support for connection marks, used by the + `CONNMARK' target and `connmark' match. Similar to the mark value + of packets, but this mark value is kept in the conntrack session + instead of the individual packets. + +config NF_CONNTRACK_EVENTS + bool "Connection tracking events" + depends on NF_CONNTRACK + help + If this option is enabled, the connection tracking code will + provide a notifier chain that can be used by other kernel code + to get notified aboutchanges in the connection tracking state. + + If unsure, say `N'. + +config NF_CT_PROTO_SCTP + tristate 'SCTP protocol on new connection tracking support (EXPERIMENTAL)' + depends on EXPERIMENTAL && NF_CONNTRACK + default n + help + With this option enabled, the layer 3 independent connection + tracking code will be able to do state tracking on SCTP connections. + + If you want to compile it as a module, say M here and read + Documentation/modules.txt. If unsure, say `N'. + +config NF_CONNTRACK_FTP + tristate "FTP support on new connection tracking (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + help + Tracking FTP connections is problematic: special helpers are + required for tracking them, and doing masquerading and other forms + of Network Address Translation on them. + + This is FTP support on Layer 3 independent connection tracking. + Layer 3 independent connection tracking is experimental scheme + which generalize ip_conntrack to support other layer 3 protocols. + + To compile it as a module, choose M here. If unsure, say N. + +endmenu diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index b3b44f8b415a..55f019ad2c08 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -5,3 +5,11 @@ obj-$(CONFIG_NETFILTER) = netfilter.o obj-$(CONFIG_NETFILTER_NETLINK) += nfnetlink.o obj-$(CONFIG_NETFILTER_NETLINK_QUEUE) += nfnetlink_queue.o obj-$(CONFIG_NETFILTER_NETLINK_LOG) += nfnetlink_log.o + +nf_conntrack-objs := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o + +obj-$(CONFIG_NF_CONNTRACK) += nf_conntrack.o +obj-$(CONFIG_NF_CONNTRACK_FTP) += nf_conntrack_ftp.o + +# SCTP protocol connection tracking +obj-$(CONFIG_NF_CT_PROTO_SCTP) += nf_conntrack_proto_sctp.o diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c new file mode 100644 index 000000000000..9a67c796b385 --- /dev/null +++ b/net/netfilter/nf_conntrack_core.c @@ -0,0 +1,1538 @@ +/* Connection state tracking for netfilter. This is separated from, + but required by, the NAT layer; it can also be used by an iptables + extension. */ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2005 Netfilter Core Team + * (C) 2003,2004 USAGI/WIDE Project + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 23 Apr 2001: Harald Welte + * - new API and handling of conntrack/nat helpers + * - now capable of multiple expectations for one master + * 16 Jul 2002: Harald Welte + * - add usage/reference counts to ip_conntrack_expect + * - export ip_conntrack[_expect]_{find_get,put} functions + * 16 Dec 2003: Yasuyuki Kozakai @USAGI + * - generalize L3 protocol denendent part. + * 23 Mar 2004: Yasuyuki Kozakai @USAGI + * - add support various size of conntrack structures. + * + * Derived from net/ipv4/netfilter/ip_conntrack_core.c + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* This rwlock protects the main hash table, protocol/helper/expected + registrations, conntrack timers*/ +#define ASSERT_READ_LOCK(x) +#define ASSERT_WRITE_LOCK(x) + +#include +#include +#include +#include +#include +#include + +#define NF_CONNTRACK_VERSION "0.4.1" + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +DEFINE_RWLOCK(nf_conntrack_lock); + +/* nf_conntrack_standalone needs this */ +atomic_t nf_conntrack_count = ATOMIC_INIT(0); + +void (*nf_conntrack_destroyed)(struct nf_conn *conntrack) = NULL; +LIST_HEAD(nf_conntrack_expect_list); +struct nf_conntrack_protocol **nf_ct_protos[PF_MAX]; +struct nf_conntrack_l3proto *nf_ct_l3protos[PF_MAX]; +static LIST_HEAD(helpers); +unsigned int nf_conntrack_htable_size = 0; +int nf_conntrack_max; +struct list_head *nf_conntrack_hash; +static kmem_cache_t *nf_conntrack_expect_cachep; +struct nf_conn nf_conntrack_untracked; +unsigned int nf_ct_log_invalid; +static LIST_HEAD(unconfirmed); +static int nf_conntrack_vmalloc; + +#ifdef CONFIG_NF_CONNTRACK_EVENTS +struct notifier_block *nf_conntrack_chain; +struct notifier_block *nf_conntrack_expect_chain; + +DEFINE_PER_CPU(struct nf_conntrack_ecache, nf_conntrack_ecache); + +/* deliver cached events and clear cache entry - must be called with locally + * disabled softirqs */ +static inline void +__nf_ct_deliver_cached_events(struct nf_conntrack_ecache *ecache) +{ + DEBUGP("ecache: delivering events for %p\n", ecache->ct); + if (nf_ct_is_confirmed(ecache->ct) && !nf_ct_is_dying(ecache->ct) + && ecache->events) + notifier_call_chain(&nf_conntrack_chain, ecache->events, + ecache->ct); + + ecache->events = 0; + nf_ct_put(ecache->ct); + ecache->ct = NULL; +} + +/* Deliver all cached events for a particular conntrack. This is called + * by code prior to async packet handling for freeing the skb */ +void nf_ct_deliver_cached_events(const struct nf_conn *ct) +{ + struct nf_conntrack_ecache *ecache; + + local_bh_disable(); + ecache = &__get_cpu_var(nf_conntrack_ecache); + if (ecache->ct == ct) + __nf_ct_deliver_cached_events(ecache); + local_bh_enable(); +} + +/* Deliver cached events for old pending events, if current conntrack != old */ +void __nf_ct_event_cache_init(struct nf_conn *ct) +{ + struct nf_conntrack_ecache *ecache; + + /* take care of delivering potentially old events */ + ecache = &__get_cpu_var(nf_conntrack_ecache); + BUG_ON(ecache->ct == ct); + if (ecache->ct) + __nf_ct_deliver_cached_events(ecache); + /* initialize for this conntrack/packet */ + ecache->ct = ct; + nf_conntrack_get(&ct->ct_general); +} + +/* flush the event cache - touches other CPU's data and must not be called + * while packets are still passing through the code */ +static void nf_ct_event_cache_flush(void) +{ + struct nf_conntrack_ecache *ecache; + int cpu; + + for_each_cpu(cpu) { + ecache = &per_cpu(nf_conntrack_ecache, cpu); + if (ecache->ct) + nf_ct_put(ecache->ct); + } +} +#else +static inline void nf_ct_event_cache_flush(void) {} +#endif /* CONFIG_NF_CONNTRACK_EVENTS */ + +DEFINE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); +EXPORT_PER_CPU_SYMBOL(nf_conntrack_stat); + +/* + * This scheme offers various size of "struct nf_conn" dependent on + * features(helper, nat, ...) + */ + +#define NF_CT_FEATURES_NAMELEN 256 +static struct { + /* name of slab cache. printed in /proc/slabinfo */ + char *name; + + /* size of slab cache */ + size_t size; + + /* slab cache pointer */ + kmem_cache_t *cachep; + + /* allocated slab cache + modules which uses this slab cache */ + int use; + + /* Initialization */ + int (*init_conntrack)(struct nf_conn *, u_int32_t); + +} nf_ct_cache[NF_CT_F_NUM]; + +/* protect members of nf_ct_cache except of "use" */ +DEFINE_RWLOCK(nf_ct_cache_lock); + +/* This avoids calling kmem_cache_create() with same name simultaneously */ +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) +{ + if (unlikely(nf_ct_protos[l3proto] == NULL)) + return &nf_conntrack_generic_protocol; + + return nf_ct_protos[l3proto][protocol]; +} + +static int nf_conntrack_hash_rnd_initted; +static unsigned int nf_conntrack_hash_rnd; + +static u_int32_t __hash_conntrack(const struct nf_conntrack_tuple *tuple, + unsigned int size, unsigned int rnd) +{ + unsigned int a, b; + a = jhash((void *)tuple->src.u3.all, sizeof(tuple->src.u3.all), + ((tuple->src.l3num) << 16) | tuple->dst.protonum); + b = jhash((void *)tuple->dst.u3.all, sizeof(tuple->dst.u3.all), + (tuple->src.u.all << 16) | tuple->dst.u.all); + + return jhash_2words(a, b, rnd) % size; +} + +static inline u_int32_t hash_conntrack(const struct nf_conntrack_tuple *tuple) +{ + return __hash_conntrack(tuple, nf_conntrack_htable_size, + nf_conntrack_hash_rnd); +} + +/* Initialize "struct nf_conn" which has spaces for helper */ +static int +init_conntrack_for_helper(struct nf_conn *conntrack, u_int32_t features) +{ + + conntrack->help = (union nf_conntrack_help *) + (((unsigned long)conntrack->data + + (__alignof__(union nf_conntrack_help) - 1)) + & (~((unsigned long)(__alignof__(union nf_conntrack_help) -1)))); + return 0; +} + +int nf_conntrack_register_cache(u_int32_t features, const char *name, + size_t size, + int (*init)(struct nf_conn *, u_int32_t)) +{ + int ret = 0; + char *cache_name; + kmem_cache_t *cachep; + + DEBUGP("nf_conntrack_register_cache: features=0x%x, name=%s, size=%d\n", + features, name, size); + + if (features < NF_CT_F_BASIC || features >= NF_CT_F_NUM) { + DEBUGP("nf_conntrack_register_cache: invalid features.: 0x%x\n", + features); + return -EINVAL; + } + + down(&nf_ct_cache_mutex); + + write_lock_bh(&nf_ct_cache_lock); + /* e.g: multiple helpers are loaded */ + if (nf_ct_cache[features].use > 0) { + DEBUGP("nf_conntrack_register_cache: already resisterd.\n"); + if ((!strncmp(nf_ct_cache[features].name, name, + NF_CT_FEATURES_NAMELEN)) + && nf_ct_cache[features].size == size + && nf_ct_cache[features].init_conntrack == init) { + DEBUGP("nf_conntrack_register_cache: reusing.\n"); + nf_ct_cache[features].use++; + ret = 0; + } else + ret = -EBUSY; + + write_unlock_bh(&nf_ct_cache_lock); + up(&nf_ct_cache_mutex); + return ret; + } + write_unlock_bh(&nf_ct_cache_lock); + + /* + * The memory space for name of slab cache must be alive until + * cache is destroyed. + */ + cache_name = kmalloc(sizeof(char)*NF_CT_FEATURES_NAMELEN, GFP_ATOMIC); + if (cache_name == NULL) { + DEBUGP("nf_conntrack_register_cache: can't alloc cache_name\n"); + ret = -ENOMEM; + goto out_up_mutex; + } + + if (strlcpy(cache_name, name, NF_CT_FEATURES_NAMELEN) + >= NF_CT_FEATURES_NAMELEN) { + printk("nf_conntrack_register_cache: name too long\n"); + ret = -EINVAL; + goto out_free_name; + } + + cachep = kmem_cache_create(cache_name, size, 0, 0, + NULL, NULL); + if (!cachep) { + printk("nf_conntrack_register_cache: Can't create slab cache " + "for the features = 0x%x\n", features); + ret = -ENOMEM; + goto out_free_name; + } + + write_lock_bh(&nf_ct_cache_lock); + nf_ct_cache[features].use = 1; + nf_ct_cache[features].size = size; + nf_ct_cache[features].init_conntrack = init; + nf_ct_cache[features].cachep = cachep; + nf_ct_cache[features].name = cache_name; + write_unlock_bh(&nf_ct_cache_lock); + + goto out_up_mutex; + +out_free_name: + kfree(cache_name); +out_up_mutex: + up(&nf_ct_cache_mutex); + return ret; +} + +/* FIXME: In the current, only nf_conntrack_cleanup() can call this function. */ +void nf_conntrack_unregister_cache(u_int32_t features) +{ + kmem_cache_t *cachep; + char *name; + + /* + * This assures that kmem_cache_create() isn't called before destroying + * slab cache. + */ + DEBUGP("nf_conntrack_unregister_cache: 0x%04x\n", features); + down(&nf_ct_cache_mutex); + + write_lock_bh(&nf_ct_cache_lock); + if (--nf_ct_cache[features].use > 0) { + write_unlock_bh(&nf_ct_cache_lock); + up(&nf_ct_cache_mutex); + return; + } + cachep = nf_ct_cache[features].cachep; + name = nf_ct_cache[features].name; + nf_ct_cache[features].cachep = NULL; + nf_ct_cache[features].name = NULL; + nf_ct_cache[features].init_conntrack = NULL; + nf_ct_cache[features].size = 0; + write_unlock_bh(&nf_ct_cache_lock); + + synchronize_net(); + + kmem_cache_destroy(cachep); + kfree(name); + + up(&nf_ct_cache_mutex); +} + +int +nf_ct_get_tuple(const struct sk_buff *skb, + unsigned int nhoff, + unsigned int dataoff, + u_int16_t l3num, + u_int8_t protonum, + struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_l3proto *l3proto, + const struct nf_conntrack_protocol *protocol) +{ + NF_CT_TUPLE_U_BLANK(tuple); + + tuple->src.l3num = l3num; + if (l3proto->pkt_to_tuple(skb, nhoff, tuple) == 0) + return 0; + + tuple->dst.protonum = protonum; + tuple->dst.dir = IP_CT_DIR_ORIGINAL; + + return protocol->pkt_to_tuple(skb, dataoff, tuple); +} + +int +nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, + const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_l3proto *l3proto, + const struct nf_conntrack_protocol *protocol) +{ + NF_CT_TUPLE_U_BLANK(inverse); + + inverse->src.l3num = orig->src.l3num; + if (l3proto->invert_tuple(inverse, orig) == 0) + return 0; + + inverse->dst.dir = !orig->dst.dir; + + inverse->dst.protonum = orig->dst.protonum; + return protocol->invert_tuple(inverse, orig); +} + +/* nf_conntrack_expect helper functions */ +static void nf_ct_unlink_expect(struct nf_conntrack_expect *exp) +{ + ASSERT_WRITE_LOCK(&nf_conntrack_lock); + NF_CT_ASSERT(!timer_pending(&exp_timeout)); + list_del(&exp->list); + NF_CT_STAT_INC(expect_delete); + exp->master->expecting--; + nf_conntrack_expect_put(exp); +} + +static void expectation_timed_out(unsigned long ul_expect) +{ + struct nf_conntrack_expect *exp = (void *)ul_expect; + + write_lock_bh(&nf_conntrack_lock); + nf_ct_unlink_expect(exp); + write_unlock_bh(&nf_conntrack_lock); + nf_conntrack_expect_put(exp); +} + +/* If an expectation for this connection is found, it gets delete from + * global list then returned. */ +static struct nf_conntrack_expect * +find_expectation(const struct nf_conntrack_tuple *tuple) +{ + struct nf_conntrack_expect *i; + + list_for_each_entry(i, &nf_conntrack_expect_list, list) { + /* If master is not in hash table yet (ie. packet hasn't left + this machine yet), how can other end know about expected? + Hence these are not the droids you are looking for (if + master ct never got confirmed, we'd hold a reference to it + and weird things would happen to future packets). */ + if (nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask) + && nf_ct_is_confirmed(i->master)) { + if (i->flags & NF_CT_EXPECT_PERMANENT) { + atomic_inc(&i->use); + return i; + } else if (del_timer(&i->timeout)) { + nf_ct_unlink_expect(i); + return i; + } + } + } + return NULL; +} + +/* delete all expectations for this conntrack */ +static void remove_expectations(struct nf_conn *ct) +{ + struct nf_conntrack_expect *i, *tmp; + + /* Optimization: most connection never expect any others. */ + if (ct->expecting == 0) + return; + + list_for_each_entry_safe(i, tmp, &nf_conntrack_expect_list, list) { + if (i->master == ct && del_timer(&i->timeout)) { + nf_ct_unlink_expect(i); + nf_conntrack_expect_put(i); + } + } +} + +static void +clean_from_lists(struct nf_conn *ct) +{ + unsigned int ho, hr; + + DEBUGP("clean_from_lists(%p)\n", ct); + ASSERT_WRITE_LOCK(&nf_conntrack_lock); + + ho = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + hr = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); + LIST_DELETE(&nf_conntrack_hash[ho], &ct->tuplehash[IP_CT_DIR_ORIGINAL]); + LIST_DELETE(&nf_conntrack_hash[hr], &ct->tuplehash[IP_CT_DIR_REPLY]); + + /* Destroy all pending expectations */ + remove_expectations(ct); +} + +static void +destroy_conntrack(struct nf_conntrack *nfct) +{ + struct nf_conn *ct = (struct nf_conn *)nfct; + struct nf_conntrack_l3proto *l3proto; + struct nf_conntrack_protocol *proto; + + DEBUGP("destroy_conntrack(%p)\n", ct); + NF_CT_ASSERT(atomic_read(&nfct->use) == 0); + NF_CT_ASSERT(!timer_pending(&ct->timeout)); + + nf_conntrack_event(IPCT_DESTROY, ct); + set_bit(IPS_DYING_BIT, &ct->status); + + /* 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); + 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); + if (proto && proto->destroy) + proto->destroy(ct); + + if (nf_conntrack_destroyed) + nf_conntrack_destroyed(ct); + + write_lock_bh(&nf_conntrack_lock); + /* Expectations will have been removed in clean_from_lists, + * 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); + + /* We overload first tuple to link into unconfirmed list. */ + if (!nf_ct_is_confirmed(ct)) { + BUG_ON(list_empty(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list)); + list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list); + } + + NF_CT_STAT_INC(delete); + write_unlock_bh(&nf_conntrack_lock); + + if (ct->master) + nf_ct_put(ct->master); + + DEBUGP("destroy_conntrack: returning ct=%p to slab\n", ct); + nf_conntrack_free(ct); +} + +static void death_by_timeout(unsigned long ul_conntrack) +{ + struct nf_conn *ct = (void *)ul_conntrack; + + write_lock_bh(&nf_conntrack_lock); + /* Inside lock so preempt is disabled on module removal path. + * Otherwise we can get spurious warnings. */ + NF_CT_STAT_INC(delete_list); + clean_from_lists(ct); + write_unlock_bh(&nf_conntrack_lock); + nf_ct_put(ct); +} + +static inline int +conntrack_tuple_cmp(const struct nf_conntrack_tuple_hash *i, + const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack) +{ + ASSERT_READ_LOCK(&nf_conntrack_lock); + return nf_ct_tuplehash_to_ctrack(i) != ignored_conntrack + && nf_ct_tuple_equal(tuple, &i->tuple); +} + +static struct nf_conntrack_tuple_hash * +__nf_conntrack_find(const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack) +{ + struct nf_conntrack_tuple_hash *h; + unsigned int hash = hash_conntrack(tuple); + + ASSERT_READ_LOCK(&nf_conntrack_lock); + list_for_each_entry(h, &nf_conntrack_hash[hash], list) { + if (conntrack_tuple_cmp(h, tuple, ignored_conntrack)) { + NF_CT_STAT_INC(found); + return h; + } + NF_CT_STAT_INC(searched); + } + + return NULL; +} + +/* Find a connection corresponding to a tuple. */ +struct nf_conntrack_tuple_hash * +nf_conntrack_find_get(const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack) +{ + struct nf_conntrack_tuple_hash *h; + + read_lock_bh(&nf_conntrack_lock); + h = __nf_conntrack_find(tuple, ignored_conntrack); + if (h) + atomic_inc(&nf_ct_tuplehash_to_ctrack(h)->ct_general.use); + read_unlock_bh(&nf_conntrack_lock); + + return h; +} + +/* Confirm a connection given skb; places it in hash table */ +int +__nf_conntrack_confirm(struct sk_buff **pskb) +{ + unsigned int hash, repl_hash; + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + + ct = nf_ct_get(*pskb, &ctinfo); + + /* ipt_REJECT uses nf_conntrack_attach to attach related + ICMP/TCP RST packets in other direction. Actual packet + which created connection will be IP_CT_NEW or for an + expected connection, IP_CT_RELATED. */ + if (CTINFO2DIR(ctinfo) != IP_CT_DIR_ORIGINAL) + return NF_ACCEPT; + + hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + repl_hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); + + /* We're not in hash table, and we refuse to set up related + connections for unconfirmed conns. But packet copies and + REJECT will give spurious warnings here. */ + /* NF_CT_ASSERT(atomic_read(&ct->ct_general.use) == 1); */ + + /* No external references means noone else could have + confirmed us. */ + NF_CT_ASSERT(!nf_ct_is_confirmed(ct)); + DEBUGP("Confirming conntrack %p\n", ct); + + write_lock_bh(&nf_conntrack_lock); + + /* See if there's one in the list already, including reverse: + NAT could have grabbed it without realizing, since we're + not in the hash. If there is, we lost race. */ + if (!LIST_FIND(&nf_conntrack_hash[hash], + conntrack_tuple_cmp, + struct nf_conntrack_tuple_hash *, + &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, NULL) + && !LIST_FIND(&nf_conntrack_hash[repl_hash], + conntrack_tuple_cmp, + struct nf_conntrack_tuple_hash *, + &ct->tuplehash[IP_CT_DIR_REPLY].tuple, NULL)) { + /* 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]); + /* Timer relative to confirmation time, not original + setting time, otherwise we'd get timer wrap in + weird delay cases. */ + ct->timeout.expires += jiffies; + add_timer(&ct->timeout); + atomic_inc(&ct->ct_general.use); + set_bit(IPS_CONFIRMED_BIT, &ct->status); + NF_CT_STAT_INC(insert); + write_unlock_bh(&nf_conntrack_lock); + if (ct->helper) + nf_conntrack_event_cache(IPCT_HELPER, *pskb); +#ifdef CONFIG_NF_NAT_NEEDED + if (test_bit(IPS_SRC_NAT_DONE_BIT, &ct->status) || + test_bit(IPS_DST_NAT_DONE_BIT, &ct->status)) + nf_conntrack_event_cache(IPCT_NATINFO, *pskb); +#endif + nf_conntrack_event_cache(master_ct(ct) ? + IPCT_RELATED : IPCT_NEW, *pskb); + return NF_ACCEPT; + } + + NF_CT_STAT_INC(insert_failed); + write_unlock_bh(&nf_conntrack_lock); + return NF_DROP; +} + +/* Returns true if a connection correspondings to the tuple (required + for NAT). */ +int +nf_conntrack_tuple_taken(const struct nf_conntrack_tuple *tuple, + const struct nf_conn *ignored_conntrack) +{ + struct nf_conntrack_tuple_hash *h; + + read_lock_bh(&nf_conntrack_lock); + h = __nf_conntrack_find(tuple, ignored_conntrack); + read_unlock_bh(&nf_conntrack_lock); + + return h != NULL; +} + +/* There's a small race here where we may free a just-assured + connection. Too bad: we're in trouble anyway. */ +static inline int unreplied(const struct nf_conntrack_tuple_hash *i) +{ + return !(test_bit(IPS_ASSURED_BIT, + &nf_ct_tuplehash_to_ctrack(i)->status)); +} + +static int early_drop(struct list_head *chain) +{ + /* Traverse backwards: gives us oldest, which is roughly LRU */ + struct nf_conntrack_tuple_hash *h; + struct nf_conn *ct = NULL; + int dropped = 0; + + read_lock_bh(&nf_conntrack_lock); + h = LIST_FIND_B(chain, unreplied, struct nf_conntrack_tuple_hash *); + if (h) { + ct = nf_ct_tuplehash_to_ctrack(h); + atomic_inc(&ct->ct_general.use); + } + read_unlock_bh(&nf_conntrack_lock); + + if (!ct) + return dropped; + + if (del_timer(&ct->timeout)) { + death_by_timeout((unsigned long)ct); + dropped = 1; + NF_CT_STAT_INC(early_drop); + } + nf_ct_put(ct); + return dropped; +} + +static inline int helper_cmp(const struct nf_conntrack_helper *i, + const struct nf_conntrack_tuple *rtuple) +{ + return nf_ct_tuple_mask_cmp(rtuple, &i->tuple, &i->mask); +} + +static struct nf_conntrack_helper * +nf_ct_find_helper(const struct nf_conntrack_tuple *tuple) +{ + return LIST_FIND(&helpers, helper_cmp, + struct nf_conntrack_helper *, + tuple); +} + +static struct nf_conn * +__nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_tuple *repl, + const struct nf_conntrack_l3proto *l3proto) +{ + struct nf_conn *conntrack = NULL; + u_int32_t features = 0; + + if (!nf_conntrack_hash_rnd_initted) { + get_random_bytes(&nf_conntrack_hash_rnd, 4); + nf_conntrack_hash_rnd_initted = 1; + } + + if (nf_conntrack_max + && atomic_read(&nf_conntrack_count) >= nf_conntrack_max) { + unsigned int hash = hash_conntrack(orig); + /* Try dropping from this hash chain. */ + if (!early_drop(&nf_conntrack_hash[hash])) { + if (net_ratelimit()) + printk(KERN_WARNING + "nf_conntrack: table full, dropping" + " packet.\n"); + return ERR_PTR(-ENOMEM); + } + } + + /* find features needed by this conntrack. */ + features = l3proto->get_features(orig); + read_lock_bh(&nf_conntrack_lock); + if (nf_ct_find_helper(repl) != NULL) + features |= NF_CT_F_HELP; + read_unlock_bh(&nf_conntrack_lock); + + DEBUGP("nf_conntrack_alloc: features=0x%x\n", features); + + read_lock_bh(&nf_ct_cache_lock); + + if (!nf_ct_cache[features].use) { + DEBUGP("nf_conntrack_alloc: not supported features = 0x%x\n", + features); + goto out; + } + + conntrack = kmem_cache_alloc(nf_ct_cache[features].cachep, GFP_ATOMIC); + if (conntrack == NULL) { + DEBUGP("nf_conntrack_alloc: Can't alloc conntrack from cache\n"); + goto out; + } + + memset(conntrack, 0, nf_ct_cache[features].size); + conntrack->features = features; + if (nf_ct_cache[features].init_conntrack && + nf_ct_cache[features].init_conntrack(conntrack, features) < 0) { + DEBUGP("nf_conntrack_alloc: failed to init\n"); + kmem_cache_free(nf_ct_cache[features].cachep, conntrack); + conntrack = NULL; + goto out; + } + + atomic_set(&conntrack->ct_general.use, 1); + conntrack->ct_general.destroy = destroy_conntrack; + conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple = *orig; + conntrack->tuplehash[IP_CT_DIR_REPLY].tuple = *repl; + /* Don't set timer yet: wait for confirmation */ + init_timer(&conntrack->timeout); + conntrack->timeout.data = (unsigned long)conntrack; + conntrack->timeout.function = death_by_timeout; + + atomic_inc(&nf_conntrack_count); +out: + read_unlock_bh(&nf_ct_cache_lock); + return conntrack; +} + +struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_tuple *repl) +{ + struct nf_conntrack_l3proto *l3proto; + + l3proto = nf_ct_find_l3proto(orig->src.l3num); + return __nf_conntrack_alloc(orig, repl, l3proto); +} + +void nf_conntrack_free(struct nf_conn *conntrack) +{ + u_int32_t features = conntrack->features; + NF_CT_ASSERT(features >= NF_CT_F_BASIC && features < NF_CT_F_NUM); + DEBUGP("nf_conntrack_free: features = 0x%x, conntrack=%p\n", features, + conntrack); + kmem_cache_free(nf_ct_cache[features].cachep, conntrack); + atomic_dec(&nf_conntrack_count); +} + +/* Allocate a new conntrack: we return -ENOMEM if classification + failed due to stress. Otherwise it really is unclassifiable. */ +static struct nf_conntrack_tuple_hash * +init_conntrack(const struct nf_conntrack_tuple *tuple, + struct nf_conntrack_l3proto *l3proto, + struct nf_conntrack_protocol *protocol, + struct sk_buff *skb, + unsigned int dataoff) +{ + struct nf_conn *conntrack; + struct nf_conntrack_tuple repl_tuple; + struct nf_conntrack_expect *exp; + + if (!nf_ct_invert_tuple(&repl_tuple, tuple, l3proto, protocol)) { + DEBUGP("Can't invert tuple.\n"); + return NULL; + } + + conntrack = __nf_conntrack_alloc(tuple, &repl_tuple, l3proto); + if (conntrack == NULL || IS_ERR(conntrack)) { + DEBUGP("Can't allocate conntrack.\n"); + return (struct nf_conntrack_tuple_hash *)conntrack; + } + + if (!protocol->new(conntrack, skb, dataoff)) { + nf_conntrack_free(conntrack); + DEBUGP("init conntrack: can't track with proto module\n"); + return NULL; + } + + write_lock_bh(&nf_conntrack_lock); + exp = find_expectation(tuple); + + if (exp) { + DEBUGP("conntrack: expectation arrives ct=%p exp=%p\n", + conntrack, exp); + /* Welcome, Mr. Bond. We've been expecting you... */ + __set_bit(IPS_EXPECTED_BIT, &conntrack->status); + conntrack->master = exp->master; +#ifdef CONFIG_NF_CONNTRACK_MARK + conntrack->mark = exp->master->mark; +#endif + nf_conntrack_get(&conntrack->master->ct_general); + NF_CT_STAT_INC(expect_new); + } else { + conntrack->helper = nf_ct_find_helper(&repl_tuple); + + NF_CT_STAT_INC(new); + } + + /* Overload tuple linked list to put us in unconfirmed list. */ + list_add(&conntrack->tuplehash[IP_CT_DIR_ORIGINAL].list, &unconfirmed); + + write_unlock_bh(&nf_conntrack_lock); + + if (exp) { + if (exp->expectfn) + exp->expectfn(conntrack, exp); + nf_conntrack_expect_put(exp); + } + + return &conntrack->tuplehash[IP_CT_DIR_ORIGINAL]; +} + +/* On success, returns conntrack ptr, sets skb->nfct and ctinfo */ +static inline struct nf_conn * +resolve_normal_ct(struct sk_buff *skb, + unsigned int dataoff, + u_int16_t l3num, + u_int8_t protonum, + struct nf_conntrack_l3proto *l3proto, + struct nf_conntrack_protocol *proto, + int *set_reply, + enum ip_conntrack_info *ctinfo) +{ + struct nf_conntrack_tuple tuple; + struct nf_conntrack_tuple_hash *h; + struct nf_conn *ct; + + if (!nf_ct_get_tuple(skb, (unsigned int)(skb->nh.raw - skb->data), + dataoff, l3num, protonum, &tuple, l3proto, + proto)) { + DEBUGP("resolve_normal_ct: Can't get tuple\n"); + return NULL; + } + + /* look for tuple match */ + h = nf_conntrack_find_get(&tuple, NULL); + if (!h) { + h = init_conntrack(&tuple, l3proto, proto, skb, dataoff); + if (!h) + return NULL; + if (IS_ERR(h)) + return (void *)h; + } + ct = nf_ct_tuplehash_to_ctrack(h); + + /* It exists; we have (non-exclusive) reference. */ + if (NF_CT_DIRECTION(h) == IP_CT_DIR_REPLY) { + *ctinfo = IP_CT_ESTABLISHED + IP_CT_IS_REPLY; + /* Please set reply bit if this packet OK */ + *set_reply = 1; + } else { + /* Once we've had two way comms, always ESTABLISHED. */ + if (test_bit(IPS_SEEN_REPLY_BIT, &ct->status)) { + DEBUGP("nf_conntrack_in: normal packet for %p\n", ct); + *ctinfo = IP_CT_ESTABLISHED; + } else if (test_bit(IPS_EXPECTED_BIT, &ct->status)) { + DEBUGP("nf_conntrack_in: related packet for %p\n", ct); + *ctinfo = IP_CT_RELATED; + } else { + DEBUGP("nf_conntrack_in: new packet for %p\n", ct); + *ctinfo = IP_CT_NEW; + } + *set_reply = 0; + } + skb->nfct = &ct->ct_general; + skb->nfctinfo = *ctinfo; + return ct; +} + +unsigned int +nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff **pskb) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + struct nf_conntrack_l3proto *l3proto; + struct nf_conntrack_protocol *proto; + unsigned int dataoff; + u_int8_t protonum; + int set_reply = 0; + int ret; + + /* Previously seen (loopback or untracked)? Ignore. */ + if ((*pskb)->nfct) { + NF_CT_STAT_INC(ignore); + return NF_ACCEPT; + } + + l3proto = nf_ct_find_l3proto((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); + + /* It may be an special packet, error, unclean... + * inverse of the return code tells to the netfilter + * core what to do with the packet. */ + if (proto->error != NULL && + (ret = proto->error(*pskb, dataoff, &ctinfo, pf, hooknum)) <= 0) { + NF_CT_STAT_INC(error); + NF_CT_STAT_INC(invalid); + return -ret; + } + + ct = resolve_normal_ct(*pskb, dataoff, pf, protonum, l3proto, proto, + &set_reply, &ctinfo); + if (!ct) { + /* Not valid part of a connection */ + NF_CT_STAT_INC(invalid); + return NF_ACCEPT; + } + + if (IS_ERR(ct)) { + /* Too stressed to deal. */ + NF_CT_STAT_INC(drop); + return NF_DROP; + } + + NF_CT_ASSERT((*pskb)->nfct); + + ret = proto->packet(ct, *pskb, dataoff, ctinfo, pf, hooknum); + if (ret < 0) { + /* Invalid: inverse of the return code tells + * the netfilter core what to do */ + DEBUGP("nf_conntrack_in: Can't track with proto module\n"); + nf_conntrack_put((*pskb)->nfct); + (*pskb)->nfct = NULL; + NF_CT_STAT_INC(invalid); + return -ret; + } + + if (set_reply && !test_and_set_bit(IPS_SEEN_REPLY_BIT, &ct->status)) + nf_conntrack_event_cache(IPCT_STATUS, *pskb); + + return ret; +} + +int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, + 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)); +} + +/* Would two expected things clash? */ +static inline int expect_clash(const struct nf_conntrack_expect *a, + const struct nf_conntrack_expect *b) +{ + /* Part covered by intersection of masks must be unequal, + otherwise they clash */ + struct nf_conntrack_tuple intersect_mask; + int count; + + intersect_mask.src.l3num = a->mask.src.l3num & b->mask.src.l3num; + intersect_mask.src.u.all = a->mask.src.u.all & b->mask.src.u.all; + intersect_mask.dst.u.all = a->mask.dst.u.all & b->mask.dst.u.all; + intersect_mask.dst.protonum = a->mask.dst.protonum + & b->mask.dst.protonum; + + for (count = 0; count < NF_CT_TUPLE_L3SIZE; count++){ + intersect_mask.src.u3.all[count] = + a->mask.src.u3.all[count] & b->mask.src.u3.all[count]; + } + + for (count = 0; count < NF_CT_TUPLE_L3SIZE; count++){ + intersect_mask.dst.u3.all[count] = + a->mask.dst.u3.all[count] & b->mask.dst.u3.all[count]; + } + + return nf_ct_tuple_mask_cmp(&a->tuple, &b->tuple, &intersect_mask); +} + +static inline int expect_matches(const struct nf_conntrack_expect *a, + const struct nf_conntrack_expect *b) +{ + return a->master == b->master + && nf_ct_tuple_equal(&a->tuple, &b->tuple) + && nf_ct_tuple_equal(&a->mask, &b->mask); +} + +/* Generally a bad idea to call this: could have matched already. */ +void nf_conntrack_unexpect_related(struct nf_conntrack_expect *exp) +{ + struct nf_conntrack_expect *i; + + write_lock_bh(&nf_conntrack_lock); + /* choose the the oldest expectation to evict */ + list_for_each_entry_reverse(i, &nf_conntrack_expect_list, list) { + if (expect_matches(i, exp) && del_timer(&i->timeout)) { + nf_ct_unlink_expect(i); + write_unlock_bh(&nf_conntrack_lock); + nf_conntrack_expect_put(i); + return; + } + } + write_unlock_bh(&nf_conntrack_lock); +} + +/* We don't increase the master conntrack refcount for non-fulfilled + * conntracks. During the conntrack destruction, the expectations are + * always killed before the conntrack itself */ +struct nf_conntrack_expect *nf_conntrack_expect_alloc(struct nf_conn *me) +{ + struct nf_conntrack_expect *new; + + new = kmem_cache_alloc(nf_conntrack_expect_cachep, GFP_ATOMIC); + if (!new) { + DEBUGP("expect_related: OOM allocating expect\n"); + return NULL; + } + new->master = me; + atomic_set(&new->use, 1); + return new; +} + +void nf_conntrack_expect_put(struct nf_conntrack_expect *exp) +{ + if (atomic_dec_and_test(&exp->use)) + kmem_cache_free(nf_conntrack_expect_cachep, exp); +} + +static void nf_conntrack_expect_insert(struct nf_conntrack_expect *exp) +{ + atomic_inc(&exp->use); + exp->master->expecting++; + list_add(&exp->list, &nf_conntrack_expect_list); + + init_timer(&exp->timeout); + exp->timeout.data = (unsigned long)exp; + exp->timeout.function = expectation_timed_out; + exp->timeout.expires = jiffies + exp->master->helper->timeout * HZ; + add_timer(&exp->timeout); + + atomic_inc(&exp->use); + NF_CT_STAT_INC(expect_create); +} + +/* Race with expectations being used means we could have none to find; OK. */ +static void evict_oldest_expect(struct nf_conn *master) +{ + struct nf_conntrack_expect *i; + + list_for_each_entry_reverse(i, &nf_conntrack_expect_list, list) { + if (i->master == master) { + if (del_timer(&i->timeout)) { + nf_ct_unlink_expect(i); + nf_conntrack_expect_put(i); + } + break; + } + } +} + +static inline int refresh_timer(struct nf_conntrack_expect *i) +{ + if (!del_timer(&i->timeout)) + return 0; + + i->timeout.expires = jiffies + i->master->helper->timeout*HZ; + add_timer(&i->timeout); + return 1; +} + +int nf_conntrack_expect_related(struct nf_conntrack_expect *expect) +{ + struct nf_conntrack_expect *i; + int ret; + + DEBUGP("nf_conntrack_expect_related %p\n", related_to); + DEBUGP("tuple: "); NF_CT_DUMP_TUPLE(&expect->tuple); + DEBUGP("mask: "); NF_CT_DUMP_TUPLE(&expect->mask); + + write_lock_bh(&nf_conntrack_lock); + list_for_each_entry(i, &nf_conntrack_expect_list, list) { + if (expect_matches(i, expect)) { + /* Refresh timer: if it's dying, ignore.. */ + if (refresh_timer(i)) { + ret = 0; + goto out; + } + } else if (expect_clash(i, expect)) { + ret = -EBUSY; + goto out; + } + } + /* Will be over limit? */ + if (expect->master->helper->max_expected && + expect->master->expecting >= expect->master->helper->max_expected) + evict_oldest_expect(expect->master); + + nf_conntrack_expect_insert(expect); + nf_conntrack_expect_event(IPEXP_NEW, expect); + ret = 0; +out: + write_unlock_bh(&nf_conntrack_lock); + return ret; +} + +/* Alter reply tuple (maybe alter helper). This is for NAT, and is + implicitly racy: see __nf_conntrack_confirm */ +void nf_conntrack_alter_reply(struct nf_conn *conntrack, + const struct nf_conntrack_tuple *newreply) +{ + write_lock_bh(&nf_conntrack_lock); + /* Should be unconfirmed, so not in hash table yet */ + NF_CT_ASSERT(!nf_ct_is_confirmed(conntrack)); + + DEBUGP("Altering reply tuple of %p to ", conntrack); + NF_CT_DUMP_TUPLE(newreply); + + conntrack->tuplehash[IP_CT_DIR_REPLY].tuple = *newreply; + if (!conntrack->master && conntrack->expecting == 0) + conntrack->helper = nf_ct_find_helper(newreply); + write_unlock_bh(&nf_conntrack_lock); +} + +int nf_conntrack_helper_register(struct nf_conntrack_helper *me) +{ + int ret; + BUG_ON(me->timeout == 0); + + ret = nf_conntrack_register_cache(NF_CT_F_HELP, "nf_conntrack:help", + sizeof(struct nf_conn) + + sizeof(union nf_conntrack_help) + + __alignof__(union nf_conntrack_help), + init_conntrack_for_helper); + if (ret < 0) { + printk(KERN_ERR "nf_conntrack_helper_reigster: Unable to create slab cache for conntracks\n"); + return ret; + } + write_lock_bh(&nf_conntrack_lock); + list_prepend(&helpers, me); + write_unlock_bh(&nf_conntrack_lock); + + return 0; +} + +static inline int unhelp(struct nf_conntrack_tuple_hash *i, + const struct nf_conntrack_helper *me) +{ + if (nf_ct_tuplehash_to_ctrack(i)->helper == me) { + nf_conntrack_event(IPCT_HELPER, nf_ct_tuplehash_to_ctrack(i)); + nf_ct_tuplehash_to_ctrack(i)->helper = NULL; + } + return 0; +} + +void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) +{ + unsigned int i; + struct nf_conntrack_expect *exp, *tmp; + + /* Need write lock here, to delete helper. */ + write_lock_bh(&nf_conntrack_lock); + LIST_DELETE(&helpers, me); + + /* Get rid of expectations */ + list_for_each_entry_safe(exp, tmp, &nf_conntrack_expect_list, list) { + if (exp->master->helper == me && del_timer(&exp->timeout)) { + nf_ct_unlink_expect(exp); + nf_conntrack_expect_put(exp); + } + } + + /* Get rid of expecteds, set helpers to NULL. */ + LIST_FIND_W(&unconfirmed, unhelp, struct nf_conntrack_tuple_hash*, me); + for (i = 0; i < nf_conntrack_htable_size; i++) + LIST_FIND_W(&nf_conntrack_hash[i], unhelp, + struct nf_conntrack_tuple_hash *, me); + write_unlock_bh(&nf_conntrack_lock); + + /* Someone could be still looking at the helper in a bh. */ + synchronize_net(); +} + +/* Refresh conntrack for this many jiffies and do accounting if do_acct is 1 */ +void __nf_ct_refresh_acct(struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + const struct sk_buff *skb, + unsigned long extra_jiffies, + int do_acct) +{ + int event = 0; + + NF_CT_ASSERT(ct->timeout.data == (unsigned long)ct); + NF_CT_ASSERT(skb); + + write_lock_bh(&nf_conntrack_lock); + + /* If not in hash table, timer will not be active yet */ + if (!nf_ct_is_confirmed(ct)) { + ct->timeout.expires = extra_jiffies; + event = IPCT_REFRESH; + } else { + /* Need del_timer for race avoidance (may already be dying). */ + if (del_timer(&ct->timeout)) { + ct->timeout.expires = jiffies + extra_jiffies; + add_timer(&ct->timeout); + event = IPCT_REFRESH; + } + } + +#ifdef CONFIG_NF_CT_ACCT + if (do_acct) { + ct->counters[CTINFO2DIR(ctinfo)].packets++; + ct->counters[CTINFO2DIR(ctinfo)].bytes += + skb->len - (unsigned int)(skb->nh.raw - skb->data); + if ((ct->counters[CTINFO2DIR(ctinfo)].packets & 0x80000000) + || (ct->counters[CTINFO2DIR(ctinfo)].bytes & 0x80000000)) + event |= IPCT_COUNTER_FILLING; + } +#endif + + write_unlock_bh(&nf_conntrack_lock); + + /* must be unlocked when calling event cache */ + if (event) + nf_conntrack_event_cache(event, skb); +} + +/* Used by ipt_REJECT and ip6t_REJECT. */ +void __nf_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + + /* This ICMP is in reverse direction to the packet which caused it */ + ct = nf_ct_get(skb, &ctinfo); + if (CTINFO2DIR(ctinfo) == IP_CT_DIR_ORIGINAL) + ctinfo = IP_CT_RELATED + IP_CT_IS_REPLY; + else + ctinfo = IP_CT_RELATED; + + /* Attach to new skbuff, and increment count */ + nskb->nfct = &ct->ct_general; + nskb->nfctinfo = ctinfo; + nf_conntrack_get(nskb->nfct); +} + +static inline int +do_iter(const struct nf_conntrack_tuple_hash *i, + int (*iter)(struct nf_conn *i, void *data), + void *data) +{ + return iter(nf_ct_tuplehash_to_ctrack(i), data); +} + +/* Bring out ya dead! */ +static struct nf_conntrack_tuple_hash * +get_next_corpse(int (*iter)(struct nf_conn *i, void *data), + void *data, unsigned int *bucket) +{ + struct nf_conntrack_tuple_hash *h = NULL; + + write_lock_bh(&nf_conntrack_lock); + for (; *bucket < nf_conntrack_htable_size; (*bucket)++) { + h = LIST_FIND_W(&nf_conntrack_hash[*bucket], do_iter, + struct nf_conntrack_tuple_hash *, iter, data); + if (h) + break; + } + if (!h) + h = LIST_FIND_W(&unconfirmed, do_iter, + struct nf_conntrack_tuple_hash *, iter, data); + if (h) + atomic_inc(&nf_ct_tuplehash_to_ctrack(h)->ct_general.use); + write_unlock_bh(&nf_conntrack_lock); + + return h; +} + +void +nf_ct_iterate_cleanup(int (*iter)(struct nf_conn *i, void *data), void *data) +{ + struct nf_conntrack_tuple_hash *h; + unsigned int bucket = 0; + + while ((h = get_next_corpse(iter, data, &bucket)) != NULL) { + struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h); + /* Time to push up daises... */ + if (del_timer(&ct->timeout)) + death_by_timeout((unsigned long)ct); + /* ... else the timer will get him soon. */ + + nf_ct_put(ct); + } +} + +static int kill_all(struct nf_conn *i, void *data) +{ + return 1; +} + +static void free_conntrack_hash(struct list_head *hash, int vmalloced, int size) +{ + if (vmalloced) + vfree(hash); + else + free_pages((unsigned long)hash, + get_order(sizeof(struct list_head) * size)); +} + +/* Mishearing the voices in his head, our hero wonders how he's + supposed to kill the mall. */ +void nf_conntrack_cleanup(void) +{ + int i; + + /* This makes sure all current packets have passed through + netfilter framework. Roll on, two-stage module + delete... */ + synchronize_net(); + + nf_ct_event_cache_flush(); + i_see_dead_people: + nf_ct_iterate_cleanup(kill_all, NULL); + if (atomic_read(&nf_conntrack_count) != 0) { + schedule(); + goto i_see_dead_people; + } + + for (i = 0; i < NF_CT_F_NUM; i++) { + if (nf_ct_cache[i].use == 0) + continue; + + NF_CT_ASSERT(nf_ct_cache[i].use == 1); + nf_ct_cache[i].use = 1; + nf_conntrack_unregister_cache(i); + } + kmem_cache_destroy(nf_conntrack_expect_cachep); + free_conntrack_hash(nf_conntrack_hash, nf_conntrack_vmalloc, + nf_conntrack_htable_size); +} + +static struct list_head *alloc_hashtable(int size, int *vmalloced) +{ + struct list_head *hash; + unsigned int i; + + *vmalloced = 0; + hash = (void*)__get_free_pages(GFP_KERNEL, + get_order(sizeof(struct list_head) + * size)); + if (!hash) { + *vmalloced = 1; + printk(KERN_WARNING "nf_conntrack: falling back to vmalloc.\n"); + hash = vmalloc(sizeof(struct list_head) * size); + } + + if (hash) + for (i = 0; i < size; i++) + INIT_LIST_HEAD(&hash[i]); + + return hash; +} + +int set_hashsize(const char *val, struct kernel_param *kp) +{ + int i, bucket, hashsize, vmalloced; + int old_vmalloced, old_size; + int rnd; + struct list_head *hash, *old_hash; + struct nf_conntrack_tuple_hash *h; + + /* On boot, we can set this without any fancy locking. */ + if (!nf_conntrack_htable_size) + return param_set_uint(val, kp); + + hashsize = simple_strtol(val, NULL, 0); + if (!hashsize) + return -EINVAL; + + hash = alloc_hashtable(hashsize, &vmalloced); + if (!hash) + return -ENOMEM; + + /* We have to rehahs for the new table anyway, so we also can + * use a newrandom seed */ + get_random_bytes(&rnd, 4); + + write_lock_bh(&nf_conntrack_lock); + for (i = 0; i < nf_conntrack_htable_size; i++) { + while (!list_empty(&nf_conntrack_hash[i])) { + h = list_entry(nf_conntrack_hash[i].next, + struct nf_conntrack_tuple_hash, list); + list_del(&h->list); + bucket = __hash_conntrack(&h->tuple, hashsize, rnd); + list_add_tail(&h->list, &hash[bucket]); + } + } + old_size = nf_conntrack_htable_size; + old_vmalloced = nf_conntrack_vmalloc; + old_hash = nf_conntrack_hash; + + nf_conntrack_htable_size = hashsize; + nf_conntrack_vmalloc = vmalloced; + nf_conntrack_hash = hash; + nf_conntrack_hash_rnd = rnd; + write_unlock_bh(&nf_conntrack_lock); + + free_conntrack_hash(old_hash, old_vmalloced, old_size); + return 0; +} + +module_param_call(hashsize, set_hashsize, param_get_uint, + &nf_conntrack_htable_size, 0600); + +int __init nf_conntrack_init(void) +{ + unsigned int i; + int ret; + + /* Idea from tcp.c: use 1/16384 of memory. On i386: 32MB + * machine has 256 buckets. >= 1GB machines have 8192 buckets. */ + if (!nf_conntrack_htable_size) { + nf_conntrack_htable_size + = (((num_physpages << PAGE_SHIFT) / 16384) + / sizeof(struct list_head)); + if (num_physpages > (1024 * 1024 * 1024 / PAGE_SIZE)) + nf_conntrack_htable_size = 8192; + if (nf_conntrack_htable_size < 16) + nf_conntrack_htable_size = 16; + } + nf_conntrack_max = 8 * nf_conntrack_htable_size; + + printk("nf_conntrack version %s (%u buckets, %d max)\n", + NF_CONNTRACK_VERSION, nf_conntrack_htable_size, + nf_conntrack_max); + + nf_conntrack_hash = alloc_hashtable(nf_conntrack_htable_size, + &nf_conntrack_vmalloc); + if (!nf_conntrack_hash) { + printk(KERN_ERR "Unable to create nf_conntrack_hash\n"); + goto err_out; + } + + ret = nf_conntrack_register_cache(NF_CT_F_BASIC, "nf_conntrack:basic", + sizeof(struct nf_conn), NULL); + if (ret < 0) { + printk(KERN_ERR "Unable to create nf_conn slab cache\n"); + goto err_free_hash; + } + + nf_conntrack_expect_cachep = kmem_cache_create("nf_conntrack_expect", + sizeof(struct nf_conntrack_expect), + 0, 0, NULL, NULL); + if (!nf_conntrack_expect_cachep) { + printk(KERN_ERR "Unable to create nf_expect slab cache\n"); + goto err_free_conntrack_slab; + } + + /* Don't NEED lock here, but good form anyway. */ + write_lock_bh(&nf_conntrack_lock); + for (i = 0; i < PF_MAX; i++) + nf_ct_l3protos[i] = &nf_conntrack_generic_l3proto; + write_unlock_bh(&nf_conntrack_lock); + + /* Set up fake conntrack: + - to never be deleted, not in any hashes */ + atomic_set(&nf_conntrack_untracked.ct_general.use, 1); + /* - and look it like as a confirmed connection */ + set_bit(IPS_CONFIRMED_BIT, &nf_conntrack_untracked.status); + + return ret; + +err_free_conntrack_slab: + nf_conntrack_unregister_cache(NF_CT_F_BASIC); +err_free_hash: + free_conntrack_hash(nf_conntrack_hash, nf_conntrack_vmalloc, + nf_conntrack_htable_size); +err_out: + return -ENOMEM; +} diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c new file mode 100644 index 000000000000..65080e269f27 --- /dev/null +++ b/net/netfilter/nf_conntrack_ftp.c @@ -0,0 +1,698 @@ +/* FTP extension for connection tracking. */ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team + * (C) 2003,2004 USAGI/WIDE Project + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI + * - enable working with Layer 3 protocol independent connection tracking. + * - track EPRT and EPSV commands with IPv6 address. + * + * Derived from net/ipv4/netfilter/ip_conntrack_ftp.c + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Rusty Russell "); +MODULE_DESCRIPTION("ftp connection tracking helper"); + +/* This is slow, but it's simple. --RR */ +static char *ftp_buffer; + +static DEFINE_SPINLOCK(nf_ftp_lock); + +#define MAX_PORTS 8 +static u_int16_t ports[MAX_PORTS]; +static unsigned int ports_c; +module_param_array(ports, ushort, &ports_c, 0400); + +static int loose; +module_param(loose, int, 0600); + +unsigned int (*nf_nat_ftp_hook)(struct sk_buff **pskb, + enum ip_conntrack_info ctinfo, + enum ip_ct_ftp_type type, + unsigned int matchoff, + unsigned int matchlen, + struct nf_conntrack_expect *exp, + u32 *seq); +EXPORT_SYMBOL_GPL(nf_nat_ftp_hook); + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +static int try_rfc959(const char *, size_t, struct nf_conntrack_man *, char); +static int try_eprt(const char *, size_t, struct nf_conntrack_man *, char); +static int try_epsv_response(const char *, size_t, struct nf_conntrack_man *, + char); + +static struct ftp_search { + enum ip_conntrack_dir dir; + const char *pattern; + size_t plen; + char skip; + char term; + enum ip_ct_ftp_type ftptype; + int (*getnum)(const char *, size_t, struct nf_conntrack_man *, char); +} search[] = { + { + IP_CT_DIR_ORIGINAL, + "PORT", sizeof("PORT") - 1, ' ', '\r', + IP_CT_FTP_PORT, + try_rfc959, + }, + { + IP_CT_DIR_REPLY, + "227 ", sizeof("227 ") - 1, '(', ')', + IP_CT_FTP_PASV, + try_rfc959, + }, + { + IP_CT_DIR_ORIGINAL, + "EPRT", sizeof("EPRT") - 1, ' ', '\r', + IP_CT_FTP_EPRT, + try_eprt, + }, + { + IP_CT_DIR_REPLY, + "229 ", sizeof("229 ") - 1, '(', ')', + IP_CT_FTP_EPSV, + try_epsv_response, + }, +}; + +/* This code is based on inet_pton() in glibc-2.2.4 */ +static int +get_ipv6_addr(const char *src, size_t dlen, struct in6_addr *dst, u_int8_t term) +{ + static const char xdigits[] = "0123456789abcdef"; + u_int8_t tmp[16], *tp, *endp, *colonp; + int ch, saw_xdigit; + u_int32_t val; + size_t clen = 0; + + tp = memset(tmp, '\0', sizeof(tmp)); + endp = tp + sizeof(tmp); + colonp = NULL; + + /* Leading :: requires some special handling. */ + if (*src == ':'){ + if (*++src != ':') { + DEBUGP("invalid \":\" at the head of addr\n"); + return 0; + } + clen++; + } + + saw_xdigit = 0; + val = 0; + while ((clen < dlen) && (*src != term)) { + const char *pch; + + ch = tolower(*src++); + clen++; + + pch = strchr(xdigits, ch); + if (pch != NULL) { + val <<= 4; + val |= (pch - xdigits); + if (val > 0xffff) + return 0; + + saw_xdigit = 1; + continue; + } + if (ch != ':') { + DEBUGP("get_ipv6_addr: invalid char. \'%c\'\n", ch); + return 0; + } + + if (!saw_xdigit) { + if (colonp) { + DEBUGP("invalid location of \"::\".\n"); + return 0; + } + colonp = tp; + continue; + } else if (*src == term) { + DEBUGP("trancated IPv6 addr\n"); + return 0; + } + + if (tp + 2 > endp) + return 0; + *tp++ = (u_int8_t) (val >> 8) & 0xff; + *tp++ = (u_int8_t) val & 0xff; + + saw_xdigit = 0; + val = 0; + continue; + } + if (saw_xdigit) { + if (tp + 2 > endp) + return 0; + *tp++ = (u_int8_t) (val >> 8) & 0xff; + *tp++ = (u_int8_t) val & 0xff; + } + if (colonp != NULL) { + /* + * Since some memmove()'s erroneously fail to handle + * overlapping regions, we'll do the shift by hand. + */ + const int n = tp - colonp; + int i; + + if (tp == endp) + return 0; + + for (i = 1; i <= n; i++) { + endp[- i] = colonp[n - i]; + colonp[n - i] = 0; + } + tp = endp; + } + if (tp != endp || (*src != term)) + return 0; + + memcpy(dst->s6_addr, tmp, sizeof(dst->s6_addr)); + return clen; +} + +static int try_number(const char *data, size_t dlen, u_int32_t array[], + int array_size, char sep, char term) +{ + u_int32_t i, len; + + memset(array, 0, sizeof(array[0])*array_size); + + /* Keep data pointing at next char. */ + for (i = 0, len = 0; len < dlen && i < array_size; len++, data++) { + if (*data >= '0' && *data <= '9') { + array[i] = array[i]*10 + *data - '0'; + } + else if (*data == sep) + i++; + else { + /* Unexpected character; true if it's the + terminator and we're finished. */ + if (*data == term && i == array_size - 1) + return len; + + DEBUGP("Char %u (got %u nums) `%u' unexpected\n", + len, i, *data); + return 0; + } + } + DEBUGP("Failed to fill %u numbers separated by %c\n", array_size, sep); + + return 0; +} + +/* Returns 0, or length of numbers: 192,168,1,1,5,6 */ +static int try_rfc959(const char *data, size_t dlen, + struct nf_conntrack_man *cmd, char term) +{ + int length; + u_int32_t array[6]; + + length = try_number(data, dlen, array, 6, ',', term); + if (length == 0) + return 0; + + cmd->u3.ip = htonl((array[0] << 24) | (array[1] << 16) | + (array[2] << 8) | array[3]); + cmd->u.tcp.port = htons((array[4] << 8) | array[5]); + return length; +} + +/* Grab port: number up to delimiter */ +static int get_port(const char *data, int start, size_t dlen, char delim, + u_int16_t *port) +{ + u_int16_t tmp_port = 0; + int i; + + for (i = start; i < dlen; i++) { + /* Finished? */ + if (data[i] == delim) { + if (tmp_port == 0) + break; + *port = htons(tmp_port); + DEBUGP("get_port: return %d\n", tmp_port); + return i + 1; + } + else if (data[i] >= '0' && data[i] <= '9') + tmp_port = tmp_port*10 + data[i] - '0'; + else { /* Some other crap */ + DEBUGP("get_port: invalid char.\n"); + break; + } + } + return 0; +} + +/* Returns 0, or length of numbers: |1|132.235.1.2|6275| or |2|3ffe::1|6275| */ +static int try_eprt(const char *data, size_t dlen, struct nf_conntrack_man *cmd, + char term) +{ + char delim; + int length; + + /* First character is delimiter, then "1" for IPv4 or "2" for IPv6, + then delimiter again. */ + if (dlen <= 3) { + DEBUGP("EPRT: too short\n"); + return 0; + } + delim = data[0]; + if (isdigit(delim) || delim < 33 || delim > 126 || data[2] != delim) { + DEBUGP("try_eprt: invalid delimitter.\n"); + return 0; + } + + if ((cmd->l3num == PF_INET && data[1] != '1') || + (cmd->l3num == PF_INET6 && data[1] != '2')) { + DEBUGP("EPRT: invalid protocol number.\n"); + return 0; + } + + DEBUGP("EPRT: Got %c%c%c\n", delim, data[1], delim); + + if (data[1] == '1') { + u_int32_t array[4]; + + /* Now we have IP address. */ + length = try_number(data + 3, dlen - 3, array, 4, '.', delim); + if (length != 0) + cmd->u3.ip = htonl((array[0] << 24) | (array[1] << 16) + | (array[2] << 8) | array[3]); + } else { + /* Now we have IPv6 address. */ + length = get_ipv6_addr(data + 3, dlen - 3, + (struct in6_addr *)cmd->u3.ip6, delim); + } + + if (length == 0) + return 0; + DEBUGP("EPRT: Got IP address!\n"); + /* Start offset includes initial "|1|", and trailing delimiter */ + return get_port(data, 3 + length + 1, dlen, delim, &cmd->u.tcp.port); +} + +/* Returns 0, or length of numbers: |||6446| */ +static int try_epsv_response(const char *data, size_t dlen, + struct nf_conntrack_man *cmd, char term) +{ + char delim; + + /* Three delimiters. */ + if (dlen <= 3) return 0; + delim = data[0]; + if (isdigit(delim) || delim < 33 || delim > 126 + || data[1] != delim || data[2] != delim) + return 0; + + return get_port(data, 3, dlen, delim, &cmd->u.tcp.port); +} + +/* Return 1 for match, 0 for accept, -1 for partial. */ +static int find_pattern(const char *data, size_t dlen, + const char *pattern, size_t plen, + char skip, char term, + unsigned int *numoff, + unsigned int *numlen, + struct nf_conntrack_man *cmd, + int (*getnum)(const char *, size_t, + struct nf_conntrack_man *, char)) +{ + size_t i; + + DEBUGP("find_pattern `%s': dlen = %u\n", pattern, dlen); + if (dlen == 0) + return 0; + + if (dlen <= plen) { + /* Short packet: try for partial? */ + if (strnicmp(data, pattern, dlen) == 0) + return -1; + else return 0; + } + + if (strnicmp(data, pattern, plen) != 0) { +#if 0 + size_t i; + + DEBUGP("ftp: string mismatch\n"); + for (i = 0; i < plen; i++) { + DEBUGP("ftp:char %u `%c'(%u) vs `%c'(%u)\n", + i, data[i], data[i], + pattern[i], pattern[i]); + } +#endif + return 0; + } + + DEBUGP("Pattern matches!\n"); + /* Now we've found the constant string, try to skip + to the 'skip' character */ + for (i = plen; data[i] != skip; i++) + if (i == dlen - 1) return -1; + + /* Skip over the last character */ + i++; + + DEBUGP("Skipped up to `%c'!\n", skip); + + *numoff = i; + *numlen = getnum(data + i, dlen - i, cmd, term); + if (!*numlen) + return -1; + + DEBUGP("Match succeeded!\n"); + return 1; +} + +/* Look up to see if we're just after a \n. */ +static int find_nl_seq(u32 seq, const struct ip_ct_ftp_master *info, int dir) +{ + unsigned int i; + + for (i = 0; i < info->seq_aft_nl_num[dir]; i++) + if (info->seq_aft_nl[dir][i] == seq) + return 1; + return 0; +} + +/* We don't update if it's older than what we have. */ +static void update_nl_seq(u32 nl_seq, struct ip_ct_ftp_master *info, int dir, + struct sk_buff *skb) +{ + unsigned int i, oldest = NUM_SEQ_TO_REMEMBER; + + /* Look for oldest: if we find exact match, we're done. */ + for (i = 0; i < info->seq_aft_nl_num[dir]; i++) { + if (info->seq_aft_nl[dir][i] == nl_seq) + return; + + if (oldest == info->seq_aft_nl_num[dir] + || before(info->seq_aft_nl[dir][i], oldest)) + oldest = i; + } + + if (info->seq_aft_nl_num[dir] < NUM_SEQ_TO_REMEMBER) { + info->seq_aft_nl[dir][info->seq_aft_nl_num[dir]++] = nl_seq; + nf_conntrack_event_cache(IPCT_HELPINFO_VOLATILE, skb); + } else if (oldest != NUM_SEQ_TO_REMEMBER) { + info->seq_aft_nl[dir][oldest] = nl_seq; + nf_conntrack_event_cache(IPCT_HELPINFO_VOLATILE, skb); + } +} + +static int help(struct sk_buff **pskb, + unsigned int protoff, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo) +{ + unsigned int dataoff, datalen; + struct tcphdr _tcph, *th; + char *fb_ptr; + int ret; + u32 seq; + int dir = CTINFO2DIR(ctinfo); + unsigned int matchlen, matchoff; + struct ip_ct_ftp_master *ct_ftp_info = &ct->help->ct_ftp_info; + struct nf_conntrack_expect *exp; + struct nf_conntrack_man cmd = {}; + + unsigned int i; + int found = 0, ends_in_nl; + + /* Until there's been traffic both ways, don't look in packets. */ + if (ctinfo != IP_CT_ESTABLISHED + && ctinfo != IP_CT_ESTABLISHED+IP_CT_IS_REPLY) { + DEBUGP("ftp: Conntrackinfo = %u\n", ctinfo); + return NF_ACCEPT; + } + + th = skb_header_pointer(*pskb, protoff, sizeof(_tcph), &_tcph); + if (th == NULL) + return NF_ACCEPT; + + dataoff = protoff + th->doff * 4; + /* No data? */ + if (dataoff >= (*pskb)->len) { + DEBUGP("ftp: dataoff(%u) >= skblen(%u)\n", dataoff, + (*pskb)->len); + return NF_ACCEPT; + } + datalen = (*pskb)->len - dataoff; + + spin_lock_bh(&nf_ftp_lock); + fb_ptr = skb_header_pointer(*pskb, dataoff, datalen, ftp_buffer); + BUG_ON(fb_ptr == NULL); + + ends_in_nl = (fb_ptr[datalen - 1] == '\n'); + seq = ntohl(th->seq) + datalen; + + /* Look up to see if we're just after a \n. */ + if (!find_nl_seq(ntohl(th->seq), ct_ftp_info, dir)) { + /* Now if this ends in \n, update ftp info. */ + DEBUGP("nf_conntrack_ftp_help: wrong seq pos %s(%u) or %s(%u)\n", + ct_ftp_info->seq_aft_nl_num[dir] > 0 ? "" : "(UNSET)", + ct_ftp_info->seq_aft_nl[dir][0], + ct_ftp_info->seq_aft_nl_num[dir] > 1 ? "" : "(UNSET)", + ct_ftp_info->seq_aft_nl[dir][1]); + ret = NF_ACCEPT; + goto out_update_nl; + } + + /* Initialize IP/IPv6 addr to expected address (it's not mentioned + in EPSV responses) */ + cmd.l3num = ct->tuplehash[dir].tuple.src.l3num; + memcpy(cmd.u3.all, &ct->tuplehash[dir].tuple.src.u3.all, + sizeof(cmd.u3.all)); + + for (i = 0; i < ARRAY_SIZE(search); i++) { + if (search[i].dir != dir) continue; + + found = find_pattern(fb_ptr, datalen, + search[i].pattern, + search[i].plen, + search[i].skip, + search[i].term, + &matchoff, &matchlen, + &cmd, + search[i].getnum); + if (found) break; + } + if (found == -1) { + /* We don't usually drop packets. After all, this is + connection tracking, not packet filtering. + However, it is necessary for accurate tracking in + this case. */ + if (net_ratelimit()) + printk("conntrack_ftp: partial %s %u+%u\n", + search[i].pattern, + ntohl(th->seq), datalen); + ret = NF_DROP; + goto out; + } else if (found == 0) { /* No match */ + ret = NF_ACCEPT; + goto out_update_nl; + } + + DEBUGP("conntrack_ftp: match `%.*s' (%u bytes at %u)\n", + (int)matchlen, fb_ptr + matchoff, + matchlen, ntohl(th->seq) + matchoff); + + exp = nf_conntrack_expect_alloc(ct); + if (exp == NULL) { + ret = NF_DROP; + goto out; + } + + /* We refer to the reverse direction ("!dir") tuples here, + * because we're expecting something in the other direction. + * Doesn't matter unless NAT is happening. */ + exp->tuple.dst.u3 = ct->tuplehash[!dir].tuple.dst.u3; + + /* Update the ftp info */ + if ((cmd.l3num == ct->tuplehash[dir].tuple.src.l3num) && + memcmp(&cmd.u3.all, &ct->tuplehash[dir].tuple.src.u3.all, + sizeof(cmd.u3.all))) { + /* Enrico Scholz's passive FTP to partially RNAT'd ftp + server: it really wants us to connect to a + different IP address. Simply don't record it for + NAT. */ + if (cmd.l3num == PF_INET) { + DEBUGP("conntrack_ftp: NOT RECORDING: %u,%u,%u,%u != %u.%u.%u.%u\n", + NIPQUAD(cmd.u3.ip), + NIPQUAD(ct->tuplehash[dir].tuple.src.u3.ip)); + } else { + DEBUGP("conntrack_ftp: NOT RECORDING: %x:%x:%x:%x:%x:%x:%x:%x != %x:%x:%x:%x:%x:%x:%x:%x\n", + NIP6(*((struct in6_addr *)cmd.u3.ip6)), + NIP6(*((struct in6_addr *)ct->tuplehash[dir] + .tuple.src.u3.ip6))); + } + + /* Thanks to Cristiano Lincoln Mattos + for reporting this potential + problem (DMZ machines opening holes to internal + networks, or the packet filter itself). */ + if (!loose) { + ret = NF_ACCEPT; + goto out_put_expect; + } + memcpy(&exp->tuple.dst.u3, &cmd.u3.all, + sizeof(exp->tuple.dst.u3)); + } + + exp->tuple.src.u3 = ct->tuplehash[!dir].tuple.src.u3; + exp->tuple.src.l3num = cmd.l3num; + exp->tuple.src.u.tcp.port = 0; + exp->tuple.dst.u.tcp.port = cmd.u.tcp.port; + exp->tuple.dst.protonum = IPPROTO_TCP; + + exp->mask = (struct nf_conntrack_tuple) + { .src = { .l3num = 0xFFFF, + .u = { .tcp = { 0 }}, + }, + .dst = { .protonum = 0xFF, + .u = { .tcp = { 0xFFFF }}, + }, + }; + if (cmd.l3num == PF_INET) { + exp->mask.src.u3.ip = 0xFFFFFFFF; + exp->mask.dst.u3.ip = 0xFFFFFFFF; + } else { + memset(exp->mask.src.u3.ip6, 0xFF, + sizeof(exp->mask.src.u3.ip6)); + memset(exp->mask.dst.u3.ip6, 0xFF, + sizeof(exp->mask.src.u3.ip6)); + } + + exp->expectfn = NULL; + exp->flags = 0; + + /* Now, NAT might want to mangle the packet, and register the + * (possibly changed) expectation itself. */ + if (nf_nat_ftp_hook) + ret = nf_nat_ftp_hook(pskb, ctinfo, search[i].ftptype, + matchoff, matchlen, exp, &seq); + else { + /* Can't expect this? Best to drop packet now. */ + if (nf_conntrack_expect_related(exp) != 0) + ret = NF_DROP; + else + ret = NF_ACCEPT; + } + +out_put_expect: + nf_conntrack_expect_put(exp); + +out_update_nl: + /* Now if this ends in \n, update ftp info. Seq may have been + * adjusted by NAT code. */ + if (ends_in_nl) + update_nl_seq(seq, ct_ftp_info, dir, *pskb); + out: + spin_unlock_bh(&nf_ftp_lock); + return ret; +} + +static struct nf_conntrack_helper ftp[MAX_PORTS][2]; +static char ftp_names[MAX_PORTS][2][sizeof("ftp-65535")]; + +/* don't make this __exit, since it's called from __init ! */ +static void fini(void) +{ + int i, j; + for (i = 0; i < ports_c; i++) { + for (j = 0; j < 2; j++) { + if (ftp[i][j].me == NULL) + continue; + + DEBUGP("nf_ct_ftp: unregistering helper for pf: %d " + "port: %d\n", + ftp[i][j].tuple.src.l3num, ports[i]); + nf_conntrack_helper_unregister(&ftp[i][j]); + } + } + + kfree(ftp_buffer); +} + +static int __init init(void) +{ + int i, j = -1, ret = 0; + char *tmpname; + + ftp_buffer = kmalloc(65536, GFP_KERNEL); + if (!ftp_buffer) + return -ENOMEM; + + if (ports_c == 0) + ports[ports_c++] = FTP_PORT; + + /* FIXME should be configurable whether IPv4 and IPv6 FTP connections + are tracked or not - YK */ + for (i = 0; i < ports_c; i++) { + memset(&ftp[i], 0, sizeof(struct nf_conntrack_helper)); + + ftp[i][0].tuple.src.l3num = PF_INET; + ftp[i][1].tuple.src.l3num = PF_INET6; + for (j = 0; j < 2; j++) { + ftp[i][j].tuple.src.u.tcp.port = htons(ports[i]); + ftp[i][j].tuple.dst.protonum = IPPROTO_TCP; + ftp[i][j].mask.src.u.tcp.port = 0xFFFF; + ftp[i][j].mask.dst.protonum = 0xFF; + ftp[i][j].max_expected = 1; + ftp[i][j].timeout = 5 * 60; /* 5 Minutes */ + ftp[i][j].me = THIS_MODULE; + ftp[i][j].help = help; + tmpname = &ftp_names[i][j][0]; + if (ports[i] == FTP_PORT) + sprintf(tmpname, "ftp"); + else + sprintf(tmpname, "ftp-%d", ports[i]); + ftp[i][j].name = tmpname; + + DEBUGP("nf_ct_ftp: registering helper for pf: %d " + "port: %d\n", + ftp[i][j].tuple.src.l3num, ports[i]); + ret = nf_conntrack_helper_register(&ftp[i][j]); + if (ret) { + printk("nf_ct_ftp: failed to register helper " + " for pf: %d port: %d\n", + ftp[i][j].tuple.src.l3num, ports[i]); + fini(); + return ret; + } + } + } + + return 0; +} + +module_init(init); +module_exit(fini); diff --git a/net/netfilter/nf_conntrack_l3proto_generic.c b/net/netfilter/nf_conntrack_l3proto_generic.c new file mode 100644 index 000000000000..7de4f06c63c5 --- /dev/null +++ b/net/netfilter/nf_conntrack_l3proto_generic.c @@ -0,0 +1,98 @@ +/* + * (C) 2003,2004 USAGI/WIDE Project + * + * Based largely upon the original ip_conntrack code which + * had the following copyright information: + * + * (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Author: + * Yasuyuki Kozakai @USAGI + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +DECLARE_PER_CPU(struct nf_conntrack_stat, nf_conntrack_stat); + +static int generic_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, + struct nf_conntrack_tuple *tuple) +{ + memset(&tuple->src.u3, 0, sizeof(tuple->src.u3)); + memset(&tuple->dst.u3, 0, sizeof(tuple->dst.u3)); + + return 1; +} + +static int generic_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + memset(&tuple->src.u3, 0, sizeof(tuple->src.u3)); + memset(&tuple->dst.u3, 0, sizeof(tuple->dst.u3)); + + return 1; +} + +static int generic_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return 0; +} + +static int generic_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + return 0; +} + +static int +generic_prepare(struct sk_buff **pskb, unsigned int hooknum, + unsigned int *dataoff, u_int8_t *protonum) +{ + /* Never track !!! */ + return -NF_ACCEPT; +} + + +static u_int32_t generic_get_features(const struct nf_conntrack_tuple *tuple) + +{ + return NF_CT_F_BASIC; +} + +struct nf_conntrack_l3proto nf_conntrack_generic_l3proto = { + .l3proto = PF_UNSPEC, + .name = "unknown", + .pkt_to_tuple = generic_pkt_to_tuple, + .invert_tuple = generic_invert_tuple, + .print_tuple = generic_print_tuple, + .print_conntrack = generic_print_conntrack, + .prepare = generic_prepare, + .get_features = generic_get_features, + .me = THIS_MODULE, +}; diff --git a/net/netfilter/nf_conntrack_proto_generic.c b/net/netfilter/nf_conntrack_proto_generic.c new file mode 100644 index 000000000000..36425f6c833f --- /dev/null +++ b/net/netfilter/nf_conntrack_proto_generic.c @@ -0,0 +1,85 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI + * - enable working with L3 protocol independent connection tracking. + * + * Derived from net/ipv4/netfilter/ip_conntrack_proto_generic.c + */ + +#include +#include +#include +#include +#include + +unsigned long nf_ct_generic_timeout = 600*HZ; + +static int generic_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + tuple->src.u.all = 0; + tuple->dst.u.all = 0; + + return 1; +} + +static int generic_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + tuple->src.u.all = 0; + tuple->dst.u.all = 0; + + return 1; +} + +/* Print out the per-protocol part of the tuple. */ +static int generic_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return 0; +} + +/* Print out the private part of the conntrack. */ +static int generic_print_conntrack(struct seq_file *s, + const struct nf_conn *state) +{ + return 0; +} + +/* Returns verdict for packet, or -1 for invalid. */ +static int packet(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + nf_ct_refresh_acct(conntrack, ctinfo, skb, nf_ct_generic_timeout); + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int new(struct nf_conn *conntrack, const struct sk_buff *skb, + unsigned int dataoff) +{ + return 1; +} + +struct nf_conntrack_protocol nf_conntrack_generic_protocol = +{ + .l3proto = PF_UNSPEC, + .proto = 0, + .name = "unknown", + .pkt_to_tuple = generic_pkt_to_tuple, + .invert_tuple = generic_invert_tuple, + .print_tuple = generic_print_tuple, + .print_conntrack = generic_print_conntrack, + .packet = packet, + .new = new, +}; diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c new file mode 100644 index 000000000000..3a600f77b4e0 --- /dev/null +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -0,0 +1,670 @@ +/* + * Connection tracking protocol helper module for SCTP. + * + * SCTP is defined in RFC 2960. References to various sections in this code + * are to this RFC. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 17 Oct 2004: Yasuyuki Kozakai @USAGI + * - enable working with L3 protocol independent connection tracking. + * + * Derived from net/ipv4/ip_conntrack_sctp.c + */ + +/* + * Added support for proc manipulation of timeouts. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +#if 0 +#define DEBUGP(format, ...) printk(format, ## __VA_ARGS__) +#else +#define DEBUGP(format, args...) +#endif + +/* Protects conntrack->proto.sctp */ +static DEFINE_RWLOCK(sctp_lock); + +/* FIXME: Examine ipfilter's timeouts and conntrack transitions more + closely. They're more complex. --RR + + And so for me for SCTP :D -Kiran */ + +static const char *sctp_conntrack_names[] = { + "NONE", + "CLOSED", + "COOKIE_WAIT", + "COOKIE_ECHOED", + "ESTABLISHED", + "SHUTDOWN_SENT", + "SHUTDOWN_RECD", + "SHUTDOWN_ACK_SENT", +}; + +#define SECS * HZ +#define MINS * 60 SECS +#define HOURS * 60 MINS +#define DAYS * 24 HOURS + +static unsigned long nf_ct_sctp_timeout_closed = 10 SECS; +static unsigned long nf_ct_sctp_timeout_cookie_wait = 3 SECS; +static unsigned long nf_ct_sctp_timeout_cookie_echoed = 3 SECS; +static unsigned long nf_ct_sctp_timeout_established = 5 DAYS; +static unsigned long nf_ct_sctp_timeout_shutdown_sent = 300 SECS / 1000; +static unsigned long nf_ct_sctp_timeout_shutdown_recd = 300 SECS / 1000; +static unsigned long nf_ct_sctp_timeout_shutdown_ack_sent = 3 SECS; + +static unsigned long * sctp_timeouts[] += { NULL, /* SCTP_CONNTRACK_NONE */ + &nf_ct_sctp_timeout_closed, /* SCTP_CONNTRACK_CLOSED */ + &nf_ct_sctp_timeout_cookie_wait, /* SCTP_CONNTRACK_COOKIE_WAIT */ + &nf_ct_sctp_timeout_cookie_echoed, /* SCTP_CONNTRACK_COOKIE_ECHOED */ + &nf_ct_sctp_timeout_established, /* SCTP_CONNTRACK_ESTABLISHED */ + &nf_ct_sctp_timeout_shutdown_sent, /* SCTP_CONNTRACK_SHUTDOWN_SENT */ + &nf_ct_sctp_timeout_shutdown_recd, /* SCTP_CONNTRACK_SHUTDOWN_RECD */ + &nf_ct_sctp_timeout_shutdown_ack_sent /* SCTP_CONNTRACK_SHUTDOWN_ACK_SENT */ + }; + +#define sNO SCTP_CONNTRACK_NONE +#define sCL SCTP_CONNTRACK_CLOSED +#define sCW SCTP_CONNTRACK_COOKIE_WAIT +#define sCE SCTP_CONNTRACK_COOKIE_ECHOED +#define sES SCTP_CONNTRACK_ESTABLISHED +#define sSS SCTP_CONNTRACK_SHUTDOWN_SENT +#define sSR SCTP_CONNTRACK_SHUTDOWN_RECD +#define sSA SCTP_CONNTRACK_SHUTDOWN_ACK_SENT +#define sIV SCTP_CONNTRACK_MAX + +/* + These are the descriptions of the states: + +NOTE: These state names are tantalizingly similar to the states of an +SCTP endpoint. But the interpretation of the states is a little different, +considering that these are the states of the connection and not of an end +point. Please note the subtleties. -Kiran + +NONE - Nothing so far. +COOKIE WAIT - We have seen an INIT chunk in the original direction, or also + an INIT_ACK chunk in the reply direction. +COOKIE ECHOED - We have seen a COOKIE_ECHO chunk in the original direction. +ESTABLISHED - We have seen a COOKIE_ACK in the reply direction. +SHUTDOWN_SENT - We have seen a SHUTDOWN chunk in the original direction. +SHUTDOWN_RECD - We have seen a SHUTDOWN chunk in the reply directoin. +SHUTDOWN_ACK_SENT - We have seen a SHUTDOWN_ACK chunk in the direction opposite + to that of the SHUTDOWN chunk. +CLOSED - We have seen a SHUTDOWN_COMPLETE chunk in the direction of + the SHUTDOWN chunk. Connection is closed. +*/ + +/* TODO + - I have assumed that the first INIT is in the original direction. + This messes things when an INIT comes in the reply direction in CLOSED + state. + - Check the error type in the reply dir before transitioning from +cookie echoed to closed. + - Sec 5.2.4 of RFC 2960 + - Multi Homing support. +*/ + +/* SCTP conntrack state transitions */ +static enum sctp_conntrack sctp_conntracks[2][9][SCTP_CONNTRACK_MAX] = { + { +/* ORIGINAL */ +/* sNO, sCL, sCW, sCE, sES, sSS, sSR, sSA */ +/* init */ {sCW, sCW, sCW, sCE, sES, sSS, sSR, sSA}, +/* init_ack */ {sCL, sCL, sCW, sCE, sES, sSS, sSR, sSA}, +/* abort */ {sCL, sCL, sCL, sCL, sCL, sCL, sCL, sCL}, +/* shutdown */ {sCL, sCL, sCW, sCE, sSS, sSS, sSR, sSA}, +/* shutdown_ack */ {sSA, sCL, sCW, sCE, sES, sSA, sSA, sSA}, +/* error */ {sCL, sCL, sCW, sCE, sES, sSS, sSR, sSA},/* Cant have Stale cookie*/ +/* cookie_echo */ {sCL, sCL, sCE, sCE, sES, sSS, sSR, sSA},/* 5.2.4 - Big TODO */ +/* cookie_ack */ {sCL, sCL, sCW, sCE, sES, sSS, sSR, sSA},/* Cant come in orig dir */ +/* shutdown_comp*/ {sCL, sCL, sCW, sCE, sES, sSS, sSR, sCL} + }, + { +/* REPLY */ +/* sNO, sCL, sCW, sCE, sES, sSS, sSR, sSA */ +/* init */ {sIV, sCL, sCW, sCE, sES, sSS, sSR, sSA},/* INIT in sCL Big TODO */ +/* init_ack */ {sIV, sCL, sCW, sCE, sES, sSS, sSR, sSA}, +/* abort */ {sIV, sCL, sCL, sCL, sCL, sCL, sCL, sCL}, +/* shutdown */ {sIV, sCL, sCW, sCE, sSR, sSS, sSR, sSA}, +/* shutdown_ack */ {sIV, sCL, sCW, sCE, sES, sSA, sSA, sSA}, +/* error */ {sIV, sCL, sCW, sCL, sES, sSS, sSR, sSA}, +/* cookie_echo */ {sIV, sCL, sCW, sCE, sES, sSS, sSR, sSA},/* Cant come in reply dir */ +/* cookie_ack */ {sIV, sCL, sCW, sES, sES, sSS, sSR, sSA}, +/* shutdown_comp*/ {sIV, sCL, sCW, sCE, sES, sSS, sSR, sCL} + } +}; + +static int sctp_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + sctp_sctphdr_t _hdr, *hp; + + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + /* Actually only need first 8 bytes. */ + hp = skb_header_pointer(skb, dataoff, 8, &_hdr); + if (hp == NULL) + return 0; + + tuple->src.u.sctp.port = hp->source; + tuple->dst.u.sctp.port = hp->dest; + return 1; +} + +static int sctp_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + tuple->src.u.sctp.port = orig->dst.u.sctp.port; + tuple->dst.u.sctp.port = orig->src.u.sctp.port; + return 1; +} + +/* Print out the per-protocol part of the tuple. */ +static int sctp_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + return seq_printf(s, "sport=%hu dport=%hu ", + ntohs(tuple->src.u.sctp.port), + ntohs(tuple->dst.u.sctp.port)); +} + +/* Print out the private part of the conntrack. */ +static int sctp_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + enum sctp_conntrack state; + + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + read_lock_bh(&sctp_lock); + state = conntrack->proto.sctp.state; + read_unlock_bh(&sctp_lock); + + return seq_printf(s, "%s ", sctp_conntrack_names[state]); +} + +#define for_each_sctp_chunk(skb, sch, _sch, offset, dataoff, count) \ +for (offset = dataoff + sizeof(sctp_sctphdr_t), count = 0; \ + offset < skb->len && \ + (sch = skb_header_pointer(skb, offset, sizeof(_sch), &_sch)); \ + offset += (htons(sch->length) + 3) & ~3, count++) + +/* Some validity checks to make sure the chunks are fine */ +static int do_basic_checks(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff, + char *map) +{ + u_int32_t offset, count; + sctp_chunkhdr_t _sch, *sch; + int flag; + + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + flag = 0; + + for_each_sctp_chunk (skb, sch, _sch, offset, dataoff, count) { + DEBUGP("Chunk Num: %d Type: %d\n", count, sch->type); + + if (sch->type == SCTP_CID_INIT + || sch->type == SCTP_CID_INIT_ACK + || sch->type == SCTP_CID_SHUTDOWN_COMPLETE) { + flag = 1; + } + + /* Cookie Ack/Echo chunks not the first OR + Init / Init Ack / Shutdown compl chunks not the only chunks */ + if ((sch->type == SCTP_CID_COOKIE_ACK + || sch->type == SCTP_CID_COOKIE_ECHO + || flag) + && count !=0 ) { + DEBUGP("Basic checks failed\n"); + return 1; + } + + if (map) { + set_bit(sch->type, (void *)map); + } + } + + DEBUGP("Basic checks passed\n"); + return 0; +} + +static int new_state(enum ip_conntrack_dir dir, + enum sctp_conntrack cur_state, + int chunk_type) +{ + int i; + + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + DEBUGP("Chunk type: %d\n", chunk_type); + + switch (chunk_type) { + case SCTP_CID_INIT: + DEBUGP("SCTP_CID_INIT\n"); + i = 0; break; + case SCTP_CID_INIT_ACK: + DEBUGP("SCTP_CID_INIT_ACK\n"); + i = 1; break; + case SCTP_CID_ABORT: + DEBUGP("SCTP_CID_ABORT\n"); + i = 2; break; + case SCTP_CID_SHUTDOWN: + DEBUGP("SCTP_CID_SHUTDOWN\n"); + i = 3; break; + case SCTP_CID_SHUTDOWN_ACK: + DEBUGP("SCTP_CID_SHUTDOWN_ACK\n"); + i = 4; break; + case SCTP_CID_ERROR: + DEBUGP("SCTP_CID_ERROR\n"); + i = 5; break; + case SCTP_CID_COOKIE_ECHO: + DEBUGP("SCTP_CID_COOKIE_ECHO\n"); + i = 6; break; + case SCTP_CID_COOKIE_ACK: + DEBUGP("SCTP_CID_COOKIE_ACK\n"); + i = 7; break; + case SCTP_CID_SHUTDOWN_COMPLETE: + DEBUGP("SCTP_CID_SHUTDOWN_COMPLETE\n"); + i = 8; break; + default: + /* Other chunks like DATA, SACK, HEARTBEAT and + its ACK do not cause a change in state */ + DEBUGP("Unknown chunk type, Will stay in %s\n", + sctp_conntrack_names[cur_state]); + return cur_state; + } + + DEBUGP("dir: %d cur_state: %s chunk_type: %d new_state: %s\n", + dir, sctp_conntrack_names[cur_state], chunk_type, + sctp_conntrack_names[sctp_conntracks[dir][i][cur_state]]); + + return sctp_conntracks[dir][i][cur_state]; +} + +/* Returns verdict for packet, or -1 for invalid. */ +static int sctp_packet(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + enum sctp_conntrack newconntrack, oldsctpstate; + sctp_sctphdr_t _sctph, *sh; + sctp_chunkhdr_t _sch, *sch; + u_int32_t offset, count; + char map[256 / sizeof (char)] = {0}; + + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + sh = skb_header_pointer(skb, dataoff, sizeof(_sctph), &_sctph); + if (sh == NULL) + return -1; + + if (do_basic_checks(conntrack, skb, dataoff, map) != 0) + return -1; + + /* Check the verification tag (Sec 8.5) */ + if (!test_bit(SCTP_CID_INIT, (void *)map) + && !test_bit(SCTP_CID_SHUTDOWN_COMPLETE, (void *)map) + && !test_bit(SCTP_CID_COOKIE_ECHO, (void *)map) + && !test_bit(SCTP_CID_ABORT, (void *)map) + && !test_bit(SCTP_CID_SHUTDOWN_ACK, (void *)map) + && (sh->vtag != conntrack->proto.sctp.vtag[CTINFO2DIR(ctinfo)])) { + DEBUGP("Verification tag check failed\n"); + return -1; + } + + oldsctpstate = newconntrack = SCTP_CONNTRACK_MAX; + for_each_sctp_chunk (skb, sch, _sch, offset, dataoff, count) { + write_lock_bh(&sctp_lock); + + /* Special cases of Verification tag check (Sec 8.5.1) */ + if (sch->type == SCTP_CID_INIT) { + /* Sec 8.5.1 (A) */ + if (sh->vtag != 0) { + write_unlock_bh(&sctp_lock); + return -1; + } + } else if (sch->type == SCTP_CID_ABORT) { + /* Sec 8.5.1 (B) */ + if (!(sh->vtag == conntrack->proto.sctp.vtag[CTINFO2DIR(ctinfo)]) + && !(sh->vtag == conntrack->proto.sctp.vtag + [1 - CTINFO2DIR(ctinfo)])) { + write_unlock_bh(&sctp_lock); + return -1; + } + } else if (sch->type == SCTP_CID_SHUTDOWN_COMPLETE) { + /* Sec 8.5.1 (C) */ + if (!(sh->vtag == conntrack->proto.sctp.vtag[CTINFO2DIR(ctinfo)]) + && !(sh->vtag == conntrack->proto.sctp.vtag + [1 - CTINFO2DIR(ctinfo)] + && (sch->flags & 1))) { + write_unlock_bh(&sctp_lock); + return -1; + } + } else if (sch->type == SCTP_CID_COOKIE_ECHO) { + /* Sec 8.5.1 (D) */ + if (!(sh->vtag == conntrack->proto.sctp.vtag[CTINFO2DIR(ctinfo)])) { + write_unlock_bh(&sctp_lock); + return -1; + } + } + + oldsctpstate = conntrack->proto.sctp.state; + newconntrack = new_state(CTINFO2DIR(ctinfo), oldsctpstate, sch->type); + + /* Invalid */ + if (newconntrack == SCTP_CONNTRACK_MAX) { + DEBUGP("nf_conntrack_sctp: Invalid dir=%i ctype=%u conntrack=%u\n", + CTINFO2DIR(ctinfo), sch->type, oldsctpstate); + write_unlock_bh(&sctp_lock); + return -1; + } + + /* If it is an INIT or an INIT ACK note down the vtag */ + if (sch->type == SCTP_CID_INIT + || sch->type == SCTP_CID_INIT_ACK) { + sctp_inithdr_t _inithdr, *ih; + + ih = skb_header_pointer(skb, offset + sizeof(sctp_chunkhdr_t), + sizeof(_inithdr), &_inithdr); + if (ih == NULL) { + write_unlock_bh(&sctp_lock); + return -1; + } + DEBUGP("Setting vtag %x for dir %d\n", + ih->init_tag, !CTINFO2DIR(ctinfo)); + conntrack->proto.sctp.vtag[!CTINFO2DIR(ctinfo)] = ih->init_tag; + } + + conntrack->proto.sctp.state = newconntrack; + if (oldsctpstate != newconntrack) + nf_conntrack_event_cache(IPCT_PROTOINFO, skb); + write_unlock_bh(&sctp_lock); + } + + nf_ct_refresh_acct(conntrack, ctinfo, skb, *sctp_timeouts[newconntrack]); + + if (oldsctpstate == SCTP_CONNTRACK_COOKIE_ECHOED + && CTINFO2DIR(ctinfo) == IP_CT_DIR_REPLY + && newconntrack == SCTP_CONNTRACK_ESTABLISHED) { + DEBUGP("Setting assured bit\n"); + set_bit(IPS_ASSURED_BIT, &conntrack->status); + nf_conntrack_event_cache(IPCT_STATUS, skb); + } + + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int sctp_new(struct nf_conn *conntrack, const struct sk_buff *skb, + unsigned int dataoff) +{ + enum sctp_conntrack newconntrack; + sctp_sctphdr_t _sctph, *sh; + sctp_chunkhdr_t _sch, *sch; + u_int32_t offset, count; + char map[256 / sizeof (char)] = {0}; + + DEBUGP(__FUNCTION__); + DEBUGP("\n"); + + sh = skb_header_pointer(skb, dataoff, sizeof(_sctph), &_sctph); + if (sh == NULL) + return 0; + + if (do_basic_checks(conntrack, skb, dataoff, map) != 0) + return 0; + + /* If an OOTB packet has any of these chunks discard (Sec 8.4) */ + if ((test_bit (SCTP_CID_ABORT, (void *)map)) + || (test_bit (SCTP_CID_SHUTDOWN_COMPLETE, (void *)map)) + || (test_bit (SCTP_CID_COOKIE_ACK, (void *)map))) { + return 0; + } + + newconntrack = SCTP_CONNTRACK_MAX; + for_each_sctp_chunk (skb, sch, _sch, offset, dataoff, count) { + /* Don't need lock here: this conntrack not in circulation yet */ + newconntrack = new_state(IP_CT_DIR_ORIGINAL, + SCTP_CONNTRACK_NONE, sch->type); + + /* Invalid: delete conntrack */ + if (newconntrack == SCTP_CONNTRACK_MAX) { + DEBUGP("nf_conntrack_sctp: invalid new deleting.\n"); + return 0; + } + + /* Copy the vtag into the state info */ + if (sch->type == SCTP_CID_INIT) { + if (sh->vtag == 0) { + sctp_inithdr_t _inithdr, *ih; + + ih = skb_header_pointer(skb, offset + sizeof(sctp_chunkhdr_t), + sizeof(_inithdr), &_inithdr); + if (ih == NULL) + return 0; + + DEBUGP("Setting vtag %x for new conn\n", + ih->init_tag); + + conntrack->proto.sctp.vtag[IP_CT_DIR_REPLY] = + ih->init_tag; + } else { + /* Sec 8.5.1 (A) */ + return 0; + } + } + /* If it is a shutdown ack OOTB packet, we expect a return + shutdown complete, otherwise an ABORT Sec 8.4 (5) and (8) */ + else { + DEBUGP("Setting vtag %x for new conn OOTB\n", + sh->vtag); + conntrack->proto.sctp.vtag[IP_CT_DIR_REPLY] = sh->vtag; + } + + conntrack->proto.sctp.state = newconntrack; + } + + return 1; +} + +struct nf_conntrack_protocol nf_conntrack_protocol_sctp4 = { + .l3proto = PF_INET, + .proto = IPPROTO_SCTP, + .name = "sctp", + .pkt_to_tuple = sctp_pkt_to_tuple, + .invert_tuple = sctp_invert_tuple, + .print_tuple = sctp_print_tuple, + .print_conntrack = sctp_print_conntrack, + .packet = sctp_packet, + .new = sctp_new, + .destroy = NULL, + .me = THIS_MODULE +}; + +struct nf_conntrack_protocol nf_conntrack_protocol_sctp6 = { + .l3proto = PF_INET6, + .proto = IPPROTO_SCTP, + .name = "sctp", + .pkt_to_tuple = sctp_pkt_to_tuple, + .invert_tuple = sctp_invert_tuple, + .print_tuple = sctp_print_tuple, + .print_conntrack = sctp_print_conntrack, + .packet = sctp_packet, + .new = sctp_new, + .destroy = NULL, + .me = THIS_MODULE +}; + +#ifdef CONFIG_SYSCTL +static ctl_table nf_ct_sysctl_table[] = { + { + .ctl_name = NET_NF_CONNTRACK_SCTP_TIMEOUT_CLOSED, + .procname = "nf_conntrack_sctp_timeout_closed", + .data = &nf_ct_sctp_timeout_closed, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_WAIT, + .procname = "nf_conntrack_sctp_timeout_cookie_wait", + .data = &nf_ct_sctp_timeout_cookie_wait, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_SCTP_TIMEOUT_COOKIE_ECHOED, + .procname = "nf_conntrack_sctp_timeout_cookie_echoed", + .data = &nf_ct_sctp_timeout_cookie_echoed, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_SCTP_TIMEOUT_ESTABLISHED, + .procname = "nf_conntrack_sctp_timeout_established", + .data = &nf_ct_sctp_timeout_established, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_SENT, + .procname = "nf_conntrack_sctp_timeout_shutdown_sent", + .data = &nf_ct_sctp_timeout_shutdown_sent, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_RECD, + .procname = "nf_conntrack_sctp_timeout_shutdown_recd", + .data = &nf_ct_sctp_timeout_shutdown_recd, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_ACK_SENT, + .procname = "nf_conntrack_sctp_timeout_shutdown_ack_sent", + .data = &nf_ct_sctp_timeout_shutdown_ack_sent, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_netfilter_table[] = { + { + .ctl_name = NET_NETFILTER, + .procname = "netfilter", + .mode = 0555, + .child = nf_ct_sysctl_table, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_net_table[] = { + { + .ctl_name = CTL_NET, + .procname = "net", + .mode = 0555, + .child = nf_ct_netfilter_table, + }, + { .ctl_name = 0 } +}; + +static struct ctl_table_header *nf_ct_sysctl_header; +#endif + +int __init init(void) +{ + int ret; + + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_sctp4); + if (ret) { + printk("nf_conntrack_proto_sctp4: protocol register failed\n"); + goto out; + } + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_sctp6); + if (ret) { + printk("nf_conntrack_proto_sctp6: protocol register failed\n"); + goto cleanup_sctp4; + } + +#ifdef CONFIG_SYSCTL + nf_ct_sysctl_header = register_sysctl_table(nf_ct_net_table, 0); + if (nf_ct_sysctl_header == NULL) { + printk("nf_conntrack_proto_sctp: can't register to sysctl.\n"); + goto cleanup; + } +#endif + + return ret; + +#ifdef CONFIG_SYSCTL + cleanup: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_sctp6); +#endif + cleanup_sctp4: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_sctp4); + out: + DEBUGP("SCTP conntrack module loading %s\n", + ret ? "failed": "succeeded"); + return ret; +} + +void __exit fini(void) +{ + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_sctp6); + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_sctp4); +#ifdef CONFIG_SYSCTL + unregister_sysctl_table(nf_ct_sysctl_header); +#endif + DEBUGP("SCTP conntrack module unloaded\n"); +} + +module_init(init); +module_exit(fini); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Kiran Kumar Immidi"); +MODULE_DESCRIPTION("Netfilter connection tracking protocol helper for SCTP"); diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c new file mode 100644 index 000000000000..83d90dd624f0 --- /dev/null +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -0,0 +1,1162 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Jozsef Kadlecsik : + * - Real stateful connection tracking + * - Modified state transitions table + * - Window scaling support added + * - SACK support added + * + * Willy Tarreau: + * - State table bugfixes + * - More robust state changes + * - Tuning timer parameters + * + * 27 Oct 2004: Yasuyuki Kozakai @USAGI + * - genelized Layer 3 protocol part. + * + * Derived from net/ipv4/netfilter/ip_conntrack_proto_tcp.c + * + * version 2.2 + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#define DEBUGP_VARS +#else +#define DEBUGP(format, args...) +#endif + +/* Protects conntrack->proto.tcp */ +static DEFINE_RWLOCK(tcp_lock); + +/* "Be conservative in what you do, + be liberal in what you accept from others." + If it's non-zero, we mark only out of window RST segments as INVALID. */ +int nf_ct_tcp_be_liberal = 0; + +/* When connection is picked up from the middle, how many packets are required + to pass in each direction when we assume we are in sync - if any side uses + window scaling, we lost the game. + If it is set to zero, we disable picking up already established + connections. */ +int nf_ct_tcp_loose = 3; + +/* Max number of the retransmitted packets without receiving an (acceptable) + ACK from the destination. If this number is reached, a shorter timer + will be started. */ +int nf_ct_tcp_max_retrans = 3; + + /* FIXME: Examine ipfilter's timeouts and conntrack transitions more + closely. They're more complex. --RR */ + +static const char *tcp_conntrack_names[] = { + "NONE", + "SYN_SENT", + "SYN_RECV", + "ESTABLISHED", + "FIN_WAIT", + "CLOSE_WAIT", + "LAST_ACK", + "TIME_WAIT", + "CLOSE", + "LISTEN" +}; + +#define SECS * HZ +#define MINS * 60 SECS +#define HOURS * 60 MINS +#define DAYS * 24 HOURS + +unsigned long nf_ct_tcp_timeout_syn_sent = 2 MINS; +unsigned long nf_ct_tcp_timeout_syn_recv = 60 SECS; +unsigned long nf_ct_tcp_timeout_established = 5 DAYS; +unsigned long nf_ct_tcp_timeout_fin_wait = 2 MINS; +unsigned long nf_ct_tcp_timeout_close_wait = 60 SECS; +unsigned long nf_ct_tcp_timeout_last_ack = 30 SECS; +unsigned long nf_ct_tcp_timeout_time_wait = 2 MINS; +unsigned long nf_ct_tcp_timeout_close = 10 SECS; + +/* RFC1122 says the R2 limit should be at least 100 seconds. + Linux uses 15 packets as limit, which corresponds + to ~13-30min depending on RTO. */ +unsigned long nf_ct_tcp_timeout_max_retrans = 5 MINS; + +static unsigned long * tcp_timeouts[] += { NULL, /* TCP_CONNTRACK_NONE */ + &nf_ct_tcp_timeout_syn_sent, /* TCP_CONNTRACK_SYN_SENT, */ + &nf_ct_tcp_timeout_syn_recv, /* TCP_CONNTRACK_SYN_RECV, */ + &nf_ct_tcp_timeout_established, /* TCP_CONNTRACK_ESTABLISHED, */ + &nf_ct_tcp_timeout_fin_wait, /* TCP_CONNTRACK_FIN_WAIT, */ + &nf_ct_tcp_timeout_close_wait, /* TCP_CONNTRACK_CLOSE_WAIT, */ + &nf_ct_tcp_timeout_last_ack, /* TCP_CONNTRACK_LAST_ACK, */ + &nf_ct_tcp_timeout_time_wait, /* TCP_CONNTRACK_TIME_WAIT, */ + &nf_ct_tcp_timeout_close, /* TCP_CONNTRACK_CLOSE, */ + NULL, /* TCP_CONNTRACK_LISTEN */ + }; + +#define sNO TCP_CONNTRACK_NONE +#define sSS TCP_CONNTRACK_SYN_SENT +#define sSR TCP_CONNTRACK_SYN_RECV +#define sES TCP_CONNTRACK_ESTABLISHED +#define sFW TCP_CONNTRACK_FIN_WAIT +#define sCW TCP_CONNTRACK_CLOSE_WAIT +#define sLA TCP_CONNTRACK_LAST_ACK +#define sTW TCP_CONNTRACK_TIME_WAIT +#define sCL TCP_CONNTRACK_CLOSE +#define sLI TCP_CONNTRACK_LISTEN +#define sIV TCP_CONNTRACK_MAX +#define sIG TCP_CONNTRACK_IGNORE + +/* What TCP flags are set from RST/SYN/FIN/ACK. */ +enum tcp_bit_set { + TCP_SYN_SET, + TCP_SYNACK_SET, + TCP_FIN_SET, + TCP_ACK_SET, + TCP_RST_SET, + TCP_NONE_SET, +}; + +/* + * The TCP state transition table needs a few words... + * + * We are the man in the middle. All the packets go through us + * but might get lost in transit to the destination. + * It is assumed that the destinations can't receive segments + * we haven't seen. + * + * The checked segment is in window, but our windows are *not* + * equivalent with the ones of the sender/receiver. We always + * try to guess the state of the current sender. + * + * The meaning of the states are: + * + * NONE: initial state + * SYN_SENT: SYN-only packet seen + * SYN_RECV: SYN-ACK packet seen + * ESTABLISHED: ACK packet seen + * FIN_WAIT: FIN packet seen + * CLOSE_WAIT: ACK seen (after FIN) + * LAST_ACK: FIN seen (after FIN) + * TIME_WAIT: last ACK seen + * CLOSE: closed connection + * + * LISTEN state is not used. + * + * Packets marked as IGNORED (sIG): + * if they may be either invalid or valid + * and the receiver may send back a connection + * closing RST or a SYN/ACK. + * + * Packets marked as INVALID (sIV): + * if they are invalid + * or we do not support the request (simultaneous open) + */ +static enum tcp_conntrack tcp_conntracks[2][6][TCP_CONNTRACK_MAX] = { + { +/* ORIGINAL */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*syn*/ { sSS, sSS, sIG, sIG, sIG, sIG, sIG, sSS, sSS, sIV }, +/* + * sNO -> sSS Initialize a new connection + * sSS -> sSS Retransmitted SYN + * sSR -> sIG Late retransmitted SYN? + * sES -> sIG Error: SYNs in window outside the SYN_SENT state + * are errors. Receiver will reply with RST + * and close the connection. + * Or we are not in sync and hold a dead connection. + * sFW -> sIG + * sCW -> sIG + * sLA -> sIG + * sTW -> sSS Reopened connection (RFC 1122). + * sCL -> sSS + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*synack*/ { sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV }, +/* + * A SYN/ACK from the client is always invalid: + * - either it tries to set up a simultaneous open, which is + * not supported; + * - or the firewall has just been inserted between the two hosts + * during the session set-up. The SYN will be retransmitted + * by the true client (or it'll time out). + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*fin*/ { sIV, sIV, sFW, sFW, sLA, sLA, sLA, sTW, sCL, sIV }, +/* + * sNO -> sIV Too late and no reason to do anything... + * sSS -> sIV Client migth not send FIN in this state: + * we enforce waiting for a SYN/ACK reply first. + * sSR -> sFW Close started. + * sES -> sFW + * sFW -> sLA FIN seen in both directions, waiting for + * the last ACK. + * Migth be a retransmitted FIN as well... + * sCW -> sLA + * sLA -> sLA Retransmitted FIN. Remain in the same state. + * sTW -> sTW + * sCL -> sCL + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*ack*/ { sES, sIV, sES, sES, sCW, sCW, sTW, sTW, sCL, sIV }, +/* + * sNO -> sES Assumed. + * sSS -> sIV ACK is invalid: we haven't seen a SYN/ACK yet. + * sSR -> sES Established state is reached. + * sES -> sES :-) + * sFW -> sCW Normal close request answered by ACK. + * sCW -> sCW + * sLA -> sTW Last ACK detected. + * sTW -> sTW Retransmitted last ACK. Remain in the same state. + * sCL -> sCL + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*rst*/ { sIV, sCL, sCL, sCL, sCL, sCL, sCL, sCL, sCL, sIV }, +/*none*/ { sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV } + }, + { +/* REPLY */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*syn*/ { sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV }, +/* + * sNO -> sIV Never reached. + * sSS -> sIV Simultaneous open, not supported + * sSR -> sIV Simultaneous open, not supported. + * sES -> sIV Server may not initiate a connection. + * sFW -> sIV + * sCW -> sIV + * sLA -> sIV + * sTW -> sIV Reopened connection, but server may not do it. + * sCL -> sIV + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*synack*/ { sIV, sSR, sSR, sIG, sIG, sIG, sIG, sIG, sIG, sIV }, +/* + * sSS -> sSR Standard open. + * sSR -> sSR Retransmitted SYN/ACK. + * sES -> sIG Late retransmitted SYN/ACK? + * sFW -> sIG Might be SYN/ACK answering ignored SYN + * sCW -> sIG + * sLA -> sIG + * sTW -> sIG + * sCL -> sIG + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*fin*/ { sIV, sIV, sFW, sFW, sLA, sLA, sLA, sTW, sCL, sIV }, +/* + * sSS -> sIV Server might not send FIN in this state. + * sSR -> sFW Close started. + * sES -> sFW + * sFW -> sLA FIN seen in both directions. + * sCW -> sLA + * sLA -> sLA Retransmitted FIN. + * sTW -> sTW + * sCL -> sCL + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*ack*/ { sIV, sIV, sSR, sES, sCW, sCW, sTW, sTW, sCL, sIV }, +/* + * sSS -> sIV Might be a half-open connection. + * sSR -> sSR Might answer late resent SYN. + * sES -> sES :-) + * sFW -> sCW Normal close request answered by ACK. + * sCW -> sCW + * sLA -> sTW Last ACK detected. + * sTW -> sTW Retransmitted last ACK. + * sCL -> sCL + */ +/* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sLI */ +/*rst*/ { sIV, sCL, sCL, sCL, sCL, sCL, sCL, sCL, sCL, sIV }, +/*none*/ { sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV } + } +}; + +static int tcp_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + struct tcphdr _hdr, *hp; + + /* Actually only need first 8 bytes. */ + hp = skb_header_pointer(skb, dataoff, 8, &_hdr); + if (hp == NULL) + return 0; + + tuple->src.u.tcp.port = hp->source; + tuple->dst.u.tcp.port = hp->dest; + + return 1; +} + +static int tcp_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + tuple->src.u.tcp.port = orig->dst.u.tcp.port; + tuple->dst.u.tcp.port = orig->src.u.tcp.port; + return 1; +} + +/* Print out the per-protocol part of the tuple. */ +static int tcp_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "sport=%hu dport=%hu ", + ntohs(tuple->src.u.tcp.port), + ntohs(tuple->dst.u.tcp.port)); +} + +/* Print out the private part of the conntrack. */ +static int tcp_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + enum tcp_conntrack state; + + read_lock_bh(&tcp_lock); + state = conntrack->proto.tcp.state; + read_unlock_bh(&tcp_lock); + + return seq_printf(s, "%s ", tcp_conntrack_names[state]); +} + +static unsigned int get_conntrack_index(const struct tcphdr *tcph) +{ + if (tcph->rst) return TCP_RST_SET; + else if (tcph->syn) return (tcph->ack ? TCP_SYNACK_SET : TCP_SYN_SET); + else if (tcph->fin) return TCP_FIN_SET; + else if (tcph->ack) return TCP_ACK_SET; + else return TCP_NONE_SET; +} + +/* TCP connection tracking based on 'Real Stateful TCP Packet Filtering + in IP Filter' by Guido van Rooij. + + http://www.nluug.nl/events/sane2000/papers.html + http://www.iae.nl/users/guido/papers/tcp_filtering.ps.gz + + The boundaries and the conditions are changed according to RFC793: + the packet must intersect the window (i.e. segments may be + after the right or before the left edge) and thus receivers may ACK + segments after the right edge of the window. + + td_maxend = max(sack + max(win,1)) seen in reply packets + td_maxwin = max(max(win, 1)) + (sack - ack) seen in sent packets + td_maxwin += seq + len - sender.td_maxend + if seq + len > sender.td_maxend + td_end = max(seq + len) seen in sent packets + + I. Upper bound for valid data: seq <= sender.td_maxend + II. Lower bound for valid data: seq + len >= sender.td_end - receiver.td_maxwin + III. Upper bound for valid ack: sack <= receiver.td_end + IV. Lower bound for valid ack: ack >= receiver.td_end - MAXACKWINDOW + + where sack is the highest right edge of sack block found in the packet. + + The upper bound limit for a valid ack is not ignored - + we doesn't have to deal with fragments. +*/ + +static inline __u32 segment_seq_plus_len(__u32 seq, + size_t len, + unsigned int dataoff, + struct tcphdr *tcph) +{ + /* XXX Should I use payload length field in IP/IPv6 header ? + * - YK */ + return (seq + len - dataoff - tcph->doff*4 + + (tcph->syn ? 1 : 0) + (tcph->fin ? 1 : 0)); +} + +/* Fixme: what about big packets? */ +#define MAXACKWINCONST 66000 +#define MAXACKWINDOW(sender) \ + ((sender)->td_maxwin > MAXACKWINCONST ? (sender)->td_maxwin \ + : MAXACKWINCONST) + +/* + * Simplified tcp_parse_options routine from tcp_input.c + */ +static void tcp_options(const struct sk_buff *skb, + unsigned int dataoff, + struct tcphdr *tcph, + struct ip_ct_tcp_state *state) +{ + unsigned char buff[(15 * 4) - sizeof(struct tcphdr)]; + unsigned char *ptr; + int length = (tcph->doff*4) - sizeof(struct tcphdr); + + if (!length) + return; + + ptr = skb_header_pointer(skb, dataoff + sizeof(struct tcphdr), + length, buff); + BUG_ON(ptr == NULL); + + state->td_scale = + state->flags = 0; + + while (length > 0) { + int opcode=*ptr++; + int opsize; + + switch (opcode) { + case TCPOPT_EOL: + return; + case TCPOPT_NOP: /* Ref: RFC 793 section 3.1 */ + length--; + continue; + default: + opsize=*ptr++; + if (opsize < 2) /* "silly options" */ + return; + if (opsize > length) + break; /* don't parse partial options */ + + if (opcode == TCPOPT_SACK_PERM + && opsize == TCPOLEN_SACK_PERM) + state->flags |= IP_CT_TCP_FLAG_SACK_PERM; + else if (opcode == TCPOPT_WINDOW + && opsize == TCPOLEN_WINDOW) { + state->td_scale = *(u_int8_t *)ptr; + + if (state->td_scale > 14) { + /* See RFC1323 */ + state->td_scale = 14; + } + state->flags |= + IP_CT_TCP_FLAG_WINDOW_SCALE; + } + ptr += opsize - 2; + length -= opsize; + } + } +} + +static void tcp_sack(const struct sk_buff *skb, unsigned int dataoff, + struct tcphdr *tcph, __u32 *sack) +{ + unsigned char buff[(15 * 4) - sizeof(struct tcphdr)]; + unsigned char *ptr; + int length = (tcph->doff*4) - sizeof(struct tcphdr); + __u32 tmp; + + if (!length) + return; + + ptr = skb_header_pointer(skb, dataoff + sizeof(struct tcphdr), + length, buff); + BUG_ON(ptr == NULL); + + /* Fast path for timestamp-only option */ + if (length == TCPOLEN_TSTAMP_ALIGNED*4 + && *(__u32 *)ptr == + __constant_ntohl((TCPOPT_NOP << 24) + | (TCPOPT_NOP << 16) + | (TCPOPT_TIMESTAMP << 8) + | TCPOLEN_TIMESTAMP)) + return; + + while (length > 0) { + int opcode = *ptr++; + int opsize, i; + + switch (opcode) { + case TCPOPT_EOL: + return; + case TCPOPT_NOP: /* Ref: RFC 793 section 3.1 */ + length--; + continue; + default: + opsize = *ptr++; + if (opsize < 2) /* "silly options" */ + return; + if (opsize > length) + break; /* don't parse partial options */ + + if (opcode == TCPOPT_SACK + && opsize >= (TCPOLEN_SACK_BASE + + TCPOLEN_SACK_PERBLOCK) + && !((opsize - TCPOLEN_SACK_BASE) + % TCPOLEN_SACK_PERBLOCK)) { + for (i = 0; + i < (opsize - TCPOLEN_SACK_BASE); + i += TCPOLEN_SACK_PERBLOCK) { + memcpy(&tmp, (__u32 *)(ptr + i) + 1, + sizeof(__u32)); + tmp = ntohl(tmp); + + if (after(tmp, *sack)) + *sack = tmp; + } + return; + } + ptr += opsize - 2; + length -= opsize; + } + } +} + +static int tcp_in_window(struct ip_ct_tcp *state, + enum ip_conntrack_dir dir, + unsigned int index, + const struct sk_buff *skb, + unsigned int dataoff, + struct tcphdr *tcph, + int pf) +{ + struct ip_ct_tcp_state *sender = &state->seen[dir]; + struct ip_ct_tcp_state *receiver = &state->seen[!dir]; + __u32 seq, ack, sack, end, win, swin; + int res; + + /* + * Get the required data from the packet. + */ + seq = ntohl(tcph->seq); + ack = sack = ntohl(tcph->ack_seq); + win = ntohs(tcph->window); + end = segment_seq_plus_len(seq, skb->len, dataoff, tcph); + + if (receiver->flags & IP_CT_TCP_FLAG_SACK_PERM) + tcp_sack(skb, dataoff, tcph, &sack); + + DEBUGP("tcp_in_window: START\n"); + DEBUGP("tcp_in_window: src=%u.%u.%u.%u:%hu dst=%u.%u.%u.%u:%hu " + "seq=%u ack=%u sack=%u win=%u end=%u\n", + NIPQUAD(iph->saddr), ntohs(tcph->source), + NIPQUAD(iph->daddr), ntohs(tcph->dest), + seq, ack, sack, win, end); + DEBUGP("tcp_in_window: sender end=%u maxend=%u maxwin=%u scale=%i " + "receiver end=%u maxend=%u maxwin=%u scale=%i\n", + sender->td_end, sender->td_maxend, sender->td_maxwin, + sender->td_scale, + receiver->td_end, receiver->td_maxend, receiver->td_maxwin, + receiver->td_scale); + + if (sender->td_end == 0) { + /* + * Initialize sender data. + */ + if (tcph->syn && tcph->ack) { + /* + * Outgoing SYN-ACK in reply to a SYN. + */ + sender->td_end = + sender->td_maxend = end; + sender->td_maxwin = (win == 0 ? 1 : win); + + tcp_options(skb, dataoff, tcph, sender); + /* + * RFC 1323: + * Both sides must send the Window Scale option + * to enable window scaling in either direction. + */ + if (!(sender->flags & IP_CT_TCP_FLAG_WINDOW_SCALE + && receiver->flags & IP_CT_TCP_FLAG_WINDOW_SCALE)) + sender->td_scale = + receiver->td_scale = 0; + } else { + /* + * We are in the middle of a connection, + * its history is lost for us. + * Let's try to use the data from the packet. + */ + sender->td_end = end; + sender->td_maxwin = (win == 0 ? 1 : win); + sender->td_maxend = end + sender->td_maxwin; + } + } else if (((state->state == TCP_CONNTRACK_SYN_SENT + && dir == IP_CT_DIR_ORIGINAL) + || (state->state == TCP_CONNTRACK_SYN_RECV + && dir == IP_CT_DIR_REPLY)) + && after(end, sender->td_end)) { + /* + * RFC 793: "if a TCP is reinitialized ... then it need + * not wait at all; it must only be sure to use sequence + * numbers larger than those recently used." + */ + sender->td_end = + sender->td_maxend = end; + sender->td_maxwin = (win == 0 ? 1 : win); + + tcp_options(skb, dataoff, tcph, sender); + } + + if (!(tcph->ack)) { + /* + * If there is no ACK, just pretend it was set and OK. + */ + ack = sack = receiver->td_end; + } else if (((tcp_flag_word(tcph) & (TCP_FLAG_ACK|TCP_FLAG_RST)) == + (TCP_FLAG_ACK|TCP_FLAG_RST)) + && (ack == 0)) { + /* + * Broken TCP stacks, that set ACK in RST packets as well + * with zero ack value. + */ + ack = sack = receiver->td_end; + } + + if (seq == end + && (!tcph->rst + || (seq == 0 && state->state == TCP_CONNTRACK_SYN_SENT))) + /* + * Packets contains no data: we assume it is valid + * and check the ack value only. + * However RST segments are always validated by their + * SEQ number, except when seq == 0 (reset sent answering + * SYN. + */ + seq = end = sender->td_end; + + DEBUGP("tcp_in_window: src=%u.%u.%u.%u:%hu dst=%u.%u.%u.%u:%hu " + "seq=%u ack=%u sack =%u win=%u end=%u\n", + NIPQUAD(iph->saddr), ntohs(tcph->source), + NIPQUAD(iph->daddr), ntohs(tcph->dest), + seq, ack, sack, win, end); + DEBUGP("tcp_in_window: sender end=%u maxend=%u maxwin=%u scale=%i " + "receiver end=%u maxend=%u maxwin=%u scale=%i\n", + sender->td_end, sender->td_maxend, sender->td_maxwin, + sender->td_scale, + receiver->td_end, receiver->td_maxend, receiver->td_maxwin, + receiver->td_scale); + + DEBUGP("tcp_in_window: I=%i II=%i III=%i IV=%i\n", + before(seq, sender->td_maxend + 1), + after(end, sender->td_end - receiver->td_maxwin - 1), + before(sack, receiver->td_end + 1), + after(ack, receiver->td_end - MAXACKWINDOW(sender))); + + if (sender->loose || receiver->loose || + (before(seq, sender->td_maxend + 1) && + after(end, sender->td_end - receiver->td_maxwin - 1) && + before(sack, receiver->td_end + 1) && + after(ack, receiver->td_end - MAXACKWINDOW(sender)))) { + /* + * Take into account window scaling (RFC 1323). + */ + if (!tcph->syn) + win <<= sender->td_scale; + + /* + * Update sender data. + */ + swin = win + (sack - ack); + if (sender->td_maxwin < swin) + sender->td_maxwin = swin; + if (after(end, sender->td_end)) + sender->td_end = end; + /* + * Update receiver data. + */ + if (after(end, sender->td_maxend)) + receiver->td_maxwin += end - sender->td_maxend; + if (after(sack + win, receiver->td_maxend - 1)) { + receiver->td_maxend = sack + win; + if (win == 0) + receiver->td_maxend++; + } + + /* + * Check retransmissions. + */ + if (index == TCP_ACK_SET) { + if (state->last_dir == dir + && state->last_seq == seq + && state->last_ack == ack + && state->last_end == end) + state->retrans++; + else { + state->last_dir = dir; + state->last_seq = seq; + state->last_ack = ack; + state->last_end = end; + state->retrans = 0; + } + } + /* + * Close the window of disabled window tracking :-) + */ + if (sender->loose) + sender->loose--; + + res = 1; + } else { + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: %s ", + before(seq, sender->td_maxend + 1) ? + after(end, sender->td_end - receiver->td_maxwin - 1) ? + before(sack, receiver->td_end + 1) ? + after(ack, receiver->td_end - MAXACKWINDOW(sender)) ? "BUG" + : "ACK is under the lower bound (possible overly delayed ACK)" + : "ACK is over the upper bound (ACKed data not seen yet)" + : "SEQ is under the lower bound (already ACKed data retransmitted)" + : "SEQ is over the upper bound (over the window of the receiver)"); + + res = nf_ct_tcp_be_liberal; + } + + DEBUGP("tcp_in_window: res=%i sender end=%u maxend=%u maxwin=%u " + "receiver end=%u maxend=%u maxwin=%u\n", + res, sender->td_end, sender->td_maxend, sender->td_maxwin, + receiver->td_end, receiver->td_maxend, receiver->td_maxwin); + + return res; +} + +#ifdef CONFIG_IP_NF_NAT_NEEDED +/* Update sender->td_end after NAT successfully mangled the packet */ +/* Caller must linearize skb at tcp header. */ +void nf_conntrack_tcp_update(struct sk_buff *skb, + unsigned int dataoff, + struct nf_conn *conntrack, + int dir) +{ + struct tcphdr *tcph = (void *)skb->data + dataoff; + __u32 end; +#ifdef DEBUGP_VARS + struct ip_ct_tcp_state *sender = &conntrack->proto.tcp.seen[dir]; + struct ip_ct_tcp_state *receiver = &conntrack->proto.tcp.seen[!dir]; +#endif + + end = segment_seq_plus_len(ntohl(tcph->seq), skb->len, dataoff, tcph); + + write_lock_bh(&tcp_lock); + /* + * We have to worry for the ack in the reply packet only... + */ + if (after(end, conntrack->proto.tcp.seen[dir].td_end)) + conntrack->proto.tcp.seen[dir].td_end = end; + conntrack->proto.tcp.last_end = end; + write_unlock_bh(&tcp_lock); + DEBUGP("tcp_update: sender end=%u maxend=%u maxwin=%u scale=%i " + "receiver end=%u maxend=%u maxwin=%u scale=%i\n", + sender->td_end, sender->td_maxend, sender->td_maxwin, + sender->td_scale, + receiver->td_end, receiver->td_maxend, receiver->td_maxwin, + receiver->td_scale); +} + +#endif + +#define TH_FIN 0x01 +#define TH_SYN 0x02 +#define TH_RST 0x04 +#define TH_PUSH 0x08 +#define TH_ACK 0x10 +#define TH_URG 0x20 +#define TH_ECE 0x40 +#define TH_CWR 0x80 + +/* table of valid flag combinations - ECE and CWR are always valid */ +static u8 tcp_valid_flags[(TH_FIN|TH_SYN|TH_RST|TH_PUSH|TH_ACK|TH_URG) + 1] = +{ + [TH_SYN] = 1, + [TH_SYN|TH_ACK] = 1, + [TH_SYN|TH_ACK|TH_PUSH] = 1, + [TH_RST] = 1, + [TH_RST|TH_ACK] = 1, + [TH_RST|TH_ACK|TH_PUSH] = 1, + [TH_FIN|TH_ACK] = 1, + [TH_ACK] = 1, + [TH_ACK|TH_PUSH] = 1, + [TH_ACK|TH_URG] = 1, + [TH_ACK|TH_URG|TH_PUSH] = 1, + [TH_FIN|TH_ACK|TH_PUSH] = 1, + [TH_FIN|TH_ACK|TH_URG] = 1, + [TH_FIN|TH_ACK|TH_URG|TH_PUSH] = 1, +}; + +/* Protect conntrack agaist broken packets. Code taken from ipt_unclean.c. */ +static int tcp_error(struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info *ctinfo, + int pf, + unsigned int hooknum, + int(*csum)(const struct sk_buff *,unsigned int)) +{ + struct tcphdr _tcph, *th; + unsigned int tcplen = skb->len - dataoff; + u_int8_t tcpflags; + + /* Smaller that minimal TCP header? */ + th = skb_header_pointer(skb, dataoff, sizeof(_tcph), &_tcph); + if (th == NULL) { + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: short packet "); + return -NF_ACCEPT; + } + + /* Not whole TCP header or malformed packet */ + if (th->doff*4 < sizeof(struct tcphdr) || tcplen < th->doff*4) { + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: truncated/malformed packet "); + return -NF_ACCEPT; + } + + /* Checksum invalid? Ignore. + * We skip checking packets on the outgoing path + * because the semantic of CHECKSUM_HW is different there + * and moreover root might send raw packets. + */ + /* FIXME: Source route IP option packets --RR */ + if (((pf == PF_INET && hooknum == NF_IP_PRE_ROUTING) || + (pf == PF_INET6 && hooknum == NF_IP6_PRE_ROUTING)) + && skb->ip_summed != CHECKSUM_UNNECESSARY + && csum(skb, dataoff)) { + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: bad TCP checksum "); + return -NF_ACCEPT; + } + + /* Check TCP flags. */ + tcpflags = (((u_int8_t *)th)[13] & ~(TH_ECE|TH_CWR)); + if (!tcp_valid_flags[tcpflags]) { + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: invalid TCP flag combination "); + return -NF_ACCEPT; + } + + return NF_ACCEPT; +} + +static int csum4(const struct sk_buff *skb, unsigned int dataoff) +{ + return csum_tcpudp_magic(skb->nh.iph->saddr, skb->nh.iph->daddr, + skb->len - dataoff, IPPROTO_TCP, + skb->ip_summed == CHECKSUM_HW ? skb->csum + : skb_checksum(skb, dataoff, + skb->len - dataoff, 0)); +} + +static int csum6(const struct sk_buff *skb, unsigned int dataoff) +{ + return csum_ipv6_magic(&skb->nh.ipv6h->saddr, &skb->nh.ipv6h->daddr, + skb->len - dataoff, IPPROTO_TCP, + skb->ip_summed == CHECKSUM_HW ? skb->csum + : skb_checksum(skb, dataoff, skb->len - dataoff, + 0)); +} + +static int tcp_error4(struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info *ctinfo, + int pf, + unsigned int hooknum) +{ + return tcp_error(skb, dataoff, ctinfo, pf, hooknum, csum4); +} + +static int tcp_error6(struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info *ctinfo, + int pf, + unsigned int hooknum) +{ + return tcp_error(skb, dataoff, ctinfo, pf, hooknum, csum6); +} + +/* Returns verdict for packet, or -1 for invalid. */ +static int tcp_packet(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + enum tcp_conntrack new_state, old_state; + enum ip_conntrack_dir dir; + struct tcphdr *th, _tcph; + unsigned long timeout; + unsigned int index; + + th = skb_header_pointer(skb, dataoff, sizeof(_tcph), &_tcph); + BUG_ON(th == NULL); + + write_lock_bh(&tcp_lock); + old_state = conntrack->proto.tcp.state; + dir = CTINFO2DIR(ctinfo); + index = get_conntrack_index(th); + new_state = tcp_conntracks[dir][index][old_state]; + + switch (new_state) { + case TCP_CONNTRACK_IGNORE: + /* Either SYN in ORIGINAL + * or SYN/ACK in REPLY. */ + if (index == TCP_SYNACK_SET + && conntrack->proto.tcp.last_index == TCP_SYN_SET + && conntrack->proto.tcp.last_dir != dir + && ntohl(th->ack_seq) == + conntrack->proto.tcp.last_end) { + /* This SYN/ACK acknowledges a SYN that we earlier + * ignored as invalid. This means that the client and + * the server are both in sync, while the firewall is + * not. We kill this session and block the SYN/ACK so + * that the client cannot but retransmit its SYN and + * thus initiate a clean new session. + */ + write_unlock_bh(&tcp_lock); + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: killing out of sync session "); + if (del_timer(&conntrack->timeout)) + conntrack->timeout.function((unsigned long) + conntrack); + return -NF_DROP; + } + conntrack->proto.tcp.last_index = index; + conntrack->proto.tcp.last_dir = dir; + conntrack->proto.tcp.last_seq = ntohl(th->seq); + conntrack->proto.tcp.last_end = + segment_seq_plus_len(ntohl(th->seq), skb->len, dataoff, th); + + write_unlock_bh(&tcp_lock); + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: invalid packed ignored "); + return NF_ACCEPT; + case TCP_CONNTRACK_MAX: + /* Invalid packet */ + DEBUGP("nf_ct_tcp: Invalid dir=%i index=%u ostate=%u\n", + dir, get_conntrack_index(th), + old_state); + write_unlock_bh(&tcp_lock); + if (LOG_INVALID(IPPROTO_TCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_tcp: invalid state "); + return -NF_ACCEPT; + case TCP_CONNTRACK_SYN_SENT: + if (old_state < TCP_CONNTRACK_TIME_WAIT) + break; + if ((conntrack->proto.tcp.seen[dir].flags & + IP_CT_TCP_FLAG_CLOSE_INIT) + || after(ntohl(th->seq), + conntrack->proto.tcp.seen[dir].td_end)) { + /* Attempt to reopen a closed connection. + * Delete this connection and look up again. */ + write_unlock_bh(&tcp_lock); + if (del_timer(&conntrack->timeout)) + conntrack->timeout.function((unsigned long) + conntrack); + return -NF_REPEAT; + } + case TCP_CONNTRACK_CLOSE: + if (index == TCP_RST_SET + && test_bit(IPS_SEEN_REPLY_BIT, &conntrack->status) + && conntrack->proto.tcp.last_index == TCP_SYN_SET + && ntohl(th->ack_seq) == conntrack->proto.tcp.last_end) { + /* RST sent to invalid SYN we had let trough + * SYN was in window then, tear down connection. + * We skip window checking, because packet might ACK + * segments we ignored in the SYN. */ + goto in_window; + } + /* Just fall trough */ + default: + /* Keep compilers happy. */ + break; + } + + if (!tcp_in_window(&conntrack->proto.tcp, dir, index, + skb, dataoff, th, pf)) { + write_unlock_bh(&tcp_lock); + return -NF_ACCEPT; + } + in_window: + /* From now on we have got in-window packets */ + conntrack->proto.tcp.last_index = index; + + DEBUGP("tcp_conntracks: src=%u.%u.%u.%u:%hu dst=%u.%u.%u.%u:%hu " + "syn=%i ack=%i fin=%i rst=%i old=%i new=%i\n", + NIPQUAD(iph->saddr), ntohs(th->source), + NIPQUAD(iph->daddr), ntohs(th->dest), + (th->syn ? 1 : 0), (th->ack ? 1 : 0), + (th->fin ? 1 : 0), (th->rst ? 1 : 0), + old_state, new_state); + + conntrack->proto.tcp.state = new_state; + if (old_state != new_state + && (new_state == TCP_CONNTRACK_FIN_WAIT + || new_state == TCP_CONNTRACK_CLOSE)) + conntrack->proto.tcp.seen[dir].flags |= IP_CT_TCP_FLAG_CLOSE_INIT; + timeout = conntrack->proto.tcp.retrans >= nf_ct_tcp_max_retrans + && *tcp_timeouts[new_state] > nf_ct_tcp_timeout_max_retrans + ? nf_ct_tcp_timeout_max_retrans : *tcp_timeouts[new_state]; + write_unlock_bh(&tcp_lock); + + nf_conntrack_event_cache(IPCT_PROTOINFO_VOLATILE, skb); + if (new_state != old_state) + nf_conntrack_event_cache(IPCT_PROTOINFO, skb); + + if (!test_bit(IPS_SEEN_REPLY_BIT, &conntrack->status)) { + /* If only reply is a RST, we can consider ourselves not to + have an established connection: this is a fairly common + problem case, so we can delete the conntrack + immediately. --RR */ + if (th->rst) { + if (del_timer(&conntrack->timeout)) + conntrack->timeout.function((unsigned long) + conntrack); + return NF_ACCEPT; + } + } else if (!test_bit(IPS_ASSURED_BIT, &conntrack->status) + && (old_state == TCP_CONNTRACK_SYN_RECV + || old_state == TCP_CONNTRACK_ESTABLISHED) + && new_state == TCP_CONNTRACK_ESTABLISHED) { + /* Set ASSURED if we see see valid ack in ESTABLISHED + after SYN_RECV or a valid answer for a picked up + connection. */ + set_bit(IPS_ASSURED_BIT, &conntrack->status); + nf_conntrack_event_cache(IPCT_STATUS, skb); + } + nf_ct_refresh_acct(conntrack, ctinfo, skb, timeout); + + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int tcp_new(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff) +{ + enum tcp_conntrack new_state; + struct tcphdr *th, _tcph; +#ifdef DEBUGP_VARS + struct ip_ct_tcp_state *sender = &conntrack->proto.tcp.seen[0]; + struct ip_ct_tcp_state *receiver = &conntrack->proto.tcp.seen[1]; +#endif + + th = skb_header_pointer(skb, dataoff, sizeof(_tcph), &_tcph); + BUG_ON(th == NULL); + + /* Don't need lock here: this conntrack not in circulation yet */ + new_state + = tcp_conntracks[0][get_conntrack_index(th)] + [TCP_CONNTRACK_NONE]; + + /* Invalid: delete conntrack */ + if (new_state >= TCP_CONNTRACK_MAX) { + DEBUGP("nf_ct_tcp: invalid new deleting.\n"); + return 0; + } + + if (new_state == TCP_CONNTRACK_SYN_SENT) { + /* SYN packet */ + conntrack->proto.tcp.seen[0].td_end = + segment_seq_plus_len(ntohl(th->seq), skb->len, + dataoff, th); + conntrack->proto.tcp.seen[0].td_maxwin = ntohs(th->window); + if (conntrack->proto.tcp.seen[0].td_maxwin == 0) + conntrack->proto.tcp.seen[0].td_maxwin = 1; + conntrack->proto.tcp.seen[0].td_maxend = + conntrack->proto.tcp.seen[0].td_end; + + tcp_options(skb, dataoff, th, &conntrack->proto.tcp.seen[0]); + conntrack->proto.tcp.seen[1].flags = 0; + conntrack->proto.tcp.seen[0].loose = + conntrack->proto.tcp.seen[1].loose = 0; + } else if (nf_ct_tcp_loose == 0) { + /* Don't try to pick up connections. */ + return 0; + } else { + /* + * We are in the middle of a connection, + * its history is lost for us. + * Let's try to use the data from the packet. + */ + conntrack->proto.tcp.seen[0].td_end = + segment_seq_plus_len(ntohl(th->seq), skb->len, + dataoff, th); + conntrack->proto.tcp.seen[0].td_maxwin = ntohs(th->window); + if (conntrack->proto.tcp.seen[0].td_maxwin == 0) + conntrack->proto.tcp.seen[0].td_maxwin = 1; + conntrack->proto.tcp.seen[0].td_maxend = + conntrack->proto.tcp.seen[0].td_end + + conntrack->proto.tcp.seen[0].td_maxwin; + conntrack->proto.tcp.seen[0].td_scale = 0; + + /* We assume SACK. Should we assume window scaling too? */ + conntrack->proto.tcp.seen[0].flags = + conntrack->proto.tcp.seen[1].flags = IP_CT_TCP_FLAG_SACK_PERM; + conntrack->proto.tcp.seen[0].loose = + conntrack->proto.tcp.seen[1].loose = nf_ct_tcp_loose; + } + + conntrack->proto.tcp.seen[1].td_end = 0; + conntrack->proto.tcp.seen[1].td_maxend = 0; + conntrack->proto.tcp.seen[1].td_maxwin = 1; + conntrack->proto.tcp.seen[1].td_scale = 0; + + /* tcp_packet will set them */ + conntrack->proto.tcp.state = TCP_CONNTRACK_NONE; + conntrack->proto.tcp.last_index = TCP_NONE_SET; + + DEBUGP("tcp_new: sender end=%u maxend=%u maxwin=%u scale=%i " + "receiver end=%u maxend=%u maxwin=%u scale=%i\n", + sender->td_end, sender->td_maxend, sender->td_maxwin, + sender->td_scale, + receiver->td_end, receiver->td_maxend, receiver->td_maxwin, + receiver->td_scale); + return 1; +} + +struct nf_conntrack_protocol nf_conntrack_protocol_tcp4 = +{ + .l3proto = PF_INET, + .proto = IPPROTO_TCP, + .name = "tcp", + .pkt_to_tuple = tcp_pkt_to_tuple, + .invert_tuple = tcp_invert_tuple, + .print_tuple = tcp_print_tuple, + .print_conntrack = tcp_print_conntrack, + .packet = tcp_packet, + .new = tcp_new, + .error = tcp_error4, +}; + +struct nf_conntrack_protocol nf_conntrack_protocol_tcp6 = +{ + .l3proto = PF_INET6, + .proto = IPPROTO_TCP, + .name = "tcp", + .pkt_to_tuple = tcp_pkt_to_tuple, + .invert_tuple = tcp_invert_tuple, + .print_tuple = tcp_print_tuple, + .print_conntrack = tcp_print_conntrack, + .packet = tcp_packet, + .new = tcp_new, + .error = tcp_error6, +}; + +EXPORT_SYMBOL(nf_conntrack_protocol_tcp4); +EXPORT_SYMBOL(nf_conntrack_protocol_tcp6); diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c new file mode 100644 index 000000000000..3cae7ce420dd --- /dev/null +++ b/net/netfilter/nf_conntrack_proto_udp.c @@ -0,0 +1,216 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI + * - enable working with Layer 3 protocol independent connection tracking. + * + * Derived from net/ipv4/netfilter/ip_conntrack_proto_udp.c + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +unsigned long nf_ct_udp_timeout = 30*HZ; +unsigned long nf_ct_udp_timeout_stream = 180*HZ; + +static int udp_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + struct udphdr _hdr, *hp; + + /* Actually only need first 8 bytes. */ + hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); + if (hp == NULL) + return 0; + + tuple->src.u.udp.port = hp->source; + tuple->dst.u.udp.port = hp->dest; + + return 1; +} + +static int udp_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + tuple->src.u.udp.port = orig->dst.u.udp.port; + tuple->dst.u.udp.port = orig->src.u.udp.port; + return 1; +} + +/* Print out the per-protocol part of the tuple. */ +static int udp_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "sport=%hu dport=%hu ", + ntohs(tuple->src.u.udp.port), + ntohs(tuple->dst.u.udp.port)); +} + +/* Print out the private part of the conntrack. */ +static int udp_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + return 0; +} + +/* Returns verdict for packet, and may modify conntracktype */ +static int udp_packet(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + /* If we've seen traffic both ways, this is some kind of UDP + stream. Extend timeout. */ + if (test_bit(IPS_SEEN_REPLY_BIT, &conntrack->status)) { + nf_ct_refresh_acct(conntrack, ctinfo, skb, + nf_ct_udp_timeout_stream); + /* Also, more likely to be important, and not a probe */ + if (!test_and_set_bit(IPS_ASSURED_BIT, &conntrack->status)) + nf_conntrack_event_cache(IPCT_STATUS, skb); + } else + nf_ct_refresh_acct(conntrack, ctinfo, skb, nf_ct_udp_timeout); + + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int udp_new(struct nf_conn *conntrack, const struct sk_buff *skb, + unsigned int dataoff) +{ + return 1; +} + +static int udp_error(struct sk_buff *skb, unsigned int dataoff, + enum ip_conntrack_info *ctinfo, + int pf, + unsigned int hooknum, + int (*csum)(const struct sk_buff *, unsigned int)) +{ + unsigned int udplen = skb->len - dataoff; + struct udphdr _hdr, *hdr; + + /* Header is too small? */ + hdr = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); + if (hdr == NULL) { + if (LOG_INVALID(IPPROTO_UDP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_udp: short packet "); + return -NF_ACCEPT; + } + + /* Truncated/malformed packets */ + if (ntohs(hdr->len) > udplen || ntohs(hdr->len) < sizeof(*hdr)) { + if (LOG_INVALID(IPPROTO_UDP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_udp: truncated/malformed packet "); + return -NF_ACCEPT; + } + + /* Packet with no checksum */ + if (!hdr->check) + return NF_ACCEPT; + + /* Checksum invalid? Ignore. + * We skip checking packets on the outgoing path + * because the semantic of CHECKSUM_HW is different there + * and moreover root might send raw packets. + * FIXME: Source route IP option packets --RR */ + if (((pf == PF_INET && hooknum == NF_IP_PRE_ROUTING) || + (pf == PF_INET6 && hooknum == NF_IP6_PRE_ROUTING)) + && skb->ip_summed != CHECKSUM_UNNECESSARY + && csum(skb, dataoff)) { + if (LOG_INVALID(IPPROTO_UDP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_udp: bad UDP checksum "); + return -NF_ACCEPT; + } + + return NF_ACCEPT; +} + +static int csum4(const struct sk_buff *skb, unsigned int dataoff) +{ + return csum_tcpudp_magic(skb->nh.iph->saddr, skb->nh.iph->daddr, + skb->len - dataoff, IPPROTO_UDP, + skb->ip_summed == CHECKSUM_HW ? skb->csum + : skb_checksum(skb, dataoff, + skb->len - dataoff, 0)); +} + +static int csum6(const struct sk_buff *skb, unsigned int dataoff) +{ + return csum_ipv6_magic(&skb->nh.ipv6h->saddr, &skb->nh.ipv6h->daddr, + skb->len - dataoff, IPPROTO_UDP, + skb->ip_summed == CHECKSUM_HW ? skb->csum + : skb_checksum(skb, dataoff, skb->len - dataoff, + 0)); +} + +static int udp_error4(struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info *ctinfo, + int pf, + unsigned int hooknum) +{ + return udp_error(skb, dataoff, ctinfo, pf, hooknum, csum4); +} + +static int udp_error6(struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info *ctinfo, + int pf, + unsigned int hooknum) +{ + return udp_error(skb, dataoff, ctinfo, pf, hooknum, csum6); +} + +struct nf_conntrack_protocol nf_conntrack_protocol_udp4 = +{ + .l3proto = PF_INET, + .proto = IPPROTO_UDP, + .name = "udp", + .pkt_to_tuple = udp_pkt_to_tuple, + .invert_tuple = udp_invert_tuple, + .print_tuple = udp_print_tuple, + .print_conntrack = udp_print_conntrack, + .packet = udp_packet, + .new = udp_new, + .error = udp_error4, +}; + +struct nf_conntrack_protocol nf_conntrack_protocol_udp6 = +{ + .l3proto = PF_INET6, + .proto = IPPROTO_UDP, + .name = "udp", + .pkt_to_tuple = udp_pkt_to_tuple, + .invert_tuple = udp_invert_tuple, + .print_tuple = udp_print_tuple, + .print_conntrack = udp_print_conntrack, + .packet = udp_packet, + .new = udp_new, + .error = udp_error6, +}; + +EXPORT_SYMBOL(nf_conntrack_protocol_udp4); +EXPORT_SYMBOL(nf_conntrack_protocol_udp6); diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c new file mode 100644 index 000000000000..45224db4fe2f --- /dev/null +++ b/net/netfilter/nf_conntrack_standalone.c @@ -0,0 +1,869 @@ +/* This file contains all the functions required for the standalone + nf_conntrack module. + + These are not required by the compatibility layer. +*/ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI + * - generalize L3 protocol dependent part. + * + * Derived from net/ipv4/netfilter/ip_conntrack_standalone.c + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#ifdef CONFIG_SYSCTL +#include +#endif + +#define ASSERT_READ_LOCK(x) +#define ASSERT_WRITE_LOCK(x) + +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +MODULE_LICENSE("GPL"); + +extern atomic_t nf_conntrack_count; +DECLARE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); + +static int kill_l3proto(struct nf_conn *i, void *data) +{ + return (i->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num == + ((struct nf_conntrack_l3proto *)data)->l3proto); +} + +static int kill_proto(struct nf_conn *i, void *data) +{ + struct nf_conntrack_protocol *proto; + proto = (struct nf_conntrack_protocol *)data; + return (i->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum == + proto->proto) && + (i->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num == + proto->l3proto); +} + +#ifdef CONFIG_PROC_FS +static int +print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple, + struct nf_conntrack_l3proto *l3proto, + struct nf_conntrack_protocol *proto) +{ + return l3proto->print_tuple(s, tuple) || proto->print_tuple(s, tuple); +} + +#ifdef CONFIG_NF_CT_ACCT +static unsigned int +seq_print_counters(struct seq_file *s, + const struct ip_conntrack_counter *counter) +{ + return seq_printf(s, "packets=%llu bytes=%llu ", + (unsigned long long)counter->packets, + (unsigned long long)counter->bytes); +} +#else +#define seq_print_counters(x, y) 0 +#endif + +struct ct_iter_state { + unsigned int bucket; +}; + +static struct list_head *ct_get_first(struct seq_file *seq) +{ + struct ct_iter_state *st = seq->private; + + for (st->bucket = 0; + st->bucket < nf_conntrack_htable_size; + st->bucket++) { + if (!list_empty(&nf_conntrack_hash[st->bucket])) + return nf_conntrack_hash[st->bucket].next; + } + return NULL; +} + +static struct list_head *ct_get_next(struct seq_file *seq, struct list_head *head) +{ + struct ct_iter_state *st = seq->private; + + head = head->next; + while (head == &nf_conntrack_hash[st->bucket]) { + if (++st->bucket >= nf_conntrack_htable_size) + return NULL; + head = nf_conntrack_hash[st->bucket].next; + } + return head; +} + +static struct list_head *ct_get_idx(struct seq_file *seq, loff_t pos) +{ + struct list_head *head = ct_get_first(seq); + + if (head) + while (pos && (head = ct_get_next(seq, head))) + pos--; + return pos ? NULL : head; +} + +static void *ct_seq_start(struct seq_file *seq, loff_t *pos) +{ + read_lock_bh(&nf_conntrack_lock); + return ct_get_idx(seq, *pos); +} + +static void *ct_seq_next(struct seq_file *s, void *v, loff_t *pos) +{ + (*pos)++; + return ct_get_next(s, v); +} + +static void ct_seq_stop(struct seq_file *s, void *v) +{ + read_unlock_bh(&nf_conntrack_lock); +} + +/* return 0 on success, 1 in case of error */ +static int ct_seq_show(struct seq_file *s, void *v) +{ + const struct nf_conntrack_tuple_hash *hash = v; + const struct nf_conn *conntrack = nf_ct_tuplehash_to_ctrack(hash); + struct nf_conntrack_l3proto *l3proto; + struct nf_conntrack_protocol *proto; + + ASSERT_READ_LOCK(&nf_conntrack_lock); + NF_CT_ASSERT(conntrack); + + /* we only want to print DIR_ORIGINAL */ + if (NF_CT_DIRECTION(hash)) + return 0; + + l3proto = nf_ct_find_l3proto(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); + NF_CT_ASSERT(proto); + + if (seq_printf(s, "%-8s %u %-8s %u %ld ", + l3proto->name, + conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num, + proto->name, + conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum, + timer_pending(&conntrack->timeout) + ? (long)(conntrack->timeout.expires - jiffies)/HZ : 0) != 0) + return -ENOSPC; + + if (l3proto->print_conntrack(s, conntrack)) + return -ENOSPC; + + if (proto->print_conntrack(s, conntrack)) + return -ENOSPC; + + if (print_tuple(s, &conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple, + l3proto, proto)) + return -ENOSPC; + + if (seq_print_counters(s, &conntrack->counters[IP_CT_DIR_ORIGINAL])) + return -ENOSPC; + + if (!(test_bit(IPS_SEEN_REPLY_BIT, &conntrack->status))) + if (seq_printf(s, "[UNREPLIED] ")) + return -ENOSPC; + + if (print_tuple(s, &conntrack->tuplehash[IP_CT_DIR_REPLY].tuple, + l3proto, proto)) + return -ENOSPC; + + if (seq_print_counters(s, &conntrack->counters[IP_CT_DIR_REPLY])) + return -ENOSPC; + + if (test_bit(IPS_ASSURED_BIT, &conntrack->status)) + if (seq_printf(s, "[ASSURED] ")) + return -ENOSPC; + +#if defined(CONFIG_NF_CONNTRACK_MARK) + if (seq_printf(s, "mark=%u ", conntrack->mark)) + return -ENOSPC; +#endif + + if (seq_printf(s, "use=%u\n", atomic_read(&conntrack->ct_general.use))) + return -ENOSPC; + + return 0; +} + +static struct seq_operations ct_seq_ops = { + .start = ct_seq_start, + .next = ct_seq_next, + .stop = ct_seq_stop, + .show = ct_seq_show +}; + +static int ct_open(struct inode *inode, struct file *file) +{ + struct seq_file *seq; + struct ct_iter_state *st; + int ret; + + st = kmalloc(sizeof(struct ct_iter_state), GFP_KERNEL); + if (st == NULL) + return -ENOMEM; + ret = seq_open(file, &ct_seq_ops); + if (ret) + goto out_free; + seq = file->private_data; + seq->private = st; + memset(st, 0, sizeof(struct ct_iter_state)); + return ret; +out_free: + kfree(st); + return ret; +} + +static struct file_operations ct_file_ops = { + .owner = THIS_MODULE, + .open = ct_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; + +/* expects */ +static void *exp_seq_start(struct seq_file *s, loff_t *pos) +{ + struct list_head *e = &nf_conntrack_expect_list; + loff_t i; + + /* strange seq_file api calls stop even if we fail, + * thus we need to grab lock since stop unlocks */ + read_lock_bh(&nf_conntrack_lock); + + if (list_empty(e)) + return NULL; + + for (i = 0; i <= *pos; i++) { + e = e->next; + if (e == &nf_conntrack_expect_list) + return NULL; + } + return e; +} + +static void *exp_seq_next(struct seq_file *s, void *v, loff_t *pos) +{ + struct list_head *e = v; + + ++*pos; + e = e->next; + + if (e == &nf_conntrack_expect_list) + return NULL; + + return e; +} + +static void exp_seq_stop(struct seq_file *s, void *v) +{ + read_unlock_bh(&nf_conntrack_lock); +} + +static int exp_seq_show(struct seq_file *s, void *v) +{ + struct nf_conntrack_expect *expect = v; + + if (expect->timeout.function) + seq_printf(s, "%ld ", timer_pending(&expect->timeout) + ? (long)(expect->timeout.expires - jiffies)/HZ : 0); + else + seq_printf(s, "- "); + seq_printf(s, "l3proto = %u proto=%u ", + 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)); + return seq_putc(s, '\n'); +} + +static struct seq_operations exp_seq_ops = { + .start = exp_seq_start, + .next = exp_seq_next, + .stop = exp_seq_stop, + .show = exp_seq_show +}; + +static int exp_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &exp_seq_ops); +} + +static struct file_operations exp_file_ops = { + .owner = THIS_MODULE, + .open = exp_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release +}; + +static void *ct_cpu_seq_start(struct seq_file *seq, loff_t *pos) +{ + int cpu; + + if (*pos == 0) + return SEQ_START_TOKEN; + + for (cpu = *pos-1; cpu < NR_CPUS; ++cpu) { + if (!cpu_possible(cpu)) + continue; + *pos = cpu + 1; + return &per_cpu(nf_conntrack_stat, cpu); + } + + return NULL; +} + +static void *ct_cpu_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + int cpu; + + for (cpu = *pos; cpu < NR_CPUS; ++cpu) { + if (!cpu_possible(cpu)) + continue; + *pos = cpu + 1; + return &per_cpu(nf_conntrack_stat, cpu); + } + + return NULL; +} + +static void ct_cpu_seq_stop(struct seq_file *seq, void *v) +{ +} + +static int ct_cpu_seq_show(struct seq_file *seq, void *v) +{ + unsigned int nr_conntracks = atomic_read(&nf_conntrack_count); + struct ip_conntrack_stat *st = v; + + if (v == SEQ_START_TOKEN) { + seq_printf(seq, "entries searched found new invalid ignore delete delete_list insert insert_failed drop early_drop icmp_error expect_new expect_create expect_delete\n"); + return 0; + } + + seq_printf(seq, "%08x %08x %08x %08x %08x %08x %08x %08x " + "%08x %08x %08x %08x %08x %08x %08x %08x \n", + nr_conntracks, + st->searched, + st->found, + st->new, + st->invalid, + st->ignore, + st->delete, + st->delete_list, + st->insert, + st->insert_failed, + st->drop, + st->early_drop, + st->error, + + st->expect_new, + st->expect_create, + st->expect_delete + ); + return 0; +} + +static struct seq_operations ct_cpu_seq_ops = { + .start = ct_cpu_seq_start, + .next = ct_cpu_seq_next, + .stop = ct_cpu_seq_stop, + .show = ct_cpu_seq_show, +}; + +static int ct_cpu_seq_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &ct_cpu_seq_ops); +} + +static struct file_operations ct_cpu_seq_fops = { + .owner = THIS_MODULE, + .open = ct_cpu_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release_private, +}; +#endif /* CONFIG_PROC_FS */ + +/* Sysctl support */ + +#ifdef CONFIG_SYSCTL + +/* From nf_conntrack_core.c */ +extern int nf_conntrack_max; +extern unsigned int nf_conntrack_htable_size; + +/* From nf_conntrack_proto_tcp.c */ +extern unsigned long nf_ct_tcp_timeout_syn_sent; +extern unsigned long nf_ct_tcp_timeout_syn_recv; +extern unsigned long nf_ct_tcp_timeout_established; +extern unsigned long nf_ct_tcp_timeout_fin_wait; +extern unsigned long nf_ct_tcp_timeout_close_wait; +extern unsigned long nf_ct_tcp_timeout_last_ack; +extern unsigned long nf_ct_tcp_timeout_time_wait; +extern unsigned long nf_ct_tcp_timeout_close; +extern unsigned long nf_ct_tcp_timeout_max_retrans; +extern int nf_ct_tcp_loose; +extern int nf_ct_tcp_be_liberal; +extern int nf_ct_tcp_max_retrans; + +/* From nf_conntrack_proto_udp.c */ +extern unsigned long nf_ct_udp_timeout; +extern unsigned long nf_ct_udp_timeout_stream; + +/* From nf_conntrack_proto_generic.c */ +extern unsigned long nf_ct_generic_timeout; + +/* Log invalid packets of a given protocol */ +static int log_invalid_proto_min = 0; +static int log_invalid_proto_max = 255; + +static struct ctl_table_header *nf_ct_sysctl_header; + +static ctl_table nf_ct_sysctl_table[] = { + { + .ctl_name = NET_NF_CONNTRACK_MAX, + .procname = "nf_conntrack_max", + .data = &nf_conntrack_max, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_NF_CONNTRACK_COUNT, + .procname = "nf_conntrack_count", + .data = &nf_conntrack_count, + .maxlen = sizeof(int), + .mode = 0444, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_NF_CONNTRACK_BUCKETS, + .procname = "nf_conntrack_buckets", + .data = &nf_conntrack_htable_size, + .maxlen = sizeof(unsigned int), + .mode = 0444, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_SYN_SENT, + .procname = "nf_conntrack_tcp_timeout_syn_sent", + .data = &nf_ct_tcp_timeout_syn_sent, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_SYN_RECV, + .procname = "nf_conntrack_tcp_timeout_syn_recv", + .data = &nf_ct_tcp_timeout_syn_recv, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_ESTABLISHED, + .procname = "nf_conntrack_tcp_timeout_established", + .data = &nf_ct_tcp_timeout_established, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_FIN_WAIT, + .procname = "nf_conntrack_tcp_timeout_fin_wait", + .data = &nf_ct_tcp_timeout_fin_wait, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_CLOSE_WAIT, + .procname = "nf_conntrack_tcp_timeout_close_wait", + .data = &nf_ct_tcp_timeout_close_wait, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_LAST_ACK, + .procname = "nf_conntrack_tcp_timeout_last_ack", + .data = &nf_ct_tcp_timeout_last_ack, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_TIME_WAIT, + .procname = "nf_conntrack_tcp_timeout_time_wait", + .data = &nf_ct_tcp_timeout_time_wait, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_CLOSE, + .procname = "nf_conntrack_tcp_timeout_close", + .data = &nf_ct_tcp_timeout_close, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_UDP_TIMEOUT, + .procname = "nf_conntrack_udp_timeout", + .data = &nf_ct_udp_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_UDP_TIMEOUT_STREAM, + .procname = "nf_conntrack_udp_timeout_stream", + .data = &nf_ct_udp_timeout_stream, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_GENERIC_TIMEOUT, + .procname = "nf_conntrack_generic_timeout", + .data = &nf_ct_generic_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_LOG_INVALID, + .procname = "nf_conntrack_log_invalid", + .data = &nf_ct_log_invalid, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_minmax, + .strategy = &sysctl_intvec, + .extra1 = &log_invalid_proto_min, + .extra2 = &log_invalid_proto_max, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_TIMEOUT_MAX_RETRANS, + .procname = "nf_conntrack_tcp_timeout_max_retrans", + .data = &nf_ct_tcp_timeout_max_retrans, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_LOOSE, + .procname = "nf_conntrack_tcp_loose", + .data = &nf_ct_tcp_loose, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_BE_LIBERAL, + .procname = "nf_conntrack_tcp_be_liberal", + .data = &nf_ct_tcp_be_liberal, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_NF_CONNTRACK_TCP_MAX_RETRANS, + .procname = "nf_conntrack_tcp_max_retrans", + .data = &nf_ct_tcp_max_retrans, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + + { .ctl_name = 0 } +}; + +#define NET_NF_CONNTRACK_MAX 2089 + +static ctl_table nf_ct_netfilter_table[] = { + { + .ctl_name = NET_NETFILTER, + .procname = "netfilter", + .mode = 0555, + .child = nf_ct_sysctl_table, + }, + { + .ctl_name = NET_NF_CONNTRACK_MAX, + .procname = "nf_conntrack_max", + .data = &nf_conntrack_max, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_net_table[] = { + { + .ctl_name = CTL_NET, + .procname = "net", + .mode = 0555, + .child = nf_ct_netfilter_table, + }, + { .ctl_name = 0 } +}; +EXPORT_SYMBOL(nf_ct_log_invalid); +#endif /* CONFIG_SYSCTL */ + +static int init_or_cleanup(int init) +{ +#ifdef CONFIG_PROC_FS + struct proc_dir_entry *proc, *proc_exp, *proc_stat; +#endif + int ret = 0; + + if (!init) goto cleanup; + + ret = nf_conntrack_init(); + if (ret < 0) + goto cleanup_nothing; + +#ifdef CONFIG_PROC_FS + proc = proc_net_fops_create("nf_conntrack", 0440, &ct_file_ops); + if (!proc) goto cleanup_init; + + proc_exp = proc_net_fops_create("nf_conntrack_expect", 0440, + &exp_file_ops); + if (!proc_exp) goto cleanup_proc; + + proc_stat = create_proc_entry("nf_conntrack", S_IRUGO, proc_net_stat); + if (!proc_stat) + goto cleanup_proc_exp; + + proc_stat->proc_fops = &ct_cpu_seq_fops; + proc_stat->owner = THIS_MODULE; +#endif +#ifdef CONFIG_SYSCTL + nf_ct_sysctl_header = register_sysctl_table(nf_ct_net_table, 0); + if (nf_ct_sysctl_header == NULL) { + printk("nf_conntrack: can't register to sysctl.\n"); + ret = -ENOMEM; + goto cleanup_proc_stat; + } +#endif + + return ret; + + cleanup: +#ifdef CONFIG_SYSCTL + unregister_sysctl_table(nf_ct_sysctl_header); + cleanup_proc_stat: +#endif +#ifdef CONFIG_PROC_FS + proc_net_remove("nf_conntrack_stat"); + cleanup_proc_exp: + proc_net_remove("nf_conntrack_expect"); + cleanup_proc: + proc_net_remove("nf_conntrack"); + cleanup_init: +#endif /* CNFIG_PROC_FS */ + nf_conntrack_cleanup(); + cleanup_nothing: + return ret; +} + +int nf_conntrack_l3proto_register(struct nf_conntrack_l3proto *proto) +{ + int ret = 0; + + write_lock_bh(&nf_conntrack_lock); + if (nf_ct_l3protos[proto->l3proto] != &nf_conntrack_generic_l3proto) { + ret = -EBUSY; + goto out; + } + nf_ct_l3protos[proto->l3proto] = proto; +out: + write_unlock_bh(&nf_conntrack_lock); + + return ret; +} + +void nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto) +{ + write_lock_bh(&nf_conntrack_lock); + nf_ct_l3protos[proto->l3proto] = &nf_conntrack_generic_l3proto; + write_unlock_bh(&nf_conntrack_lock); + + /* Somebody could be still looking at the proto in bh. */ + synchronize_net(); + + /* Remove all contrack entries for this protocol */ + nf_ct_iterate_cleanup(kill_l3proto, proto); +} + +/* FIXME: Allow NULL functions and sub in pointers to generic for + them. --RR */ +int nf_conntrack_protocol_register(struct nf_conntrack_protocol *proto) +{ + int ret = 0; + +retry: + write_lock_bh(&nf_conntrack_lock); + if (nf_ct_protos[proto->l3proto]) { + if (nf_ct_protos[proto->l3proto][proto->proto] + != &nf_conntrack_generic_protocol) { + ret = -EBUSY; + goto out_unlock; + } + } else { + /* l3proto may be loaded latter. */ + struct nf_conntrack_protocol **proto_array; + int i; + + write_unlock_bh(&nf_conntrack_lock); + + proto_array = (struct nf_conntrack_protocol **) + kmalloc(MAX_NF_CT_PROTO * + sizeof(struct nf_conntrack_protocol *), + GFP_KERNEL); + if (proto_array == NULL) { + ret = -ENOMEM; + goto out; + } + for (i = 0; i < MAX_NF_CT_PROTO; i++) + proto_array[i] = &nf_conntrack_generic_protocol; + + write_lock_bh(&nf_conntrack_lock); + if (nf_ct_protos[proto->l3proto]) { + /* bad timing, but no problem */ + write_unlock_bh(&nf_conntrack_lock); + kfree(proto_array); + } else { + nf_ct_protos[proto->l3proto] = proto_array; + write_unlock_bh(&nf_conntrack_lock); + } + + /* + * Just once because array is never freed until unloading + * nf_conntrack.ko + */ + goto retry; + } + + nf_ct_protos[proto->l3proto][proto->proto] = proto; + +out_unlock: + write_unlock_bh(&nf_conntrack_lock); +out: + return ret; +} + +void nf_conntrack_protocol_unregister(struct nf_conntrack_protocol *proto) +{ + write_lock_bh(&nf_conntrack_lock); + nf_ct_protos[proto->l3proto][proto->proto] + = &nf_conntrack_generic_protocol; + write_unlock_bh(&nf_conntrack_lock); + + /* Somebody could be still looking at the proto in bh. */ + synchronize_net(); + + /* Remove all contrack entries for this protocol */ + nf_ct_iterate_cleanup(kill_proto, proto); +} + +static int __init init(void) +{ + return init_or_cleanup(1); +} + +static void __exit fini(void) +{ + init_or_cleanup(0); +} + +module_init(init); +module_exit(fini); + +/* Some modules need us, but don't depend directly on any symbol. + They should call this. */ +void need_nf_conntrack(void) +{ +} + +#ifdef CONFIG_NF_CONNTRACK_EVENTS +EXPORT_SYMBOL_GPL(nf_conntrack_chain); +EXPORT_SYMBOL_GPL(nf_conntrack_expect_chain); +EXPORT_SYMBOL_GPL(nf_conntrack_register_notifier); +EXPORT_SYMBOL_GPL(nf_conntrack_unregister_notifier); +EXPORT_SYMBOL_GPL(__nf_ct_event_cache_init); +EXPORT_PER_CPU_SYMBOL_GPL(nf_conntrack_ecache); +EXPORT_SYMBOL_GPL(nf_ct_deliver_cached_events); +#endif +EXPORT_SYMBOL(nf_conntrack_l3proto_register); +EXPORT_SYMBOL(nf_conntrack_l3proto_unregister); +EXPORT_SYMBOL(nf_conntrack_protocol_register); +EXPORT_SYMBOL(nf_conntrack_protocol_unregister); +EXPORT_SYMBOL(nf_ct_invert_tuplepr); +EXPORT_SYMBOL(nf_conntrack_alter_reply); +EXPORT_SYMBOL(nf_conntrack_destroyed); +EXPORT_SYMBOL(need_nf_conntrack); +EXPORT_SYMBOL(nf_conntrack_helper_register); +EXPORT_SYMBOL(nf_conntrack_helper_unregister); +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_l3protos); +EXPORT_SYMBOL(nf_conntrack_expect_alloc); +EXPORT_SYMBOL(nf_conntrack_expect_put); +EXPORT_SYMBOL(nf_conntrack_expect_related); +EXPORT_SYMBOL(nf_conntrack_unexpect_related); +EXPORT_SYMBOL(nf_conntrack_tuple_taken); +EXPORT_SYMBOL(nf_conntrack_htable_size); +EXPORT_SYMBOL(nf_conntrack_lock); +EXPORT_SYMBOL(nf_conntrack_hash); +EXPORT_SYMBOL(nf_conntrack_untracked); +EXPORT_SYMBOL_GPL(nf_conntrack_find_get); +#ifdef CONFIG_IP_NF_NAT_NEEDED +EXPORT_SYMBOL(nf_conntrack_tcp_update); +#endif +EXPORT_SYMBOL(__nf_conntrack_confirm); +EXPORT_SYMBOL(nf_ct_get_tuple); +EXPORT_SYMBOL(nf_ct_invert_tuple); +EXPORT_SYMBOL(nf_conntrack_in); +EXPORT_SYMBOL(__nf_conntrack_attach); -- cgit v1.2.2 From bfa83a9e03cf8d501c6272999843470afecb32ed Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 10 Nov 2005 02:25:51 +0100 Subject: [NETLINK]: Type-safe netlink messages/attributes interface Introduces a new type-safe interface for netlink message and attributes handling. The interface is fully binary compatible with the old interface towards userspace. Besides type safety, this interface features attribute validation capabilities, simplified message contstruction, and documentation. The resulting netlink code should be smaller, less error prone and easier to understand. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/netlink/Makefile | 2 +- net/netlink/attr.c | 328 +++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 329 insertions(+), 1 deletion(-) create mode 100644 net/netlink/attr.c (limited to 'net') diff --git a/net/netlink/Makefile b/net/netlink/Makefile index 39d9c2dcd03c..68179015bd39 100644 --- a/net/netlink/Makefile +++ b/net/netlink/Makefile @@ -2,4 +2,4 @@ # Makefile for the netlink driver. # -obj-y := af_netlink.o +obj-y := af_netlink.o attr.o diff --git a/net/netlink/attr.c b/net/netlink/attr.c new file mode 100644 index 000000000000..fffef4ab276f --- /dev/null +++ b/net/netlink/attr.c @@ -0,0 +1,328 @@ +/* + * NETLINK Netlink attributes + * + * Authors: Thomas Graf + * Alexey Kuznetsov + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +static u16 nla_attr_minlen[NLA_TYPE_MAX+1] __read_mostly = { + [NLA_U8] = sizeof(u8), + [NLA_U16] = sizeof(u16), + [NLA_U32] = sizeof(u32), + [NLA_U64] = sizeof(u64), + [NLA_STRING] = 1, + [NLA_NESTED] = NLA_HDRLEN, +}; + +static int validate_nla(struct nlattr *nla, int maxtype, + struct nla_policy *policy) +{ + struct nla_policy *pt; + int minlen = 0; + + if (nla->nla_type <= 0 || nla->nla_type > maxtype) + return 0; + + pt = &policy[nla->nla_type]; + + BUG_ON(pt->type > NLA_TYPE_MAX); + + if (pt->minlen) + minlen = pt->minlen; + else if (pt->type != NLA_UNSPEC) + minlen = nla_attr_minlen[pt->type]; + + if (pt->type == NLA_FLAG && nla_len(nla) > 0) + return -ERANGE; + + if (nla_len(nla) < minlen) + return -ERANGE; + + return 0; +} + +/** + * nla_validate - Validate a stream of attributes + * @head: head of attribute stream + * @len: length of attribute stream + * @maxtype: maximum attribute type to be expected + * @policy: validation policy + * + * Validates all attributes in the specified attribute stream against the + * specified policy. Attributes with a type exceeding maxtype will be + * ignored. See documenation of struct nla_policy for more details. + * + * Returns 0 on success or a negative error code. + */ +int nla_validate(struct nlattr *head, int len, int maxtype, + struct nla_policy *policy) +{ + struct nlattr *nla; + int rem, err; + + nla_for_each_attr(nla, head, len, rem) { + err = validate_nla(nla, maxtype, policy); + if (err < 0) + goto errout; + } + + err = 0; +errout: + return err; +} + +/** + * nla_parse - Parse a stream of attributes into a tb buffer + * @tb: destination array with maxtype+1 elements + * @maxtype: maximum attribute type to be expected + * @head: head of attribute stream + * @len: length of attribute stream + * + * Parses a stream of attributes and stores a pointer to each attribute in + * the tb array accessable via the attribute type. Attributes with a type + * exceeding maxtype will be silently ignored for backwards compatibility + * reasons. policy may be set to NULL if no validation is required. + * + * Returns 0 on success or a negative error code. + */ +int nla_parse(struct nlattr *tb[], int maxtype, struct nlattr *head, int len, + struct nla_policy *policy) +{ + struct nlattr *nla; + int rem, err; + + memset(tb, 0, sizeof(struct nlattr *) * (maxtype + 1)); + + nla_for_each_attr(nla, head, len, rem) { + u16 type = nla->nla_type; + + if (type > 0 && type <= maxtype) { + if (policy) { + err = validate_nla(nla, maxtype, policy); + if (err < 0) + goto errout; + } + + tb[type] = nla; + } + } + + if (unlikely(rem > 0)) + printk(KERN_WARNING "netlink: %d bytes leftover after parsing " + "attributes.\n", rem); + + err = 0; +errout: + return err; +} + +/** + * nla_find - Find a specific attribute in a stream of attributes + * @head: head of attribute stream + * @len: length of attribute stream + * @attrtype: type of attribute to look for + * + * Returns the first attribute in the stream matching the specified type. + */ +struct nlattr *nla_find(struct nlattr *head, int len, int attrtype) +{ + struct nlattr *nla; + int rem; + + nla_for_each_attr(nla, head, len, rem) + if (nla->nla_type == attrtype) + return nla; + + return NULL; +} + +/** + * nla_strlcpy - Copy string attribute payload into a sized buffer + * @dst: where to copy the string to + * @src: attribute to copy the string from + * @dstsize: size of destination buffer + * + * Copies at most dstsize - 1 bytes into the destination buffer. + * The result is always a valid NUL-terminated string. Unlike + * strlcpy the destination buffer is always padded out. + * + * Returns the length of the source buffer. + */ +size_t nla_strlcpy(char *dst, const struct nlattr *nla, size_t dstsize) +{ + size_t srclen = nla_len(nla); + char *src = nla_data(nla); + + if (srclen > 0 && src[srclen - 1] == '\0') + srclen--; + + if (dstsize > 0) { + size_t len = (srclen >= dstsize) ? dstsize - 1 : srclen; + + memset(dst, 0, dstsize); + memcpy(dst, src, len); + } + + return srclen; +} + +/** + * nla_memcpy - Copy a netlink attribute into another memory area + * @dest: where to copy to memcpy + * @src: netlink attribute to copy from + * @count: size of the destination area + * + * Note: The number of bytes copied is limited by the length of + * attribute's payload. memcpy + * + * Returns the number of bytes copied. + */ +int nla_memcpy(void *dest, struct nlattr *src, int count) +{ + int minlen = min_t(int, count, nla_len(src)); + + memcpy(dest, nla_data(src), minlen); + + return minlen; +} + +/** + * nla_memcmp - Compare an attribute with sized memory area + * @nla: netlink attribute + * @data: memory area + * @size: size of memory area + */ +int nla_memcmp(const struct nlattr *nla, const void *data, + size_t size) +{ + int d = nla_len(nla) - size; + + if (d == 0) + d = memcmp(nla_data(nla), data, size); + + return d; +} + +/** + * nla_strcmp - Compare a string attribute against a string + * @nla: netlink string attribute + * @str: another string + */ +int nla_strcmp(const struct nlattr *nla, const char *str) +{ + int len = strlen(str) + 1; + int d = nla_len(nla) - len; + + if (d == 0) + d = memcmp(nla_data(nla), str, len); + + return d; +} + +/** + * __nla_reserve - reserve room for attribute on the skb + * @skb: socket buffer to reserve room on + * @attrtype: attribute type + * @attrlen: length of attribute payload + * + * Adds a netlink attribute header to a socket buffer and reserves + * room for the payload but does not copy it. + * + * The caller is responsible to ensure that the skb provides enough + * tailroom for the attribute header and payload. + */ +struct nlattr *__nla_reserve(struct sk_buff *skb, int attrtype, int attrlen) +{ + struct nlattr *nla; + + nla = (struct nlattr *) skb_put(skb, nla_total_size(attrlen)); + nla->nla_type = attrtype; + nla->nla_len = nla_attr_size(attrlen); + + memset((unsigned char *) nla + nla->nla_len, 0, nla_padlen(attrlen)); + + return nla; +} + +/** + * nla_reserve - reserve room for attribute on the skb + * @skb: socket buffer to reserve room on + * @attrtype: attribute type + * @attrlen: length of attribute payload + * + * Adds a netlink attribute header to a socket buffer and reserves + * room for the payload but does not copy it. + * + * Returns NULL if the tailroom of the skb is insufficient to store + * the attribute header and payload. + */ +struct nlattr *nla_reserve(struct sk_buff *skb, int attrtype, int attrlen) +{ + if (unlikely(skb_tailroom(skb) < nla_total_size(attrlen))) + return NULL; + + return __nla_reserve(skb, attrtype, attrlen); +} + +/** + * __nla_put - Add a netlink attribute to a socket buffer + * @skb: socket buffer to add attribute to + * @attrtype: attribute type + * @attrlen: length of attribute payload + * @data: head of attribute payload + * + * The caller is responsible to ensure that the skb provides enough + * tailroom for the attribute header and payload. + */ +void __nla_put(struct sk_buff *skb, int attrtype, int attrlen, + const void *data) +{ + struct nlattr *nla; + + nla = __nla_reserve(skb, attrtype, attrlen); + memcpy(nla_data(nla), data, attrlen); +} + + +/** + * nla_put - Add a netlink attribute to a socket buffer + * @skb: socket buffer to add attribute to + * @attrtype: attribute type + * @attrlen: length of attribute payload + * @data: head of attribute payload + * + * Returns -1 if the tailroom of the skb is insufficient to store + * the attribute header and payload. + */ +int nla_put(struct sk_buff *skb, int attrtype, int attrlen, const void *data) +{ + if (unlikely(skb_tailroom(skb) < nla_total_size(attrlen))) + return -1; + + __nla_put(skb, attrtype, attrlen, data); + return 0; +} + + +EXPORT_SYMBOL(nla_validate); +EXPORT_SYMBOL(nla_parse); +EXPORT_SYMBOL(nla_find); +EXPORT_SYMBOL(nla_strlcpy); +EXPORT_SYMBOL(__nla_reserve); +EXPORT_SYMBOL(nla_reserve); +EXPORT_SYMBOL(__nla_put); +EXPORT_SYMBOL(nla_put); +EXPORT_SYMBOL(nla_memcpy); +EXPORT_SYMBOL(nla_memcmp); +EXPORT_SYMBOL(nla_strcmp); -- cgit v1.2.2 From a8f74b228826eef1cbe04a05647d61e896f5fd63 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 10 Nov 2005 02:25:52 +0100 Subject: [NETLINK]: Make netlink_callback->done() optional Most netlink families make no use of the done() callback, making it optional gets rid of all unnecessary dummy implementations. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/core/rtnetlink.c | 8 +------- net/ipv4/inet_diag.c | 9 +-------- net/ipv6/route.c | 2 +- net/netlink/af_netlink.c | 6 ++++-- net/xfrm/xfrm_user.c | 8 +------- 5 files changed, 8 insertions(+), 25 deletions(-) (limited to 'net') diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 9bed7569ce3f..193fd8637f9f 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -462,11 +462,6 @@ void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change) netlink_broadcast(rtnl, skb, 0, RTNLGRP_LINK, GFP_KERNEL); } -static int rtnetlink_done(struct netlink_callback *cb) -{ - return 0; -} - /* Protected by RTNL sempahore. */ static struct rtattr **rta_buf; static int rtattr_max; @@ -533,8 +528,7 @@ rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, int *errp) goto err_inval; if ((*errp = netlink_dump_start(rtnl, skb, nlh, - link->dumpit, - rtnetlink_done)) != 0) { + link->dumpit, NULL)) != 0) { return -1; } rlen = NLMSG_ALIGN(nlh->nlmsg_len); diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index 71f3c7350c6e..39061ed53cfd 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -724,12 +724,6 @@ done: return skb->len; } -static int inet_diag_dump_done(struct netlink_callback *cb) -{ - return 0; -} - - static __inline__ int inet_diag_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) { @@ -760,8 +754,7 @@ inet_diag_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) goto err_inval; } return netlink_dump_start(idiagnl, skb, nlh, - inet_diag_dump, - inet_diag_dump_done); + inet_diag_dump, NULL); } else { return inet_diag_get_exact(skb, nlh); } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 227e99ed510c..f7f42c3e96cb 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1710,7 +1710,7 @@ static void fib6_dump_end(struct netlink_callback *cb) static int fib6_dump_done(struct netlink_callback *cb) { fib6_dump_end(cb); - return cb->done(cb); + return cb->done ? cb->done(cb) : 0; } int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 5ca283537bc6..f3fb7e575816 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -427,7 +427,8 @@ static int netlink_release(struct socket *sock) spin_lock(&nlk->cb_lock); if (nlk->cb) { - nlk->cb->done(nlk->cb); + if (nlk->cb->done) + nlk->cb->done(nlk->cb); netlink_destroy_callback(nlk->cb); nlk->cb = NULL; } @@ -1322,7 +1323,8 @@ static int netlink_dump(struct sock *sk) skb_queue_tail(&sk->sk_receive_queue, skb); sk->sk_data_ready(sk, skb->len); - cb->done(cb); + if (cb->done) + cb->done(cb); nlk->cb = NULL; spin_unlock(&nlk->cb_lock); diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index c35336a0f71b..6996b7e85665 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -948,11 +948,6 @@ static struct xfrm_link { [XFRM_MSG_FLUSHPOLICY - XFRM_MSG_BASE] = { .doit = xfrm_flush_policy }, }; -static int xfrm_done(struct netlink_callback *cb) -{ - return 0; -} - static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, int *errp) { struct rtattr *xfrma[XFRMA_MAX]; @@ -990,8 +985,7 @@ static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, int *err goto err_einval; if ((*errp = netlink_dump_start(xfrm_nl, skb, nlh, - link->dump, - xfrm_done)) != 0) { + link->dump, NULL)) != 0) { return -1; } rlen = NLMSG_ALIGN(nlh->nlmsg_len); -- cgit v1.2.2 From 82ace47a7256fd39d370a6442e0649f75961b831 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 10 Nov 2005 02:25:53 +0100 Subject: [NETLINK]: Generic netlink receive queue processor Introduces netlink_run_queue() to handle the receive queue of a netlink socket in a generic way. Processes as much as there was in the queue upon entry and invokes a callback function for each netlink message found. The callback function may refuse a message by returning a negative error code but setting the error pointer to 0 in which case netlink_run_queue() will return with a qlen != 0. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/netlink/af_netlink.c | 91 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 91 insertions(+) (limited to 'net') diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index f3fb7e575816..8c38ee6d255e 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -58,6 +58,7 @@ #include #include +#include #define Nprintk(a...) #define NLGRPSZ(x) (ALIGN(x, sizeof(unsigned long) * 8) / 8) @@ -1411,6 +1412,94 @@ void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err) netlink_unicast(in_skb->sk, skb, NETLINK_CB(in_skb).pid, MSG_DONTWAIT); } +static int netlink_rcv_skb(struct sk_buff *skb, int (*cb)(struct sk_buff *, + struct nlmsghdr *, int *)) +{ + unsigned int total_len; + struct nlmsghdr *nlh; + int err; + + while (skb->len >= nlmsg_total_size(0)) { + nlh = (struct nlmsghdr *) skb->data; + + if (skb->len < nlh->nlmsg_len) + return 0; + + total_len = min(NLMSG_ALIGN(nlh->nlmsg_len), skb->len); + + if (cb(skb, nlh, &err) < 0) { + /* Not an error, but we have to interrupt processing + * here. Note: that in this case we do not pull + * message from skb, it will be processed later. + */ + if (err == 0) + return -1; + netlink_ack(skb, nlh, err); + } else if (nlh->nlmsg_flags & NLM_F_ACK) + netlink_ack(skb, nlh, 0); + + skb_pull(skb, total_len); + } + + return 0; +} + +/** + * nelink_run_queue - Process netlink receive queue. + * @sk: Netlink socket containing the queue + * @qlen: Place to store queue length upon entry + * @cb: Callback function invoked for each netlink message found + * + * Processes as much as there was in the queue upon entry and invokes + * a callback function for each netlink message found. The callback + * function may refuse a message by returning a negative error code + * but setting the error pointer to 0 in which case this function + * returns with a qlen != 0. + * + * qlen must be initialized to 0 before the initial entry, afterwards + * the function may be called repeatedly until qlen reaches 0. + */ +void netlink_run_queue(struct sock *sk, unsigned int *qlen, + int (*cb)(struct sk_buff *, struct nlmsghdr *, int *)) +{ + struct sk_buff *skb; + + if (!*qlen || *qlen > skb_queue_len(&sk->sk_receive_queue)) + *qlen = skb_queue_len(&sk->sk_receive_queue); + + for (; *qlen; (*qlen)--) { + skb = skb_dequeue(&sk->sk_receive_queue); + if (netlink_rcv_skb(skb, cb)) { + if (skb->len) + skb_queue_head(&sk->sk_receive_queue, skb); + else { + kfree_skb(skb); + (*qlen)--; + } + break; + } + + kfree_skb(skb); + } +} + +/** + * netlink_queue_skip - Skip netlink message while processing queue. + * @nlh: Netlink message to be skipped + * @skb: Socket buffer containing the netlink messages. + * + * Pulls the given netlink message off the socket buffer so the next + * call to netlink_queue_run() will not reconsider the message. + */ +void netlink_queue_skip(struct nlmsghdr *nlh, struct sk_buff *skb) +{ + int msglen = NLMSG_ALIGN(nlh->nlmsg_len); + + if (msglen > skb->len) + msglen = skb->len; + + skb_pull(skb, msglen); +} #ifdef CONFIG_PROC_FS struct nl_seq_iter { @@ -1659,6 +1748,8 @@ out: core_initcall(netlink_proto_init); EXPORT_SYMBOL(netlink_ack); +EXPORT_SYMBOL(netlink_run_queue); +EXPORT_SYMBOL(netlink_queue_skip); EXPORT_SYMBOL(netlink_broadcast); EXPORT_SYMBOL(netlink_dump_start); EXPORT_SYMBOL(netlink_kernel_create); -- cgit v1.2.2 From 88fc2c84312d095545c08a9f871ad1888a688cf6 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 10 Nov 2005 02:25:54 +0100 Subject: [XFRM]: Use generic netlink receive queue processor Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/xfrm/xfrm_user.c | 61 +++++----------------------------------------------- 1 file changed, 5 insertions(+), 56 deletions(-) (limited to 'net') diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 6996b7e85665..0cdd9a07e043 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -18,7 +18,6 @@ #include #include #include -#include #include #include #include @@ -26,6 +25,7 @@ #include #include #include +#include #include static struct sock *xfrm_nl; @@ -979,8 +979,6 @@ static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, int *err if ((type == (XFRM_MSG_GETSA - XFRM_MSG_BASE) || type == (XFRM_MSG_GETPOLICY - XFRM_MSG_BASE)) && (nlh->nlmsg_flags & NLM_F_DUMP)) { - u32 rlen; - if (link->dump == NULL) goto err_einval; @@ -988,10 +986,8 @@ static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, int *err link->dump, NULL)) != 0) { return -1; } - rlen = NLMSG_ALIGN(nlh->nlmsg_len); - if (rlen > skb->len) - rlen = skb->len; - skb_pull(skb, rlen); + + netlink_queue_skip(nlh, skb); return -1; } @@ -1026,60 +1022,13 @@ err_einval: return -1; } -static int xfrm_user_rcv_skb(struct sk_buff *skb) -{ - int err; - struct nlmsghdr *nlh; - - while (skb->len >= NLMSG_SPACE(0)) { - u32 rlen; - - nlh = (struct nlmsghdr *) skb->data; - if (nlh->nlmsg_len < sizeof(*nlh) || - skb->len < nlh->nlmsg_len) - return 0; - rlen = NLMSG_ALIGN(nlh->nlmsg_len); - if (rlen > skb->len) - rlen = skb->len; - if (xfrm_user_rcv_msg(skb, nlh, &err) < 0) { - if (err == 0) - return -1; - netlink_ack(skb, nlh, err); - } else if (nlh->nlmsg_flags & NLM_F_ACK) - netlink_ack(skb, nlh, 0); - skb_pull(skb, rlen); - } - - return 0; -} - static void xfrm_netlink_rcv(struct sock *sk, int len) { - unsigned int qlen = skb_queue_len(&sk->sk_receive_queue); + unsigned int qlen = 0; do { - struct sk_buff *skb; - down(&xfrm_cfg_sem); - - if (qlen > skb_queue_len(&sk->sk_receive_queue)) - qlen = skb_queue_len(&sk->sk_receive_queue); - - for (; qlen; qlen--) { - skb = skb_dequeue(&sk->sk_receive_queue); - if (xfrm_user_rcv_skb(skb)) { - if (skb->len) - skb_queue_head(&sk->sk_receive_queue, - skb); - else { - kfree_skb(skb); - qlen--; - } - break; - } - kfree_skb(skb); - } - + netlink_run_queue(sk, &qlen, &xfrm_user_rcv_msg); up(&xfrm_cfg_sem); } while (qlen); -- cgit v1.2.2 From 9ac4a16983ea4edf719c390a1a234d956947688d Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 10 Nov 2005 02:25:55 +0100 Subject: [RTNETLINK]: Use generic netlink receive queue processor Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/core/rtnetlink.c | 75 ++++------------------------------------------------ 1 file changed, 5 insertions(+), 70 deletions(-) (limited to 'net') diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 193fd8637f9f..8700379685e0 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -49,6 +49,7 @@ #include #include #include +#include DECLARE_MUTEX(rtnl_sem); @@ -519,8 +520,6 @@ rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, int *errp) } if (kind == 2 && nlh->nlmsg_flags&NLM_F_DUMP) { - u32 rlen; - if (link->dumpit == NULL) link = &(rtnetlink_links[PF_UNSPEC][type]); @@ -531,10 +530,8 @@ rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, int *errp) link->dumpit, NULL)) != 0) { return -1; } - rlen = NLMSG_ALIGN(nlh->nlmsg_len); - if (rlen > skb->len) - rlen = skb->len; - skb_pull(skb, rlen); + + netlink_queue_skip(nlh, skb); return -1; } @@ -573,75 +570,13 @@ err_inval: return -1; } -/* - * Process one packet of messages. - * Malformed skbs with wrong lengths of messages are discarded silently. - */ - -static inline int rtnetlink_rcv_skb(struct sk_buff *skb) -{ - int err; - struct nlmsghdr * nlh; - - while (skb->len >= NLMSG_SPACE(0)) { - u32 rlen; - - nlh = (struct nlmsghdr *)skb->data; - if (nlh->nlmsg_len < sizeof(*nlh) || skb->len < nlh->nlmsg_len) - return 0; - rlen = NLMSG_ALIGN(nlh->nlmsg_len); - if (rlen > skb->len) - rlen = skb->len; - if (rtnetlink_rcv_msg(skb, nlh, &err)) { - /* Not error, but we must interrupt processing here: - * Note, that in this case we do not pull message - * from skb, it will be processed later. - */ - if (err == 0) - return -1; - netlink_ack(skb, nlh, err); - } else if (nlh->nlmsg_flags&NLM_F_ACK) - netlink_ack(skb, nlh, 0); - skb_pull(skb, rlen); - } - - return 0; -} - -/* - * rtnetlink input queue processing routine: - * - process as much as there was in the queue upon entry. - * - feed skbs to rtnetlink_rcv_skb, until it refuse a message, - * that will occur, when a dump started. - */ - static void rtnetlink_rcv(struct sock *sk, int len) { - unsigned int qlen = skb_queue_len(&sk->sk_receive_queue); + unsigned int qlen = 0; do { - struct sk_buff *skb; - rtnl_lock(); - - if (qlen > skb_queue_len(&sk->sk_receive_queue)) - qlen = skb_queue_len(&sk->sk_receive_queue); - - for (; qlen; qlen--) { - skb = skb_dequeue(&sk->sk_receive_queue); - if (rtnetlink_rcv_skb(skb)) { - if (skb->len) - skb_queue_head(&sk->sk_receive_queue, - skb); - else { - kfree_skb(skb); - qlen--; - } - break; - } - kfree_skb(skb); - } - + netlink_run_queue(sk, &qlen, &rtnetlink_rcv_msg); up(&rtnl_sem); netdev_run_todo(); -- cgit v1.2.2 From 482a8524f85a7d8c40c6fb5d072e85bc2fef327f Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 10 Nov 2005 02:25:56 +0100 Subject: [NETLINK]: Generic netlink family The generic netlink family builds on top of netlink and provides simplifies access for the less demanding netlink users. It solves the problem of protocol numbers running out by introducing a so called controller taking care of id management and name resolving. Generic netlink modules register themself after filling out their id card (struct genl_family), after successful registration the modules are able to register callbacks to command numbers by filling out a struct genl_ops and calling genl_register_op(). The registered callbacks are invoked with attributes parsed making life of simple modules a lot easier. Although generic netlink modules can request static identifiers, it is recommended to use GENL_ID_GENERATE and to let the controller assign a unique identifier to the module. Userspace applications will then ask the controller and lookup the idenfier by the module name. Due to the current multicast implementation of netlink, the number of generic netlink modules is restricted to 1024 to avoid wasting memory for the per socket multiacst subscription bitmask. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- net/netlink/Makefile | 2 +- net/netlink/genetlink.c | 579 ++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 580 insertions(+), 1 deletion(-) create mode 100644 net/netlink/genetlink.c (limited to 'net') diff --git a/net/netlink/Makefile b/net/netlink/Makefile index 68179015bd39..e3589c2de49e 100644 --- a/net/netlink/Makefile +++ b/net/netlink/Makefile @@ -2,4 +2,4 @@ # Makefile for the netlink driver. # -obj-y := af_netlink.o attr.o +obj-y := af_netlink.o attr.o genetlink.o diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c new file mode 100644 index 000000000000..287cfcc56951 --- /dev/null +++ b/net/netlink/genetlink.c @@ -0,0 +1,579 @@ +/* + * NETLINK Generic Netlink Family + * + * Authors: Jamal Hadi Salim + * Thomas Graf + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +struct sock *genl_sock = NULL; + +static DECLARE_MUTEX(genl_sem); /* serialization of message processing */ + +static void genl_lock(void) +{ + down(&genl_sem); +} + +static int genl_trylock(void) +{ + return down_trylock(&genl_sem); +} + +static void genl_unlock(void) +{ + up(&genl_sem); + + if (genl_sock && genl_sock->sk_receive_queue.qlen) + genl_sock->sk_data_ready(genl_sock, 0); +} + +#define GENL_FAM_TAB_SIZE 16 +#define GENL_FAM_TAB_MASK (GENL_FAM_TAB_SIZE - 1) + +static struct list_head family_ht[GENL_FAM_TAB_SIZE]; + +static int genl_ctrl_event(int event, void *data); + +static inline unsigned int genl_family_hash(unsigned int id) +{ + return id & GENL_FAM_TAB_MASK; +} + +static inline struct list_head *genl_family_chain(unsigned int id) +{ + return &family_ht[genl_family_hash(id)]; +} + +static struct genl_family *genl_family_find_byid(unsigned int id) +{ + struct genl_family *f; + + list_for_each_entry(f, genl_family_chain(id), family_list) + if (f->id == id) + return f; + + return NULL; +} + +static struct genl_family *genl_family_find_byname(char *name) +{ + struct genl_family *f; + int i; + + for (i = 0; i < GENL_FAM_TAB_SIZE; i++) + list_for_each_entry(f, genl_family_chain(i), family_list) + if (strcmp(f->name, name) == 0) + return f; + + return NULL; +} + +static struct genl_ops *genl_get_cmd(u8 cmd, struct genl_family *family) +{ + struct genl_ops *ops; + + list_for_each_entry(ops, &family->ops_list, ops_list) + if (ops->cmd == cmd) + return ops; + + return NULL; +} + +/* Of course we are going to have problems once we hit + * 2^16 alive types, but that can only happen by year 2K +*/ +static inline u16 genl_generate_id(void) +{ + static u16 id_gen_idx; + int overflowed = 0; + + do { + if (id_gen_idx == 0) + id_gen_idx = GENL_MIN_ID; + + if (++id_gen_idx > GENL_MAX_ID) { + if (!overflowed) { + overflowed = 1; + id_gen_idx = 0; + continue; + } else + return 0; + } + + } while (genl_family_find_byid(id_gen_idx)); + + return id_gen_idx; +} + +/** + * genl_register_ops - register generic netlink operations + * @family: generic netlink family + * @ops: operations to be registered + * + * Registers the specified operations and assigns them to the specified + * family. Either a doit or dumpit callback must be specified or the + * operation will fail. Only one operation structure per command + * identifier may be registered. + * + * See include/net/genetlink.h for more documenation on the operations + * structure. + * + * Returns 0 on success or a negative error code. + */ +int genl_register_ops(struct genl_family *family, struct genl_ops *ops) +{ + int err = -EINVAL; + + if (ops->dumpit == NULL && ops->doit == NULL) + goto errout; + + if (genl_get_cmd(ops->cmd, family)) { + err = -EEXIST; + goto errout; + } + + genl_lock(); + list_add_tail(&ops->ops_list, &family->ops_list); + genl_unlock(); + + genl_ctrl_event(CTRL_CMD_NEWOPS, ops); + err = 0; +errout: + return err; +} + +/** + * genl_unregister_ops - unregister generic netlink operations + * @family: generic netlink family + * @ops: operations to be unregistered + * + * Unregisters the specified operations and unassigns them from the + * specified family. The operation blocks until the current message + * processing has finished and doesn't start again until the + * unregister process has finished. + * + * Note: It is not necessary to unregister all operations before + * unregistering the family, unregistering the family will cause + * all assigned operations to be unregistered automatically. + * + * Returns 0 on success or a negative error code. + */ +int genl_unregister_ops(struct genl_family *family, struct genl_ops *ops) +{ + struct genl_ops *rc; + + genl_lock(); + list_for_each_entry(rc, &family->ops_list, ops_list) { + if (rc == ops) { + list_del(&ops->ops_list); + genl_unlock(); + genl_ctrl_event(CTRL_CMD_DELOPS, ops); + return 0; + } + } + genl_unlock(); + + return -ENOENT; +} + +/** + * genl_register_family - register a generic netlink family + * @family: generic netlink family + * + * Registers the specified family after validating it first. Only one + * family may be registered with the same family name or identifier. + * The family id may equal GENL_ID_GENERATE causing an unique id to + * be automatically generated and assigned. + * + * Return 0 on success or a negative error code. + */ +int genl_register_family(struct genl_family *family) +{ + int err = -EINVAL; + + if (family->id && family->id < GENL_MIN_ID) + goto errout; + + if (family->id > GENL_MAX_ID) + goto errout; + + INIT_LIST_HEAD(&family->ops_list); + + genl_lock(); + + if (genl_family_find_byname(family->name)) { + err = -EEXIST; + goto errout_locked; + } + + if (genl_family_find_byid(family->id)) { + err = -EEXIST; + goto errout_locked; + } + + if (!try_module_get(family->owner)) { + err = -EBUSY; + goto errout_locked; + } + + if (family->id == GENL_ID_GENERATE) { + u16 newid = genl_generate_id(); + + if (!newid) { + err = -ENOMEM; + goto errout_locked; + } + + family->id = newid; + } + + if (family->maxattr) { + family->attrbuf = kmalloc((family->maxattr+1) * + sizeof(struct nlattr *), GFP_KERNEL); + if (family->attrbuf == NULL) { + err = -ENOMEM; + goto errout; + } + } else + family->attrbuf = NULL; + + list_add_tail(&family->family_list, genl_family_chain(family->id)); + genl_unlock(); + + genl_ctrl_event(CTRL_CMD_NEWFAMILY, family); + + return 0; + +errout_locked: + genl_unlock(); +errout: + return err; +} + +/** + * genl_unregister_family - unregister generic netlink family + * @family: generic netlink family + * + * Unregisters the specified family. + * + * Returns 0 on success or a negative error code. + */ +int genl_unregister_family(struct genl_family *family) +{ + struct genl_family *rc; + + genl_lock(); + + list_for_each_entry(rc, genl_family_chain(family->id), family_list) { + if (family->id != rc->id || strcmp(rc->name, family->name)) + continue; + + list_del(&rc->family_list); + INIT_LIST_HEAD(&family->ops_list); + genl_unlock(); + + module_put(family->owner); + kfree(family->attrbuf); + genl_ctrl_event(CTRL_CMD_DELFAMILY, family); + return 0; + } + + genl_unlock(); + + return -ENOENT; +} + +static inline int genl_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, + int *errp) +{ + struct genl_ops *ops; + struct genl_family *family; + struct genl_info info; + struct genlmsghdr *hdr = nlmsg_data(nlh); + int hdrlen, err = -EINVAL; + + if (!(nlh->nlmsg_flags & NLM_F_REQUEST)) + goto ignore; + + if (nlh->nlmsg_type < NLMSG_MIN_TYPE) + goto ignore; + + family = genl_family_find_byid(nlh->nlmsg_type); + if (family == NULL) { + err = -ENOENT; + goto errout; + } + + hdrlen = GENL_HDRLEN + family->hdrsize; + if (nlh->nlmsg_len < nlmsg_msg_size(hdrlen)) + goto errout; + + ops = genl_get_cmd(hdr->cmd, family); + if (ops == NULL) { + err = -EOPNOTSUPP; + goto errout; + } + + if ((ops->flags & GENL_ADMIN_PERM) && security_netlink_recv(skb)) { + err = -EPERM; + goto errout; + } + + if (nlh->nlmsg_flags & NLM_F_DUMP) { + if (ops->dumpit == NULL) { + err = -EOPNOTSUPP; + goto errout; + } + + *errp = err = netlink_dump_start(genl_sock, skb, nlh, + ops->dumpit, NULL); + if (err == 0) + skb_pull(skb, min(NLMSG_ALIGN(nlh->nlmsg_len), + skb->len)); + return -1; + } + + if (ops->doit == NULL) { + err = -EOPNOTSUPP; + goto errout; + } + + if (family->attrbuf) { + err = nlmsg_parse(nlh, hdrlen, family->attrbuf, family->maxattr, + ops->policy); + if (err < 0) + goto errout; + } + + info.snd_seq = nlh->nlmsg_seq; + info.snd_pid = NETLINK_CB(skb).pid; + info.nlhdr = nlh; + info.genlhdr = nlmsg_data(nlh); + info.userhdr = nlmsg_data(nlh) + GENL_HDRLEN; + info.attrs = family->attrbuf; + + *errp = err = ops->doit(skb, &info); + return err; + +ignore: + return 0; + +errout: + *errp = err; + return -1; +} + +static void genl_rcv(struct sock *sk, int len) +{ + unsigned int qlen = 0; + + do { + if (genl_trylock()) + return; + netlink_run_queue(sk, &qlen, &genl_rcv_msg); + genl_unlock(); + } while (qlen && genl_sock && genl_sock->sk_receive_queue.qlen); +} + +/************************************************************************** + * Controller + **************************************************************************/ + +static int ctrl_fill_info(struct genl_family *family, u32 pid, u32 seq, + u32 flags, struct sk_buff *skb, u8 cmd) +{ + void *hdr; + + hdr = genlmsg_put(skb, pid, seq, GENL_ID_CTRL, 0, flags, cmd, + family->version); + if (hdr == NULL) + return -1; + + NLA_PUT_STRING(skb, CTRL_ATTR_FAMILY_NAME, family->name); + NLA_PUT_U16(skb, CTRL_ATTR_FAMILY_ID, family->id); + + return genlmsg_end(skb, hdr); + +nla_put_failure: + return genlmsg_cancel(skb, hdr); +} + +static int ctrl_dumpfamily(struct sk_buff *skb, struct netlink_callback *cb) +{ + + int i, n = 0; + struct genl_family *rt; + int chains_to_skip = cb->args[0]; + int fams_to_skip = cb->args[1]; + + for (i = 0; i < GENL_FAM_TAB_SIZE; i++) { + if (i < chains_to_skip) + continue; + n = 0; + list_for_each_entry(rt, genl_family_chain(i), family_list) { + if (++n < fams_to_skip) + continue; + if (ctrl_fill_info(rt, NETLINK_CB(cb->skb).pid, + cb->nlh->nlmsg_seq, NLM_F_MULTI, + skb, CTRL_CMD_NEWFAMILY) < 0) + goto errout; + } + + fams_to_skip = 0; + } + +errout: + cb->args[0] = i; + cb->args[1] = n; + + return skb->len; +} + +static struct sk_buff *ctrl_build_msg(struct genl_family *family, u32 pid, + int seq, int cmd) +{ + struct sk_buff *skb; + int err; + + skb = nlmsg_new(NLMSG_GOODSIZE); + if (skb == NULL) + return ERR_PTR(-ENOBUFS); + + err = ctrl_fill_info(family, pid, seq, 0, skb, cmd); + if (err < 0) { + nlmsg_free(skb); + return ERR_PTR(err); + } + + return skb; +} + +static struct nla_policy ctrl_policy[CTRL_ATTR_MAX+1] __read_mostly = { + [CTRL_ATTR_FAMILY_ID] = { .type = NLA_U16 }, + [CTRL_ATTR_FAMILY_NAME] = { .type = NLA_STRING }, +}; + +static int ctrl_getfamily(struct sk_buff *skb, struct genl_info *info) +{ + struct sk_buff *msg; + struct genl_family *res = NULL; + int err = -EINVAL; + + if (info->attrs[CTRL_ATTR_FAMILY_ID]) { + u16 id = nla_get_u16(info->attrs[CTRL_ATTR_FAMILY_ID]); + res = genl_family_find_byid(id); + } + + if (info->attrs[CTRL_ATTR_FAMILY_NAME]) { + char name[GENL_NAMSIZ]; + + if (nla_strlcpy(name, info->attrs[CTRL_ATTR_FAMILY_NAME], + GENL_NAMSIZ) >= GENL_NAMSIZ) + goto errout; + + res = genl_family_find_byname(name); + } + + if (res == NULL) { + err = -ENOENT; + goto errout; + } + + msg = ctrl_build_msg(res, info->snd_pid, info->snd_seq, + CTRL_CMD_NEWFAMILY); + if (IS_ERR(msg)) { + err = PTR_ERR(msg); + goto errout; + } + + err = genlmsg_unicast(msg, info->snd_pid); +errout: + return err; +} + +static int genl_ctrl_event(int event, void *data) +{ + struct sk_buff *msg; + + if (genl_sock == NULL) + return 0; + + switch (event) { + case CTRL_CMD_NEWFAMILY: + case CTRL_CMD_DELFAMILY: + msg = ctrl_build_msg(data, 0, 0, event); + if (IS_ERR(msg)) + return PTR_ERR(msg); + + genlmsg_multicast(msg, 0, GENL_ID_CTRL); + break; + } + + return 0; +} + +static struct genl_ops genl_ctrl_ops = { + .cmd = CTRL_CMD_GETFAMILY, + .doit = ctrl_getfamily, + .dumpit = ctrl_dumpfamily, + .policy = ctrl_policy, +}; + +static struct genl_family genl_ctrl = { + .id = GENL_ID_CTRL, + .name = "nlctrl", + .version = 0x1, + .maxattr = CTRL_ATTR_MAX, + .owner = THIS_MODULE, +}; + +static int __init genl_init(void) +{ + int i, err; + + for (i = 0; i < GENL_FAM_TAB_SIZE; i++) + INIT_LIST_HEAD(&family_ht[i]); + + err = genl_register_family(&genl_ctrl); + if (err < 0) + goto errout; + + err = genl_register_ops(&genl_ctrl, &genl_ctrl_ops); + if (err < 0) + goto errout_register; + + netlink_set_nonroot(NETLINK_GENERIC, NL_NONROOT_RECV); + genl_sock = netlink_kernel_create(NETLINK_GENERIC, GENL_MAX_ID, + genl_rcv, THIS_MODULE); + if (genl_sock == NULL) { + panic("GENL: Cannot initialize generic netlink\n"); + return -ENOMEM; + } + + return 0; + +errout_register: + genl_unregister_family(&genl_ctrl); +errout: + panic("GENL: Cannot register controller: %d\n", err); + return err; +} + +subsys_initcall(genl_init); + +EXPORT_SYMBOL(genl_sock); +EXPORT_SYMBOL(genl_register_ops); +EXPORT_SYMBOL(genl_unregister_ops); +EXPORT_SYMBOL(genl_register_family); +EXPORT_SYMBOL(genl_unregister_family); -- cgit v1.2.2