From e5fc9e7a666e5964b60e05903b90aa832354b68c Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Fri, 12 Nov 2010 17:33:17 +0100 Subject: netfilter: nf_conntrack: don't always initialize ct->proto ct->proto is big(60 bytes) due to structure ip_ct_tcp, and we don't need to initialize the whole for all the other protocols. This patch moves proto to the end of structure nf_conn, and pushes the initialization down to the individual protocols. Signed-off-by: Changli Gao Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_core.c | 3 ++- net/netfilter/nf_conntrack_netlink.c | 1 + net/netfilter/nf_conntrack_proto_dccp.c | 3 +++ net/netfilter/nf_conntrack_proto_sctp.c | 1 + net/netfilter/nf_conntrack_proto_tcp.c | 14 +++----------- 5 files changed, 10 insertions(+), 12 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 27a5ea6b6a0f..0ba7d4801daf 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -651,7 +651,8 @@ __nf_conntrack_alloc(struct net *net, u16 zone, * and ct->tuplehash[IP_CT_DIR_REPLY].hnnode.next unchanged. */ memset(&ct->tuplehash[IP_CT_DIR_MAX], 0, - sizeof(*ct) - offsetof(struct nf_conn, tuplehash[IP_CT_DIR_MAX])); + offsetof(struct nf_conn, proto) - + offsetof(struct nf_conn, tuplehash[IP_CT_DIR_MAX])); spin_lock_init(&ct->lock); ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple = *orig; ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode.pprev = NULL; diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index b729ace1dcc1..7f59be82449f 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -1375,6 +1375,7 @@ ctnetlink_create_conntrack(struct net *net, u16 zone, } #endif + memset(&ct->proto, 0, sizeof(ct->proto)); if (cda[CTA_PROTOINFO]) { err = ctnetlink_change_protoinfo(ct, cda); if (err < 0) diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c index 5292560d6d4a..9ae57c57c50e 100644 --- a/net/netfilter/nf_conntrack_proto_dccp.c +++ b/net/netfilter/nf_conntrack_proto_dccp.c @@ -452,6 +452,9 @@ static bool dccp_new(struct nf_conn *ct, const struct sk_buff *skb, ct->proto.dccp.role[IP_CT_DIR_ORIGINAL] = CT_DCCP_ROLE_CLIENT; ct->proto.dccp.role[IP_CT_DIR_REPLY] = CT_DCCP_ROLE_SERVER; ct->proto.dccp.state = CT_DCCP_NONE; + ct->proto.dccp.last_pkt = DCCP_PKT_REQUEST; + ct->proto.dccp.last_dir = IP_CT_DIR_ORIGINAL; + ct->proto.dccp.handshake_seq = 0; return true; out_invalid: diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c index c6049c2d5ea8..6f4ee70f460b 100644 --- a/net/netfilter/nf_conntrack_proto_sctp.c +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -413,6 +413,7 @@ static bool sctp_new(struct nf_conn *ct, const struct sk_buff *skb, test_bit(SCTP_CID_COOKIE_ACK, map)) return false; + memset(&ct->proto.sctp, 0, sizeof(ct->proto.sctp)); new_state = SCTP_CONNTRACK_MAX; for_each_sctp_chunk (skb, sch, _sch, offset, dataoff, count) { /* Don't need lock here: this conntrack not in circulation yet */ diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 3fb2b73b24dc..6f38d0e2ea4a 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -1066,9 +1066,7 @@ static bool tcp_new(struct nf_conn *ct, const struct sk_buff *skb, 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]; + new_state = tcp_conntracks[0][get_conntrack_index(th)][TCP_CONNTRACK_NONE]; /* Invalid: delete conntrack */ if (new_state >= TCP_CONNTRACK_MAX) { @@ -1077,6 +1075,7 @@ static bool tcp_new(struct nf_conn *ct, const struct sk_buff *skb, } if (new_state == TCP_CONNTRACK_SYN_SENT) { + memset(&ct->proto.tcp, 0, sizeof(ct->proto.tcp)); /* SYN packet */ ct->proto.tcp.seen[0].td_end = segment_seq_plus_len(ntohl(th->seq), skb->len, @@ -1088,11 +1087,11 @@ static bool tcp_new(struct nf_conn *ct, const struct sk_buff *skb, ct->proto.tcp.seen[0].td_end; tcp_options(skb, dataoff, th, &ct->proto.tcp.seen[0]); - ct->proto.tcp.seen[1].flags = 0; } else if (nf_ct_tcp_loose == 0) { /* Don't try to pick up connections. */ return false; } else { + memset(&ct->proto.tcp, 0, sizeof(ct->proto.tcp)); /* * We are in the middle of a connection, * its history is lost for us. @@ -1107,7 +1106,6 @@ static bool tcp_new(struct nf_conn *ct, const struct sk_buff *skb, ct->proto.tcp.seen[0].td_maxend = ct->proto.tcp.seen[0].td_end + ct->proto.tcp.seen[0].td_maxwin; - ct->proto.tcp.seen[0].td_scale = 0; /* We assume SACK and liberal window checking to handle * window scaling */ @@ -1116,13 +1114,7 @@ static bool tcp_new(struct nf_conn *ct, const struct sk_buff *skb, IP_CT_TCP_FLAG_BE_LIBERAL; } - ct->proto.tcp.seen[1].td_end = 0; - ct->proto.tcp.seen[1].td_maxend = 0; - ct->proto.tcp.seen[1].td_maxwin = 0; - ct->proto.tcp.seen[1].td_scale = 0; - /* tcp_packet will set them */ - ct->proto.tcp.state = TCP_CONNTRACK_NONE; ct->proto.tcp.last_index = TCP_NONE_SET; pr_debug("tcp_new: sender end=%u maxend=%u maxwin=%u scale=%i " -- cgit v1.2.3 From ca36181050a523f6c0af3ef7cb509bbbc4ede276 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Fri, 12 Nov 2010 17:34:17 +0100 Subject: netfilter: xt_NFQUEUE: remove modulo operations Signed-off-by: Changli Gao Acked-by: Eric Dumazet Signed-off-by: Patrick McHardy --- net/netfilter/xt_NFQUEUE.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_NFQUEUE.c b/net/netfilter/xt_NFQUEUE.c index 039cce1bde3d..39627706aac6 100644 --- a/net/netfilter/xt_NFQUEUE.c +++ b/net/netfilter/xt_NFQUEUE.c @@ -72,10 +72,12 @@ nfqueue_tg_v1(struct sk_buff *skb, const struct xt_action_param *par) if (info->queues_total > 1) { if (par->family == NFPROTO_IPV4) - queue = hash_v4(skb) % info->queues_total + queue; + queue = (((u64) hash_v4(skb) * info->queues_total) >> + 32) + queue; #if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE) else if (par->family == NFPROTO_IPV6) - queue = hash_v6(skb) % info->queues_total + queue; + queue = (((u64) hash_v6(skb) * info->queues_total) >> + 32) + queue; #endif } return NF_QUEUE_NR(queue); -- cgit v1.2.3 From 3b2368806915e1e69ac3bcc0d6a7cfde64307655 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Mon, 15 Nov 2010 11:47:52 +0100 Subject: netfilter: ct_extend: fix the wrong alloc_size In function update_alloc_size(), sizeof(struct nf_ct_ext) is added twice wrongly. Signed-off-by: Changli Gao Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_extend.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/nf_conntrack_extend.c b/net/netfilter/nf_conntrack_extend.c index bd82450c193f..920f9244388b 100644 --- a/net/netfilter/nf_conntrack_extend.c +++ b/net/netfilter/nf_conntrack_extend.c @@ -144,9 +144,8 @@ static void update_alloc_size(struct nf_ct_ext_type *type) if (!t1) continue; - t1->alloc_size = sizeof(struct nf_ct_ext) - + ALIGN(sizeof(struct nf_ct_ext), t1->align) - + t1->len; + t1->alloc_size = ALIGN(sizeof(struct nf_ct_ext), t1->align) + + t1->len; for (j = 0; j < NF_CT_EXT_NUM; j++) { t2 = nf_ct_ext_types[j]; if (t2 == NULL || t2 == t1 || -- cgit v1.2.3 From 9811600f7c1f18152430c6b93b0a76fdd88a59ce Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Fr=E9d=E9ric=20Leroy?= Date: Mon, 15 Nov 2010 13:57:56 +0100 Subject: netfilter: xt_CLASSIFY: add ARP support, allow CLASSIFY target on any table MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Signed-off-by: Fr�d�ric Leroy Signed-off-by: Patrick McHardy --- net/netfilter/xt_CLASSIFY.c | 36 ++++++++++++++++++++++++------------ 1 file changed, 24 insertions(+), 12 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_CLASSIFY.c b/net/netfilter/xt_CLASSIFY.c index c2c0e4abeb99..af9c4dadf816 100644 --- a/net/netfilter/xt_CLASSIFY.c +++ b/net/netfilter/xt_CLASSIFY.c @@ -19,12 +19,14 @@ #include #include #include +#include MODULE_AUTHOR("Patrick McHardy "); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("Xtables: Qdisc classification"); MODULE_ALIAS("ipt_CLASSIFY"); MODULE_ALIAS("ip6t_CLASSIFY"); +MODULE_ALIAS("arpt_CLASSIFY"); static unsigned int classify_tg(struct sk_buff *skb, const struct xt_action_param *par) @@ -35,26 +37,36 @@ classify_tg(struct sk_buff *skb, const struct xt_action_param *par) return XT_CONTINUE; } -static struct xt_target classify_tg_reg __read_mostly = { - .name = "CLASSIFY", - .revision = 0, - .family = NFPROTO_UNSPEC, - .table = "mangle", - .hooks = (1 << NF_INET_LOCAL_OUT) | (1 << NF_INET_FORWARD) | - (1 << NF_INET_POST_ROUTING), - .target = classify_tg, - .targetsize = sizeof(struct xt_classify_target_info), - .me = THIS_MODULE, +static struct xt_target classify_tg_reg[] __read_mostly = { + { + .name = "CLASSIFY", + .revision = 0, + .family = NFPROTO_UNSPEC, + .hooks = (1 << NF_INET_LOCAL_OUT) | (1 << NF_INET_FORWARD) | + (1 << NF_INET_POST_ROUTING), + .target = classify_tg, + .targetsize = sizeof(struct xt_classify_target_info), + .me = THIS_MODULE, + }, + { + .name = "CLASSIFY", + .revision = 0, + .family = NFPROTO_ARP, + .hooks = (1 << NF_ARP_OUT) | (1 << NF_ARP_FORWARD), + .target = classify_tg, + .targetsize = sizeof(struct xt_classify_target_info), + .me = THIS_MODULE, + }, }; static int __init classify_tg_init(void) { - return xt_register_target(&classify_tg_reg); + return xt_register_targets(classify_tg_reg, ARRAY_SIZE(classify_tg_reg)); } static void __exit classify_tg_exit(void) { - xt_unregister_target(&classify_tg_reg); + xt_unregister_targets(classify_tg_reg, ARRAY_SIZE(classify_tg_reg)); } module_init(classify_tg_init); -- cgit v1.2.3 From 0e60ebe04c51807db972d03665651ae6b5c26d7e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 15 Nov 2010 18:17:21 +0100 Subject: netfilter: add __rcu annotations Add some __rcu annotations and use helpers to reduce number of sparse warnings (CONFIG_SPARSE_RCU_POINTER=y) Signed-off-by: Eric Dumazet Signed-off-by: Patrick McHardy --- net/netfilter/core.c | 4 ++-- net/netfilter/nf_conntrack_expect.c | 6 +++--- net/netfilter/nf_conntrack_proto.c | 20 +++++++++++++++----- net/netfilter/nf_conntrack_standalone.c | 9 ++++++--- net/netfilter/nf_log.c | 6 ++++-- net/netfilter/nf_queue.c | 18 ++++++++++++++---- net/netfilter/nfnetlink_log.c | 6 +++--- 7 files changed, 47 insertions(+), 22 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/core.c b/net/netfilter/core.c index 85dabb86be6f..5faec4fd8193 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -212,7 +212,7 @@ EXPORT_SYMBOL(skb_make_writable); /* This does not belong here, but locally generated errors need it if connection tracking in use: without this, connection may not be in hash table, and hence manufactured ICMP or RST packets will not be associated with it. */ -void (*ip_ct_attach)(struct sk_buff *, struct sk_buff *); +void (*ip_ct_attach)(struct sk_buff *, struct sk_buff *) __rcu __read_mostly; EXPORT_SYMBOL(ip_ct_attach); void nf_ct_attach(struct sk_buff *new, struct sk_buff *skb) @@ -229,7 +229,7 @@ void nf_ct_attach(struct sk_buff *new, struct sk_buff *skb) } EXPORT_SYMBOL(nf_ct_attach); -void (*nf_ct_destroy)(struct nf_conntrack *); +void (*nf_ct_destroy)(struct nf_conntrack *) __rcu __read_mostly; EXPORT_SYMBOL(nf_ct_destroy); void nf_conntrack_destroy(struct nf_conntrack *nfct) diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 46e8966912b1..cab196cf428c 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -482,7 +482,7 @@ static struct hlist_node *ct_expect_get_first(struct seq_file *seq) struct hlist_node *n; for (st->bucket = 0; st->bucket < nf_ct_expect_hsize; st->bucket++) { - n = rcu_dereference(net->ct.expect_hash[st->bucket].first); + n = rcu_dereference(hlist_first_rcu(&net->ct.expect_hash[st->bucket])); if (n) return n; } @@ -495,11 +495,11 @@ static struct hlist_node *ct_expect_get_next(struct seq_file *seq, struct net *net = seq_file_net(seq); struct ct_expect_iter_state *st = seq->private; - head = rcu_dereference(head->next); + head = rcu_dereference(hlist_next_rcu(head)); while (head == NULL) { if (++st->bucket >= nf_ct_expect_hsize) return NULL; - head = rcu_dereference(net->ct.expect_hash[st->bucket].first); + head = rcu_dereference(hlist_first_rcu(&net->ct.expect_hash[st->bucket])); } return head; } diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index dc7bb74110df..03b56a0fff30 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -166,6 +166,7 @@ static void nf_ct_l3proto_unregister_sysctl(struct nf_conntrack_l3proto *l3proto int nf_conntrack_l3proto_register(struct nf_conntrack_l3proto *proto) { int ret = 0; + struct nf_conntrack_l3proto *old; if (proto->l3proto >= AF_MAX) return -EBUSY; @@ -174,7 +175,9 @@ int nf_conntrack_l3proto_register(struct nf_conntrack_l3proto *proto) return -EINVAL; mutex_lock(&nf_ct_proto_mutex); - if (nf_ct_l3protos[proto->l3proto] != &nf_conntrack_l3proto_generic) { + old = rcu_dereference_protected(nf_ct_l3protos[proto->l3proto], + lockdep_is_held(&nf_ct_proto_mutex)); + if (old != &nf_conntrack_l3proto_generic) { ret = -EBUSY; goto out_unlock; } @@ -201,7 +204,9 @@ void nf_conntrack_l3proto_unregister(struct nf_conntrack_l3proto *proto) BUG_ON(proto->l3proto >= AF_MAX); mutex_lock(&nf_ct_proto_mutex); - BUG_ON(nf_ct_l3protos[proto->l3proto] != proto); + BUG_ON(rcu_dereference_protected(nf_ct_l3protos[proto->l3proto], + lockdep_is_held(&nf_ct_proto_mutex) + ) != proto); rcu_assign_pointer(nf_ct_l3protos[proto->l3proto], &nf_conntrack_l3proto_generic); nf_ct_l3proto_unregister_sysctl(proto); @@ -299,8 +304,10 @@ int nf_conntrack_l4proto_register(struct nf_conntrack_l4proto *l4proto) smp_wmb(); nf_ct_protos[l4proto->l3proto] = proto_array; - } else if (nf_ct_protos[l4proto->l3proto][l4proto->l4proto] != - &nf_conntrack_l4proto_generic) { + } else if (rcu_dereference_protected( + nf_ct_protos[l4proto->l3proto][l4proto->l4proto], + lockdep_is_held(&nf_ct_proto_mutex) + ) != &nf_conntrack_l4proto_generic) { ret = -EBUSY; goto out_unlock; } @@ -331,7 +338,10 @@ void nf_conntrack_l4proto_unregister(struct nf_conntrack_l4proto *l4proto) BUG_ON(l4proto->l3proto >= PF_MAX); mutex_lock(&nf_ct_proto_mutex); - BUG_ON(nf_ct_protos[l4proto->l3proto][l4proto->l4proto] != l4proto); + BUG_ON(rcu_dereference_protected( + nf_ct_protos[l4proto->l3proto][l4proto->l4proto], + lockdep_is_held(&nf_ct_proto_mutex) + ) != l4proto); rcu_assign_pointer(nf_ct_protos[l4proto->l3proto][l4proto->l4proto], &nf_conntrack_l4proto_generic); nf_ct_l4proto_unregister_sysctl(l4proto); diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 0fb65705b44b..328f1d2a51f8 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -29,6 +29,7 @@ #include #include #include +#include MODULE_LICENSE("GPL"); @@ -56,7 +57,7 @@ static struct hlist_nulls_node *ct_get_first(struct seq_file *seq) for (st->bucket = 0; st->bucket < net->ct.htable_size; st->bucket++) { - n = rcu_dereference(net->ct.hash[st->bucket].first); + n = rcu_dereference(hlist_nulls_first_rcu(&net->ct.hash[st->bucket])); if (!is_a_nulls(n)) return n; } @@ -69,13 +70,15 @@ static struct hlist_nulls_node *ct_get_next(struct seq_file *seq, struct net *net = seq_file_net(seq); struct ct_iter_state *st = seq->private; - head = rcu_dereference(head->next); + head = rcu_dereference(hlist_nulls_next_rcu(head)); while (is_a_nulls(head)) { if (likely(get_nulls_value(head) == st->bucket)) { if (++st->bucket >= net->ct.htable_size) return NULL; } - head = rcu_dereference(net->ct.hash[st->bucket].first); + head = rcu_dereference( + hlist_nulls_first_rcu( + &net->ct.hash[st->bucket])); } return head; } diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c index b07393eab88e..20c775cff2a8 100644 --- a/net/netfilter/nf_log.c +++ b/net/netfilter/nf_log.c @@ -161,7 +161,8 @@ static int seq_show(struct seq_file *s, void *v) struct nf_logger *t; int ret; - logger = nf_loggers[*pos]; + logger = rcu_dereference_protected(nf_loggers[*pos], + lockdep_is_held(&nf_log_mutex)); if (!logger) ret = seq_printf(s, "%2lld NONE (", *pos); @@ -249,7 +250,8 @@ static int nf_log_proc_dostring(ctl_table *table, int write, mutex_unlock(&nf_log_mutex); } else { mutex_lock(&nf_log_mutex); - logger = nf_loggers[tindex]; + logger = rcu_dereference_protected(nf_loggers[tindex], + lockdep_is_held(&nf_log_mutex)); if (!logger) table->data = "NONE"; else diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index 74aebed5bd28..1876f7411561 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -27,14 +27,17 @@ static DEFINE_MUTEX(queue_handler_mutex); int nf_register_queue_handler(u_int8_t pf, const struct nf_queue_handler *qh) { int ret; + const struct nf_queue_handler *old; if (pf >= ARRAY_SIZE(queue_handler)) return -EINVAL; mutex_lock(&queue_handler_mutex); - if (queue_handler[pf] == qh) + old = rcu_dereference_protected(queue_handler[pf], + lockdep_is_held(&queue_handler_mutex)); + if (old == qh) ret = -EEXIST; - else if (queue_handler[pf]) + else if (old) ret = -EBUSY; else { rcu_assign_pointer(queue_handler[pf], qh); @@ -49,11 +52,15 @@ EXPORT_SYMBOL(nf_register_queue_handler); /* The caller must flush their queue before this */ int nf_unregister_queue_handler(u_int8_t pf, const struct nf_queue_handler *qh) { + const struct nf_queue_handler *old; + if (pf >= ARRAY_SIZE(queue_handler)) return -EINVAL; mutex_lock(&queue_handler_mutex); - if (queue_handler[pf] && queue_handler[pf] != qh) { + old = rcu_dereference_protected(queue_handler[pf], + lockdep_is_held(&queue_handler_mutex)); + if (old && old != qh) { mutex_unlock(&queue_handler_mutex); return -EINVAL; } @@ -73,7 +80,10 @@ void nf_unregister_queue_handlers(const struct nf_queue_handler *qh) mutex_lock(&queue_handler_mutex); for (pf = 0; pf < ARRAY_SIZE(queue_handler); pf++) { - if (queue_handler[pf] == qh) + if (rcu_dereference_protected( + queue_handler[pf], + lockdep_is_held(&queue_handler_mutex) + ) == qh) rcu_assign_pointer(queue_handler[pf], NULL); } mutex_unlock(&queue_handler_mutex); diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index 6a1572b0ab41..91592da504b9 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -874,19 +874,19 @@ static struct hlist_node *get_first(struct iter_state *st) for (st->bucket = 0; st->bucket < INSTANCE_BUCKETS; st->bucket++) { if (!hlist_empty(&instance_table[st->bucket])) - return rcu_dereference_bh(instance_table[st->bucket].first); + return rcu_dereference_bh(hlist_first_rcu(&instance_table[st->bucket])); } return NULL; } static struct hlist_node *get_next(struct iter_state *st, struct hlist_node *h) { - h = rcu_dereference_bh(h->next); + h = rcu_dereference_bh(hlist_next_rcu(h)); while (!h) { if (++st->bucket >= INSTANCE_BUCKETS) return NULL; - h = rcu_dereference_bh(instance_table[st->bucket].first); + h = rcu_dereference_bh(hlist_first_rcu(&instance_table[st->bucket])); } return h; } -- cgit v1.2.3 From c5d277d29ad1ae9add8d6984025ccd2e835971ce Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 15 Nov 2010 19:45:13 +0100 Subject: netfilter: rcu sparse cleanups Use RCU helpers to reduce number of sparse warnings (CONFIG_SPARSE_RCU_POINTER=y), and adds lockdep checks. Signed-off-by: Eric Dumazet Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_expect.c | 15 ++++++++++++--- net/netfilter/nf_conntrack_extend.c | 6 ++++-- net/netfilter/nf_conntrack_helper.c | 10 ++++++++-- net/netfilter/nf_conntrack_proto.c | 4 ++-- 4 files changed, 26 insertions(+), 9 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index cab196cf428c..bbb21402596d 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -337,7 +337,10 @@ static void nf_ct_expect_insert(struct nf_conntrack_expect *exp) setup_timer(&exp->timeout, nf_ct_expectation_timed_out, (unsigned long)exp); if (master_help) { - p = &master_help->helper->expect_policy[exp->class]; + p = &rcu_dereference_protected( + master_help->helper, + lockdep_is_held(&nf_conntrack_lock) + )->expect_policy[exp->class]; exp->timeout.expires = jiffies + p->timeout * HZ; } add_timer(&exp->timeout); @@ -373,7 +376,10 @@ static inline int refresh_timer(struct nf_conntrack_expect *i) if (!del_timer(&i->timeout)) return 0; - p = &master_help->helper->expect_policy[i->class]; + p = &rcu_dereference_protected( + master_help->helper, + lockdep_is_held(&nf_conntrack_lock) + )->expect_policy[i->class]; i->timeout.expires = jiffies + p->timeout * HZ; add_timer(&i->timeout); return 1; @@ -411,7 +417,10 @@ static inline int __nf_ct_expect_check(struct nf_conntrack_expect *expect) } /* Will be over limit? */ if (master_help) { - p = &master_help->helper->expect_policy[expect->class]; + p = &rcu_dereference_protected( + master_help->helper, + lockdep_is_held(&nf_conntrack_lock) + )->expect_policy[expect->class]; if (p->max_expected && master_help->expecting[expect->class] >= p->max_expected) { evict_oldest_expect(master, expect); diff --git a/net/netfilter/nf_conntrack_extend.c b/net/netfilter/nf_conntrack_extend.c index 920f9244388b..80a23ed62bb0 100644 --- a/net/netfilter/nf_conntrack_extend.c +++ b/net/netfilter/nf_conntrack_extend.c @@ -140,14 +140,16 @@ static void update_alloc_size(struct nf_ct_ext_type *type) /* This assumes that extended areas in conntrack for the types whose NF_CT_EXT_F_PREALLOC bit set are allocated in order */ for (i = min; i <= max; i++) { - t1 = nf_ct_ext_types[i]; + t1 = rcu_dereference_protected(nf_ct_ext_types[i], + lockdep_is_held(&nf_ct_ext_type_mutex)); if (!t1) continue; t1->alloc_size = ALIGN(sizeof(struct nf_ct_ext), t1->align) + t1->len; for (j = 0; j < NF_CT_EXT_NUM; j++) { - t2 = nf_ct_ext_types[j]; + t2 = rcu_dereference_protected(nf_ct_ext_types[j], + lockdep_is_held(&nf_ct_ext_type_mutex)); if (t2 == NULL || t2 == t1 || (t2->flags & NF_CT_EXT_F_PREALLOC) == 0) continue; diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index 59e1a4cd4e8b..767bbe98a0f0 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -158,7 +158,10 @@ static inline int unhelp(struct nf_conntrack_tuple_hash *i, struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(i); struct nf_conn_help *help = nfct_help(ct); - if (help && help->helper == me) { + if (help && rcu_dereference_protected( + help->helper, + lockdep_is_held(&nf_conntrack_lock) + ) == me) { nf_conntrack_event(IPCT_HELPER, ct); rcu_assign_pointer(help->helper, NULL); } @@ -210,7 +213,10 @@ static void __nf_conntrack_helper_unregister(struct nf_conntrack_helper *me, hlist_for_each_entry_safe(exp, n, next, &net->ct.expect_hash[i], hnode) { struct nf_conn_help *help = nfct_help(exp->master); - if ((help->helper == me || exp->helper == me) && + if ((rcu_dereference_protected( + help->helper, + lockdep_is_held(&nf_conntrack_lock) + ) == me || exp->helper == me) && del_timer(&exp->timeout)) { nf_ct_unlink_expect(exp); nf_ct_expect_put(exp); diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index 03b56a0fff30..5701c8dd783c 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -284,7 +284,7 @@ int nf_conntrack_l4proto_register(struct nf_conntrack_l4proto *l4proto) mutex_lock(&nf_ct_proto_mutex); if (!nf_ct_protos[l4proto->l3proto]) { /* l3proto may be loaded latter. */ - struct nf_conntrack_l4proto **proto_array; + struct nf_conntrack_l4proto __rcu **proto_array; int i; proto_array = kmalloc(MAX_NF_CT_PROTO * @@ -296,7 +296,7 @@ int nf_conntrack_l4proto_register(struct nf_conntrack_l4proto *l4proto) } for (i = 0; i < MAX_NF_CT_PROTO; i++) - proto_array[i] = &nf_conntrack_l4proto_generic; + RCU_INIT_POINTER(proto_array[i], &nf_conntrack_l4proto_generic); /* Before making proto_array visible to lockless readers, * we must make sure its content is committed to memory. -- cgit v1.2.3 From e9e5eee8733739f13a204132b502494b3f494f3b Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Mon, 8 Nov 2010 20:05:57 +0900 Subject: IPVS: Add persistence engine to connection entry The dest of a connection may not exist if it has been created as the result of connection synchronisation. But in order for connection entries for templates with persistence engine data created through connection synchronisation to be valid access to the persistence engine pointer is required. So add the persistence engine to the connection itself. Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_conn.c | 19 ++++++++++--------- net/netfilter/ipvs/ip_vs_ctl.c | 4 ++-- net/netfilter/ipvs/ip_vs_pe.c | 14 ++++---------- 3 files changed, 16 insertions(+), 21 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index e9adecdc8ca4..64a9ca314100 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -176,8 +176,8 @@ static unsigned int ip_vs_conn_hashkey_conn(const struct ip_vs_conn *cp) ip_vs_conn_fill_param(cp->af, cp->protocol, &cp->caddr, cp->cport, NULL, 0, &p); - if (cp->dest && cp->dest->svc->pe) { - p.pe = cp->dest->svc->pe; + if (cp->pe) { + p.pe = cp->pe; p.pe_data = cp->pe_data; p.pe_data_len = cp->pe_data_len; } @@ -765,6 +765,7 @@ static void ip_vs_conn_expire(unsigned long data) if (cp->flags & IP_VS_CONN_F_NFCT) ip_vs_conn_drop_conntrack(cp); + ip_vs_pe_put(cp->pe); kfree(cp->pe_data); if (unlikely(cp->app != NULL)) ip_vs_unbind_app(cp); @@ -826,7 +827,9 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p, &cp->daddr, daddr); cp->dport = dport; cp->flags = flags; - if (flags & IP_VS_CONN_F_TEMPLATE && p->pe_data) { + if (flags & IP_VS_CONN_F_TEMPLATE && p->pe) { + ip_vs_pe_get(p->pe); + cp->pe = p->pe; cp->pe_data = p->pe_data; cp->pe_data_len = p->pe_data_len; } @@ -958,15 +961,13 @@ static int ip_vs_conn_seq_show(struct seq_file *seq, void *v) char pe_data[IP_VS_PENAME_MAXLEN + IP_VS_PEDATA_MAXLEN + 3]; size_t len = 0; - if (cp->dest && cp->pe_data && - cp->dest->svc->pe->show_pe_data) { + if (cp->pe_data) { pe_data[0] = ' '; - len = strlen(cp->dest->svc->pe->name); - memcpy(pe_data + 1, cp->dest->svc->pe->name, len); + len = strlen(cp->pe->name); + memcpy(pe_data + 1, cp->pe->name, len); pe_data[len + 1] = ' '; len += 2; - len += cp->dest->svc->pe->show_pe_data(cp, - pe_data + len); + len += cp->pe->show_pe_data(cp, pe_data + len); } pe_data[len] = '\0'; diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 5f5daa30b0af..3e92558dfcc2 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -1139,7 +1139,7 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u, } if (u->pe_name && *u->pe_name) { - pe = ip_vs_pe_get(u->pe_name); + pe = ip_vs_pe_getbyname(u->pe_name); if (pe == NULL) { pr_info("persistence engine module ip_vs_pe_%s " "not found\n", u->pe_name); @@ -1250,7 +1250,7 @@ ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u) old_sched = sched; if (u->pe_name && *u->pe_name) { - pe = ip_vs_pe_get(u->pe_name); + pe = ip_vs_pe_getbyname(u->pe_name); if (pe == NULL) { pr_info("persistence engine module ip_vs_pe_%s " "not found\n", u->pe_name); diff --git a/net/netfilter/ipvs/ip_vs_pe.c b/net/netfilter/ipvs/ip_vs_pe.c index 3414af70ee12..e99f920b93d1 100644 --- a/net/netfilter/ipvs/ip_vs_pe.c +++ b/net/netfilter/ipvs/ip_vs_pe.c @@ -30,7 +30,7 @@ void ip_vs_unbind_pe(struct ip_vs_service *svc) /* Get pe in the pe list by name */ static struct ip_vs_pe * -ip_vs_pe_getbyname(const char *pe_name) +__ip_vs_pe_getbyname(const char *pe_name) { struct ip_vs_pe *pe; @@ -60,28 +60,22 @@ ip_vs_pe_getbyname(const char *pe_name) } /* Lookup pe and try to load it if it doesn't exist */ -struct ip_vs_pe *ip_vs_pe_get(const char *name) +struct ip_vs_pe *ip_vs_pe_getbyname(const char *name) { struct ip_vs_pe *pe; /* Search for the pe by name */ - pe = ip_vs_pe_getbyname(name); + pe = __ip_vs_pe_getbyname(name); /* If pe not found, load the module and search again */ if (!pe) { request_module("ip_vs_pe_%s", name); - pe = ip_vs_pe_getbyname(name); + pe = __ip_vs_pe_getbyname(name); } return pe; } -void ip_vs_pe_put(struct ip_vs_pe *pe) -{ - if (pe && pe->module) - module_put(pe->module); -} - /* Register a pe in the pe list */ int register_ip_vs_pe(struct ip_vs_pe *pe) { -- cgit v1.2.3 From ea2c73afc23db3084fd857b027446c38fc7ff2c9 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Mon, 8 Nov 2010 20:06:30 +0900 Subject: IPVS: Only match pe_data created by the same pe Only match persistence engine data if it was created by the same persistence engine. Reported-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_conn.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 64a9ca314100..261db1a17633 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -354,7 +354,7 @@ struct ip_vs_conn *ip_vs_ct_in_get(const struct ip_vs_conn_param *p) list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { if (p->pe_data && p->pe->ct_match) { - if (p->pe->ct_match(p, cp)) + if (p->pe == cp->pe && p->pe->ct_match(p, cp)) goto out; continue; } -- cgit v1.2.3 From d494262b8a0f3507b62104a565849124abe29827 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Tue, 9 Nov 2010 09:33:15 +0900 Subject: IPVS: Make the cp argument to ip_vs_sync_conn() static Acked-by: Hans Schillstrom Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_sync.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index ab85aedea17e..a4dccbc4f1bb 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -236,7 +236,7 @@ get_curr_sync_buff(unsigned long time) * Add an ip_vs_conn information into the current sync_buff. * Called by ip_vs_in. */ -void ip_vs_sync_conn(struct ip_vs_conn *cp) +void ip_vs_sync_conn(const struct ip_vs_conn *cp) { struct ip_vs_sync_mesg *m; struct ip_vs_sync_conn *s; -- cgit v1.2.3 From 7ae246a15a5c9d26cfb572d36794325db0400b18 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Tue, 9 Nov 2010 09:33:25 +0900 Subject: IPVS: Remove useless { } block from ip_vs_process_message() Acked-by: Hans Schillstrom Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_sync.c | 24 +++++++++++------------- 1 file changed, 11 insertions(+), 13 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index a4dccbc4f1bb..72b3d88d35f4 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -381,20 +381,18 @@ static void ip_vs_process_message(const char *buffer, const size_t buflen) } } - { - if (ip_vs_conn_fill_param_sync(AF_INET, s->protocol, - (union nf_inet_addr *)&s->caddr, - s->cport, - (union nf_inet_addr *)&s->vaddr, - s->vport, ¶m)) { - pr_err("ip_vs_conn_fill_param_sync failed"); - return; - } - if (!(flags & IP_VS_CONN_F_TEMPLATE)) - cp = ip_vs_conn_in_get(¶m); - else - cp = ip_vs_ct_in_get(¶m); + if (ip_vs_conn_fill_param_sync(AF_INET, s->protocol, + (union nf_inet_addr *)&s->caddr, + s->cport, + (union nf_inet_addr *)&s->vaddr, + s->vport, ¶m)) { + pr_err("ip_vs_conn_fill_param_sync failed"); + return; } + if (!(flags & IP_VS_CONN_F_TEMPLATE)) + cp = ip_vs_conn_in_get(¶m); + else + cp = ip_vs_ct_in_get(¶m); if (!cp) { /* * Find the appropriate destination for the connection. -- cgit v1.2.3 From 8aadf93c9c1ff1a53aafd18d038be0d709b5ebc0 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Tue, 9 Nov 2010 09:33:28 +0900 Subject: IPVS: buffer argument to ip_vs_process_message() should not be const It is assigned to a non-const variable and its contents are modified. Acked-by: Hans Schillstrom Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_sync.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index 72b3d88d35f4..3897d6bf3b29 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -303,7 +303,7 @@ ip_vs_conn_fill_param_sync(int af, int protocol, * Process received multicast message and create the corresponding * ip_vs_conn entries. */ -static void ip_vs_process_message(const char *buffer, const size_t buflen) +static void ip_vs_process_message(char *buffer, const size_t buflen) { struct ip_vs_sync_mesg *m = (struct ip_vs_sync_mesg *)buffer; struct ip_vs_sync_conn *s; -- cgit v1.2.3 From 4ecd29447e6b9c12190e21c3e44ed5b12693c467 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 15 Nov 2010 18:38:52 +0100 Subject: ipvs: add static and read_mostly attributes ip_vs_conn_tab_bits & ip_vs_conn_tab_mask are static to ipvs/ip_vs_conn.c ip_vs_conn_tab_size, ip_vs_conn_tab_mask, ip_vs_conn_tab [the pointer], ip_vs_conn_rnd are mostly read. Signed-off-by: Eric Dumazet Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_conn.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 261db1a17633..7615f9e3d955 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -48,18 +48,18 @@ /* * Connection hash size. Default is what was selected at compile time. */ -int ip_vs_conn_tab_bits = CONFIG_IP_VS_TAB_BITS; +static int ip_vs_conn_tab_bits = CONFIG_IP_VS_TAB_BITS; module_param_named(conn_tab_bits, ip_vs_conn_tab_bits, int, 0444); MODULE_PARM_DESC(conn_tab_bits, "Set connections' hash size"); /* size and mask values */ -int ip_vs_conn_tab_size; -int ip_vs_conn_tab_mask; +int ip_vs_conn_tab_size __read_mostly; +static int ip_vs_conn_tab_mask __read_mostly; /* * Connection hash table: for input and output packets lookups of IPVS */ -static struct list_head *ip_vs_conn_tab; +static struct list_head *ip_vs_conn_tab __read_mostly; /* SLAB cache for IPVS connections */ static struct kmem_cache *ip_vs_conn_cachep __read_mostly; @@ -71,7 +71,7 @@ static atomic_t ip_vs_conn_count = ATOMIC_INIT(0); static atomic_t ip_vs_conn_no_cport_cnt = ATOMIC_INIT(0); /* random value for IPVS connection hash */ -static unsigned int ip_vs_conn_rnd; +static unsigned int ip_vs_conn_rnd __read_mostly; /* * Fine locking granularity for big connection hash table -- cgit v1.2.3 From a333e2ec05791bb866086274ac9749315900a0a6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 15 Nov 2010 19:46:33 +0100 Subject: ipvs: remove shadow rt variable Remove a sparse warning about rt variable. Signed-off-by: Eric Dumazet Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_xmit.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 10bd39c0ae2d..50b131c28b85 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -188,7 +188,6 @@ __ip_vs_reroute_locally(struct sk_buff *skb) }, .mark = skb->mark, }; - struct rtable *rt; if (ip_route_output_key(net, &rt, &fl)) return 0; -- cgit v1.2.3 From 8f1b03a4c18e8f3f0801447b62330faa8ed3bb37 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Tue, 9 Nov 2010 10:08:49 +0900 Subject: ipvs: allow transmit of GRO aggregated skbs Attempt at allowing LVS to transmit skbs of greater than MTU length that have been aggregated by GRO and can thus be deaggregated by GSO. Cc: Julian Anastasov Cc: Herbert Xu Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_xmit.c | 25 +++++++++++++++---------- 1 file changed, 15 insertions(+), 10 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 50b131c28b85..fb2a445ddc59 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -407,7 +407,8 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, /* MTU checking */ mtu = dst_mtu(&rt->dst); - if ((skb->len > mtu) && (iph->frag_off & htons(IP_DF))) { + if ((skb->len > mtu) && (iph->frag_off & htons(IP_DF)) && + !skb_is_gso(skb)) { ip_rt_put(rt); icmp_send(skb, ICMP_DEST_UNREACH,ICMP_FRAG_NEEDED, htonl(mtu)); IP_VS_DBG_RL("%s(): frag needed\n", __func__); @@ -460,7 +461,7 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, /* MTU checking */ mtu = dst_mtu(&rt->dst); - if (skb->len > mtu) { + if (skb->len > mtu && !skb_is_gso(skb)) { if (!skb->dev) { struct net *net = dev_net(skb_dst(skb)->dev); @@ -560,7 +561,8 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, /* MTU checking */ mtu = dst_mtu(&rt->dst); - if ((skb->len > mtu) && (iph->frag_off & htons(IP_DF))) { + if ((skb->len > mtu) && (iph->frag_off & htons(IP_DF)) && + !skb_is_gso(skb)) { icmp_send(skb, ICMP_DEST_UNREACH,ICMP_FRAG_NEEDED, htonl(mtu)); IP_VS_DBG_RL_PKT(0, AF_INET, pp, skb, 0, "ip_vs_nat_xmit(): frag needed for"); @@ -675,7 +677,7 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, /* MTU checking */ mtu = dst_mtu(&rt->dst); - if (skb->len > mtu) { + if (skb->len > mtu && !skb_is_gso(skb)) { if (!skb->dev) { struct net *net = dev_net(skb_dst(skb)->dev); @@ -790,8 +792,8 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, df |= (old_iph->frag_off & htons(IP_DF)); - if ((old_iph->frag_off & htons(IP_DF)) - && mtu < ntohs(old_iph->tot_len)) { + if ((old_iph->frag_off & htons(IP_DF) && + mtu < ntohs(old_iph->tot_len) && !skb_is_gso(skb))) { icmp_send(skb, ICMP_DEST_UNREACH,ICMP_FRAG_NEEDED, htonl(mtu)); IP_VS_DBG_RL("%s(): frag needed\n", __func__); goto tx_error_put; @@ -903,7 +905,8 @@ ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, if (skb_dst(skb)) skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu); - if (mtu < ntohs(old_iph->payload_len) + sizeof(struct ipv6hdr)) { + if (mtu < ntohs(old_iph->payload_len) + sizeof(struct ipv6hdr) && + !skb_is_gso(skb)) { if (!skb->dev) { struct net *net = dev_net(skb_dst(skb)->dev); @@ -1008,7 +1011,8 @@ ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, /* MTU checking */ mtu = dst_mtu(&rt->dst); - if ((iph->frag_off & htons(IP_DF)) && skb->len > mtu) { + if ((iph->frag_off & htons(IP_DF)) && skb->len > mtu && + !skb_is_gso(skb)) { icmp_send(skb, ICMP_DEST_UNREACH,ICMP_FRAG_NEEDED, htonl(mtu)); ip_rt_put(rt); IP_VS_DBG_RL("%s(): frag needed\n", __func__); @@ -1175,7 +1179,8 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, /* MTU checking */ mtu = dst_mtu(&rt->dst); - if ((skb->len > mtu) && (ip_hdr(skb)->frag_off & htons(IP_DF))) { + if ((skb->len > mtu) && (ip_hdr(skb)->frag_off & htons(IP_DF)) && + !skb_is_gso(skb)) { icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu)); IP_VS_DBG_RL("%s(): frag needed\n", __func__); goto tx_error_put; @@ -1289,7 +1294,7 @@ ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, /* MTU checking */ mtu = dst_mtu(&rt->dst); - if (skb->len > mtu) { + if (skb->len > mtu && !skb_is_gso(skb)) { if (!skb->dev) { struct net *net = dev_net(skb_dst(skb)->dev); -- cgit v1.2.3 From 3bfd45f93c8bca7a5dc955235ff083602d95aa43 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 16 Nov 2010 10:19:18 +0100 Subject: netfilter: nf_conntrack: one less atomic op in nf_ct_expect_insert() Instead of doing atomic_inc(&exp->use) twice, call atomic_add(2, &exp->use); Signed-off-by: Eric Dumazet Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_expect.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index bbb21402596d..774f32ba2ac9 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -323,7 +323,8 @@ static void nf_ct_expect_insert(struct nf_conntrack_expect *exp) const struct nf_conntrack_expect_policy *p; unsigned int h = nf_ct_expect_dst_hash(&exp->tuple); - atomic_inc(&exp->use); + /* two references : one for hash insert, one for the timer */ + atomic_add(2, &exp->use); if (master_help) { hlist_add_head(&exp->lnode, &master_help->expectations); @@ -345,7 +346,6 @@ static void nf_ct_expect_insert(struct nf_conntrack_expect *exp) } add_timer(&exp->timeout); - atomic_inc(&exp->use); NF_CT_STAT_INC(net, expect_create); } -- cgit v1.2.3 From 0e051e683ba4acb4e67c272c6a89707d974099d1 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Fri, 19 Nov 2010 14:25:07 +0100 Subject: IPVS: Backup, Prepare for transferring firewall marks (fwmark) to the backup daemon. One struct will have fwmark added: * ip_vs_conn ip_vs_conn_new() and ip_vs_find_dest() will have an extra param - fwmark The effects of that, is in this patch. Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_conn.c | 5 +++-- net/netfilter/ipvs/ip_vs_core.c | 8 ++++---- net/netfilter/ipvs/ip_vs_ctl.c | 4 ++-- net/netfilter/ipvs/ip_vs_ftp.c | 5 +++-- net/netfilter/ipvs/ip_vs_sync.c | 4 ++-- 5 files changed, 14 insertions(+), 12 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 7615f9e3d955..66e4662925d5 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -613,7 +613,7 @@ struct ip_vs_dest *ip_vs_try_bind_dest(struct ip_vs_conn *cp) if ((cp) && (!cp->dest)) { dest = ip_vs_find_dest(cp->af, &cp->daddr, cp->dport, &cp->vaddr, cp->vport, - cp->protocol); + cp->protocol, cp->fwmark); ip_vs_bind_dest(cp, dest); return dest; } else @@ -803,7 +803,7 @@ void ip_vs_conn_expire_now(struct ip_vs_conn *cp) struct ip_vs_conn * ip_vs_conn_new(const struct ip_vs_conn_param *p, const union nf_inet_addr *daddr, __be16 dport, unsigned flags, - struct ip_vs_dest *dest) + struct ip_vs_dest *dest, __u32 fwmark) { struct ip_vs_conn *cp; struct ip_vs_protocol *pp = ip_vs_proto_get(p->protocol); @@ -827,6 +827,7 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p, &cp->daddr, daddr); cp->dport = dport; cp->flags = flags; + cp->fwmark = fwmark; if (flags & IP_VS_CONN_F_TEMPLATE && p->pe) { ip_vs_pe_get(p->pe); cp->pe = p->pe; diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index b4e51e9c5a04..e2bb3cd41c07 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -293,7 +293,7 @@ ip_vs_sched_persist(struct ip_vs_service *svc, * and thus param.pe_data will be destroyed * when the template expires */ ct = ip_vs_conn_new(¶m, &dest->addr, dport, - IP_VS_CONN_F_TEMPLATE, dest); + IP_VS_CONN_F_TEMPLATE, dest, skb->mark); if (ct == NULL) { kfree(param.pe_data); return NULL; @@ -319,7 +319,7 @@ ip_vs_sched_persist(struct ip_vs_service *svc, */ ip_vs_conn_fill_param(svc->af, iph.protocol, &iph.saddr, ports[0], &iph.daddr, ports[1], ¶m); - cp = ip_vs_conn_new(¶m, &dest->addr, dport, flags, dest); + cp = ip_vs_conn_new(¶m, &dest->addr, dport, flags, dest, skb->mark); if (cp == NULL) { ip_vs_conn_put(ct); return NULL; @@ -423,7 +423,7 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, pptr[0], &iph.daddr, pptr[1], &p); cp = ip_vs_conn_new(&p, &dest->addr, dest->port ? dest->port : pptr[1], - flags, dest); + flags, dest, skb->mark); if (!cp) return NULL; } @@ -489,7 +489,7 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, &iph.daddr, pptr[1], &p); cp = ip_vs_conn_new(&p, &daddr, 0, IP_VS_CONN_F_BYPASS | flags, - NULL); + NULL, skb->mark); if (!cp) return NF_DROP; } diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 3e92558dfcc2..a5bd00279047 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -657,12 +657,12 @@ ip_vs_lookup_dest(struct ip_vs_service *svc, const union nf_inet_addr *daddr, struct ip_vs_dest *ip_vs_find_dest(int af, const union nf_inet_addr *daddr, __be16 dport, const union nf_inet_addr *vaddr, - __be16 vport, __u16 protocol) + __be16 vport, __u16 protocol, __u32 fwmark) { struct ip_vs_dest *dest; struct ip_vs_service *svc; - svc = ip_vs_service_get(af, 0, protocol, vaddr, vport); + svc = ip_vs_service_get(af, fwmark, protocol, vaddr, vport); if (!svc) return NULL; dest = ip_vs_lookup_dest(svc, daddr, dport); diff --git a/net/netfilter/ipvs/ip_vs_ftp.c b/net/netfilter/ipvs/ip_vs_ftp.c index 75455000ad1c..84aef65b37d1 100644 --- a/net/netfilter/ipvs/ip_vs_ftp.c +++ b/net/netfilter/ipvs/ip_vs_ftp.c @@ -208,7 +208,7 @@ static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp, n_cp = ip_vs_conn_new(&p, &from, port, IP_VS_CONN_F_NO_CPORT | IP_VS_CONN_F_NFCT, - cp->dest); + cp->dest, skb->mark); if (!n_cp) return 0; @@ -365,7 +365,8 @@ static int ip_vs_ftp_in(struct ip_vs_app *app, struct ip_vs_conn *cp, if (!n_cp) { n_cp = ip_vs_conn_new(&p, &cp->daddr, htons(ntohs(cp->dport)-1), - IP_VS_CONN_F_NFCT, cp->dest); + IP_VS_CONN_F_NFCT, cp->dest, + skb->mark); if (!n_cp) return 0; diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index 3897d6bf3b29..47eed672dc08 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -404,7 +404,7 @@ static void ip_vs_process_message(char *buffer, const size_t buflen) s->dport, (union nf_inet_addr *)&s->vaddr, s->vport, - s->protocol); + s->protocol, 0); /* Set the approprite ativity flag */ if (s->protocol == IPPROTO_TCP) { if (state != IP_VS_TCP_S_ESTABLISHED) @@ -419,7 +419,7 @@ static void ip_vs_process_message(char *buffer, const size_t buflen) } cp = ip_vs_conn_new(¶m, (union nf_inet_addr *)&s->daddr, - s->dport, flags, dest); + s->dport, flags, dest, 0); if (dest) atomic_dec(&dest->refcnt); if (!cp) { -- cgit v1.2.3 From ce144f249f3f21a095a093d5d1ebd845177858da Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Fri, 19 Nov 2010 14:25:08 +0100 Subject: IPVS: Split ports[2] into src_port and dst_port Avoid sending invalid pointer due to skb_linearize() call. This patch prepares for next patch where skb_linearize is a part. In ip_vs_sched_persist() params the ports ptr will be replaced by src and dst port. Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index e2bb3cd41c07..9acdd79a4a05 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -200,7 +200,7 @@ ip_vs_conn_fill_param_persist(const struct ip_vs_service *svc, static struct ip_vs_conn * ip_vs_sched_persist(struct ip_vs_service *svc, struct sk_buff *skb, - __be16 ports[2]) + __be16 src_port, __be16 dst_port) { struct ip_vs_conn *cp = NULL; struct ip_vs_iphdr iph; @@ -224,8 +224,8 @@ ip_vs_sched_persist(struct ip_vs_service *svc, IP_VS_DBG_BUF(6, "p-schedule: src %s:%u dest %s:%u " "mnet %s\n", - IP_VS_DBG_ADDR(svc->af, &iph.saddr), ntohs(ports[0]), - IP_VS_DBG_ADDR(svc->af, &iph.daddr), ntohs(ports[1]), + IP_VS_DBG_ADDR(svc->af, &iph.saddr), ntohs(src_port), + IP_VS_DBG_ADDR(svc->af, &iph.daddr), ntohs(dst_port), IP_VS_DBG_ADDR(svc->af, &snet)); /* @@ -247,14 +247,14 @@ ip_vs_sched_persist(struct ip_vs_service *svc, const union nf_inet_addr fwmark = { .ip = htonl(svc->fwmark) }; __be16 vport = 0; - if (ports[1] == svc->port) { + if (dst_port == svc->port) { /* non-FTP template: * * FTP template: * */ if (svc->port != FTPPORT) - vport = ports[1]; + vport = dst_port; } else { /* Note: persistent fwmark-based services and * persistent port zero service are handled here. @@ -285,7 +285,7 @@ ip_vs_sched_persist(struct ip_vs_service *svc, return NULL; } - if (ports[1] == svc->port && svc->port != FTPPORT) + if (dst_port == svc->port && svc->port != FTPPORT) dport = dest->port; /* Create a template @@ -306,7 +306,7 @@ ip_vs_sched_persist(struct ip_vs_service *svc, kfree(param.pe_data); } - dport = ports[1]; + dport = dst_port; if (dport == svc->port && dest->port) dport = dest->port; @@ -317,8 +317,9 @@ ip_vs_sched_persist(struct ip_vs_service *svc, /* * Create a new connection according to the template */ - ip_vs_conn_fill_param(svc->af, iph.protocol, &iph.saddr, ports[0], - &iph.daddr, ports[1], ¶m); + ip_vs_conn_fill_param(svc->af, iph.protocol, &iph.saddr, src_port, + &iph.daddr, dst_port, ¶m); + cp = ip_vs_conn_new(¶m, &dest->addr, dport, flags, dest, skb->mark); if (cp == NULL) { ip_vs_conn_put(ct); @@ -388,7 +389,7 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, */ if (svc->flags & IP_VS_SVC_F_PERSISTENT) { *ignored = 0; - return ip_vs_sched_persist(svc, skb, pptr); + return ip_vs_sched_persist(svc, skb, pptr[0], pptr[1]); } /* -- cgit v1.2.3 From 3716522653a79b724b02ee911f1b60c41932f847 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Fri, 19 Nov 2010 14:25:09 +0100 Subject: IPVS: skb defrag in L7 helpers L7 helpers like sip needs skb defrag since L7 data can be fragmented. This patch requires "IPVS Break ports-2 into src_port and dst_port" patch Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_pe_sip.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_pe_sip.c b/net/netfilter/ipvs/ip_vs_pe_sip.c index b8b4e9620f3e..0d83bc01fed4 100644 --- a/net/netfilter/ipvs/ip_vs_pe_sip.c +++ b/net/netfilter/ipvs/ip_vs_pe_sip.c @@ -71,6 +71,7 @@ ip_vs_sip_fill_param(struct ip_vs_conn_param *p, struct sk_buff *skb) struct ip_vs_iphdr iph; unsigned int dataoff, datalen, matchoff, matchlen; const char *dptr; + int retc; ip_vs_fill_iphdr(p->af, skb_network_header(skb), &iph); @@ -83,6 +84,8 @@ ip_vs_sip_fill_param(struct ip_vs_conn_param *p, struct sk_buff *skb) if (dataoff >= skb->len) return -EINVAL; + if ((retc=skb_linearize(skb)) < 0) + return retc; dptr = skb->data + dataoff; datalen = skb->len - dataoff; -- cgit v1.2.3 From a5959d53d6048a56103ee0ade1eb6f2c0c733b1d Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Fri, 19 Nov 2010 14:25:10 +0100 Subject: IPVS: Handle Scheduling errors. If ip_vs_conn_fill_param_persist return an error to ip_vs_sched_persist, this error must propagate as ignored=-1 to ip_vs_schedule(). Errors from ip_vs_conn_new() in ip_vs_sched_persist() and ip_vs_schedule() should also return *ignored=-1; This patch just relies on the fact that ignored is 1 before calling ip_vs_sched_persist(). Sent from Julian: "The new case when ip_vs_conn_fill_param_persist fails should set *ignored = -1, so that we can use NF_DROP, see below. *ignored = -1 should be also used for ip_vs_conn_new failure in ip_vs_sched_persist() and ip_vs_schedule(). The new negative value should be handled in tcp,udp,sctp" "To summarize: - *ignored = 1: protocol tried to schedule (eg. on SYN), found svc but the svc/scheduler decides that this packet should be accepted with NF_ACCEPT because it must not be scheduled. - *ignored = 0: scheduler can not find destination, so try bypass or return ICMP and then NF_DROP (ip_vs_leave). - *ignored = -1: scheduler tried to schedule but fatal error occurred, eg. ip_vs_conn_new failure (ENOMEM) or ip_vs_sip_fill_param failure such as missing Call-ID, ENOMEM on skb_linearize or pe_data. In this case we should return NF_DROP without any attempts to send ICMP with ip_vs_leave." More or less all ideas and input to this patch is work from Julian Anastasov Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 56 +++++++++++++++++++++++++---------- net/netfilter/ipvs/ip_vs_proto_sctp.c | 11 +++++-- net/netfilter/ipvs/ip_vs_proto_tcp.c | 10 +++++-- net/netfilter/ipvs/ip_vs_proto_udp.c | 10 +++++-- 4 files changed, 64 insertions(+), 23 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 9acdd79a4a05..3445da6e8c95 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -177,7 +177,7 @@ ip_vs_set_state(struct ip_vs_conn *cp, int direction, return pp->state_transition(cp, direction, skb, pp); } -static inline void +static inline int ip_vs_conn_fill_param_persist(const struct ip_vs_service *svc, struct sk_buff *skb, int protocol, const union nf_inet_addr *caddr, __be16 cport, @@ -187,7 +187,9 @@ ip_vs_conn_fill_param_persist(const struct ip_vs_service *svc, ip_vs_conn_fill_param(svc->af, protocol, caddr, cport, vaddr, vport, p); p->pe = svc->pe; if (p->pe && p->pe->fill_param) - p->pe->fill_param(p, skb); + return p->pe->fill_param(p, skb); + + return 0; } /* @@ -200,7 +202,7 @@ ip_vs_conn_fill_param_persist(const struct ip_vs_service *svc, static struct ip_vs_conn * ip_vs_sched_persist(struct ip_vs_service *svc, struct sk_buff *skb, - __be16 src_port, __be16 dst_port) + __be16 src_port, __be16 dst_port, int *ignored) { struct ip_vs_conn *cp = NULL; struct ip_vs_iphdr iph; @@ -268,20 +270,27 @@ ip_vs_sched_persist(struct ip_vs_service *svc, vaddr = &fwmark; } } - ip_vs_conn_fill_param_persist(svc, skb, protocol, &snet, 0, - vaddr, vport, ¶m); + /* return *ignored = -1 so NF_DROP can be used */ + if (ip_vs_conn_fill_param_persist(svc, skb, protocol, &snet, 0, + vaddr, vport, ¶m) < 0) { + *ignored = -1; + return NULL; + } } /* Check if a template already exists */ ct = ip_vs_ct_in_get(¶m); if (!ct || !ip_vs_check_template(ct)) { - /* No template found or the dest of the connection + /* + * No template found or the dest of the connection * template is not available. + * return *ignored=0 i.e. ICMP and NF_DROP */ dest = svc->scheduler->schedule(svc, skb); if (!dest) { IP_VS_DBG(1, "p-schedule: no dest found.\n"); kfree(param.pe_data); + *ignored = 0; return NULL; } @@ -296,6 +305,7 @@ ip_vs_sched_persist(struct ip_vs_service *svc, IP_VS_CONN_F_TEMPLATE, dest, skb->mark); if (ct == NULL) { kfree(param.pe_data); + *ignored = -1; return NULL; } @@ -323,6 +333,7 @@ ip_vs_sched_persist(struct ip_vs_service *svc, cp = ip_vs_conn_new(¶m, &dest->addr, dport, flags, dest, skb->mark); if (cp == NULL) { ip_vs_conn_put(ct); + *ignored = -1; return NULL; } @@ -342,6 +353,21 @@ ip_vs_sched_persist(struct ip_vs_service *svc, * It selects a server according to the virtual service, and * creates a connection entry. * Protocols supported: TCP, UDP + * + * Usage of *ignored + * + * 1 : protocol tried to schedule (eg. on SYN), found svc but the + * svc/scheduler decides that this packet should be accepted with + * NF_ACCEPT because it must not be scheduled. + * + * 0 : scheduler can not find destination, so try bypass or + * return ICMP and then NF_DROP (ip_vs_leave). + * + * -1 : scheduler tried to schedule but fatal error occurred, eg. + * ip_vs_conn_new failure (ENOMEM) or ip_vs_sip_fill_param + * failure such as missing Call-ID, ENOMEM on skb_linearize + * or pe_data. In this case we should return NF_DROP without + * any attempts to send ICMP with ip_vs_leave. */ struct ip_vs_conn * ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, @@ -372,11 +398,9 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, } /* - * Do not schedule replies from local real server. It is risky - * for fwmark services but mostly for persistent services. + * Do not schedule replies from local real server. */ if ((!skb->dev || skb->dev->flags & IFF_LOOPBACK) && - (svc->flags & IP_VS_SVC_F_PERSISTENT || svc->fwmark) && (cp = pp->conn_in_get(svc->af, skb, pp, &iph, iph.len, 1))) { IP_VS_DBG_PKT(12, svc->af, pp, skb, 0, "Not scheduling reply for existing connection"); @@ -387,10 +411,10 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, /* * Persistent service */ - if (svc->flags & IP_VS_SVC_F_PERSISTENT) { - *ignored = 0; - return ip_vs_sched_persist(svc, skb, pptr[0], pptr[1]); - } + if (svc->flags & IP_VS_SVC_F_PERSISTENT) + return ip_vs_sched_persist(svc, skb, pptr[0], pptr[1], ignored); + + *ignored = 0; /* * Non-persistent service @@ -403,8 +427,6 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, return NULL; } - *ignored = 0; - dest = svc->scheduler->schedule(svc, skb); if (dest == NULL) { IP_VS_DBG(1, "Schedule: no dest found.\n"); @@ -425,8 +447,10 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, cp = ip_vs_conn_new(&p, &dest->addr, dest->port ? dest->port : pptr[1], flags, dest, skb->mark); - if (!cp) + if (!cp) { + *ignored = -1; return NULL; + } } IP_VS_DBG_BUF(6, "Schedule fwd:%c c:%s:%u v:%s:%u " diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c index 1ea96bcd342b..a315159983ad 100644 --- a/net/netfilter/ipvs/ip_vs_proto_sctp.c +++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c @@ -47,13 +47,18 @@ sctp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, * incoming connection, and create a connection entry. */ *cpp = ip_vs_schedule(svc, skb, pp, &ignored); - if (!*cpp && !ignored) { - *verdict = ip_vs_leave(svc, skb, pp); + if (!*cpp && ignored <= 0) { + if (!ignored) + *verdict = ip_vs_leave(svc, skb, pp); + else { + ip_vs_service_put(svc); + *verdict = NF_DROP; + } return 0; } ip_vs_service_put(svc); } - + /* NF_ACCEPT */ return 1; } diff --git a/net/netfilter/ipvs/ip_vs_proto_tcp.c b/net/netfilter/ipvs/ip_vs_proto_tcp.c index f6c5200e2146..1cdab12abfef 100644 --- a/net/netfilter/ipvs/ip_vs_proto_tcp.c +++ b/net/netfilter/ipvs/ip_vs_proto_tcp.c @@ -64,12 +64,18 @@ tcp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, * incoming connection, and create a connection entry. */ *cpp = ip_vs_schedule(svc, skb, pp, &ignored); - if (!*cpp && !ignored) { - *verdict = ip_vs_leave(svc, skb, pp); + if (!*cpp && ignored <= 0) { + if (!ignored) + *verdict = ip_vs_leave(svc, skb, pp); + else { + ip_vs_service_put(svc); + *verdict = NF_DROP; + } return 0; } ip_vs_service_put(svc); } + /* NF_ACCEPT */ return 1; } diff --git a/net/netfilter/ipvs/ip_vs_proto_udp.c b/net/netfilter/ipvs/ip_vs_proto_udp.c index 9d106a06bb0a..cd398de010cc 100644 --- a/net/netfilter/ipvs/ip_vs_proto_udp.c +++ b/net/netfilter/ipvs/ip_vs_proto_udp.c @@ -63,12 +63,18 @@ udp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, * incoming connection, and create a connection entry. */ *cpp = ip_vs_schedule(svc, skb, pp, &ignored); - if (!*cpp && !ignored) { - *verdict = ip_vs_leave(svc, skb, pp); + if (!*cpp && ignored <= 0) { + if (!ignored) + *verdict = ip_vs_leave(svc, skb, pp); + else { + ip_vs_service_put(svc); + *verdict = NF_DROP; + } return 0; } ip_vs_service_put(svc); } + /* NF_ACCEPT */ return 1; } -- cgit v1.2.3 From 2981bc9a63456500037ca1f434b93a561e63f384 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Fri, 19 Nov 2010 14:25:11 +0100 Subject: IPVS: Backup, Adding structs for new sync format New structs defined for version 1 of sync. * ip_vs_sync_v4 Ipv4 base format struct * ip_vs_sync_v6 Ipv6 base format struct Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_sync.c | 154 ++++++++++++++++++++++++++++++++++++---- 1 file changed, 142 insertions(+), 12 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index 47eed672dc08..566482f227fa 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -43,11 +43,13 @@ #define IP_VS_SYNC_GROUP 0xe0000051 /* multicast addr - 224.0.0.81 */ #define IP_VS_SYNC_PORT 8848 /* multicast port */ +#define SYNC_PROTO_VER 1 /* Protocol version in header */ /* * IPVS sync connection entry + * Version 0, i.e. original version. */ -struct ip_vs_sync_conn { +struct ip_vs_sync_conn_v0 { __u8 reserved; /* Protocol, addresses and port numbers */ @@ -71,40 +73,157 @@ struct ip_vs_sync_conn_options { struct ip_vs_seq out_seq; /* outgoing seq. struct */ }; +/* + Sync Connection format (sync_conn) + + 0 1 2 3 + 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | Type | Protocol | Ver. | Size | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | Flags | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | State | cport | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | vport | dport | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | fwmark | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | timeout (in sec.) | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | ... | + | IP-Addresses (v4 or v6) | + | ... | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + Optional Parameters. + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | Param. Type | Param. Length | Param. data | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | + | ... | + | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | | Param Type | Param. Length | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | Param data | + | Last Param data should be padded for 32 bit alignment | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +*/ + +/* + * Type 0, IPv4 sync connection format + */ +struct ip_vs_sync_v4 { + __u8 type; + __u8 protocol; /* Which protocol (TCP/UDP) */ + __be16 ver_size; /* Version msb 4 bits */ + /* Flags and state transition */ + __be32 flags; /* status flags */ + __be16 state; /* state info */ + /* Protocol, addresses and port numbers */ + __be16 cport; + __be16 vport; + __be16 dport; + __be32 fwmark; /* Firewall mark from skb */ + __be32 timeout; /* cp timeout */ + __be32 caddr; /* client address */ + __be32 vaddr; /* virtual address */ + __be32 daddr; /* destination address */ + /* The sequence options start here */ + /* PE data padded to 32bit alignment after seq. options */ +}; +/* + * Type 2 messages IPv6 + */ +struct ip_vs_sync_v6 { + __u8 type; + __u8 protocol; /* Which protocol (TCP/UDP) */ + __be16 ver_size; /* Version msb 4 bits */ + /* Flags and state transition */ + __be32 flags; /* status flags */ + __be16 state; /* state info */ + /* Protocol, addresses and port numbers */ + __be16 cport; + __be16 vport; + __be16 dport; + __be32 fwmark; /* Firewall mark from skb */ + __be32 timeout; /* cp timeout */ + struct in6_addr caddr; /* client address */ + struct in6_addr vaddr; /* virtual address */ + struct in6_addr daddr; /* destination address */ + /* The sequence options start here */ + /* PE data padded to 32bit alignment after seq. options */ +}; + +union ip_vs_sync_conn { + struct ip_vs_sync_v4 v4; + struct ip_vs_sync_v6 v6; +}; + +/* Bits in Type field in above */ +#define STYPE_INET6 0 +#define STYPE_F_INET6 (1 << STYPE_INET6) + +#define SVER_SHIFT 12 /* Shift to get version */ +#define SVER_MASK 0x0fff /* Mask to strip version */ + +#define IPVS_OPT_SEQ_DATA 1 +#define IPVS_OPT_PE_DATA 2 +#define IPVS_OPT_PE_NAME 3 +#define IPVS_OPT_PARAM 7 + +#define IPVS_OPT_F_SEQ_DATA (1 << (IPVS_OPT_SEQ_DATA-1)) +#define IPVS_OPT_F_PE_DATA (1 << (IPVS_OPT_PE_DATA-1)) +#define IPVS_OPT_F_PE_NAME (1 << (IPVS_OPT_PE_NAME-1)) +#define IPVS_OPT_F_PARAM (1 << (IPVS_OPT_PARAM-1)) + struct ip_vs_sync_thread_data { struct socket *sock; char *buf; }; -#define SIMPLE_CONN_SIZE (sizeof(struct ip_vs_sync_conn)) +/* Version 0 definition of packet sizes */ +#define SIMPLE_CONN_SIZE (sizeof(struct ip_vs_sync_conn_v0)) #define FULL_CONN_SIZE \ -(sizeof(struct ip_vs_sync_conn) + sizeof(struct ip_vs_sync_conn_options)) +(sizeof(struct ip_vs_sync_conn_v0) + sizeof(struct ip_vs_sync_conn_options)) /* - The master mulitcasts messages to the backup load balancers in the - following format. + The master mulitcasts messages (Datagrams) to the backup load balancers + in the following format. + + Version 1: + Note, first byte should be Zero, so ver 0 receivers will drop the packet. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ - | Count Conns | SyncID | Size | + | 0 | SyncID | Size | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | Count Conns | Version | Reserved, set to Zero | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | | IPVS Sync Connection (1) | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | . | - | . | + ~ . ~ | . | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | | IPVS Sync Connection (n) | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + + Version 0 Header + 0 1 2 3 + 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | Count Conns | SyncID | Size | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + | IPVS Sync Connection (1) | */ #define SYNC_MESG_HEADER_LEN 4 #define MAX_CONNS_PER_SYNCBUFF 255 /* nr_conns in ip_vs_sync_mesg is 8 bit */ +/* Version 0 header */ struct ip_vs_sync_mesg { __u8 nr_conns; __u8 syncid; @@ -113,6 +232,17 @@ struct ip_vs_sync_mesg { /* ip_vs_sync_conn entries start here */ }; +/* Version 1 header */ +struct ip_vs_sync_mesg_v2 { + __u8 reserved; /* must be zero */ + __u8 syncid; + __u16 size; + __u8 nr_conns; + __s8 version; /* SYNC_PROTO_VER */ + __u16 spare; + /* ip_vs_sync_conn entries start here */ +}; + /* the maximum length of sync (sending/receiving) message */ static int sync_send_mesg_maxlen; static int sync_recv_mesg_maxlen; @@ -239,7 +369,7 @@ get_curr_sync_buff(unsigned long time) void ip_vs_sync_conn(const struct ip_vs_conn *cp) { struct ip_vs_sync_mesg *m; - struct ip_vs_sync_conn *s; + struct ip_vs_sync_conn_v0 *s; int len; spin_lock(&curr_sb_lock); @@ -254,7 +384,7 @@ void ip_vs_sync_conn(const struct ip_vs_conn *cp) len = (cp->flags & IP_VS_CONN_F_SEQ_MASK) ? FULL_CONN_SIZE : SIMPLE_CONN_SIZE; m = curr_sb->mesg; - s = (struct ip_vs_sync_conn *)curr_sb->head; + s = (struct ip_vs_sync_conn_v0 *)curr_sb->head; /* copy members */ s->protocol = cp->protocol; @@ -306,7 +436,7 @@ ip_vs_conn_fill_param_sync(int af, int protocol, static void ip_vs_process_message(char *buffer, const size_t buflen) { struct ip_vs_sync_mesg *m = (struct ip_vs_sync_mesg *)buffer; - struct ip_vs_sync_conn *s; + struct ip_vs_sync_conn_v0 *s; struct ip_vs_sync_conn_options *opt; struct ip_vs_conn *cp; struct ip_vs_protocol *pp; @@ -343,7 +473,7 @@ static void ip_vs_process_message(char *buffer, const size_t buflen) IP_VS_ERR_RL("bogus conn in sync message\n"); return; } - s = (struct ip_vs_sync_conn *) p; + s = (struct ip_vs_sync_conn_v0 *) p; flags = ntohs(s->flags) | IP_VS_CONN_F_SYNC; flags &= ~IP_VS_CONN_F_HASHED; if (flags & IP_VS_CONN_F_SEQ_MASK) { @@ -849,7 +979,7 @@ int start_sync_thread(int state, char *mcast_ifn, __u8 syncid) IP_VS_DBG(7, "%s(): pid %d\n", __func__, task_pid_nr(current)); IP_VS_DBG(7, "Each ip_vs_sync_conn entry needs %Zd bytes\n", - sizeof(struct ip_vs_sync_conn)); + sizeof(struct ip_vs_sync_conn_v0)); if (state == IP_VS_STATE_MASTER) { if (sync_master_thread) -- cgit v1.2.3 From fe5e7a1efb664df0280f10377813d7099fb7eb0f Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Fri, 19 Nov 2010 14:25:12 +0100 Subject: IPVS: Backup, Adding Version 1 receive capability Functionality improvements * flags changed from 16 to 32 bits * fwmark added (32 bits) * timeout in sec. added (32 bits) * pe data added (Variable length) * IPv6 capabilities (3x16 bytes for addr.) * Version and type in every conn msg. ip_vs_process_message() now handles Version 1 messages and will call ip_vs_process_message_v0() for version 0 messages. ip_vs_proc_conn() is common for both version, and handles the update of connection hash. ip_vs_conn_fill_param_sync() - Version 1 messages only ip_vs_conn_fill_param_sync_v0() - Version 0 messages only Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_pe.c | 5 +- net/netfilter/ipvs/ip_vs_sync.c | 549 +++++++++++++++++++++++++++++++--------- 2 files changed, 431 insertions(+), 123 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_pe.c b/net/netfilter/ipvs/ip_vs_pe.c index e99f920b93d1..5cf859ccb31b 100644 --- a/net/netfilter/ipvs/ip_vs_pe.c +++ b/net/netfilter/ipvs/ip_vs_pe.c @@ -29,12 +29,11 @@ void ip_vs_unbind_pe(struct ip_vs_service *svc) } /* Get pe in the pe list by name */ -static struct ip_vs_pe * -__ip_vs_pe_getbyname(const char *pe_name) +struct ip_vs_pe *__ip_vs_pe_getbyname(const char *pe_name) { struct ip_vs_pe *pe; - IP_VS_DBG(2, "%s(): pe_name \"%s\"\n", __func__, + IP_VS_DBG(10, "%s(): pe_name \"%s\"\n", __func__, pe_name); spin_lock_bh(&ip_vs_pe_lock); diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index 566482f227fa..e071508901d1 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -35,6 +35,8 @@ #include #include +#include /* Used for ntoh_seq and hton_seq */ + #include #include @@ -286,6 +288,16 @@ static struct sockaddr_in mcast_addr = { .sin_addr.s_addr = cpu_to_be32(IP_VS_SYNC_GROUP), }; +/* + * Copy of struct ip_vs_seq + * From unaligned network order to aligned host order + */ +static void ntoh_seq(struct ip_vs_seq *no, struct ip_vs_seq *ho) +{ + ho->init_seq = get_unaligned_be32(&no->init_seq); + ho->delta = get_unaligned_be32(&no->delta); + ho->previous_delta = get_unaligned_be32(&no->previous_delta); +} static inline struct ip_vs_sync_buff *sb_dequeue(void) { @@ -418,59 +430,186 @@ void ip_vs_sync_conn(const struct ip_vs_conn *cp) ip_vs_sync_conn(cp->control); } +/* + * fill_param used by version 1 + */ static inline int -ip_vs_conn_fill_param_sync(int af, int protocol, - const union nf_inet_addr *caddr, __be16 cport, - const union nf_inet_addr *vaddr, __be16 vport, - struct ip_vs_conn_param *p) +ip_vs_conn_fill_param_sync(int af, union ip_vs_sync_conn *sc, + struct ip_vs_conn_param *p, + __u8 *pe_data, unsigned int pe_data_len, + __u8 *pe_name, unsigned int pe_name_len) { - /* XXX: Need to take into account persistence engine */ - ip_vs_conn_fill_param(af, protocol, caddr, cport, vaddr, vport, p); +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) + ip_vs_conn_fill_param(af, sc->v6.protocol, + (const union nf_inet_addr *)&sc->v6.caddr, + sc->v6.cport, + (const union nf_inet_addr *)&sc->v6.vaddr, + sc->v6.vport, p); + else +#endif + ip_vs_conn_fill_param(af, sc->v4.protocol, + (const union nf_inet_addr *)&sc->v4.caddr, + sc->v4.cport, + (const union nf_inet_addr *)&sc->v4.vaddr, + sc->v4.vport, p); + /* Handle pe data */ + if (pe_data_len) { + if (pe_name_len) { + char buff[IP_VS_PENAME_MAXLEN+1]; + + memcpy(buff, pe_name, pe_name_len); + buff[pe_name_len]=0; + p->pe = __ip_vs_pe_getbyname(buff); + if (!p->pe) { + IP_VS_DBG(3, "BACKUP, no %s engine found/loaded\n", buff); + return 1; + } + } else { + IP_VS_ERR_RL("BACKUP, Invalid PE parameters\n"); + return 1; + } + + p->pe_data = kmalloc(pe_data_len, GFP_ATOMIC); + if (!p->pe_data) { + if (p->pe->module) + module_put(p->pe->module); + return -ENOMEM; + } + memcpy(p->pe_data, pe_data, pe_data_len); + p->pe_data_len = pe_data_len; + } return 0; } /* - * Process received multicast message and create the corresponding - * ip_vs_conn entries. + * Connection Add / Update. + * Common for version 0 and 1 reception of backup sync_conns. + * Param: ... + * timeout is in sec. + */ +static void ip_vs_proc_conn(struct ip_vs_conn_param *param, unsigned flags, + unsigned state, unsigned protocol, unsigned type, + const union nf_inet_addr *daddr, __be16 dport, + unsigned long timeout, __u32 fwmark, + struct ip_vs_sync_conn_options *opt, + struct ip_vs_protocol *pp) +{ + struct ip_vs_dest *dest; + struct ip_vs_conn *cp; + + + if (!(flags & IP_VS_CONN_F_TEMPLATE)) + cp = ip_vs_conn_in_get(param); + else + cp = ip_vs_ct_in_get(param); + + if (cp && param->pe_data) /* Free pe_data */ + kfree(param->pe_data); + if (!cp) { + /* + * Find the appropriate destination for the connection. + * If it is not found the connection will remain unbound + * but still handled. + */ + dest = ip_vs_find_dest(type, daddr, dport, param->vaddr, + param->vport, protocol, fwmark); + + /* Set the approprite ativity flag */ + if (protocol == IPPROTO_TCP) { + if (state != IP_VS_TCP_S_ESTABLISHED) + flags |= IP_VS_CONN_F_INACTIVE; + else + flags &= ~IP_VS_CONN_F_INACTIVE; + } else if (protocol == IPPROTO_SCTP) { + if (state != IP_VS_SCTP_S_ESTABLISHED) + flags |= IP_VS_CONN_F_INACTIVE; + else + flags &= ~IP_VS_CONN_F_INACTIVE; + } + cp = ip_vs_conn_new(param, daddr, dport, flags, dest, fwmark); + if (dest) + atomic_dec(&dest->refcnt); + if (!cp) { + if (param->pe_data) + kfree(param->pe_data); + IP_VS_DBG(2, "BACKUP, add new conn. failed\n"); + return; + } + } else if (!cp->dest) { + dest = ip_vs_try_bind_dest(cp); + if (dest) + atomic_dec(&dest->refcnt); + } else if ((cp->dest) && (cp->protocol == IPPROTO_TCP) && + (cp->state != state)) { + /* update active/inactive flag for the connection */ + dest = cp->dest; + if (!(cp->flags & IP_VS_CONN_F_INACTIVE) && + (state != IP_VS_TCP_S_ESTABLISHED)) { + atomic_dec(&dest->activeconns); + atomic_inc(&dest->inactconns); + cp->flags |= IP_VS_CONN_F_INACTIVE; + } else if ((cp->flags & IP_VS_CONN_F_INACTIVE) && + (state == IP_VS_TCP_S_ESTABLISHED)) { + atomic_inc(&dest->activeconns); + atomic_dec(&dest->inactconns); + cp->flags &= ~IP_VS_CONN_F_INACTIVE; + } + } else if ((cp->dest) && (cp->protocol == IPPROTO_SCTP) && + (cp->state != state)) { + dest = cp->dest; + if (!(cp->flags & IP_VS_CONN_F_INACTIVE) && + (state != IP_VS_SCTP_S_ESTABLISHED)) { + atomic_dec(&dest->activeconns); + atomic_inc(&dest->inactconns); + cp->flags &= ~IP_VS_CONN_F_INACTIVE; + } + } + + if (opt) + memcpy(&cp->in_seq, opt, sizeof(*opt)); + atomic_set(&cp->in_pkts, sysctl_ip_vs_sync_threshold[0]); + cp->state = state; + cp->old_state = cp->state; + /* + * For Ver 0 messages style + * - Not possible to recover the right timeout for templates + * - can not find the right fwmark + * virtual service. If needed, we can do it for + * non-fwmark persistent services. + * Ver 1 messages style. + * - No problem. + */ + if (timeout) { + if (timeout > MAX_SCHEDULE_TIMEOUT / HZ) + timeout = MAX_SCHEDULE_TIMEOUT / HZ; + cp->timeout = timeout*HZ; + } else if (!(flags & IP_VS_CONN_F_TEMPLATE) && pp->timeout_table) + cp->timeout = pp->timeout_table[state]; + else + cp->timeout = (3*60*HZ); + ip_vs_conn_put(cp); +} + +/* + * Process received multicast message for Version 0 */ -static void ip_vs_process_message(char *buffer, const size_t buflen) +static void ip_vs_process_message_v0(const char *buffer, const size_t buflen) { struct ip_vs_sync_mesg *m = (struct ip_vs_sync_mesg *)buffer; struct ip_vs_sync_conn_v0 *s; struct ip_vs_sync_conn_options *opt; - struct ip_vs_conn *cp; struct ip_vs_protocol *pp; - struct ip_vs_dest *dest; struct ip_vs_conn_param param; char *p; int i; - if (buflen < sizeof(struct ip_vs_sync_mesg)) { - IP_VS_ERR_RL("sync message header too short\n"); - return; - } - - /* Convert size back to host byte order */ - m->size = ntohs(m->size); - - if (buflen != m->size) { - IP_VS_ERR_RL("bogus sync message size\n"); - return; - } - - /* SyncID sanity check */ - if (ip_vs_backup_syncid != 0 && m->syncid != ip_vs_backup_syncid) { - IP_VS_DBG(7, "Ignoring incoming msg with syncid = %d\n", - m->syncid); - return; - } - p = (char *)buffer + sizeof(struct ip_vs_sync_mesg); for (i=0; inr_conns; i++) { unsigned flags, state; if (p + SIMPLE_CONN_SIZE > buffer+buflen) { - IP_VS_ERR_RL("bogus conn in sync message\n"); + IP_VS_ERR_RL("BACKUP v0, bogus conn\n"); return; } s = (struct ip_vs_sync_conn_v0 *) p; @@ -480,7 +619,7 @@ static void ip_vs_process_message(char *buffer, const size_t buflen) opt = (struct ip_vs_sync_conn_options *)&s[1]; p += FULL_CONN_SIZE; if (p > buffer+buflen) { - IP_VS_ERR_RL("bogus conn options in sync message\n"); + IP_VS_ERR_RL("BACKUP v0, Dropping buffer bogus conn options\n"); return; } } else { @@ -492,12 +631,12 @@ static void ip_vs_process_message(char *buffer, const size_t buflen) if (!(flags & IP_VS_CONN_F_TEMPLATE)) { pp = ip_vs_proto_get(s->protocol); if (!pp) { - IP_VS_ERR_RL("Unsupported protocol %u in sync msg\n", + IP_VS_DBG(2, "BACKUP v0, Unsupported protocol %u\n", s->protocol); continue; } if (state >= pp->num_states) { - IP_VS_DBG(2, "Invalid %s state %u in sync msg\n", + IP_VS_DBG(2, "BACKUP v0, Invalid %s state %u\n", pp->name, state); continue; } @@ -505,103 +644,273 @@ static void ip_vs_process_message(char *buffer, const size_t buflen) /* protocol in templates is not used for state/timeout */ pp = NULL; if (state > 0) { - IP_VS_DBG(2, "Invalid template state %u in sync msg\n", + IP_VS_DBG(2, "BACKUP v0, Invalid template state %u\n", state); state = 0; } } - if (ip_vs_conn_fill_param_sync(AF_INET, s->protocol, - (union nf_inet_addr *)&s->caddr, - s->cport, - (union nf_inet_addr *)&s->vaddr, - s->vport, ¶m)) { - pr_err("ip_vs_conn_fill_param_sync failed"); - return; + ip_vs_conn_fill_param(AF_INET, s->protocol, + (const union nf_inet_addr *)&s->caddr, + s->cport, + (const union nf_inet_addr *)&s->vaddr, + s->vport, ¶m); + + /* Send timeout as Zero */ + ip_vs_proc_conn(¶m, flags, state, s->protocol, AF_INET, + (union nf_inet_addr *)&s->daddr, s->dport, + 0, 0, opt, pp); + } +} + +/* + * Handle options + */ +static inline int ip_vs_proc_seqopt(__u8 *p, unsigned int plen, + __u32 *opt_flags, + struct ip_vs_sync_conn_options *opt) +{ + struct ip_vs_sync_conn_options *topt; + + topt = (struct ip_vs_sync_conn_options *)p; + + if (plen != sizeof(struct ip_vs_sync_conn_options)) { + IP_VS_DBG(2, "BACKUP, bogus conn options length\n"); + return -EINVAL; + } + if (*opt_flags & IPVS_OPT_F_SEQ_DATA) { + IP_VS_DBG(2, "BACKUP, conn options found twice\n"); + return -EINVAL; + } + ntoh_seq(&topt->in_seq, &opt->in_seq); + ntoh_seq(&topt->out_seq, &opt->out_seq); + *opt_flags |= IPVS_OPT_F_SEQ_DATA; + return 0; +} + +static int ip_vs_proc_str(__u8 *p, unsigned int plen, unsigned int *data_len, + __u8 **data, unsigned int maxlen, + __u32 *opt_flags, __u32 flag) +{ + if (plen > maxlen) { + IP_VS_DBG(2, "BACKUP, bogus par.data len > %d\n", maxlen); + return -EINVAL; + } + if (*opt_flags & flag) { + IP_VS_DBG(2, "BACKUP, Par.data found twice 0x%x\n", flag); + return -EINVAL; + } + *data_len = plen; + *data = p; + *opt_flags |= flag; + return 0; +} +/* + * Process a Version 1 sync. connection + */ +static inline int ip_vs_proc_sync_conn(__u8 *p, __u8 *msg_end) +{ + struct ip_vs_sync_conn_options opt; + union ip_vs_sync_conn *s; + struct ip_vs_protocol *pp; + struct ip_vs_conn_param param; + __u32 flags; + unsigned int af, state, pe_data_len=0, pe_name_len=0; + __u8 *pe_data=NULL, *pe_name=NULL; + __u32 opt_flags=0; + int retc=0; + + s = (union ip_vs_sync_conn *) p; + + if (s->v6.type & STYPE_F_INET6) { +#ifdef CONFIG_IP_VS_IPV6 + af = AF_INET6; + p += sizeof(struct ip_vs_sync_v6); +#else + IP_VS_DBG(3,"BACKUP, IPv6 msg received, and IPVS is not compiled for IPv6\n"); + retc = 10; + goto out; +#endif + } else if (!s->v4.type) { + af = AF_INET; + p += sizeof(struct ip_vs_sync_v4); + } else { + return -10; + } + if (p > msg_end) + return -20; + + /* Process optional params check Type & Len. */ + while (p < msg_end) { + int ptype; + int plen; + + if (p+2 > msg_end) + return -30; + ptype = *(p++); + plen = *(p++); + + if (!plen || ((p + plen) > msg_end)) + return -40; + /* Handle seq option p = param data */ + switch (ptype & ~IPVS_OPT_F_PARAM) { + case IPVS_OPT_SEQ_DATA: + if (ip_vs_proc_seqopt(p, plen, &opt_flags, &opt)) + return -50; + break; + + case IPVS_OPT_PE_DATA: + if (ip_vs_proc_str(p, plen, &pe_data_len, &pe_data, + IP_VS_PEDATA_MAXLEN, &opt_flags, + IPVS_OPT_F_PE_DATA)) + return -60; + break; + + case IPVS_OPT_PE_NAME: + if (ip_vs_proc_str(p, plen,&pe_name_len, &pe_name, + IP_VS_PENAME_MAXLEN, &opt_flags, + IPVS_OPT_F_PE_NAME)) + return -70; + break; + + default: + /* Param data mandatory ? */ + if (!(ptype & IPVS_OPT_F_PARAM)) { + IP_VS_DBG(3, "BACKUP, Unknown mandatory param %d found\n", + ptype & ~IPVS_OPT_F_PARAM); + retc = 20; + goto out; + } } - if (!(flags & IP_VS_CONN_F_TEMPLATE)) - cp = ip_vs_conn_in_get(¶m); - else - cp = ip_vs_ct_in_get(¶m); - if (!cp) { - /* - * Find the appropriate destination for the connection. - * If it is not found the connection will remain unbound - * but still handled. - */ - dest = ip_vs_find_dest(AF_INET, - (union nf_inet_addr *)&s->daddr, - s->dport, - (union nf_inet_addr *)&s->vaddr, - s->vport, - s->protocol, 0); - /* Set the approprite ativity flag */ - if (s->protocol == IPPROTO_TCP) { - if (state != IP_VS_TCP_S_ESTABLISHED) - flags |= IP_VS_CONN_F_INACTIVE; - else - flags &= ~IP_VS_CONN_F_INACTIVE; - } else if (s->protocol == IPPROTO_SCTP) { - if (state != IP_VS_SCTP_S_ESTABLISHED) - flags |= IP_VS_CONN_F_INACTIVE; - else - flags &= ~IP_VS_CONN_F_INACTIVE; + p += plen; /* Next option */ + } + + /* Get flags and Mask off unsupported */ + flags = ntohl(s->v4.flags) & IP_VS_CONN_F_BACKUP_MASK; + flags |= IP_VS_CONN_F_SYNC; + state = ntohs(s->v4.state); + + if (!(flags & IP_VS_CONN_F_TEMPLATE)) { + pp = ip_vs_proto_get(s->v4.protocol); + if (!pp) { + IP_VS_DBG(3,"BACKUP, Unsupported protocol %u\n", + s->v4.protocol); + retc = 30; + goto out; + } + if (state >= pp->num_states) { + IP_VS_DBG(3, "BACKUP, Invalid %s state %u\n", + pp->name, state); + retc = 40; + goto out; + } + } else { + /* protocol in templates is not used for state/timeout */ + pp = NULL; + if (state > 0) { + IP_VS_DBG(3, "BACKUP, Invalid template state %u\n", + state); + state = 0; + } + } + if (ip_vs_conn_fill_param_sync(af, s, ¶m, + pe_data, pe_data_len, + pe_name, pe_name_len)) { + retc = 50; + goto out; + } + /* If only IPv4, just silent skip IPv6 */ + if (af == AF_INET) + ip_vs_proc_conn(¶m, flags, state, s->v4.protocol, af, + (union nf_inet_addr *)&s->v4.daddr, s->v4.dport, + ntohl(s->v4.timeout), ntohl(s->v4.fwmark), + (opt_flags & IPVS_OPT_F_SEQ_DATA ? &opt : NULL), + pp); +#ifdef CONFIG_IP_VS_IPV6 + else + ip_vs_proc_conn(¶m, flags, state, s->v6.protocol, af, + (union nf_inet_addr *)&s->v6.daddr, s->v6.dport, + ntohl(s->v6.timeout), ntohl(s->v6.fwmark), + (opt_flags & IPVS_OPT_F_SEQ_DATA ? &opt : NULL), + pp); +#endif + return 0; + /* Error exit */ +out: + IP_VS_DBG(2, "BACKUP, Single msg dropped err:%d\n", retc); + return retc; + +} +/* + * Process received multicast message and create the corresponding + * ip_vs_conn entries. + * Handles Version 0 & 1 + */ +static void ip_vs_process_message(__u8 *buffer, const size_t buflen) +{ + struct ip_vs_sync_mesg_v2 *m2 = (struct ip_vs_sync_mesg_v2 *)buffer; + __u8 *p, *msg_end; + unsigned int i, nr_conns; + + if (buflen < sizeof(struct ip_vs_sync_mesg)) { + IP_VS_DBG(2, "BACKUP, message header too short\n"); + return; + } + /* Convert size back to host byte order */ + m2->size = ntohs(m2->size); + + if (buflen != m2->size) { + IP_VS_DBG(2, "BACKUP, bogus message size\n"); + return; + } + /* SyncID sanity check */ + if (ip_vs_backup_syncid != 0 && m2->syncid != ip_vs_backup_syncid) { + IP_VS_DBG(7, "BACKUP, Ignoring syncid = %d\n", m2->syncid); + return; + } + /* Handle version 1 message */ + if ((m2->version == SYNC_PROTO_VER) && (m2->reserved == 0) + && (m2->spare == 0)) { + + msg_end = buffer + sizeof(struct ip_vs_sync_mesg_v2); + nr_conns = m2->nr_conns; + + for (i=0; iv4) > buffer+buflen) { + IP_VS_ERR_RL("BACKUP, Dropping buffer, to small\n"); + return; } - cp = ip_vs_conn_new(¶m, - (union nf_inet_addr *)&s->daddr, - s->dport, flags, dest, 0); - if (dest) - atomic_dec(&dest->refcnt); - if (!cp) { - pr_err("ip_vs_conn_new failed\n"); + s = (union ip_vs_sync_conn *)p; + size = ntohs(s->v4.ver_size) & SVER_MASK; + msg_end = p + size; + /* Basic sanity checks */ + if (msg_end > buffer+buflen) { + IP_VS_ERR_RL("BACKUP, Dropping buffer, msg > buffer\n"); return; } - } else if (!cp->dest) { - dest = ip_vs_try_bind_dest(cp); - if (dest) - atomic_dec(&dest->refcnt); - } else if ((cp->dest) && (cp->protocol == IPPROTO_TCP) && - (cp->state != state)) { - /* update active/inactive flag for the connection */ - dest = cp->dest; - if (!(cp->flags & IP_VS_CONN_F_INACTIVE) && - (state != IP_VS_TCP_S_ESTABLISHED)) { - atomic_dec(&dest->activeconns); - atomic_inc(&dest->inactconns); - cp->flags |= IP_VS_CONN_F_INACTIVE; - } else if ((cp->flags & IP_VS_CONN_F_INACTIVE) && - (state == IP_VS_TCP_S_ESTABLISHED)) { - atomic_inc(&dest->activeconns); - atomic_dec(&dest->inactconns); - cp->flags &= ~IP_VS_CONN_F_INACTIVE; + if (ntohs(s->v4.ver_size) >> SVER_SHIFT) { + IP_VS_ERR_RL("BACKUP, Dropping buffer, Unknown version %d\n", + ntohs(s->v4.ver_size) >> SVER_SHIFT); + return; } - } else if ((cp->dest) && (cp->protocol == IPPROTO_SCTP) && - (cp->state != state)) { - dest = cp->dest; - if (!(cp->flags & IP_VS_CONN_F_INACTIVE) && - (state != IP_VS_SCTP_S_ESTABLISHED)) { - atomic_dec(&dest->activeconns); - atomic_inc(&dest->inactconns); - cp->flags &= ~IP_VS_CONN_F_INACTIVE; + /* Process a single sync_conn */ + if ((retc=ip_vs_proc_sync_conn(p, msg_end)) < 0) { + IP_VS_ERR_RL("BACKUP, Dropping buffer, Err: %d in decoding\n", + retc); + return; } + /* Make sure we have 32 bit alignment */ + msg_end = p + ((size + 3) & ~3); } - - if (opt) - memcpy(&cp->in_seq, opt, sizeof(*opt)); - atomic_set(&cp->in_pkts, sysctl_ip_vs_sync_threshold[0]); - cp->state = state; - cp->old_state = cp->state; - /* - * We can not recover the right timeout for templates - * in all cases, we can not find the right fwmark - * virtual service. If needed, we can do it for - * non-fwmark persistent services. - */ - if (!(flags & IP_VS_CONN_F_TEMPLATE) && pp->timeout_table) - cp->timeout = pp->timeout_table[state]; - else - cp->timeout = (3*60*HZ); - ip_vs_conn_put(cp); + } else { + /* Old type of message */ + ip_vs_process_message_v0(buffer, buflen); + return; } } -- cgit v1.2.3 From 986a075795339c5ea1122ce9290dfd5504252eb0 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Fri, 19 Nov 2010 14:25:13 +0100 Subject: IPVS: Backup, Change sending to Version 1 format Enable sending and removal of version 0 sending Affected functions, ip_vs_sync_buff_create() ip_vs_sync_conn() ip_vs_core.c removal of IPv4 check. *v5 Just check cp->pe_data_len in ip_vs_sync_conn Check if padding needed before adding a new sync_conn to the buffer, i.e. avoid sending padding at the end. *v4 moved sanity check and pe_name_len after sloop. use cp->pe instead of cp->dest->svc->pe real length in each sync_conn, not padded length however total size of a sync_msg includes padding. *v3 Sending ip_vs_sync_conn_options in network order. Sending Templates for ONE_PACKET conn. Renaming of ip_vs_sync_mesg to ip_vs_sync_mesg_v0 Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 13 ++- net/netfilter/ipvs/ip_vs_sync.c | 189 +++++++++++++++++++++++++++++++--------- 2 files changed, 155 insertions(+), 47 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 3445da6e8c95..5287771d0647 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -1560,9 +1560,15 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) * * Sync connection if it is about to close to * encorage the standby servers to update the connections timeout + * + * For ONE_PKT let ip_vs_sync_conn() do the filter work. */ - pkts = atomic_add_return(1, &cp->in_pkts); - if (af == AF_INET && (ip_vs_sync_state & IP_VS_STATE_MASTER) && + if (cp->flags & IP_VS_CONN_F_ONE_PACKET) + pkts = sysctl_ip_vs_sync_threshold[0]; + else + pkts = atomic_add_return(1, &cp->in_pkts); + + if ((ip_vs_sync_state & IP_VS_STATE_MASTER) && cp->protocol == IPPROTO_SCTP) { if ((cp->state == IP_VS_SCTP_S_ESTABLISHED && (pkts % sysctl_ip_vs_sync_threshold[1] @@ -1577,8 +1583,7 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) } /* Keep this block last: TCP and others with pp->num_states <= 1 */ - else if (af == AF_INET && - (ip_vs_sync_state & IP_VS_STATE_MASTER) && + else if ((ip_vs_sync_state & IP_VS_STATE_MASTER) && (((cp->protocol != IPPROTO_TCP || cp->state == IP_VS_TCP_S_ESTABLISHED) && (pkts % sysctl_ip_vs_sync_threshold[1] diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index e071508901d1..df5abf0e25af 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -226,7 +226,7 @@ struct ip_vs_sync_thread_data { #define MAX_CONNS_PER_SYNCBUFF 255 /* nr_conns in ip_vs_sync_mesg is 8 bit */ /* Version 0 header */ -struct ip_vs_sync_mesg { +struct ip_vs_sync_mesg_v0 { __u8 nr_conns; __u8 syncid; __u16 size; @@ -235,7 +235,7 @@ struct ip_vs_sync_mesg { }; /* Version 1 header */ -struct ip_vs_sync_mesg_v2 { +struct ip_vs_sync_mesg { __u8 reserved; /* must be zero */ __u8 syncid; __u16 size; @@ -299,6 +299,17 @@ static void ntoh_seq(struct ip_vs_seq *no, struct ip_vs_seq *ho) ho->previous_delta = get_unaligned_be32(&no->previous_delta); } +/* + * Copy of struct ip_vs_seq + * From Aligned host order to unaligned network order + */ +static void hton_seq(struct ip_vs_seq *ho, struct ip_vs_seq *no) +{ + put_unaligned_be32(ho->init_seq, &no->init_seq); + put_unaligned_be32(ho->delta, &no->delta); + put_unaligned_be32(ho->previous_delta, &no->previous_delta); +} + static inline struct ip_vs_sync_buff *sb_dequeue(void) { struct ip_vs_sync_buff *sb; @@ -317,6 +328,9 @@ static inline struct ip_vs_sync_buff *sb_dequeue(void) return sb; } +/* + * Create a new sync buffer for Version 1 proto. + */ static inline struct ip_vs_sync_buff * ip_vs_sync_buff_create(void) { struct ip_vs_sync_buff *sb; @@ -328,11 +342,15 @@ static inline struct ip_vs_sync_buff * ip_vs_sync_buff_create(void) kfree(sb); return NULL; } - sb->mesg->nr_conns = 0; + sb->mesg->reserved = 0; /* old nr_conns i.e. must be zeo now */ + sb->mesg->version = SYNC_PROTO_VER; sb->mesg->syncid = ip_vs_master_syncid; - sb->mesg->size = 4; - sb->head = (unsigned char *)sb->mesg + 4; + sb->mesg->size = sizeof(struct ip_vs_sync_mesg); + sb->mesg->nr_conns = 0; + sb->mesg->spare = 0; + sb->head = (unsigned char *)sb->mesg + sizeof(struct ip_vs_sync_mesg); sb->end = (unsigned char *)sb->mesg + sync_send_mesg_maxlen; + sb->firstuse = jiffies; return sb; } @@ -373,18 +391,60 @@ get_curr_sync_buff(unsigned long time) return sb; } - /* * Add an ip_vs_conn information into the current sync_buff. * Called by ip_vs_in. + * Sending Version 1 messages */ -void ip_vs_sync_conn(const struct ip_vs_conn *cp) +void ip_vs_sync_conn(struct ip_vs_conn *cp) { struct ip_vs_sync_mesg *m; - struct ip_vs_sync_conn_v0 *s; - int len; + union ip_vs_sync_conn *s; + __u8 *p; + unsigned int len, pe_name_len, pad; + + /* Do not sync ONE PACKET */ + if (cp->flags & IP_VS_CONN_F_ONE_PACKET) + goto control; +sloop: + /* Sanity checks */ + pe_name_len = 0; + if (cp->pe_data_len) { + if (!cp->pe_data || !cp->dest) { + IP_VS_ERR_RL("SYNC, connection pe_data invalid\n"); + return; + } + pe_name_len = strnlen(cp->pe->name, IP_VS_PENAME_MAXLEN); + } spin_lock(&curr_sb_lock); + +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) + len = sizeof(struct ip_vs_sync_v6); + else +#endif + len = sizeof(struct ip_vs_sync_v4); + + if (cp->flags & IP_VS_CONN_F_SEQ_MASK) + len += sizeof(struct ip_vs_sync_conn_options) + 2; + + if (cp->pe_data_len) + len += cp->pe_data_len + 2; /* + Param hdr field */ + if (pe_name_len) + len += pe_name_len + 2; + + /* check if there is a space for this one */ + pad = 0; + if (curr_sb) { + pad = (4 - (size_t)curr_sb->head) & 3; + if (curr_sb->head + len + pad > curr_sb->end) { + sb_queue_tail(curr_sb); + curr_sb = NULL; + pad = 0; + } + } + if (!curr_sb) { if (!(curr_sb=ip_vs_sync_buff_create())) { spin_unlock(&curr_sb_lock); @@ -393,41 +453,84 @@ void ip_vs_sync_conn(const struct ip_vs_conn *cp) } } - len = (cp->flags & IP_VS_CONN_F_SEQ_MASK) ? FULL_CONN_SIZE : - SIMPLE_CONN_SIZE; m = curr_sb->mesg; - s = (struct ip_vs_sync_conn_v0 *)curr_sb->head; - - /* copy members */ - s->protocol = cp->protocol; - s->cport = cp->cport; - s->vport = cp->vport; - s->dport = cp->dport; - s->caddr = cp->caddr.ip; - s->vaddr = cp->vaddr.ip; - s->daddr = cp->daddr.ip; - s->flags = htons(cp->flags & ~IP_VS_CONN_F_HASHED); - s->state = htons(cp->state); - if (cp->flags & IP_VS_CONN_F_SEQ_MASK) { - struct ip_vs_sync_conn_options *opt = - (struct ip_vs_sync_conn_options *)&s[1]; - memcpy(opt, &cp->in_seq, sizeof(*opt)); - } - + p = curr_sb->head; + curr_sb->head += pad + len; + m->size += pad + len; + /* Add ev. padding from prev. sync_conn */ + while (pad--) + *(p++) = 0; + + s = (union ip_vs_sync_conn *)p; + + /* Set message type & copy members */ + s->v4.type = (cp->af == AF_INET6 ? STYPE_F_INET6 : 0); + s->v4.ver_size = htons(len & SVER_MASK); /* Version 0 */ + s->v4.flags = htonl(cp->flags & ~IP_VS_CONN_F_HASHED); + s->v4.state = htons(cp->state); + s->v4.protocol = cp->protocol; + s->v4.cport = cp->cport; + s->v4.vport = cp->vport; + s->v4.dport = cp->dport; + s->v4.fwmark = htonl(cp->fwmark); + s->v4.timeout = htonl(cp->timeout / HZ); m->nr_conns++; - m->size += len; - curr_sb->head += len; - /* check if there is a space for next one */ - if (curr_sb->head+FULL_CONN_SIZE > curr_sb->end) { - sb_queue_tail(curr_sb); - curr_sb = NULL; +#ifdef CONFIG_IP_VS_IPV6 + if (cp->af == AF_INET6) { + p += sizeof(struct ip_vs_sync_v6); + ipv6_addr_copy(&s->v6.caddr, &cp->caddr.in6); + ipv6_addr_copy(&s->v6.vaddr, &cp->vaddr.in6); + ipv6_addr_copy(&s->v6.daddr, &cp->daddr.in6); + } else +#endif + { + p += sizeof(struct ip_vs_sync_v4); /* options ptr */ + s->v4.caddr = cp->caddr.ip; + s->v4.vaddr = cp->vaddr.ip; + s->v4.daddr = cp->daddr.ip; + } + if (cp->flags & IP_VS_CONN_F_SEQ_MASK) { + *(p++) = IPVS_OPT_SEQ_DATA; + *(p++) = sizeof(struct ip_vs_sync_conn_options); + hton_seq((struct ip_vs_seq *)p, &cp->in_seq); + p += sizeof(struct ip_vs_seq); + hton_seq((struct ip_vs_seq *)p, &cp->out_seq); + p += sizeof(struct ip_vs_seq); } + /* Handle pe data */ + if (cp->pe_data_len && cp->pe_data) { + *(p++) = IPVS_OPT_PE_DATA; + *(p++) = cp->pe_data_len; + memcpy(p, cp->pe_data, cp->pe_data_len); + p += cp->pe_data_len; + if (pe_name_len) { + /* Add PE_NAME */ + *(p++) = IPVS_OPT_PE_NAME; + *(p++) = pe_name_len; + memcpy(p, cp->pe->name, pe_name_len); + p += pe_name_len; + } + } + spin_unlock(&curr_sb_lock); +control: /* synchronize its controller if it has */ - if (cp->control) - ip_vs_sync_conn(cp->control); + cp = cp->control; + if (!cp) + return; + /* + * Reduce sync rate for templates + * i.e only increment in_pkts for Templates. + */ + if (cp->flags & IP_VS_CONN_F_TEMPLATE) { + int pkts = atomic_add_return(1, &cp->in_pkts); + + if (pkts % sysctl_ip_vs_sync_threshold[1] != 1) + return; + } + goto sloop; } /* @@ -596,7 +699,7 @@ static void ip_vs_proc_conn(struct ip_vs_conn_param *param, unsigned flags, */ static void ip_vs_process_message_v0(const char *buffer, const size_t buflen) { - struct ip_vs_sync_mesg *m = (struct ip_vs_sync_mesg *)buffer; + struct ip_vs_sync_mesg_v0 *m = (struct ip_vs_sync_mesg_v0 *)buffer; struct ip_vs_sync_conn_v0 *s; struct ip_vs_sync_conn_options *opt; struct ip_vs_protocol *pp; @@ -604,7 +707,7 @@ static void ip_vs_process_message_v0(const char *buffer, const size_t buflen) char *p; int i; - p = (char *)buffer + sizeof(struct ip_vs_sync_mesg); + p = (char *)buffer + sizeof(struct ip_vs_sync_mesg_v0); for (i=0; inr_conns; i++) { unsigned flags, state; @@ -848,11 +951,11 @@ out: */ static void ip_vs_process_message(__u8 *buffer, const size_t buflen) { - struct ip_vs_sync_mesg_v2 *m2 = (struct ip_vs_sync_mesg_v2 *)buffer; + struct ip_vs_sync_mesg *m2 = (struct ip_vs_sync_mesg *)buffer; __u8 *p, *msg_end; - unsigned int i, nr_conns; + int i, nr_conns; - if (buflen < sizeof(struct ip_vs_sync_mesg)) { + if (buflen < sizeof(struct ip_vs_sync_mesg_v0)) { IP_VS_DBG(2, "BACKUP, message header too short\n"); return; } @@ -872,7 +975,7 @@ static void ip_vs_process_message(__u8 *buffer, const size_t buflen) if ((m2->version == SYNC_PROTO_VER) && (m2->reserved == 0) && (m2->spare == 0)) { - msg_end = buffer + sizeof(struct ip_vs_sync_mesg_v2); + msg_end = buffer + sizeof(struct ip_vs_sync_mesg); nr_conns = m2->nr_conns; for (i=0; i Date: Fri, 19 Nov 2010 14:25:14 +0100 Subject: IPVS: Backup, adding version 0 sending capabilities This patch adds a sysclt net.ipv4.vs.sync_version that can be used to send sync msg in version 0 or 1 format. sync_version value is logical, Value 1 (default) New version 0 Plain old version Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_ctl.c | 28 ++++++++- net/netfilter/ipvs/ip_vs_sync.c | 134 ++++++++++++++++++++++++++++++++++++++++ 2 files changed, 161 insertions(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index a5bd00279047..d12a13c497ba 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -92,7 +92,7 @@ int sysctl_ip_vs_nat_icmp_send = 0; int sysctl_ip_vs_conntrack; #endif int sysctl_ip_vs_snat_reroute = 1; - +int sysctl_ip_vs_sync_ver = 1; /* Default version of sync proto */ #ifdef CONFIG_IP_VS_DEBUG static int sysctl_ip_vs_debug_level = 0; @@ -1536,6 +1536,25 @@ proc_do_sync_threshold(ctl_table *table, int write, return rc; } +static int +proc_do_sync_mode(ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + int *valp = table->data; + int val = *valp; + int rc; + + rc = proc_dointvec(table, write, buffer, lenp, ppos); + if (write && (*valp != val)) { + if ((*valp < 0) || (*valp > 1)) { + /* Restore the correct value */ + *valp = val; + } else { + ip_vs_sync_switch_mode(val); + } + } + return rc; +} /* * IPVS sysctl table (under the /proc/sys/net/ipv4/vs/) @@ -1602,6 +1621,13 @@ static struct ctl_table vs_vars[] = { .mode = 0644, .proc_handler = &proc_dointvec, }, + { + .procname = "sync_version", + .data = &sysctl_ip_vs_sync_ver, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_do_sync_mode, + }, #if 0 { .procname = "timeout_established", diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index df5abf0e25af..c1c167ab73ee 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -5,6 +5,18 @@ * high-performance and highly available server based on a * cluster of servers. * + * Version 1, is capable of handling both version 0 and 1 messages. + * Version 0 is the plain old format. + * Note Version 0 receivers will just drop Ver 1 messages. + * Version 1 is capable of handle IPv6, Persistence data, + * time-outs, and firewall marks. + * In ver.1 "ip_vs_sync_conn_options" will be sent in netw. order. + * Ver. 0 can be turned on by sysctl -w net.ipv4.vs.sync_version=0 + * + * Definitions Message: is a complete datagram + * Sync_conn: is a part of a Message + * Param Data is an option to a Sync_conn. + * * Authors: Wensong Zhang * * ip_vs_sync: sync connection info from master load balancer to backups @@ -15,6 +27,8 @@ * Alexandre Cassen : Added SyncID support for incoming sync * messages filtering. * Justin Ossevoort : Fix endian problem on sync message size. + * Hans Schillstrom : Added Version 1: i.e. IPv6, + * Persistence support, fwmark and time-out. */ #define KMSG_COMPONENT "IPVS" @@ -391,6 +405,121 @@ get_curr_sync_buff(unsigned long time) return sb; } +/* + * Switch mode from sending version 0 or 1 + * - must handle sync_buf + */ +void ip_vs_sync_switch_mode(int mode) { + + if (!ip_vs_sync_state & IP_VS_STATE_MASTER) + return; + if (mode == sysctl_ip_vs_sync_ver || !curr_sb) + return; + + spin_lock_bh(&curr_sb_lock); + /* Buffer empty ? then let buf_create do the job */ + if ( curr_sb->mesg->size <= sizeof(struct ip_vs_sync_mesg)) { + kfree(curr_sb); + curr_sb = NULL; + } else { + spin_lock_bh(&ip_vs_sync_lock); + if (ip_vs_sync_state & IP_VS_STATE_MASTER) + list_add_tail(&curr_sb->list, &ip_vs_sync_queue); + else + ip_vs_sync_buff_release(curr_sb); + spin_unlock_bh(&ip_vs_sync_lock); + } + spin_unlock_bh(&curr_sb_lock); +} + +/* + * Create a new sync buffer for Version 0 proto. + */ +static inline struct ip_vs_sync_buff * ip_vs_sync_buff_create_v0(void) +{ + struct ip_vs_sync_buff *sb; + struct ip_vs_sync_mesg_v0 *mesg; + + if (!(sb=kmalloc(sizeof(struct ip_vs_sync_buff), GFP_ATOMIC))) + return NULL; + + if (!(sb->mesg=kmalloc(sync_send_mesg_maxlen, GFP_ATOMIC))) { + kfree(sb); + return NULL; + } + mesg = (struct ip_vs_sync_mesg_v0 *)sb->mesg; + mesg->nr_conns = 0; + mesg->syncid = ip_vs_master_syncid; + mesg->size = 4; + sb->head = (unsigned char *)mesg + 4; + sb->end = (unsigned char *)mesg + sync_send_mesg_maxlen; + sb->firstuse = jiffies; + return sb; +} + +/* + * Version 0 , could be switched in by sys_ctl. + * Add an ip_vs_conn information into the current sync_buff. + */ +void ip_vs_sync_conn_v0(struct ip_vs_conn *cp) +{ + struct ip_vs_sync_mesg_v0 *m; + struct ip_vs_sync_conn_v0 *s; + int len; + + if (unlikely(cp->af != AF_INET)) + return; + /* Do not sync ONE PACKET */ + if (cp->flags & IP_VS_CONN_F_ONE_PACKET) + return; + + spin_lock(&curr_sb_lock); + if (!curr_sb) { + if (!(curr_sb=ip_vs_sync_buff_create_v0())) { + spin_unlock(&curr_sb_lock); + pr_err("ip_vs_sync_buff_create failed.\n"); + return; + } + } + + len = (cp->flags & IP_VS_CONN_F_SEQ_MASK) ? FULL_CONN_SIZE : + SIMPLE_CONN_SIZE; + m = (struct ip_vs_sync_mesg_v0 *)curr_sb->mesg; + s = (struct ip_vs_sync_conn_v0 *)curr_sb->head; + + /* copy members */ + s->reserved = 0; + s->protocol = cp->protocol; + s->cport = cp->cport; + s->vport = cp->vport; + s->dport = cp->dport; + s->caddr = cp->caddr.ip; + s->vaddr = cp->vaddr.ip; + s->daddr = cp->daddr.ip; + s->flags = htons(cp->flags & ~IP_VS_CONN_F_HASHED); + s->state = htons(cp->state); + if (cp->flags & IP_VS_CONN_F_SEQ_MASK) { + struct ip_vs_sync_conn_options *opt = + (struct ip_vs_sync_conn_options *)&s[1]; + memcpy(opt, &cp->in_seq, sizeof(*opt)); + } + + m->nr_conns++; + m->size += len; + curr_sb->head += len; + + /* check if there is a space for next one */ + if (curr_sb->head + FULL_CONN_SIZE > curr_sb->end) { + sb_queue_tail(curr_sb); + curr_sb = NULL; + } + spin_unlock(&curr_sb_lock); + + /* synchronize its controller if it has */ + if (cp->control) + ip_vs_sync_conn(cp->control); +} + /* * Add an ip_vs_conn information into the current sync_buff. * Called by ip_vs_in. @@ -403,6 +532,11 @@ void ip_vs_sync_conn(struct ip_vs_conn *cp) __u8 *p; unsigned int len, pe_name_len, pad; + /* Handle old version of the protocol */ + if (sysctl_ip_vs_sync_ver == 0) { + ip_vs_sync_conn_v0(cp); + return; + } /* Do not sync ONE PACKET */ if (cp->flags & IP_VS_CONN_F_ONE_PACKET) goto control; -- cgit v1.2.3 From 61b1ab4583e275af216c8454b9256de680499b19 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:42 +0100 Subject: IPVS: netns, add basic init per netns. Preparation for network name-space init, in this stage some empty functions exists. In most files there is a check if it is root ns i.e. init_net if (!net_eq(net, &init_net)) return ... this will be removed by the last patch, when enabling name-space. *v3 ip_vs_conn.c merge error corrected. net_ipvs #ifdef removed as sugested by Jan Engelhardt [ horms@verge.net.au: Removed whitespace-change-only hunks ] Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_app.c | 28 +++++++++++++++--- net/netfilter/ipvs/ip_vs_conn.c | 34 ++++++++++++++++++---- net/netfilter/ipvs/ip_vs_core.c | 63 ++++++++++++++++++++++++++++++++++++++-- net/netfilter/ipvs/ip_vs_ctl.c | 49 +++++++++++++++++++++++++------ net/netfilter/ipvs/ip_vs_est.c | 20 ++++++++++++- net/netfilter/ipvs/ip_vs_ftp.c | 34 +++++++++++++++++++--- net/netfilter/ipvs/ip_vs_lblc.c | 37 +++++++++++++++++++++-- net/netfilter/ipvs/ip_vs_lblcr.c | 38 +++++++++++++++++++++--- net/netfilter/ipvs/ip_vs_proto.c | 19 ++++++++++++ net/netfilter/ipvs/ip_vs_sync.c | 27 +++++++++++++++++ 10 files changed, 316 insertions(+), 33 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_app.c b/net/netfilter/ipvs/ip_vs_app.c index a475edee0912..40b09ccc4896 100644 --- a/net/netfilter/ipvs/ip_vs_app.c +++ b/net/netfilter/ipvs/ip_vs_app.c @@ -569,15 +569,35 @@ static const struct file_operations ip_vs_app_fops = { }; #endif -int __init ip_vs_app_init(void) +static int __net_init __ip_vs_app_init(struct net *net) { - /* we will replace it with proc_net_ipvs_create() soon */ - proc_net_fops_create(&init_net, "ip_vs_app", 0, &ip_vs_app_fops); + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + + proc_net_fops_create(net, "ip_vs_app", 0, &ip_vs_app_fops); return 0; } +static void __net_exit __ip_vs_app_cleanup(struct net *net) +{ + proc_net_remove(net, "ip_vs_app"); +} + +static struct pernet_operations ip_vs_app_ops = { + .init = __ip_vs_app_init, + .exit = __ip_vs_app_cleanup, +}; + +int __init ip_vs_app_init(void) +{ + int rv; + + rv = register_pernet_subsys(&ip_vs_app_ops); + return rv; +} + void ip_vs_app_cleanup(void) { - proc_net_remove(&init_net, "ip_vs_app"); + unregister_pernet_subsys(&ip_vs_app_ops); } diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 66e4662925d5..7c1b502f8d8d 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -1201,11 +1201,36 @@ static void ip_vs_conn_flush(void) goto flush_again; } } +/* + * per netns init and exit + */ +int __net_init __ip_vs_conn_init(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + proc_net_fops_create(net, "ip_vs_conn", 0, &ip_vs_conn_fops); + proc_net_fops_create(net, "ip_vs_conn_sync", 0, &ip_vs_conn_sync_fops); + return 0; +} + +static void __net_exit __ip_vs_conn_cleanup(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return; + + proc_net_remove(net, "ip_vs_conn"); + proc_net_remove(net, "ip_vs_conn_sync"); +} +static struct pernet_operations ipvs_conn_ops = { + .init = __ip_vs_conn_init, + .exit = __ip_vs_conn_cleanup, +}; int __init ip_vs_conn_init(void) { int idx; + int retc; /* Compute size and mask */ ip_vs_conn_tab_size = 1 << ip_vs_conn_tab_bits; @@ -1243,24 +1268,21 @@ int __init ip_vs_conn_init(void) rwlock_init(&__ip_vs_conntbl_lock_array[idx].l); } - proc_net_fops_create(&init_net, "ip_vs_conn", 0, &ip_vs_conn_fops); - proc_net_fops_create(&init_net, "ip_vs_conn_sync", 0, &ip_vs_conn_sync_fops); + retc = register_pernet_subsys(&ipvs_conn_ops); /* calculate the random value for connection hash */ get_random_bytes(&ip_vs_conn_rnd, sizeof(ip_vs_conn_rnd)); - return 0; + return retc; } - void ip_vs_conn_cleanup(void) { + unregister_pernet_subsys(&ipvs_conn_ops); /* flush all the connection entries first */ ip_vs_conn_flush(); /* Release the empty cache */ kmem_cache_destroy(ip_vs_conn_cachep); - proc_net_remove(&init_net, "ip_vs_conn"); - proc_net_remove(&init_net, "ip_vs_conn_sync"); vfree(ip_vs_conn_tab); } diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 5287771d0647..206f40c548d7 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -41,6 +41,7 @@ #include /* for icmp_send */ #include #include +#include /* net_generic() */ #include #include @@ -68,6 +69,12 @@ EXPORT_SYMBOL(ip_vs_conn_put); EXPORT_SYMBOL(ip_vs_get_debug_level); #endif +int ip_vs_net_id __read_mostly; +#ifdef IP_VS_GENERIC_NETNS +EXPORT_SYMBOL(ip_vs_net_id); +#endif +/* netns cnt used for uniqueness */ +static atomic_t ipvs_netns_cnt = ATOMIC_INIT(0); /* ID used in ICMP lookups */ #define icmp_id(icmph) (((icmph)->un).echo.id) @@ -1813,6 +1820,44 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = { #endif }; +/* + * Initialize IP Virtual Server netns mem. + */ +static int __net_init __ip_vs_init(struct net *net) +{ + struct netns_ipvs *ipvs; + + if (!net_eq(net, &init_net)) { + pr_err("The final patch for enabling netns is missing\n"); + return -EPERM; + } + ipvs = net_generic(net, ip_vs_net_id); + if (ipvs == NULL) { + pr_err("%s(): no memory.\n", __func__); + return -ENOMEM; + } + /* Counters used for creating unique names */ + ipvs->gen = atomic_read(&ipvs_netns_cnt); + atomic_inc(&ipvs_netns_cnt); + net->ipvs = ipvs; + printk(KERN_INFO "IPVS: Creating netns size=%lu id=%d\n", + sizeof(struct netns_ipvs), ipvs->gen); + return 0; +} + +static void __net_exit __ip_vs_cleanup(struct net *net) +{ + struct netns_ipvs *ipvs = net_ipvs(net); + + IP_VS_DBG(10, "ipvs netns %d released\n", ipvs->gen); +} + +static struct pernet_operations ipvs_core_ops = { + .init = __ip_vs_init, + .exit = __ip_vs_cleanup, + .id = &ip_vs_net_id, + .size = sizeof(struct netns_ipvs), +}; /* * Initialize IP Virtual Server @@ -1821,8 +1866,11 @@ static int __init ip_vs_init(void) { int ret; - ip_vs_estimator_init(); + ret = register_pernet_subsys(&ipvs_core_ops); /* Alloc ip_vs struct */ + if (ret < 0) + return ret; + ip_vs_estimator_init(); ret = ip_vs_control_init(); if (ret < 0) { pr_err("can't setup control.\n"); @@ -1843,15 +1891,23 @@ static int __init ip_vs_init(void) goto cleanup_app; } + ret = ip_vs_sync_init(); + if (ret < 0) { + pr_err("can't setup sync data.\n"); + goto cleanup_conn; + } + ret = nf_register_hooks(ip_vs_ops, ARRAY_SIZE(ip_vs_ops)); if (ret < 0) { pr_err("can't register hooks.\n"); - goto cleanup_conn; + goto cleanup_sync; } pr_info("ipvs loaded.\n"); return ret; +cleanup_sync: + ip_vs_sync_cleanup(); cleanup_conn: ip_vs_conn_cleanup(); cleanup_app: @@ -1861,17 +1917,20 @@ static int __init ip_vs_init(void) ip_vs_control_cleanup(); cleanup_estimator: ip_vs_estimator_cleanup(); + unregister_pernet_subsys(&ipvs_core_ops); /* free ip_vs struct */ return ret; } static void __exit ip_vs_cleanup(void) { nf_unregister_hooks(ip_vs_ops, ARRAY_SIZE(ip_vs_ops)); + ip_vs_sync_cleanup(); ip_vs_conn_cleanup(); ip_vs_app_cleanup(); ip_vs_protocol_cleanup(); ip_vs_control_cleanup(); ip_vs_estimator_cleanup(); + unregister_pernet_subsys(&ipvs_core_ops); /* free ip_vs struct */ pr_info("ipvs unloaded.\n"); } diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index ca49e928f302..ceeef4352d34 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -3406,6 +3406,42 @@ static void ip_vs_genl_unregister(void) /* End of Generic Netlink interface definitions */ +/* + * per netns intit/exit func. + */ +int __net_init __ip_vs_control_init(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + + proc_net_fops_create(net, "ip_vs", 0, &ip_vs_info_fops); + proc_net_fops_create(net, "ip_vs_stats", 0, &ip_vs_stats_fops); + sysctl_header = register_net_sysctl_table(net, net_vs_ctl_path, + vs_vars); + if (sysctl_header == NULL) + goto err_reg; + ip_vs_new_estimator(&ip_vs_stats); + return 0; + +err_reg: + return -ENOMEM; +} + +static void __net_exit __ip_vs_control_cleanup(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return; + + ip_vs_kill_estimator(&ip_vs_stats); + unregister_net_sysctl_table(sysctl_header); + proc_net_remove(net, "ip_vs_stats"); + proc_net_remove(net, "ip_vs"); +} + +static struct pernet_operations ipvs_control_ops = { + .init = __ip_vs_control_init, + .exit = __ip_vs_control_cleanup, +}; int __init ip_vs_control_init(void) { @@ -3437,12 +3473,9 @@ int __init ip_vs_control_init(void) return ret; } - proc_net_fops_create(&init_net, "ip_vs", 0, &ip_vs_info_fops); - proc_net_fops_create(&init_net, "ip_vs_stats",0, &ip_vs_stats_fops); - - sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars); - - ip_vs_new_estimator(&ip_vs_stats); + ret = register_pernet_subsys(&ipvs_control_ops); + if (ret) + return ret; /* Hook the defense timer */ schedule_delayed_work(&defense_work, DEFENSE_TIMER_PERIOD); @@ -3459,9 +3492,7 @@ void ip_vs_control_cleanup(void) cancel_delayed_work_sync(&defense_work); cancel_work_sync(&defense_work.work); ip_vs_kill_estimator(&ip_vs_stats); - unregister_sysctl_table(sysctl_header); - proc_net_remove(&init_net, "ip_vs_stats"); - proc_net_remove(&init_net, "ip_vs"); + unregister_pernet_subsys(&ipvs_control_ops); ip_vs_genl_unregister(); nf_unregister_sockopt(&ip_vs_sockopts); LeaveFunction(2); diff --git a/net/netfilter/ipvs/ip_vs_est.c b/net/netfilter/ipvs/ip_vs_est.c index ff28801962e0..7417a0c1408b 100644 --- a/net/netfilter/ipvs/ip_vs_est.c +++ b/net/netfilter/ipvs/ip_vs_est.c @@ -157,13 +157,31 @@ void ip_vs_zero_estimator(struct ip_vs_stats *stats) est->outbps = 0; } +static int __net_init __ip_vs_estimator_init(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + + return 0; +} + +static struct pernet_operations ip_vs_app_ops = { + .init = __ip_vs_estimator_init, +}; + int __init ip_vs_estimator_init(void) { + int rv; + + rv = register_pernet_subsys(&ip_vs_app_ops); + if (rv < 0) + return rv; mod_timer(&est_timer, jiffies + 2 * HZ); - return 0; + return rv; } void ip_vs_estimator_cleanup(void) { del_timer_sync(&est_timer); + unregister_pernet_subsys(&ip_vs_app_ops); } diff --git a/net/netfilter/ipvs/ip_vs_ftp.c b/net/netfilter/ipvs/ip_vs_ftp.c index 84aef65b37d1..0e762f322aa3 100644 --- a/net/netfilter/ipvs/ip_vs_ftp.c +++ b/net/netfilter/ipvs/ip_vs_ftp.c @@ -399,15 +399,17 @@ static struct ip_vs_app ip_vs_ftp = { .pkt_in = ip_vs_ftp_in, }; - /* - * ip_vs_ftp initialization + * per netns ip_vs_ftp initialization */ -static int __init ip_vs_ftp_init(void) +static int __net_init __ip_vs_ftp_init(struct net *net) { int i, ret; struct ip_vs_app *app = &ip_vs_ftp; + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + ret = register_ip_vs_app(app); if (ret) return ret; @@ -427,14 +429,38 @@ static int __init ip_vs_ftp_init(void) return ret; } +/* + * netns exit + */ +static void __ip_vs_ftp_exit(struct net *net) +{ + struct ip_vs_app *app = &ip_vs_ftp; + + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return; + + unregister_ip_vs_app(app); +} + +static struct pernet_operations ip_vs_ftp_ops = { + .init = __ip_vs_ftp_init, + .exit = __ip_vs_ftp_exit, +}; +int __init ip_vs_ftp_init(void) +{ + int rv; + + rv = register_pernet_subsys(&ip_vs_ftp_ops); + return rv; +} /* * ip_vs_ftp finish. */ static void __exit ip_vs_ftp_exit(void) { - unregister_ip_vs_app(&ip_vs_ftp); + unregister_pernet_subsys(&ip_vs_ftp_ops); } diff --git a/net/netfilter/ipvs/ip_vs_lblc.c b/net/netfilter/ipvs/ip_vs_lblc.c index 9323f8944199..84278fb4e055 100644 --- a/net/netfilter/ipvs/ip_vs_lblc.c +++ b/net/netfilter/ipvs/ip_vs_lblc.c @@ -543,23 +543,54 @@ static struct ip_vs_scheduler ip_vs_lblc_scheduler = .schedule = ip_vs_lblc_schedule, }; +/* + * per netns init. + */ +static int __net_init __ip_vs_lblc_init(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + + sysctl_header = register_net_sysctl_table(net, net_vs_ctl_path, + vs_vars_table); + if (!sysctl_header) + return -ENOMEM; + + return 0; +} + +static void __net_exit __ip_vs_lblc_exit(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return; + + unregister_net_sysctl_table(sysctl_header); +} + +static struct pernet_operations ip_vs_lblc_ops = { + .init = __ip_vs_lblc_init, + .exit = __ip_vs_lblc_exit, +}; static int __init ip_vs_lblc_init(void) { int ret; - sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars_table); + ret = register_pernet_subsys(&ip_vs_lblc_ops); + if (ret) + return ret; + ret = register_ip_vs_scheduler(&ip_vs_lblc_scheduler); if (ret) - unregister_sysctl_table(sysctl_header); + unregister_pernet_subsys(&ip_vs_lblc_ops); return ret; } static void __exit ip_vs_lblc_cleanup(void) { - unregister_sysctl_table(sysctl_header); unregister_ip_vs_scheduler(&ip_vs_lblc_scheduler); + unregister_pernet_subsys(&ip_vs_lblc_ops); } diff --git a/net/netfilter/ipvs/ip_vs_lblcr.c b/net/netfilter/ipvs/ip_vs_lblcr.c index dbeed8ea421a..7c7396a6acbf 100644 --- a/net/netfilter/ipvs/ip_vs_lblcr.c +++ b/net/netfilter/ipvs/ip_vs_lblcr.c @@ -744,23 +744,53 @@ static struct ip_vs_scheduler ip_vs_lblcr_scheduler = .schedule = ip_vs_lblcr_schedule, }; +/* + * per netns init. + */ +static int __net_init __ip_vs_lblcr_init(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + + sysctl_header = register_net_sysctl_table(net, net_vs_ctl_path, + vs_vars_table); + if (!sysctl_header) + return -ENOMEM; + + return 0; +} + +static void __net_exit __ip_vs_lblcr_exit(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return; + + unregister_net_sysctl_table(sysctl_header); +} + +static struct pernet_operations ip_vs_lblcr_ops = { + .init = __ip_vs_lblcr_init, + .exit = __ip_vs_lblcr_exit, +}; static int __init ip_vs_lblcr_init(void) { int ret; - sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars_table); + ret = register_pernet_subsys(&ip_vs_lblcr_ops); + if (ret) + return ret; + ret = register_ip_vs_scheduler(&ip_vs_lblcr_scheduler); if (ret) - unregister_sysctl_table(sysctl_header); + unregister_pernet_subsys(&ip_vs_lblcr_ops); return ret; } - static void __exit ip_vs_lblcr_cleanup(void) { - unregister_sysctl_table(sysctl_header); unregister_ip_vs_scheduler(&ip_vs_lblcr_scheduler); + unregister_pernet_subsys(&ip_vs_lblcr_ops); } diff --git a/net/netfilter/ipvs/ip_vs_proto.c b/net/netfilter/ipvs/ip_vs_proto.c index c53998390877..45392942d0e7 100644 --- a/net/netfilter/ipvs/ip_vs_proto.c +++ b/net/netfilter/ipvs/ip_vs_proto.c @@ -236,6 +236,23 @@ ip_vs_tcpudp_debug_packet(int af, struct ip_vs_protocol *pp, ip_vs_tcpudp_debug_packet_v4(pp, skb, offset, msg); } +/* + * per network name-space init + */ +static int __net_init __ip_vs_protocol_init(struct net *net) +{ + return 0; +} + +static void __net_exit __ip_vs_protocol_cleanup(struct net *net) +{ + /* empty */ +} + +static struct pernet_operations ipvs_proto_ops = { + .init = __ip_vs_protocol_init, + .exit = __ip_vs_protocol_cleanup, +}; int __init ip_vs_protocol_init(void) { @@ -265,6 +282,7 @@ int __init ip_vs_protocol_init(void) REGISTER_PROTOCOL(&ip_vs_protocol_esp); #endif pr_info("Registered protocols (%s)\n", &protocols[2]); + return register_pernet_subsys(&ipvs_proto_ops); return 0; } @@ -275,6 +293,7 @@ void ip_vs_protocol_cleanup(void) struct ip_vs_protocol *pp; int i; + unregister_pernet_subsys(&ipvs_proto_ops); /* unregister all the ipvs protocols */ for (i = 0; i < IP_VS_PROTO_TAB_SIZE; i++) { while ((pp = ip_vs_proto_table[i]) != NULL) diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index c1c167ab73ee..3668739a6d06 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -1639,3 +1639,30 @@ int stop_sync_thread(int state) return 0; } + +/* + * Initialize data struct for each netns + */ +static int __net_init __ip_vs_sync_init(struct net *net) +{ + return 0; +} + +static void __ip_vs_sync_cleanup(struct net *net) +{ +} +static struct pernet_operations ipvs_sync_ops = { + .init = __ip_vs_sync_init, + .exit = __ip_vs_sync_cleanup, +}; + + +int __init ip_vs_sync_init(void) +{ + return register_pernet_subsys(&ipvs_sync_ops); +} + +void __exit ip_vs_sync_cleanup(void) +{ + unregister_pernet_subsys(&ipvs_sync_ops); +} -- cgit v1.2.3 From fc723250c9cb046cc19833a2b1c4309bbf59ac36 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:43 +0100 Subject: IPVS: netns to services part 1 Services hash tables got netns ptr a hash arg, While Real Servers (rs) has been moved to ipvs struct. Two new inline functions added to get net ptr from skb. Since ip_vs is called from different contexts there is two places to dig for the net ptr skb->dev or skb->sk this is handled in skb_net() and skb_sknet() Global functions, ip_vs_service_get() ip_vs_lookup_real_service() etc have got struct net *net as first param. If possible get net ptr skb etc, - if not &init_net is used at this early stage of patching. ip_vs_ctl.c procfs not ready for netns yet. *v3 Comments by Julian - __ip_vs_service_find and __ip_vs_svc_fwm_find are fast path, net_eq(svc->net, net) so the check is at the end now. - net = skb_net(skb) in ip_vs_out moved after check for skb_dst. Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_conn.c | 2 +- net/netfilter/ipvs/ip_vs_core.c | 4 +- net/netfilter/ipvs/ip_vs_ctl.c | 232 +++++++++++++++++++--------------- net/netfilter/ipvs/ip_vs_proto_sctp.c | 5 +- net/netfilter/ipvs/ip_vs_proto_tcp.c | 7 +- net/netfilter/ipvs/ip_vs_proto_udp.c | 5 +- net/netfilter/ipvs/ip_vs_sync.c | 2 +- 7 files changed, 147 insertions(+), 110 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 7c1b502f8d8d..7a0e79e3ad0f 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -611,7 +611,7 @@ struct ip_vs_dest *ip_vs_try_bind_dest(struct ip_vs_conn *cp) struct ip_vs_dest *dest; if ((cp) && (!cp->dest)) { - dest = ip_vs_find_dest(cp->af, &cp->daddr, cp->dport, + dest = ip_vs_find_dest(&init_net, cp->af, &cp->daddr, cp->dport, &cp->vaddr, cp->vport, cp->protocol, cp->fwmark); ip_vs_bind_dest(cp, dest); diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 206f40c548d7..d0616ea1eebf 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -1031,6 +1031,7 @@ drop: static unsigned int ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af) { + struct net *net = NULL; struct ip_vs_iphdr iph; struct ip_vs_protocol *pp; struct ip_vs_conn *cp; @@ -1054,6 +1055,7 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af) if (unlikely(!skb_dst(skb))) return NF_ACCEPT; + net = skb_net(skb); ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) { @@ -1119,7 +1121,7 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af) sizeof(_ports), _ports); if (pptr == NULL) return NF_ACCEPT; /* Not for me */ - if (ip_vs_lookup_real_service(af, iph.protocol, + if (ip_vs_lookup_real_service(net, af, iph.protocol, &iph.saddr, pptr[0])) { /* diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index ceeef4352d34..2d7c96bd2114 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -287,15 +287,6 @@ static struct list_head ip_vs_svc_table[IP_VS_SVC_TAB_SIZE]; /* the service table hashed by fwmark */ static struct list_head ip_vs_svc_fwm_table[IP_VS_SVC_TAB_SIZE]; -/* - * Hash table: for real service lookups - */ -#define IP_VS_RTAB_BITS 4 -#define IP_VS_RTAB_SIZE (1 << IP_VS_RTAB_BITS) -#define IP_VS_RTAB_MASK (IP_VS_RTAB_SIZE - 1) - -static struct list_head ip_vs_rtable[IP_VS_RTAB_SIZE]; - /* * Trash for destinations */ @@ -311,9 +302,9 @@ static atomic_t ip_vs_nullsvc_counter = ATOMIC_INIT(0); /* * Returns hash value for virtual service */ -static __inline__ unsigned -ip_vs_svc_hashkey(int af, unsigned proto, const union nf_inet_addr *addr, - __be16 port) +static inline unsigned +ip_vs_svc_hashkey(struct net *net, int af, unsigned proto, + const union nf_inet_addr *addr, __be16 port) { register unsigned porth = ntohs(port); __be32 addr_fold = addr->ip; @@ -323,6 +314,7 @@ ip_vs_svc_hashkey(int af, unsigned proto, const union nf_inet_addr *addr, addr_fold = addr->ip6[0]^addr->ip6[1]^ addr->ip6[2]^addr->ip6[3]; #endif + addr_fold ^= ((size_t)net>>8); return (proto^ntohl(addr_fold)^(porth>>IP_VS_SVC_TAB_BITS)^porth) & IP_VS_SVC_TAB_MASK; @@ -331,13 +323,13 @@ ip_vs_svc_hashkey(int af, unsigned proto, const union nf_inet_addr *addr, /* * Returns hash value of fwmark for virtual service lookup */ -static __inline__ unsigned ip_vs_svc_fwm_hashkey(__u32 fwmark) +static inline unsigned ip_vs_svc_fwm_hashkey(struct net *net, __u32 fwmark) { - return fwmark & IP_VS_SVC_TAB_MASK; + return (((size_t)net>>8) ^ fwmark) & IP_VS_SVC_TAB_MASK; } /* - * Hashes a service in the ip_vs_svc_table by + * Hashes a service in the ip_vs_svc_table by * or in the ip_vs_svc_fwm_table by fwmark. * Should be called with locked tables. */ @@ -353,16 +345,16 @@ static int ip_vs_svc_hash(struct ip_vs_service *svc) if (svc->fwmark == 0) { /* - * Hash it by in ip_vs_svc_table + * Hash it by in ip_vs_svc_table */ - hash = ip_vs_svc_hashkey(svc->af, svc->protocol, &svc->addr, - svc->port); + hash = ip_vs_svc_hashkey(svc->net, svc->af, svc->protocol, + &svc->addr, svc->port); list_add(&svc->s_list, &ip_vs_svc_table[hash]); } else { /* - * Hash it by fwmark in ip_vs_svc_fwm_table + * Hash it by fwmark in svc_fwm_table */ - hash = ip_vs_svc_fwm_hashkey(svc->fwmark); + hash = ip_vs_svc_fwm_hashkey(svc->net, svc->fwmark); list_add(&svc->f_list, &ip_vs_svc_fwm_table[hash]); } @@ -374,7 +366,7 @@ static int ip_vs_svc_hash(struct ip_vs_service *svc) /* - * Unhashes a service from ip_vs_svc_table/ip_vs_svc_fwm_table. + * Unhashes a service from svc_table / svc_fwm_table. * Should be called with locked tables. */ static int ip_vs_svc_unhash(struct ip_vs_service *svc) @@ -386,10 +378,10 @@ static int ip_vs_svc_unhash(struct ip_vs_service *svc) } if (svc->fwmark == 0) { - /* Remove it from the ip_vs_svc_table table */ + /* Remove it from the svc_table table */ list_del(&svc->s_list); } else { - /* Remove it from the ip_vs_svc_fwm_table table */ + /* Remove it from the svc_fwm_table table */ list_del(&svc->f_list); } @@ -400,23 +392,24 @@ static int ip_vs_svc_unhash(struct ip_vs_service *svc) /* - * Get service by {proto,addr,port} in the service table. + * Get service by {netns, proto,addr,port} in the service table. */ static inline struct ip_vs_service * -__ip_vs_service_find(int af, __u16 protocol, const union nf_inet_addr *vaddr, - __be16 vport) +__ip_vs_service_find(struct net *net, int af, __u16 protocol, + const union nf_inet_addr *vaddr, __be16 vport) { unsigned hash; struct ip_vs_service *svc; /* Check for "full" addressed entries */ - hash = ip_vs_svc_hashkey(af, protocol, vaddr, vport); + hash = ip_vs_svc_hashkey(net, af, protocol, vaddr, vport); list_for_each_entry(svc, &ip_vs_svc_table[hash], s_list){ if ((svc->af == af) && ip_vs_addr_equal(af, &svc->addr, vaddr) && (svc->port == vport) - && (svc->protocol == protocol)) { + && (svc->protocol == protocol) + && net_eq(svc->net, net)) { /* HIT */ return svc; } @@ -430,16 +423,17 @@ __ip_vs_service_find(int af, __u16 protocol, const union nf_inet_addr *vaddr, * Get service by {fwmark} in the service table. */ static inline struct ip_vs_service * -__ip_vs_svc_fwm_find(int af, __u32 fwmark) +__ip_vs_svc_fwm_find(struct net *net, int af, __u32 fwmark) { unsigned hash; struct ip_vs_service *svc; /* Check for fwmark addressed entries */ - hash = ip_vs_svc_fwm_hashkey(fwmark); + hash = ip_vs_svc_fwm_hashkey(net, fwmark); list_for_each_entry(svc, &ip_vs_svc_fwm_table[hash], f_list) { - if (svc->fwmark == fwmark && svc->af == af) { + if (svc->fwmark == fwmark && svc->af == af + && net_eq(svc->net, net)) { /* HIT */ return svc; } @@ -449,7 +443,7 @@ __ip_vs_svc_fwm_find(int af, __u32 fwmark) } struct ip_vs_service * -ip_vs_service_get(int af, __u32 fwmark, __u16 protocol, +ip_vs_service_get(struct net *net, int af, __u32 fwmark, __u16 protocol, const union nf_inet_addr *vaddr, __be16 vport) { struct ip_vs_service *svc; @@ -459,14 +453,15 @@ ip_vs_service_get(int af, __u32 fwmark, __u16 protocol, /* * Check the table hashed by fwmark first */ - if (fwmark && (svc = __ip_vs_svc_fwm_find(af, fwmark))) + svc = __ip_vs_svc_fwm_find(net, af, fwmark); + if (fwmark && svc) goto out; /* * Check the table hashed by * for "full" addressed entries */ - svc = __ip_vs_service_find(af, protocol, vaddr, vport); + svc = __ip_vs_service_find(net, af, protocol, vaddr, vport); if (svc == NULL && protocol == IPPROTO_TCP @@ -476,7 +471,7 @@ ip_vs_service_get(int af, __u32 fwmark, __u16 protocol, * Check if ftp service entry exists, the packet * might belong to FTP data connections. */ - svc = __ip_vs_service_find(af, protocol, vaddr, FTPPORT); + svc = __ip_vs_service_find(net, af, protocol, vaddr, FTPPORT); } if (svc == NULL @@ -484,7 +479,7 @@ ip_vs_service_get(int af, __u32 fwmark, __u16 protocol, /* * Check if the catch-all port (port zero) exists */ - svc = __ip_vs_service_find(af, protocol, vaddr, 0); + svc = __ip_vs_service_find(net, af, protocol, vaddr, 0); } out: @@ -545,10 +540,10 @@ static inline unsigned ip_vs_rs_hashkey(int af, } /* - * Hashes ip_vs_dest in ip_vs_rtable by . + * Hashes ip_vs_dest in rs_table by . * should be called with locked tables. */ -static int ip_vs_rs_hash(struct ip_vs_dest *dest) +static int ip_vs_rs_hash(struct netns_ipvs *ipvs, struct ip_vs_dest *dest) { unsigned hash; @@ -562,19 +557,19 @@ static int ip_vs_rs_hash(struct ip_vs_dest *dest) */ hash = ip_vs_rs_hashkey(dest->af, &dest->addr, dest->port); - list_add(&dest->d_list, &ip_vs_rtable[hash]); + list_add(&dest->d_list, &ipvs->rs_table[hash]); return 1; } /* - * UNhashes ip_vs_dest from ip_vs_rtable. + * UNhashes ip_vs_dest from rs_table. * should be called with locked tables. */ static int ip_vs_rs_unhash(struct ip_vs_dest *dest) { /* - * Remove it from the ip_vs_rtable table. + * Remove it from the rs_table table. */ if (!list_empty(&dest->d_list)) { list_del(&dest->d_list); @@ -588,10 +583,11 @@ static int ip_vs_rs_unhash(struct ip_vs_dest *dest) * Lookup real service by in the real service table. */ struct ip_vs_dest * -ip_vs_lookup_real_service(int af, __u16 protocol, +ip_vs_lookup_real_service(struct net *net, int af, __u16 protocol, const union nf_inet_addr *daddr, __be16 dport) { + struct netns_ipvs *ipvs = net_ipvs(net); unsigned hash; struct ip_vs_dest *dest; @@ -602,7 +598,7 @@ ip_vs_lookup_real_service(int af, __u16 protocol, hash = ip_vs_rs_hashkey(af, daddr, dport); read_lock(&__ip_vs_rs_lock); - list_for_each_entry(dest, &ip_vs_rtable[hash], d_list) { + list_for_each_entry(dest, &ipvs->rs_table[hash], d_list) { if ((dest->af == af) && ip_vs_addr_equal(af, &dest->addr, daddr) && (dest->port == dport) @@ -652,7 +648,8 @@ ip_vs_lookup_dest(struct ip_vs_service *svc, const union nf_inet_addr *daddr, * ip_vs_lookup_real_service() looked promissing, but * seems not working as expected. */ -struct ip_vs_dest *ip_vs_find_dest(int af, const union nf_inet_addr *daddr, +struct ip_vs_dest *ip_vs_find_dest(struct net *net, int af, + const union nf_inet_addr *daddr, __be16 dport, const union nf_inet_addr *vaddr, __be16 vport, __u16 protocol, __u32 fwmark) @@ -660,7 +657,7 @@ struct ip_vs_dest *ip_vs_find_dest(int af, const union nf_inet_addr *daddr, struct ip_vs_dest *dest; struct ip_vs_service *svc; - svc = ip_vs_service_get(af, fwmark, protocol, vaddr, vport); + svc = ip_vs_service_get(net, af, fwmark, protocol, vaddr, vport); if (!svc) return NULL; dest = ip_vs_lookup_dest(svc, daddr, dport); @@ -768,6 +765,7 @@ static void __ip_vs_update_dest(struct ip_vs_service *svc, struct ip_vs_dest *dest, struct ip_vs_dest_user_kern *udest, int add) { + struct netns_ipvs *ipvs = net_ipvs(svc->net); int conn_flags; /* set the weight and the flags */ @@ -780,11 +778,11 @@ __ip_vs_update_dest(struct ip_vs_service *svc, struct ip_vs_dest *dest, conn_flags |= IP_VS_CONN_F_NOOUTPUT; } else { /* - * Put the real service in ip_vs_rtable if not present. + * Put the real service in rs_table if not present. * For now only for NAT! */ write_lock_bh(&__ip_vs_rs_lock); - ip_vs_rs_hash(dest); + ip_vs_rs_hash(ipvs, dest); write_unlock_bh(&__ip_vs_rs_lock); } atomic_set(&dest->conn_flags, conn_flags); @@ -1117,7 +1115,7 @@ ip_vs_del_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) * Add a service into the service hash table */ static int -ip_vs_add_service(struct ip_vs_service_user_kern *u, +ip_vs_add_service(struct net *net, struct ip_vs_service_user_kern *u, struct ip_vs_service **svc_p) { int ret = 0; @@ -1172,6 +1170,7 @@ ip_vs_add_service(struct ip_vs_service_user_kern *u, svc->flags = u->flags; svc->timeout = u->timeout * HZ; svc->netmask = u->netmask; + svc->net = net; INIT_LIST_HEAD(&svc->destinations); rwlock_init(&svc->sched_lock); @@ -1428,17 +1427,19 @@ static int ip_vs_del_service(struct ip_vs_service *svc) /* * Flush all the virtual services */ -static int ip_vs_flush(void) +static int ip_vs_flush(struct net *net) { int idx; struct ip_vs_service *svc, *nxt; /* - * Flush the service table hashed by + * Flush the service table hashed by */ for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { - list_for_each_entry_safe(svc, nxt, &ip_vs_svc_table[idx], s_list) { - ip_vs_unlink_service(svc); + list_for_each_entry_safe(svc, nxt, &ip_vs_svc_table[idx], + s_list) { + if (net_eq(svc->net, net)) + ip_vs_unlink_service(svc); } } @@ -1448,7 +1449,8 @@ static int ip_vs_flush(void) for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { list_for_each_entry_safe(svc, nxt, &ip_vs_svc_fwm_table[idx], f_list) { - ip_vs_unlink_service(svc); + if (net_eq(svc->net, net)) + ip_vs_unlink_service(svc); } } @@ -1472,20 +1474,22 @@ static int ip_vs_zero_service(struct ip_vs_service *svc) return 0; } -static int ip_vs_zero_all(void) +static int ip_vs_zero_all(struct net *net) { int idx; struct ip_vs_service *svc; for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { list_for_each_entry(svc, &ip_vs_svc_table[idx], s_list) { - ip_vs_zero_service(svc); + if (net_eq(svc->net, net)) + ip_vs_zero_service(svc); } } for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { list_for_each_entry(svc, &ip_vs_svc_fwm_table[idx], f_list) { - ip_vs_zero_service(svc); + if (net_eq(svc->net, net)) + ip_vs_zero_service(svc); } } @@ -1763,6 +1767,7 @@ static struct ctl_table_header * sysctl_header; #ifdef CONFIG_PROC_FS struct ip_vs_iter { + struct seq_net_private p; /* Do not move this, netns depends upon it*/ struct list_head *table; int bucket; }; @@ -1789,6 +1794,7 @@ static inline const char *ip_vs_fwd_name(unsigned flags) /* Get the Nth entry in the two lists */ static struct ip_vs_service *ip_vs_info_array(struct seq_file *seq, loff_t pos) { + struct net *net = seq_file_net(seq); struct ip_vs_iter *iter = seq->private; int idx; struct ip_vs_service *svc; @@ -1796,7 +1802,7 @@ static struct ip_vs_service *ip_vs_info_array(struct seq_file *seq, loff_t pos) /* look in hash by protocol */ for (idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { list_for_each_entry(svc, &ip_vs_svc_table[idx], s_list) { - if (pos-- == 0){ + if (net_eq(svc->net, net) && pos-- == 0) { iter->table = ip_vs_svc_table; iter->bucket = idx; return svc; @@ -1807,7 +1813,7 @@ static struct ip_vs_service *ip_vs_info_array(struct seq_file *seq, loff_t pos) /* keep looking in fwmark */ for (idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { list_for_each_entry(svc, &ip_vs_svc_fwm_table[idx], f_list) { - if (pos-- == 0) { + if (net_eq(svc->net, net) && pos-- == 0) { iter->table = ip_vs_svc_fwm_table; iter->bucket = idx; return svc; @@ -1961,7 +1967,7 @@ static const struct seq_operations ip_vs_info_seq_ops = { static int ip_vs_info_open(struct inode *inode, struct file *file) { - return seq_open_private(file, &ip_vs_info_seq_ops, + return seq_open_net(inode, file, &ip_vs_info_seq_ops, sizeof(struct ip_vs_iter)); } @@ -2011,7 +2017,7 @@ static int ip_vs_stats_show(struct seq_file *seq, void *v) static int ip_vs_stats_seq_open(struct inode *inode, struct file *file) { - return single_open(file, ip_vs_stats_show, NULL); + return single_open_net(inode, file, ip_vs_stats_show); } static const struct file_operations ip_vs_stats_fops = { @@ -2113,6 +2119,7 @@ static void ip_vs_copy_udest_compat(struct ip_vs_dest_user_kern *udest, static int do_ip_vs_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) { + struct net *net = sock_net(sk); int ret; unsigned char arg[MAX_ARG_LEN]; struct ip_vs_service_user *usvc_compat; @@ -2147,7 +2154,7 @@ do_ip_vs_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) if (cmd == IP_VS_SO_SET_FLUSH) { /* Flush the virtual service */ - ret = ip_vs_flush(); + ret = ip_vs_flush(net); goto out_unlock; } else if (cmd == IP_VS_SO_SET_TIMEOUT) { /* Set timeout values for (tcp tcpfin udp) */ @@ -2174,7 +2181,7 @@ do_ip_vs_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) if (cmd == IP_VS_SO_SET_ZERO) { /* if no service address is set, zero counters in all */ if (!usvc.fwmark && !usvc.addr.ip && !usvc.port) { - ret = ip_vs_zero_all(); + ret = ip_vs_zero_all(net); goto out_unlock; } } @@ -2191,10 +2198,10 @@ do_ip_vs_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) /* Lookup the exact service by or fwmark */ if (usvc.fwmark == 0) - svc = __ip_vs_service_find(usvc.af, usvc.protocol, + svc = __ip_vs_service_find(net, usvc.af, usvc.protocol, &usvc.addr, usvc.port); else - svc = __ip_vs_svc_fwm_find(usvc.af, usvc.fwmark); + svc = __ip_vs_svc_fwm_find(net, usvc.af, usvc.fwmark); if (cmd != IP_VS_SO_SET_ADD && (svc == NULL || svc->protocol != usvc.protocol)) { @@ -2207,7 +2214,7 @@ do_ip_vs_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) if (svc != NULL) ret = -EEXIST; else - ret = ip_vs_add_service(&usvc, &svc); + ret = ip_vs_add_service(net, &usvc, &svc); break; case IP_VS_SO_SET_EDIT: ret = ip_vs_edit_service(svc, &usvc); @@ -2267,7 +2274,8 @@ ip_vs_copy_service(struct ip_vs_service_entry *dst, struct ip_vs_service *src) } static inline int -__ip_vs_get_service_entries(const struct ip_vs_get_services *get, +__ip_vs_get_service_entries(struct net *net, + const struct ip_vs_get_services *get, struct ip_vs_get_services __user *uptr) { int idx, count=0; @@ -2278,7 +2286,7 @@ __ip_vs_get_service_entries(const struct ip_vs_get_services *get, for (idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { list_for_each_entry(svc, &ip_vs_svc_table[idx], s_list) { /* Only expose IPv4 entries to old interface */ - if (svc->af != AF_INET) + if (svc->af != AF_INET || !net_eq(svc->net, net)) continue; if (count >= get->num_services) @@ -2297,7 +2305,7 @@ __ip_vs_get_service_entries(const struct ip_vs_get_services *get, for (idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { list_for_each_entry(svc, &ip_vs_svc_fwm_table[idx], f_list) { /* Only expose IPv4 entries to old interface */ - if (svc->af != AF_INET) + if (svc->af != AF_INET || !net_eq(svc->net, net)) continue; if (count >= get->num_services) @@ -2317,7 +2325,7 @@ __ip_vs_get_service_entries(const struct ip_vs_get_services *get, } static inline int -__ip_vs_get_dest_entries(const struct ip_vs_get_dests *get, +__ip_vs_get_dest_entries(struct net *net, const struct ip_vs_get_dests *get, struct ip_vs_get_dests __user *uptr) { struct ip_vs_service *svc; @@ -2325,9 +2333,9 @@ __ip_vs_get_dest_entries(const struct ip_vs_get_dests *get, int ret = 0; if (get->fwmark) - svc = __ip_vs_svc_fwm_find(AF_INET, get->fwmark); + svc = __ip_vs_svc_fwm_find(net, AF_INET, get->fwmark); else - svc = __ip_vs_service_find(AF_INET, get->protocol, &addr, + svc = __ip_vs_service_find(net, AF_INET, get->protocol, &addr, get->port); if (svc) { @@ -2401,7 +2409,9 @@ do_ip_vs_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) unsigned char arg[128]; int ret = 0; unsigned int copylen; + struct net *net = sock_net(sk); + BUG_ON(!net); if (!capable(CAP_NET_ADMIN)) return -EPERM; @@ -2463,7 +2473,7 @@ do_ip_vs_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) ret = -EINVAL; goto out; } - ret = __ip_vs_get_service_entries(get, user); + ret = __ip_vs_get_service_entries(net, get, user); } break; @@ -2476,10 +2486,11 @@ do_ip_vs_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) entry = (struct ip_vs_service_entry *)arg; addr.ip = entry->addr; if (entry->fwmark) - svc = __ip_vs_svc_fwm_find(AF_INET, entry->fwmark); + svc = __ip_vs_svc_fwm_find(net, AF_INET, entry->fwmark); else - svc = __ip_vs_service_find(AF_INET, entry->protocol, - &addr, entry->port); + svc = __ip_vs_service_find(net, AF_INET, + entry->protocol, &addr, + entry->port); if (svc) { ip_vs_copy_service(entry, svc); if (copy_to_user(user, entry, sizeof(*entry)) != 0) @@ -2502,7 +2513,7 @@ do_ip_vs_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) ret = -EINVAL; goto out; } - ret = __ip_vs_get_dest_entries(get, user); + ret = __ip_vs_get_dest_entries(net, get, user); } break; @@ -2722,11 +2733,12 @@ static int ip_vs_genl_dump_services(struct sk_buff *skb, int idx = 0, i; int start = cb->args[0]; struct ip_vs_service *svc; + struct net *net = skb_sknet(skb); mutex_lock(&__ip_vs_mutex); for (i = 0; i < IP_VS_SVC_TAB_SIZE; i++) { list_for_each_entry(svc, &ip_vs_svc_table[i], s_list) { - if (++idx <= start) + if (++idx <= start || !net_eq(svc->net, net)) continue; if (ip_vs_genl_dump_service(skb, svc, cb) < 0) { idx--; @@ -2737,7 +2749,7 @@ static int ip_vs_genl_dump_services(struct sk_buff *skb, for (i = 0; i < IP_VS_SVC_TAB_SIZE; i++) { list_for_each_entry(svc, &ip_vs_svc_fwm_table[i], f_list) { - if (++idx <= start) + if (++idx <= start || !net_eq(svc->net, net)) continue; if (ip_vs_genl_dump_service(skb, svc, cb) < 0) { idx--; @@ -2753,7 +2765,8 @@ nla_put_failure: return skb->len; } -static int ip_vs_genl_parse_service(struct ip_vs_service_user_kern *usvc, +static int ip_vs_genl_parse_service(struct net *net, + struct ip_vs_service_user_kern *usvc, struct nlattr *nla, int full_entry, struct ip_vs_service **ret_svc) { @@ -2796,9 +2809,9 @@ static int ip_vs_genl_parse_service(struct ip_vs_service_user_kern *usvc, } if (usvc->fwmark) - svc = __ip_vs_svc_fwm_find(usvc->af, usvc->fwmark); + svc = __ip_vs_svc_fwm_find(net, usvc->af, usvc->fwmark); else - svc = __ip_vs_service_find(usvc->af, usvc->protocol, + svc = __ip_vs_service_find(net, usvc->af, usvc->protocol, &usvc->addr, usvc->port); *ret_svc = svc; @@ -2835,13 +2848,14 @@ static int ip_vs_genl_parse_service(struct ip_vs_service_user_kern *usvc, return 0; } -static struct ip_vs_service *ip_vs_genl_find_service(struct nlattr *nla) +static struct ip_vs_service *ip_vs_genl_find_service(struct net *net, + struct nlattr *nla) { struct ip_vs_service_user_kern usvc; struct ip_vs_service *svc; int ret; - ret = ip_vs_genl_parse_service(&usvc, nla, 0, &svc); + ret = ip_vs_genl_parse_service(net, &usvc, nla, 0, &svc); return ret ? ERR_PTR(ret) : svc; } @@ -2909,6 +2923,7 @@ static int ip_vs_genl_dump_dests(struct sk_buff *skb, struct ip_vs_service *svc; struct ip_vs_dest *dest; struct nlattr *attrs[IPVS_CMD_ATTR_MAX + 1]; + struct net *net; mutex_lock(&__ip_vs_mutex); @@ -2917,7 +2932,8 @@ static int ip_vs_genl_dump_dests(struct sk_buff *skb, IPVS_CMD_ATTR_MAX, ip_vs_cmd_policy)) goto out_err; - svc = ip_vs_genl_find_service(attrs[IPVS_CMD_ATTR_SERVICE]); + net = skb_sknet(skb); + svc = ip_vs_genl_find_service(net, attrs[IPVS_CMD_ATTR_SERVICE]); if (IS_ERR(svc) || svc == NULL) goto out_err; @@ -3102,13 +3118,15 @@ static int ip_vs_genl_set_cmd(struct sk_buff *skb, struct genl_info *info) struct ip_vs_dest_user_kern udest; int ret = 0, cmd; int need_full_svc = 0, need_full_dest = 0; + struct net *net; + net = skb_sknet(skb); cmd = info->genlhdr->cmd; mutex_lock(&__ip_vs_mutex); if (cmd == IPVS_CMD_FLUSH) { - ret = ip_vs_flush(); + ret = ip_vs_flush(net); goto out; } else if (cmd == IPVS_CMD_SET_CONFIG) { ret = ip_vs_genl_set_config(info->attrs); @@ -3133,7 +3151,7 @@ static int ip_vs_genl_set_cmd(struct sk_buff *skb, struct genl_info *info) goto out; } else if (cmd == IPVS_CMD_ZERO && !info->attrs[IPVS_CMD_ATTR_SERVICE]) { - ret = ip_vs_zero_all(); + ret = ip_vs_zero_all(net); goto out; } @@ -3143,7 +3161,7 @@ static int ip_vs_genl_set_cmd(struct sk_buff *skb, struct genl_info *info) if (cmd == IPVS_CMD_NEW_SERVICE || cmd == IPVS_CMD_SET_SERVICE) need_full_svc = 1; - ret = ip_vs_genl_parse_service(&usvc, + ret = ip_vs_genl_parse_service(net, &usvc, info->attrs[IPVS_CMD_ATTR_SERVICE], need_full_svc, &svc); if (ret) @@ -3173,7 +3191,7 @@ static int ip_vs_genl_set_cmd(struct sk_buff *skb, struct genl_info *info) switch (cmd) { case IPVS_CMD_NEW_SERVICE: if (svc == NULL) - ret = ip_vs_add_service(&usvc, &svc); + ret = ip_vs_add_service(net, &usvc, &svc); else ret = -EEXIST; break; @@ -3211,7 +3229,9 @@ static int ip_vs_genl_get_cmd(struct sk_buff *skb, struct genl_info *info) struct sk_buff *msg; void *reply; int ret, cmd, reply_cmd; + struct net *net; + net = skb_sknet(skb); cmd = info->genlhdr->cmd; if (cmd == IPVS_CMD_GET_SERVICE) @@ -3240,7 +3260,8 @@ static int ip_vs_genl_get_cmd(struct sk_buff *skb, struct genl_info *info) { struct ip_vs_service *svc; - svc = ip_vs_genl_find_service(info->attrs[IPVS_CMD_ATTR_SERVICE]); + svc = ip_vs_genl_find_service(net, + info->attrs[IPVS_CMD_ATTR_SERVICE]); if (IS_ERR(svc)) { ret = PTR_ERR(svc); goto out_err; @@ -3411,9 +3432,15 @@ static void ip_vs_genl_unregister(void) */ int __net_init __ip_vs_control_init(struct net *net) { + int idx; + struct netns_ipvs *ipvs = net_ipvs(net); + if (!net_eq(net, &init_net)) /* netns not enabled yet */ return -EPERM; + for (idx = 0; idx < IP_VS_RTAB_SIZE; idx++) + INIT_LIST_HEAD(&ipvs->rs_table[idx]); + proc_net_fops_create(net, "ip_vs", 0, &ip_vs_info_fops); proc_net_fops_create(net, "ip_vs_stats", 0, &ip_vs_stats_fops); sysctl_header = register_net_sysctl_table(net, net_vs_ctl_path, @@ -3445,43 +3472,48 @@ static struct pernet_operations ipvs_control_ops = { int __init ip_vs_control_init(void) { - int ret; int idx; + int ret; EnterFunction(2); - /* Initialize ip_vs_svc_table, ip_vs_svc_fwm_table, ip_vs_rtable */ + /* Initialize svc_table, ip_vs_svc_fwm_table, rs_table */ for(idx = 0; idx < IP_VS_SVC_TAB_SIZE; idx++) { INIT_LIST_HEAD(&ip_vs_svc_table[idx]); INIT_LIST_HEAD(&ip_vs_svc_fwm_table[idx]); } - for(idx = 0; idx < IP_VS_RTAB_SIZE; idx++) { - INIT_LIST_HEAD(&ip_vs_rtable[idx]); + + ret = register_pernet_subsys(&ipvs_control_ops); + if (ret) { + pr_err("cannot register namespace.\n"); + goto err; } - smp_wmb(); + + smp_wmb(); /* Do we really need it now ? */ ret = nf_register_sockopt(&ip_vs_sockopts); if (ret) { pr_err("cannot register sockopt.\n"); - return ret; + goto err_net; } ret = ip_vs_genl_register(); if (ret) { pr_err("cannot register Generic Netlink interface.\n"); nf_unregister_sockopt(&ip_vs_sockopts); - return ret; + goto err_net; } - ret = register_pernet_subsys(&ipvs_control_ops); - if (ret) - return ret; - /* Hook the defense timer */ schedule_delayed_work(&defense_work, DEFENSE_TIMER_PERIOD); LeaveFunction(2); return 0; + +err_net: + unregister_pernet_subsys(&ipvs_control_ops); +err: + return ret; } diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c index a315159983ad..521b827083fe 100644 --- a/net/netfilter/ipvs/ip_vs_proto_sctp.c +++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c @@ -12,6 +12,7 @@ static int sctp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, int *verdict, struct ip_vs_conn **cpp) { + struct net *net; struct ip_vs_service *svc; sctp_chunkhdr_t _schunkh, *sch; sctp_sctphdr_t *sh, _sctph; @@ -27,9 +28,9 @@ sctp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, sizeof(_schunkh), &_schunkh); if (sch == NULL) return 0; - + net = skb_net(skb); if ((sch->type == SCTP_CID_INIT) && - (svc = ip_vs_service_get(af, skb->mark, iph.protocol, + (svc = ip_vs_service_get(net, af, skb->mark, iph.protocol, &iph.daddr, sh->dest))) { int ignored; diff --git a/net/netfilter/ipvs/ip_vs_proto_tcp.c b/net/netfilter/ipvs/ip_vs_proto_tcp.c index 1cdab12abfef..c175d3166263 100644 --- a/net/netfilter/ipvs/ip_vs_proto_tcp.c +++ b/net/netfilter/ipvs/ip_vs_proto_tcp.c @@ -31,6 +31,7 @@ static int tcp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, int *verdict, struct ip_vs_conn **cpp) { + struct net *net; struct ip_vs_service *svc; struct tcphdr _tcph, *th; struct ip_vs_iphdr iph; @@ -42,11 +43,11 @@ tcp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, *verdict = NF_DROP; return 0; } - + net = skb_net(skb); /* No !th->ack check to allow scheduling on SYN+ACK for Active FTP */ if (th->syn && - (svc = ip_vs_service_get(af, skb->mark, iph.protocol, &iph.daddr, - th->dest))) { + (svc = ip_vs_service_get(net, af, skb->mark, iph.protocol, + &iph.daddr, th->dest))) { int ignored; if (ip_vs_todrop()) { diff --git a/net/netfilter/ipvs/ip_vs_proto_udp.c b/net/netfilter/ipvs/ip_vs_proto_udp.c index cd398de010cc..5ab54f648654 100644 --- a/net/netfilter/ipvs/ip_vs_proto_udp.c +++ b/net/netfilter/ipvs/ip_vs_proto_udp.c @@ -31,6 +31,7 @@ static int udp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, int *verdict, struct ip_vs_conn **cpp) { + struct net *net; struct ip_vs_service *svc; struct udphdr _udph, *uh; struct ip_vs_iphdr iph; @@ -42,8 +43,8 @@ udp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, *verdict = NF_DROP; return 0; } - - svc = ip_vs_service_get(af, skb->mark, iph.protocol, + net = skb_net(skb); + svc = ip_vs_service_get(net, af, skb->mark, iph.protocol, &iph.daddr, uh->dest); if (svc) { int ignored; diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index 3668739a6d06..662aa2c22a05 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -749,7 +749,7 @@ static void ip_vs_proc_conn(struct ip_vs_conn_param *param, unsigned flags, * If it is not found the connection will remain unbound * but still handled. */ - dest = ip_vs_find_dest(type, daddr, dport, param->vaddr, + dest = ip_vs_find_dest(&init_net, type, daddr, dport, param->vaddr, param->vport, protocol, fwmark); /* Set the approprite ativity flag */ -- cgit v1.2.3 From d0a1eef9c38218af20c809b2220a960b7ed81a36 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:44 +0100 Subject: IPVS: netns awarness to lblcr sheduler var sysctl_ip_vs_lblcr_expiration moved to ipvs struct as sysctl_lblcr_expiration procfs updated to handle this. Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_lblcr.c | 54 ++++++++++++++++++++++++++-------------- 1 file changed, 36 insertions(+), 18 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_lblcr.c b/net/netfilter/ipvs/ip_vs_lblcr.c index 7c7396a6acbf..61ae8cfcf0b4 100644 --- a/net/netfilter/ipvs/ip_vs_lblcr.c +++ b/net/netfilter/ipvs/ip_vs_lblcr.c @@ -70,8 +70,6 @@ * entries that haven't been touched for a day. */ #define COUNT_FOR_FULL_EXPIRATION 30 -static int sysctl_ip_vs_lblcr_expiration = 24*60*60*HZ; - /* * for IPVS lblcr entry hash table @@ -296,7 +294,7 @@ struct ip_vs_lblcr_table { static ctl_table vs_vars_table[] = { { .procname = "lblcr_expiration", - .data = &sysctl_ip_vs_lblcr_expiration, + .data = NULL, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, @@ -304,8 +302,6 @@ static ctl_table vs_vars_table[] = { { } }; -static struct ctl_table_header * sysctl_header; - static inline void ip_vs_lblcr_free(struct ip_vs_lblcr_entry *en) { list_del(&en->list); @@ -425,14 +421,15 @@ static inline void ip_vs_lblcr_full_check(struct ip_vs_service *svc) unsigned long now = jiffies; int i, j; struct ip_vs_lblcr_entry *en, *nxt; + struct netns_ipvs *ipvs = net_ipvs(svc->net); for (i=0, j=tbl->rover; isched_lock); list_for_each_entry_safe(en, nxt, &tbl->bucket[j], list) { - if (time_after(en->lastuse+sysctl_ip_vs_lblcr_expiration, - now)) + if (time_after(en->lastuse + + ipvs->sysctl_lblcr_expiration, now)) continue; ip_vs_lblcr_free(en); @@ -664,6 +661,7 @@ ip_vs_lblcr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) read_lock(&svc->sched_lock); en = ip_vs_lblcr_get(svc->af, tbl, &iph.daddr); if (en) { + struct netns_ipvs *ipvs = net_ipvs(svc->net); /* We only hold a read lock, but this is atomic */ en->lastuse = jiffies; @@ -675,7 +673,7 @@ ip_vs_lblcr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) /* More than one destination + enough time passed by, cleanup */ if (atomic_read(&en->set.size) > 1 && time_after(jiffies, en->set.lastmod + - sysctl_ip_vs_lblcr_expiration)) { + ipvs->sysctl_lblcr_expiration)) { struct ip_vs_dest *m; write_lock(&en->set.lock); @@ -749,23 +747,43 @@ static struct ip_vs_scheduler ip_vs_lblcr_scheduler = */ static int __net_init __ip_vs_lblcr_init(struct net *net) { - if (!net_eq(net, &init_net)) /* netns not enabled yet */ - return -EPERM; - - sysctl_header = register_net_sysctl_table(net, net_vs_ctl_path, - vs_vars_table); - if (!sysctl_header) - return -ENOMEM; + struct netns_ipvs *ipvs = net_ipvs(net); + + if (!net_eq(net, &init_net)) { + ipvs->lblcr_ctl_table = kmemdup(vs_vars_table, + sizeof(vs_vars_table), + GFP_KERNEL); + if (ipvs->lblcr_ctl_table == NULL) + goto err_dup; + } else + ipvs->lblcr_ctl_table = vs_vars_table; + ipvs->sysctl_lblcr_expiration = 24*60*60*HZ; + ipvs->lblcr_ctl_table[0].data = &ipvs->sysctl_lblcr_expiration; + + ipvs->lblcr_ctl_header = + register_net_sysctl_table(net, net_vs_ctl_path, + ipvs->lblcr_ctl_table); + if (!ipvs->lblcr_ctl_header) + goto err_reg; return 0; + +err_reg: + if (!net_eq(net, &init_net)) + kfree(ipvs->lblcr_ctl_table); + +err_dup: + return -ENOMEM; } static void __net_exit __ip_vs_lblcr_exit(struct net *net) { - if (!net_eq(net, &init_net)) /* netns not enabled yet */ - return; + struct netns_ipvs *ipvs = net_ipvs(net); + + unregister_net_sysctl_table(ipvs->lblcr_ctl_header); - unregister_net_sysctl_table(sysctl_header); + if (!net_eq(net, &init_net)) + kfree(ipvs->lblcr_ctl_table); } static struct pernet_operations ip_vs_lblcr_ops = { -- cgit v1.2.3 From b6e885ddb903e681b7cbb4e68ad775154660e1f4 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:45 +0100 Subject: IPVS: netns awarness to lblc sheduler var sysctl_ip_vs_lblc_expiration moved to ipvs struct as sysctl_lblc_expiration procfs updated to handle this. Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_lblc.c | 50 ++++++++++++++++++++++++++++------------- 1 file changed, 34 insertions(+), 16 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_lblc.c b/net/netfilter/ipvs/ip_vs_lblc.c index 84278fb4e055..d5bec3371871 100644 --- a/net/netfilter/ipvs/ip_vs_lblc.c +++ b/net/netfilter/ipvs/ip_vs_lblc.c @@ -70,7 +70,6 @@ * entries that haven't been touched for a day. */ #define COUNT_FOR_FULL_EXPIRATION 30 -static int sysctl_ip_vs_lblc_expiration = 24*60*60*HZ; /* @@ -117,7 +116,7 @@ struct ip_vs_lblc_table { static ctl_table vs_vars_table[] = { { .procname = "lblc_expiration", - .data = &sysctl_ip_vs_lblc_expiration, + .data = NULL, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, @@ -125,8 +124,6 @@ static ctl_table vs_vars_table[] = { { } }; -static struct ctl_table_header * sysctl_header; - static inline void ip_vs_lblc_free(struct ip_vs_lblc_entry *en) { list_del(&en->list); @@ -248,6 +245,7 @@ static inline void ip_vs_lblc_full_check(struct ip_vs_service *svc) struct ip_vs_lblc_entry *en, *nxt; unsigned long now = jiffies; int i, j; + struct netns_ipvs *ipvs = net_ipvs(svc->net); for (i=0, j=tbl->rover; isched_lock); list_for_each_entry_safe(en, nxt, &tbl->bucket[j], list) { if (time_before(now, - en->lastuse + sysctl_ip_vs_lblc_expiration)) + en->lastuse + + ipvs->sysctl_lblc_expiration)) continue; ip_vs_lblc_free(en); @@ -548,23 +547,43 @@ static struct ip_vs_scheduler ip_vs_lblc_scheduler = */ static int __net_init __ip_vs_lblc_init(struct net *net) { - if (!net_eq(net, &init_net)) /* netns not enabled yet */ - return -EPERM; - - sysctl_header = register_net_sysctl_table(net, net_vs_ctl_path, - vs_vars_table); - if (!sysctl_header) - return -ENOMEM; + struct netns_ipvs *ipvs = net_ipvs(net); + + if (!net_eq(net, &init_net)) { + ipvs->lblc_ctl_table = kmemdup(vs_vars_table, + sizeof(vs_vars_table), + GFP_KERNEL); + if (ipvs->lblc_ctl_table == NULL) + goto err_dup; + } else + ipvs->lblc_ctl_table = vs_vars_table; + ipvs->sysctl_lblc_expiration = 24*60*60*HZ; + ipvs->lblc_ctl_table[0].data = &ipvs->sysctl_lblc_expiration; + + ipvs->lblc_ctl_header = + register_net_sysctl_table(net, net_vs_ctl_path, + ipvs->lblc_ctl_table); + if (!ipvs->lblc_ctl_header) + goto err_reg; return 0; + +err_reg: + if (!net_eq(net, &init_net)) + kfree(ipvs->lblc_ctl_table); + +err_dup: + return -ENOMEM; } static void __net_exit __ip_vs_lblc_exit(struct net *net) { - if (!net_eq(net, &init_net)) /* netns not enabled yet */ - return; + struct netns_ipvs *ipvs = net_ipvs(net); + + unregister_net_sysctl_table(ipvs->lblc_ctl_header); - unregister_net_sysctl_table(sysctl_header); + if (!net_eq(net, &init_net)) + kfree(ipvs->lblc_ctl_table); } static struct pernet_operations ip_vs_lblc_ops = { @@ -586,7 +605,6 @@ static int __init ip_vs_lblc_init(void) return ret; } - static void __exit ip_vs_lblc_cleanup(void) { unregister_ip_vs_scheduler(&ip_vs_lblc_scheduler); -- cgit v1.2.3 From 252c64103237f1841088f0f29b4f084b1c774546 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:46 +0100 Subject: IPVS: netns, prepare protocol Add support for protocol data per name-space. in struct ip_vs_protocol, appcnt will be removed when all protos are modified for network name-space. This patch causes warnings of unused functions, they will be used when next patch will be applied. Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_proto.c | 66 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 66 insertions(+) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_proto.c b/net/netfilter/ipvs/ip_vs_proto.c index 45392942d0e7..576e29648c53 100644 --- a/net/netfilter/ipvs/ip_vs_proto.c +++ b/net/netfilter/ipvs/ip_vs_proto.c @@ -60,6 +60,31 @@ static int __used __init register_ip_vs_protocol(struct ip_vs_protocol *pp) return 0; } +/* + * register an ipvs protocols netns related data + */ +static int +register_ip_vs_proto_netns(struct net *net, struct ip_vs_protocol *pp) +{ + struct netns_ipvs *ipvs = net_ipvs(net); + unsigned hash = IP_VS_PROTO_HASH(pp->protocol); + struct ip_vs_proto_data *pd = + kzalloc(sizeof(struct ip_vs_proto_data), GFP_ATOMIC); + + if (!pd) { + pr_err("%s(): no memory.\n", __func__); + return -ENOMEM; + } + pd->pp = pp; /* For speed issues */ + pd->next = ipvs->proto_data_table[hash]; + ipvs->proto_data_table[hash] = pd; + atomic_set(&pd->appcnt, 0); /* Init app counter */ + + if (pp->init_netns != NULL) + pp->init_netns(net, pd); + + return 0; +} /* * unregister an ipvs protocol @@ -82,6 +107,29 @@ static int unregister_ip_vs_protocol(struct ip_vs_protocol *pp) return -ESRCH; } +/* + * unregister an ipvs protocols netns data + */ +static int +unregister_ip_vs_proto_netns(struct net *net, struct ip_vs_proto_data *pd) +{ + struct netns_ipvs *ipvs = net_ipvs(net); + struct ip_vs_proto_data **pd_p; + unsigned hash = IP_VS_PROTO_HASH(pd->pp->protocol); + + pd_p = &ipvs->proto_data_table[hash]; + for (; *pd_p; pd_p = &(*pd_p)->next) { + if (*pd_p == pd) { + *pd_p = pd->next; + if (pd->pp->exit_netns != NULL) + pd->pp->exit_netns(net, pd); + kfree(pd); + return 0; + } + } + + return -ESRCH; +} /* * get ip_vs_protocol object by its proto. @@ -100,6 +148,24 @@ struct ip_vs_protocol * ip_vs_proto_get(unsigned short proto) } EXPORT_SYMBOL(ip_vs_proto_get); +/* + * get ip_vs_protocol object data by netns and proto + */ +struct ip_vs_proto_data * +ip_vs_proto_data_get(struct net *net, unsigned short proto) +{ + struct netns_ipvs *ipvs = net_ipvs(net); + struct ip_vs_proto_data *pd; + unsigned hash = IP_VS_PROTO_HASH(proto); + + for (pd = ipvs->proto_data_table[hash]; pd; pd = pd->next) { + if (pd->pp->protocol == proto) + return pd; + } + + return NULL; +} +EXPORT_SYMBOL(ip_vs_proto_data_get); /* * Propagate event for state change to all protocols -- cgit v1.2.3 From 4a85b96c08ef84076f84e87280223a4301988ed9 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:47 +0100 Subject: IPVS: netns preparation for proto_tcp In this phase (one), all local vars will be moved to ipvs struct. Remaining work, add param struct net *net to a couple of functions that is common for all protos and use all ip_vs_proto_data *v3 Removed unused function as sugested by Simon Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_ftp.c | 8 ++- net/netfilter/ipvs/ip_vs_proto.c | 13 ++++- net/netfilter/ipvs/ip_vs_proto_tcp.c | 97 +++++++++++++++++++----------------- 3 files changed, 70 insertions(+), 48 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ftp.c b/net/netfilter/ipvs/ip_vs_ftp.c index 0e762f322aa3..b38ae941f677 100644 --- a/net/netfilter/ipvs/ip_vs_ftp.c +++ b/net/netfilter/ipvs/ip_vs_ftp.c @@ -157,6 +157,7 @@ static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp, int ret = 0; enum ip_conntrack_info ctinfo; struct nf_conn *ct; + struct net *net; #ifdef CONFIG_IP_VS_IPV6 /* This application helper doesn't work with IPv6 yet, @@ -257,8 +258,9 @@ static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp, * would be adjusted twice. */ + net = skb_net(skb); cp->app_data = NULL; - ip_vs_tcp_conn_listen(n_cp); + ip_vs_tcp_conn_listen(net, n_cp); ip_vs_conn_put(n_cp); return ret; } @@ -287,6 +289,7 @@ static int ip_vs_ftp_in(struct ip_vs_app *app, struct ip_vs_conn *cp, union nf_inet_addr to; __be16 port; struct ip_vs_conn *n_cp; + struct net *net; #ifdef CONFIG_IP_VS_IPV6 /* This application helper doesn't work with IPv6 yet, @@ -378,7 +381,8 @@ static int ip_vs_ftp_in(struct ip_vs_app *app, struct ip_vs_conn *cp, /* * Move tunnel to listen state */ - ip_vs_tcp_conn_listen(n_cp); + net = skb_net(skb); + ip_vs_tcp_conn_listen(net, n_cp); ip_vs_conn_put(n_cp); return 1; diff --git a/net/netfilter/ipvs/ip_vs_proto.c b/net/netfilter/ipvs/ip_vs_proto.c index 576e29648c53..320c6a65f370 100644 --- a/net/netfilter/ipvs/ip_vs_proto.c +++ b/net/netfilter/ipvs/ip_vs_proto.c @@ -307,12 +307,23 @@ ip_vs_tcpudp_debug_packet(int af, struct ip_vs_protocol *pp, */ static int __net_init __ip_vs_protocol_init(struct net *net) { +#ifdef CONFIG_IP_VS_PROTO_TCP + register_ip_vs_proto_netns(net, &ip_vs_protocol_tcp); +#endif return 0; } static void __net_exit __ip_vs_protocol_cleanup(struct net *net) { - /* empty */ + struct netns_ipvs *ipvs = net_ipvs(net); + struct ip_vs_proto_data *pd; + int i; + + /* unregister all the ipvs proto data for this netns */ + for (i = 0; i < IP_VS_PROTO_TAB_SIZE; i++) { + while ((pd = ipvs->proto_data_table[i]) != NULL) + unregister_ip_vs_proto_netns(net, pd); + } } static struct pernet_operations ipvs_proto_ops = { diff --git a/net/netfilter/ipvs/ip_vs_proto_tcp.c b/net/netfilter/ipvs/ip_vs_proto_tcp.c index c175d3166263..9d9df3d61093 100644 --- a/net/netfilter/ipvs/ip_vs_proto_tcp.c +++ b/net/netfilter/ipvs/ip_vs_proto_tcp.c @@ -9,8 +9,12 @@ * as published by the Free Software Foundation; either version * 2 of the License, or (at your option) any later version. * - * Changes: + * Changes: Hans Schillstrom * + * Network name space (netns) aware. + * Global data moved to netns i.e struct netns_ipvs + * tcp_timeouts table has copy per netns in a hash table per + * protocol ip_vs_proto_data and is handled by netns */ #define KMSG_COMPONENT "IPVS" @@ -345,7 +349,7 @@ static const int tcp_state_off[IP_VS_DIR_LAST] = { /* * Timeout table[state] */ -static int tcp_timeouts[IP_VS_TCP_S_LAST+1] = { +static const int tcp_timeouts[IP_VS_TCP_S_LAST+1] = { [IP_VS_TCP_S_NONE] = 2*HZ, [IP_VS_TCP_S_ESTABLISHED] = 15*60*HZ, [IP_VS_TCP_S_SYN_SENT] = 2*60*HZ, @@ -460,13 +464,6 @@ static void tcp_timeout_change(struct ip_vs_protocol *pp, int flags) tcp_state_table = (on? tcp_states_dos : tcp_states); } -static int -tcp_set_state_timeout(struct ip_vs_protocol *pp, char *sname, int to) -{ - return ip_vs_set_state_timeout(pp->timeout_table, IP_VS_TCP_S_LAST, - tcp_state_name_table, sname, to); -} - static inline int tcp_state_idx(struct tcphdr *th) { if (th->rst) @@ -487,6 +484,7 @@ set_tcp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, int state_idx; int new_state = IP_VS_TCP_S_CLOSE; int state_off = tcp_state_off[direction]; + struct ip_vs_proto_data *pd; /* Temp fix */ /* * Update state offset to INPUT_ONLY if necessary @@ -542,10 +540,13 @@ set_tcp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, } } - cp->timeout = pp->timeout_table[cp->state = new_state]; + pd = ip_vs_proto_data_get(&init_net, pp->protocol); + if (likely(pd)) + cp->timeout = pd->timeout_table[cp->state = new_state]; + else /* What to do ? */ + cp->timeout = tcp_timeouts[cp->state = new_state]; } - /* * Handle state transitions */ @@ -573,17 +574,6 @@ tcp_state_transition(struct ip_vs_conn *cp, int direction, return 1; } - -/* - * Hash table for TCP application incarnations - */ -#define TCP_APP_TAB_BITS 4 -#define TCP_APP_TAB_SIZE (1 << TCP_APP_TAB_BITS) -#define TCP_APP_TAB_MASK (TCP_APP_TAB_SIZE - 1) - -static struct list_head tcp_apps[TCP_APP_TAB_SIZE]; -static DEFINE_SPINLOCK(tcp_app_lock); - static inline __u16 tcp_app_hashkey(__be16 port) { return (((__force u16)port >> TCP_APP_TAB_BITS) ^ (__force u16)port) @@ -597,21 +587,23 @@ static int tcp_register_app(struct ip_vs_app *inc) __u16 hash; __be16 port = inc->port; int ret = 0; + struct netns_ipvs *ipvs = net_ipvs(&init_net); + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, IPPROTO_TCP); hash = tcp_app_hashkey(port); - spin_lock_bh(&tcp_app_lock); - list_for_each_entry(i, &tcp_apps[hash], p_list) { + spin_lock_bh(&ipvs->tcp_app_lock); + list_for_each_entry(i, &ipvs->tcp_apps[hash], p_list) { if (i->port == port) { ret = -EEXIST; goto out; } } - list_add(&inc->p_list, &tcp_apps[hash]); - atomic_inc(&ip_vs_protocol_tcp.appcnt); + list_add(&inc->p_list, &ipvs->tcp_apps[hash]); + atomic_inc(&pd->pp->appcnt); out: - spin_unlock_bh(&tcp_app_lock); + spin_unlock_bh(&ipvs->tcp_app_lock); return ret; } @@ -619,16 +611,20 @@ static int tcp_register_app(struct ip_vs_app *inc) static void tcp_unregister_app(struct ip_vs_app *inc) { - spin_lock_bh(&tcp_app_lock); - atomic_dec(&ip_vs_protocol_tcp.appcnt); + struct netns_ipvs *ipvs = net_ipvs(&init_net); + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, IPPROTO_TCP); + + spin_lock_bh(&ipvs->tcp_app_lock); + atomic_dec(&pd->pp->appcnt); list_del(&inc->p_list); - spin_unlock_bh(&tcp_app_lock); + spin_unlock_bh(&ipvs->tcp_app_lock); } static int tcp_app_conn_bind(struct ip_vs_conn *cp) { + struct netns_ipvs *ipvs = net_ipvs(&init_net); int hash; struct ip_vs_app *inc; int result = 0; @@ -640,12 +636,12 @@ tcp_app_conn_bind(struct ip_vs_conn *cp) /* Lookup application incarnations and bind the right one */ hash = tcp_app_hashkey(cp->vport); - spin_lock(&tcp_app_lock); - list_for_each_entry(inc, &tcp_apps[hash], p_list) { + spin_lock(&ipvs->tcp_app_lock); + list_for_each_entry(inc, &ipvs->tcp_apps[hash], p_list) { if (inc->port == cp->vport) { if (unlikely(!ip_vs_app_inc_get(inc))) break; - spin_unlock(&tcp_app_lock); + spin_unlock(&ipvs->tcp_app_lock); IP_VS_DBG_BUF(9, "%s(): Binding conn %s:%u->" "%s:%u to app %s on port %u\n", @@ -662,7 +658,7 @@ tcp_app_conn_bind(struct ip_vs_conn *cp) goto out; } } - spin_unlock(&tcp_app_lock); + spin_unlock(&ipvs->tcp_app_lock); out: return result; @@ -672,24 +668,34 @@ tcp_app_conn_bind(struct ip_vs_conn *cp) /* * Set LISTEN timeout. (ip_vs_conn_put will setup timer) */ -void ip_vs_tcp_conn_listen(struct ip_vs_conn *cp) +void ip_vs_tcp_conn_listen(struct net *net, struct ip_vs_conn *cp) { + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(net, IPPROTO_TCP); + spin_lock(&cp->lock); cp->state = IP_VS_TCP_S_LISTEN; - cp->timeout = ip_vs_protocol_tcp.timeout_table[IP_VS_TCP_S_LISTEN]; + cp->timeout = (pd ? pd->timeout_table[IP_VS_TCP_S_LISTEN] + : tcp_timeouts[IP_VS_TCP_S_LISTEN]); spin_unlock(&cp->lock); } - -static void ip_vs_tcp_init(struct ip_vs_protocol *pp) +/* --------------------------------------------- + * timeouts is netns related now. + * --------------------------------------------- + */ +static void __ip_vs_tcp_init(struct net *net, struct ip_vs_proto_data *pd) { - IP_VS_INIT_HASH_TABLE(tcp_apps); - pp->timeout_table = tcp_timeouts; -} + struct netns_ipvs *ipvs = net_ipvs(net); + ip_vs_init_hash_table(ipvs->tcp_apps, TCP_APP_TAB_SIZE); + spin_lock_init(&ipvs->tcp_app_lock); + pd->timeout_table = ip_vs_create_timeout_table((int *)tcp_timeouts, + sizeof(tcp_timeouts)); +} -static void ip_vs_tcp_exit(struct ip_vs_protocol *pp) +static void __ip_vs_tcp_exit(struct net *net, struct ip_vs_proto_data *pd) { + kfree(pd->timeout_table); } @@ -699,8 +705,10 @@ struct ip_vs_protocol ip_vs_protocol_tcp = { .num_states = IP_VS_TCP_S_LAST, .dont_defrag = 0, .appcnt = ATOMIC_INIT(0), - .init = ip_vs_tcp_init, - .exit = ip_vs_tcp_exit, + .init = NULL, + .exit = NULL, + .init_netns = __ip_vs_tcp_init, + .exit_netns = __ip_vs_tcp_exit, .register_app = tcp_register_app, .unregister_app = tcp_unregister_app, .conn_schedule = tcp_conn_schedule, @@ -714,5 +722,4 @@ struct ip_vs_protocol ip_vs_protocol_tcp = { .app_conn_bind = tcp_app_conn_bind, .debug_packet = ip_vs_tcpudp_debug_packet, .timeout_change = tcp_timeout_change, - .set_state_timeout = tcp_set_state_timeout, }; -- cgit v1.2.3 From 78b16bde104cc74bedbf462b0ebed2990f35ff6b Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:48 +0100 Subject: IPVS: netns preparation for proto_udp In this phase (one), all local vars will be moved to ipvs struct. Remaining work, add param struct net *net to a couple of functions that is common for all protos and use ip_vs_proto_data *v3 Removed unused function set_state_timeout() Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_proto.c | 3 ++ net/netfilter/ipvs/ip_vs_proto_udp.c | 86 ++++++++++++++++++------------------ 2 files changed, 46 insertions(+), 43 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_proto.c b/net/netfilter/ipvs/ip_vs_proto.c index 320c6a65f370..cdc414238fcb 100644 --- a/net/netfilter/ipvs/ip_vs_proto.c +++ b/net/netfilter/ipvs/ip_vs_proto.c @@ -309,6 +309,9 @@ static int __net_init __ip_vs_protocol_init(struct net *net) { #ifdef CONFIG_IP_VS_PROTO_TCP register_ip_vs_proto_netns(net, &ip_vs_protocol_tcp); +#endif +#ifdef CONFIG_IP_VS_PROTO_UDP + register_ip_vs_proto_netns(net, &ip_vs_protocol_udp); #endif return 0; } diff --git a/net/netfilter/ipvs/ip_vs_proto_udp.c b/net/netfilter/ipvs/ip_vs_proto_udp.c index 5ab54f648654..71a4721a8f8a 100644 --- a/net/netfilter/ipvs/ip_vs_proto_udp.c +++ b/net/netfilter/ipvs/ip_vs_proto_udp.c @@ -9,7 +9,8 @@ * as published by the Free Software Foundation; either version * 2 of the License, or (at your option) any later version. * - * Changes: + * Changes: Hans Schillstrom + * Network name space (netns) aware. * */ @@ -345,19 +346,6 @@ udp_csum_check(int af, struct sk_buff *skb, struct ip_vs_protocol *pp) return 1; } - -/* - * Note: the caller guarantees that only one of register_app, - * unregister_app or app_conn_bind is called each time. - */ - -#define UDP_APP_TAB_BITS 4 -#define UDP_APP_TAB_SIZE (1 << UDP_APP_TAB_BITS) -#define UDP_APP_TAB_MASK (UDP_APP_TAB_SIZE - 1) - -static struct list_head udp_apps[UDP_APP_TAB_SIZE]; -static DEFINE_SPINLOCK(udp_app_lock); - static inline __u16 udp_app_hashkey(__be16 port) { return (((__force u16)port >> UDP_APP_TAB_BITS) ^ (__force u16)port) @@ -371,22 +359,24 @@ static int udp_register_app(struct ip_vs_app *inc) __u16 hash; __be16 port = inc->port; int ret = 0; + struct netns_ipvs *ipvs = net_ipvs(&init_net); + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, IPPROTO_UDP); hash = udp_app_hashkey(port); - spin_lock_bh(&udp_app_lock); - list_for_each_entry(i, &udp_apps[hash], p_list) { + spin_lock_bh(&ipvs->udp_app_lock); + list_for_each_entry(i, &ipvs->udp_apps[hash], p_list) { if (i->port == port) { ret = -EEXIST; goto out; } } - list_add(&inc->p_list, &udp_apps[hash]); - atomic_inc(&ip_vs_protocol_udp.appcnt); + list_add(&inc->p_list, &ipvs->udp_apps[hash]); + atomic_inc(&pd->pp->appcnt); out: - spin_unlock_bh(&udp_app_lock); + spin_unlock_bh(&ipvs->udp_app_lock); return ret; } @@ -394,15 +384,19 @@ static int udp_register_app(struct ip_vs_app *inc) static void udp_unregister_app(struct ip_vs_app *inc) { - spin_lock_bh(&udp_app_lock); - atomic_dec(&ip_vs_protocol_udp.appcnt); + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, IPPROTO_UDP); + struct netns_ipvs *ipvs = net_ipvs(&init_net); + + spin_lock_bh(&ipvs->udp_app_lock); + atomic_dec(&pd->pp->appcnt); list_del(&inc->p_list); - spin_unlock_bh(&udp_app_lock); + spin_unlock_bh(&ipvs->udp_app_lock); } static int udp_app_conn_bind(struct ip_vs_conn *cp) { + struct netns_ipvs *ipvs = net_ipvs(&init_net); int hash; struct ip_vs_app *inc; int result = 0; @@ -414,12 +408,12 @@ static int udp_app_conn_bind(struct ip_vs_conn *cp) /* Lookup application incarnations and bind the right one */ hash = udp_app_hashkey(cp->vport); - spin_lock(&udp_app_lock); - list_for_each_entry(inc, &udp_apps[hash], p_list) { + spin_lock(&ipvs->udp_app_lock); + list_for_each_entry(inc, &ipvs->udp_apps[hash], p_list) { if (inc->port == cp->vport) { if (unlikely(!ip_vs_app_inc_get(inc))) break; - spin_unlock(&udp_app_lock); + spin_unlock(&ipvs->udp_app_lock); IP_VS_DBG_BUF(9, "%s(): Binding conn %s:%u->" "%s:%u to app %s on port %u\n", @@ -436,14 +430,14 @@ static int udp_app_conn_bind(struct ip_vs_conn *cp) goto out; } } - spin_unlock(&udp_app_lock); + spin_unlock(&ipvs->udp_app_lock); out: return result; } -static int udp_timeouts[IP_VS_UDP_S_LAST+1] = { +static const int udp_timeouts[IP_VS_UDP_S_LAST+1] = { [IP_VS_UDP_S_NORMAL] = 5*60*HZ, [IP_VS_UDP_S_LAST] = 2*HZ, }; @@ -453,14 +447,6 @@ static const char *const udp_state_name_table[IP_VS_UDP_S_LAST+1] = { [IP_VS_UDP_S_LAST] = "BUG!", }; - -static int -udp_set_state_timeout(struct ip_vs_protocol *pp, char *sname, int to) -{ - return ip_vs_set_state_timeout(pp->timeout_table, IP_VS_UDP_S_LAST, - udp_state_name_table, sname, to); -} - static const char * udp_state_name(int state) { if (state >= IP_VS_UDP_S_LAST) @@ -473,18 +459,31 @@ udp_state_transition(struct ip_vs_conn *cp, int direction, const struct sk_buff *skb, struct ip_vs_protocol *pp) { - cp->timeout = pp->timeout_table[IP_VS_UDP_S_NORMAL]; + struct ip_vs_proto_data *pd; /* Temp fix, pp will be replaced by pd */ + + pd = ip_vs_proto_data_get(&init_net, IPPROTO_UDP); + if (unlikely(!pd)) { + pr_err("UDP no ns data\n"); + return 0; + } + + cp->timeout = pd->timeout_table[IP_VS_UDP_S_NORMAL]; return 1; } -static void udp_init(struct ip_vs_protocol *pp) +static void __udp_init(struct net *net, struct ip_vs_proto_data *pd) { - IP_VS_INIT_HASH_TABLE(udp_apps); - pp->timeout_table = udp_timeouts; + struct netns_ipvs *ipvs = net_ipvs(net); + + ip_vs_init_hash_table(ipvs->udp_apps, UDP_APP_TAB_SIZE); + spin_lock_init(&ipvs->udp_app_lock); + pd->timeout_table = ip_vs_create_timeout_table((int *)udp_timeouts, + sizeof(udp_timeouts)); } -static void udp_exit(struct ip_vs_protocol *pp) +static void __udp_exit(struct net *net, struct ip_vs_proto_data *pd) { + kfree(pd->timeout_table); } @@ -493,8 +492,10 @@ struct ip_vs_protocol ip_vs_protocol_udp = { .protocol = IPPROTO_UDP, .num_states = IP_VS_UDP_S_LAST, .dont_defrag = 0, - .init = udp_init, - .exit = udp_exit, + .init = NULL, + .exit = NULL, + .init_netns = __udp_init, + .exit_netns = __udp_exit, .conn_schedule = udp_conn_schedule, .conn_in_get = ip_vs_conn_in_get_proto, .conn_out_get = ip_vs_conn_out_get_proto, @@ -508,5 +509,4 @@ struct ip_vs_protocol ip_vs_protocol_udp = { .app_conn_bind = udp_app_conn_bind, .debug_packet = ip_vs_tcpudp_debug_packet, .timeout_change = NULL, - .set_state_timeout = udp_set_state_timeout, }; -- cgit v1.2.3 From 9d934878e7870fbbbd8eaed2e467552536877def Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:49 +0100 Subject: IPVS: netns preparation for proto_sctp In this phase (one), all local vars will be moved to ipvs struct. Remaining work, add param struct net *net to a couple of functions that is common for all protos and use ip_vs_proto_data *v3 Removed unuset function set_state_timeout() Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_proto.c | 3 + net/netfilter/ipvs/ip_vs_proto_sctp.c | 121 ++++++++++++++++------------------ 2 files changed, 61 insertions(+), 63 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_proto.c b/net/netfilter/ipvs/ip_vs_proto.c index cdc414238fcb..001b2f825043 100644 --- a/net/netfilter/ipvs/ip_vs_proto.c +++ b/net/netfilter/ipvs/ip_vs_proto.c @@ -312,6 +312,9 @@ static int __net_init __ip_vs_protocol_init(struct net *net) #endif #ifdef CONFIG_IP_VS_PROTO_UDP register_ip_vs_proto_netns(net, &ip_vs_protocol_udp); +#endif +#ifdef CONFIG_IP_VS_PROTO_SCTP + register_ip_vs_proto_netns(net, &ip_vs_protocol_sctp); #endif return 0; } diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c index 521b827083fe..f826dd1e4630 100644 --- a/net/netfilter/ipvs/ip_vs_proto_sctp.c +++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c @@ -862,7 +862,7 @@ static struct ipvs_sctp_nextstate /* * Timeout table[state] */ -static int sctp_timeouts[IP_VS_SCTP_S_LAST + 1] = { +static const int sctp_timeouts[IP_VS_SCTP_S_LAST + 1] = { [IP_VS_SCTP_S_NONE] = 2 * HZ, [IP_VS_SCTP_S_INIT_CLI] = 1 * 60 * HZ, [IP_VS_SCTP_S_INIT_SER] = 1 * 60 * HZ, @@ -906,18 +906,6 @@ static const char *sctp_state_name(int state) return "?"; } -static void sctp_timeout_change(struct ip_vs_protocol *pp, int flags) -{ -} - -static int -sctp_set_state_timeout(struct ip_vs_protocol *pp, char *sname, int to) -{ - -return ip_vs_set_state_timeout(pp->timeout_table, IP_VS_SCTP_S_LAST, - sctp_state_name_table, sname, to); -} - static inline int set_sctp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, int direction, const struct sk_buff *skb) @@ -926,6 +914,7 @@ set_sctp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, unsigned char chunk_type; int event, next_state; int ihl; + struct ip_vs_proto_data *pd; #ifdef CONFIG_IP_VS_IPV6 ihl = cp->af == AF_INET ? ip_hdrlen(skb) : sizeof(struct ipv6hdr); @@ -1001,10 +990,13 @@ set_sctp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, } } } + pd = ip_vs_proto_data_get(&init_net, pp->protocol); /* tmp fix */ + if (likely(pd)) + cp->timeout = pd->timeout_table[cp->state = next_state]; + else /* What to do ? */ + cp->timeout = sctp_timeouts[cp->state = next_state]; - cp->timeout = pp->timeout_table[cp->state = next_state]; - - return 1; + return 1; } static int @@ -1020,16 +1012,6 @@ sctp_state_transition(struct ip_vs_conn *cp, int direction, return ret; } -/* - * Hash table for SCTP application incarnations - */ -#define SCTP_APP_TAB_BITS 4 -#define SCTP_APP_TAB_SIZE (1 << SCTP_APP_TAB_BITS) -#define SCTP_APP_TAB_MASK (SCTP_APP_TAB_SIZE - 1) - -static struct list_head sctp_apps[SCTP_APP_TAB_SIZE]; -static DEFINE_SPINLOCK(sctp_app_lock); - static inline __u16 sctp_app_hashkey(__be16 port) { return (((__force u16)port >> SCTP_APP_TAB_BITS) ^ (__force u16)port) @@ -1042,34 +1024,40 @@ static int sctp_register_app(struct ip_vs_app *inc) __u16 hash; __be16 port = inc->port; int ret = 0; + struct netns_ipvs *ipvs = net_ipvs(&init_net); + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, IPPROTO_SCTP); hash = sctp_app_hashkey(port); - spin_lock_bh(&sctp_app_lock); - list_for_each_entry(i, &sctp_apps[hash], p_list) { + spin_lock_bh(&ipvs->sctp_app_lock); + list_for_each_entry(i, &ipvs->sctp_apps[hash], p_list) { if (i->port == port) { ret = -EEXIST; goto out; } } - list_add(&inc->p_list, &sctp_apps[hash]); - atomic_inc(&ip_vs_protocol_sctp.appcnt); + list_add(&inc->p_list, &ipvs->sctp_apps[hash]); + atomic_inc(&pd->pp->appcnt); out: - spin_unlock_bh(&sctp_app_lock); + spin_unlock_bh(&ipvs->sctp_app_lock); return ret; } static void sctp_unregister_app(struct ip_vs_app *inc) { - spin_lock_bh(&sctp_app_lock); - atomic_dec(&ip_vs_protocol_sctp.appcnt); + struct netns_ipvs *ipvs = net_ipvs(&init_net); + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, IPPROTO_SCTP); + + spin_lock_bh(&ipvs->sctp_app_lock); + atomic_dec(&pd->pp->appcnt); list_del(&inc->p_list); - spin_unlock_bh(&sctp_app_lock); + spin_unlock_bh(&ipvs->sctp_app_lock); } static int sctp_app_conn_bind(struct ip_vs_conn *cp) { + struct netns_ipvs *ipvs = net_ipvs(&init_net); int hash; struct ip_vs_app *inc; int result = 0; @@ -1080,12 +1068,12 @@ static int sctp_app_conn_bind(struct ip_vs_conn *cp) /* Lookup application incarnations and bind the right one */ hash = sctp_app_hashkey(cp->vport); - spin_lock(&sctp_app_lock); - list_for_each_entry(inc, &sctp_apps[hash], p_list) { + spin_lock(&ipvs->sctp_app_lock); + list_for_each_entry(inc, &ipvs->sctp_apps[hash], p_list) { if (inc->port == cp->vport) { if (unlikely(!ip_vs_app_inc_get(inc))) break; - spin_unlock(&sctp_app_lock); + spin_unlock(&ipvs->sctp_app_lock); IP_VS_DBG_BUF(9, "%s: Binding conn %s:%u->" "%s:%u to app %s on port %u\n", @@ -1101,43 +1089,50 @@ static int sctp_app_conn_bind(struct ip_vs_conn *cp) goto out; } } - spin_unlock(&sctp_app_lock); + spin_unlock(&ipvs->sctp_app_lock); out: return result; } -static void ip_vs_sctp_init(struct ip_vs_protocol *pp) +/* --------------------------------------------- + * timeouts is netns related now. + * --------------------------------------------- + */ +static void __ip_vs_sctp_init(struct net *net, struct ip_vs_proto_data *pd) { - IP_VS_INIT_HASH_TABLE(sctp_apps); - pp->timeout_table = sctp_timeouts; -} + struct netns_ipvs *ipvs = net_ipvs(net); + ip_vs_init_hash_table(ipvs->sctp_apps, SCTP_APP_TAB_SIZE); + spin_lock_init(&ipvs->tcp_app_lock); + pd->timeout_table = ip_vs_create_timeout_table((int *)sctp_timeouts, + sizeof(sctp_timeouts)); +} -static void ip_vs_sctp_exit(struct ip_vs_protocol *pp) +static void __ip_vs_sctp_exit(struct net *net, struct ip_vs_proto_data *pd) { - + kfree(pd->timeout_table); } struct ip_vs_protocol ip_vs_protocol_sctp = { - .name = "SCTP", - .protocol = IPPROTO_SCTP, - .num_states = IP_VS_SCTP_S_LAST, - .dont_defrag = 0, - .appcnt = ATOMIC_INIT(0), - .init = ip_vs_sctp_init, - .exit = ip_vs_sctp_exit, - .register_app = sctp_register_app, + .name = "SCTP", + .protocol = IPPROTO_SCTP, + .num_states = IP_VS_SCTP_S_LAST, + .dont_defrag = 0, + .init = NULL, + .exit = NULL, + .init_netns = __ip_vs_sctp_init, + .exit_netns = __ip_vs_sctp_exit, + .register_app = sctp_register_app, .unregister_app = sctp_unregister_app, - .conn_schedule = sctp_conn_schedule, - .conn_in_get = ip_vs_conn_in_get_proto, - .conn_out_get = ip_vs_conn_out_get_proto, - .snat_handler = sctp_snat_handler, - .dnat_handler = sctp_dnat_handler, - .csum_check = sctp_csum_check, - .state_name = sctp_state_name, + .conn_schedule = sctp_conn_schedule, + .conn_in_get = ip_vs_conn_in_get_proto, + .conn_out_get = ip_vs_conn_out_get_proto, + .snat_handler = sctp_snat_handler, + .dnat_handler = sctp_dnat_handler, + .csum_check = sctp_csum_check, + .state_name = sctp_state_name, .state_transition = sctp_state_transition, - .app_conn_bind = sctp_app_conn_bind, - .debug_packet = ip_vs_tcpudp_debug_packet, - .timeout_change = sctp_timeout_change, - .set_state_timeout = sctp_set_state_timeout, + .app_conn_bind = sctp_app_conn_bind, + .debug_packet = ip_vs_tcpudp_debug_packet, + .timeout_change = NULL, }; -- cgit v1.2.3 From 88fe2d372793a71ae4f6319a16f537d56a83906c Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:50 +0100 Subject: IPVS: netns preparation for proto_ah_esp In this phase (one), all local vars will be moved to ipvs struct. Remaining work, add param struct net *net to a couple of functions that common for all protos. Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_proto.c | 6 ++++++ net/netfilter/ipvs/ip_vs_proto_ah_esp.c | 20 ++++---------------- 2 files changed, 10 insertions(+), 16 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_proto.c b/net/netfilter/ipvs/ip_vs_proto.c index 001b2f825043..9f609d4d5d58 100644 --- a/net/netfilter/ipvs/ip_vs_proto.c +++ b/net/netfilter/ipvs/ip_vs_proto.c @@ -315,6 +315,12 @@ static int __net_init __ip_vs_protocol_init(struct net *net) #endif #ifdef CONFIG_IP_VS_PROTO_SCTP register_ip_vs_proto_netns(net, &ip_vs_protocol_sctp); +#endif +#ifdef CONFIG_IP_VS_PROTO_AH + register_ip_vs_proto_netns(net, &ip_vs_protocol_ah); +#endif +#ifdef CONFIG_IP_VS_PROTO_ESP + register_ip_vs_proto_netns(net, &ip_vs_protocol_esp); #endif return 0; } diff --git a/net/netfilter/ipvs/ip_vs_proto_ah_esp.c b/net/netfilter/ipvs/ip_vs_proto_ah_esp.c index 3a0461117d3f..b8b37fafc988 100644 --- a/net/netfilter/ipvs/ip_vs_proto_ah_esp.c +++ b/net/netfilter/ipvs/ip_vs_proto_ah_esp.c @@ -117,26 +117,14 @@ ah_esp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, return 0; } -static void ah_esp_init(struct ip_vs_protocol *pp) -{ - /* nothing to do now */ -} - - -static void ah_esp_exit(struct ip_vs_protocol *pp) -{ - /* nothing to do now */ -} - - #ifdef CONFIG_IP_VS_PROTO_AH struct ip_vs_protocol ip_vs_protocol_ah = { .name = "AH", .protocol = IPPROTO_AH, .num_states = 1, .dont_defrag = 1, - .init = ah_esp_init, - .exit = ah_esp_exit, + .init = NULL, + .exit = NULL, .conn_schedule = ah_esp_conn_schedule, .conn_in_get = ah_esp_conn_in_get, .conn_out_get = ah_esp_conn_out_get, @@ -159,8 +147,8 @@ struct ip_vs_protocol ip_vs_protocol_esp = { .protocol = IPPROTO_ESP, .num_states = 1, .dont_defrag = 1, - .init = ah_esp_init, - .exit = ah_esp_exit, + .init = NULL, + .exit = NULL, .conn_schedule = ah_esp_conn_schedule, .conn_in_get = ah_esp_conn_in_get, .conn_out_get = ah_esp_conn_out_get, -- cgit v1.2.3 From 9330419d9aa4f97df412ac9be9fc0388c67dd315 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:51 +0100 Subject: IPVS: netns, use ip_vs_proto_data as param. ip_vs_protocol *pp is replaced by ip_vs_proto_data *pd in function call in ip_vs_protocol struct i.e. :, - timeout_change() - state_transition() ip_vs_protocol_timeout_change() got ipvs as param, due to above and a upcoming patch - defence work Most of this changes are triggered by Julians comment: "tcp_timeout_change should work with the new struct ip_vs_proto_data so that tcp_state_table will go to pd->state_table and set_tcp_state will get pd instead of pp" *v3 Mostly comments from Julian The pp -> pd conversion should start from functions like ip_vs_out() that use pp = ip_vs_proto_get(iph.protocol), now they should use ip_vs_proto_data_get(net, iph.protocol). conn_in_get() and conn_out_get() unused param *pp, removed. *v4 ip_vs_protocol_timeout_change() walk the proto_data path. Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_conn.c | 2 - net/netfilter/ipvs/ip_vs_core.c | 77 ++++++++++++++++++++------------- net/netfilter/ipvs/ip_vs_ctl.c | 55 ++++++++++++++--------- net/netfilter/ipvs/ip_vs_proto.c | 21 ++++++--- net/netfilter/ipvs/ip_vs_proto_ah_esp.c | 10 ++--- net/netfilter/ipvs/ip_vs_proto_sctp.c | 16 +++---- net/netfilter/ipvs/ip_vs_proto_tcp.c | 27 +++++------- net/netfilter/ipvs/ip_vs_proto_udp.c | 11 ++--- net/netfilter/xt_ipvs.c | 2 +- 9 files changed, 123 insertions(+), 98 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 7a0e79e3ad0f..a7aba6a4697e 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -329,7 +329,6 @@ ip_vs_conn_fill_param_proto(int af, const struct sk_buff *skb, struct ip_vs_conn * ip_vs_conn_in_get_proto(int af, const struct sk_buff *skb, - struct ip_vs_protocol *pp, const struct ip_vs_iphdr *iph, unsigned int proto_off, int inverse) { @@ -428,7 +427,6 @@ struct ip_vs_conn *ip_vs_conn_out_get(const struct ip_vs_conn_param *p) struct ip_vs_conn * ip_vs_conn_out_get_proto(int af, const struct sk_buff *skb, - struct ip_vs_protocol *pp, const struct ip_vs_iphdr *iph, unsigned int proto_off, int inverse) { diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index d0616ea1eebf..9317affc5ea1 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -177,11 +177,11 @@ ip_vs_conn_stats(struct ip_vs_conn *cp, struct ip_vs_service *svc) static inline int ip_vs_set_state(struct ip_vs_conn *cp, int direction, const struct sk_buff *skb, - struct ip_vs_protocol *pp) + struct ip_vs_proto_data *pd) { - if (unlikely(!pp->state_transition)) + if (unlikely(!pd->pp->state_transition)) return 0; - return pp->state_transition(cp, direction, skb, pp); + return pd->pp->state_transition(cp, direction, skb, pd); } static inline int @@ -378,8 +378,9 @@ ip_vs_sched_persist(struct ip_vs_service *svc, */ struct ip_vs_conn * ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, - struct ip_vs_protocol *pp, int *ignored) + struct ip_vs_proto_data *pd, int *ignored) { + struct ip_vs_protocol *pp = pd->pp; struct ip_vs_conn *cp = NULL; struct ip_vs_iphdr iph; struct ip_vs_dest *dest; @@ -408,7 +409,7 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, * Do not schedule replies from local real server. */ if ((!skb->dev || skb->dev->flags & IFF_LOOPBACK) && - (cp = pp->conn_in_get(svc->af, skb, pp, &iph, iph.len, 1))) { + (cp = pp->conn_in_get(svc->af, skb, &iph, iph.len, 1))) { IP_VS_DBG_PKT(12, svc->af, pp, skb, 0, "Not scheduling reply for existing connection"); __ip_vs_conn_put(cp); @@ -479,11 +480,12 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, * no destination is available for a new connection. */ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, - struct ip_vs_protocol *pp) + struct ip_vs_proto_data *pd) { __be16 _ports[2], *pptr; struct ip_vs_iphdr iph; int unicast; + ip_vs_fill_iphdr(svc->af, skb_network_header(skb), &iph); pptr = skb_header_pointer(skb, iph.len, sizeof(_ports), _ports); @@ -530,10 +532,10 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, ip_vs_in_stats(cp, skb); /* set state */ - cs = ip_vs_set_state(cp, IP_VS_DIR_INPUT, skb, pp); + cs = ip_vs_set_state(cp, IP_VS_DIR_INPUT, skb, pd); /* transmit the first SYN packet */ - ret = cp->packet_xmit(skb, cp, pp); + ret = cp->packet_xmit(skb, cp, pd->pp); /* do not touch skb anymore */ atomic_inc(&cp->in_pkts); @@ -840,7 +842,7 @@ static int ip_vs_out_icmp(struct sk_buff *skb, int *related, ip_vs_fill_iphdr(AF_INET, cih, &ciph); /* The embedded headers contain source and dest in reverse order */ - cp = pp->conn_out_get(AF_INET, skb, pp, &ciph, offset, 1); + cp = pp->conn_out_get(AF_INET, skb, &ciph, offset, 1); if (!cp) return NF_ACCEPT; @@ -917,7 +919,7 @@ static int ip_vs_out_icmp_v6(struct sk_buff *skb, int *related, ip_vs_fill_iphdr(AF_INET6, cih, &ciph); /* The embedded headers contain source and dest in reverse order */ - cp = pp->conn_out_get(AF_INET6, skb, pp, &ciph, offset, 1); + cp = pp->conn_out_get(AF_INET6, skb, &ciph, offset, 1); if (!cp) return NF_ACCEPT; @@ -956,9 +958,11 @@ static inline int is_tcp_reset(const struct sk_buff *skb, int nh_len) * Used for NAT and local client. */ static unsigned int -handle_response(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, +handle_response(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, struct ip_vs_conn *cp, int ihl) { + struct ip_vs_protocol *pp = pd->pp; + IP_VS_DBG_PKT(11, af, pp, skb, 0, "Outgoing packet"); if (!skb_make_writable(skb, ihl)) @@ -1007,7 +1011,7 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, IP_VS_DBG_PKT(10, af, pp, skb, 0, "After SNAT"); ip_vs_out_stats(cp, skb); - ip_vs_set_state(cp, IP_VS_DIR_OUTPUT, skb, pp); + ip_vs_set_state(cp, IP_VS_DIR_OUTPUT, skb, pd); skb->ipvs_property = 1; if (!(cp->flags & IP_VS_CONN_F_NFCT)) ip_vs_notrack(skb); @@ -1034,6 +1038,7 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af) struct net *net = NULL; struct ip_vs_iphdr iph; struct ip_vs_protocol *pp; + struct ip_vs_proto_data *pd; struct ip_vs_conn *cp; EnterFunction(11); @@ -1079,9 +1084,10 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af) ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); } - pp = ip_vs_proto_get(iph.protocol); - if (unlikely(!pp)) + pd = ip_vs_proto_data_get(net, iph.protocol); + if (unlikely(!pd)) return NF_ACCEPT; + pp = pd->pp; /* reassemble IP fragments */ #ifdef CONFIG_IP_VS_IPV6 @@ -1107,10 +1113,10 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af) /* * Check if the packet belongs to an existing entry */ - cp = pp->conn_out_get(af, skb, pp, &iph, iph.len, 0); + cp = pp->conn_out_get(af, skb, &iph, iph.len, 0); if (likely(cp)) - return handle_response(af, skb, pp, cp, iph.len); + return handle_response(af, skb, pd, cp, iph.len); if (sysctl_ip_vs_nat_icmp_send && (pp->protocol == IPPROTO_TCP || pp->protocol == IPPROTO_UDP || @@ -1236,12 +1242,14 @@ ip_vs_local_reply6(unsigned int hooknum, struct sk_buff *skb, static int ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum) { + struct net *net = NULL; struct iphdr *iph; struct icmphdr _icmph, *ic; struct iphdr _ciph, *cih; /* The ip header contained within the ICMP */ struct ip_vs_iphdr ciph; struct ip_vs_conn *cp; struct ip_vs_protocol *pp; + struct ip_vs_proto_data *pd; unsigned int offset, ihl, verdict; union nf_inet_addr snet; @@ -1283,9 +1291,11 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum) if (cih == NULL) return NF_ACCEPT; /* The packet looks wrong, ignore */ - pp = ip_vs_proto_get(cih->protocol); - if (!pp) + net = skb_net(skb); + pd = ip_vs_proto_data_get(net, cih->protocol); + if (!pd) return NF_ACCEPT; + pp = pd->pp; /* Is the embedded protocol header present? */ if (unlikely(cih->frag_off & htons(IP_OFFSET) && @@ -1299,10 +1309,10 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum) ip_vs_fill_iphdr(AF_INET, cih, &ciph); /* The embedded headers contain source and dest in reverse order */ - cp = pp->conn_in_get(AF_INET, skb, pp, &ciph, offset, 1); + cp = pp->conn_in_get(AF_INET, skb, &ciph, offset, 1); if (!cp) { /* The packet could also belong to a local client */ - cp = pp->conn_out_get(AF_INET, skb, pp, &ciph, offset, 1); + cp = pp->conn_out_get(AF_INET, skb, &ciph, offset, 1); if (cp) { snet.ip = iph->saddr; return handle_response_icmp(AF_INET, skb, &snet, @@ -1346,6 +1356,7 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum) static int ip_vs_in_icmp_v6(struct sk_buff *skb, int *related, unsigned int hooknum) { + struct net *net = NULL; struct ipv6hdr *iph; struct icmp6hdr _icmph, *ic; struct ipv6hdr _ciph, *cih; /* The ip header contained @@ -1353,6 +1364,7 @@ ip_vs_in_icmp_v6(struct sk_buff *skb, int *related, unsigned int hooknum) struct ip_vs_iphdr ciph; struct ip_vs_conn *cp; struct ip_vs_protocol *pp; + struct ip_vs_proto_data *pd; unsigned int offset, verdict; union nf_inet_addr snet; struct rt6_info *rt; @@ -1395,9 +1407,11 @@ ip_vs_in_icmp_v6(struct sk_buff *skb, int *related, unsigned int hooknum) if (cih == NULL) return NF_ACCEPT; /* The packet looks wrong, ignore */ - pp = ip_vs_proto_get(cih->nexthdr); - if (!pp) + net = skb_net(skb); + pd = ip_vs_proto_data_get(net, cih->nexthdr); + if (!pd) return NF_ACCEPT; + pp = pd->pp; /* Is the embedded protocol header present? */ /* TODO: we don't support fragmentation at the moment anyways */ @@ -1411,10 +1425,10 @@ ip_vs_in_icmp_v6(struct sk_buff *skb, int *related, unsigned int hooknum) ip_vs_fill_iphdr(AF_INET6, cih, &ciph); /* The embedded headers contain source and dest in reverse order */ - cp = pp->conn_in_get(AF_INET6, skb, pp, &ciph, offset, 1); + cp = pp->conn_in_get(AF_INET6, skb, &ciph, offset, 1); if (!cp) { /* The packet could also belong to a local client */ - cp = pp->conn_out_get(AF_INET6, skb, pp, &ciph, offset, 1); + cp = pp->conn_out_get(AF_INET6, skb, &ciph, offset, 1); if (cp) { ipv6_addr_copy(&snet.in6, &iph->saddr); return handle_response_icmp(AF_INET6, skb, &snet, @@ -1457,8 +1471,10 @@ ip_vs_in_icmp_v6(struct sk_buff *skb, int *related, unsigned int hooknum) static unsigned int ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) { + struct net *net = NULL; struct ip_vs_iphdr iph; struct ip_vs_protocol *pp; + struct ip_vs_proto_data *pd; struct ip_vs_conn *cp; int ret, restart, pkts; @@ -1514,20 +1530,21 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) ip_vs_fill_iphdr(af, skb_network_header(skb), &iph); } + net = skb_net(skb); /* Protocol supported? */ - pp = ip_vs_proto_get(iph.protocol); - if (unlikely(!pp)) + pd = ip_vs_proto_data_get(net, iph.protocol); + if (unlikely(!pd)) return NF_ACCEPT; - + pp = pd->pp; /* * Check if the packet belongs to an existing connection entry */ - cp = pp->conn_in_get(af, skb, pp, &iph, iph.len, 0); + cp = pp->conn_in_get(af, skb, &iph, iph.len, 0); if (unlikely(!cp)) { int v; - if (!pp->conn_schedule(af, skb, pp, &v, &cp)) + if (!pp->conn_schedule(af, skb, pd, &v, &cp)) return v; } @@ -1555,7 +1572,7 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) } ip_vs_in_stats(cp, skb); - restart = ip_vs_set_state(cp, IP_VS_DIR_INPUT, skb, pp); + restart = ip_vs_set_state(cp, IP_VS_DIR_INPUT, skb, pd); if (cp->packet_xmit) ret = cp->packet_xmit(skb, cp, pp); /* do not touch skb anymore */ diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 2d7c96bd2114..88474f1e828a 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -38,6 +38,7 @@ #include #include +#include #include #ifdef CONFIG_IP_VS_IPV6 #include @@ -125,7 +126,7 @@ static int __ip_vs_addr_is_local_v6(const struct in6_addr *addr) * update_defense_level is called from keventd and from sysctl, * so it needs to protect itself from softirqs */ -static void update_defense_level(void) +static void update_defense_level(struct netns_ipvs *ipvs) { struct sysinfo i; static int old_secure_tcp = 0; @@ -239,7 +240,8 @@ static void update_defense_level(void) } old_secure_tcp = sysctl_ip_vs_secure_tcp; if (to_change >= 0) - ip_vs_protocol_timeout_change(sysctl_ip_vs_secure_tcp>1); + ip_vs_protocol_timeout_change(ipvs, + sysctl_ip_vs_secure_tcp > 1); spin_unlock(&ip_vs_securetcp_lock); local_bh_enable(); @@ -255,7 +257,10 @@ static DECLARE_DELAYED_WORK(defense_work, defense_work_handler); static void defense_work_handler(struct work_struct *work) { - update_defense_level(); + struct net *net = &init_net; + struct netns_ipvs *ipvs = net_ipvs(net); + + update_defense_level(ipvs); if (atomic_read(&ip_vs_dropentry)) ip_vs_random_dropentry(); @@ -1502,6 +1507,7 @@ static int proc_do_defense_mode(ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { + struct net *net = current->nsproxy->net_ns; int *valp = table->data; int val = *valp; int rc; @@ -1512,7 +1518,7 @@ proc_do_defense_mode(ctl_table *table, int write, /* Restore the correct value */ *valp = val; } else { - update_defense_level(); + update_defense_level(net_ipvs(net)); } } return rc; @@ -2033,8 +2039,10 @@ static const struct file_operations ip_vs_stats_fops = { /* * Set timeout values for tcp tcpfin udp in the timeout_table. */ -static int ip_vs_set_timeout(struct ip_vs_timeout_user *u) +static int ip_vs_set_timeout(struct net *net, struct ip_vs_timeout_user *u) { + struct ip_vs_proto_data *pd; + IP_VS_DBG(2, "Setting timeout tcp:%d tcpfin:%d udp:%d\n", u->tcp_timeout, u->tcp_fin_timeout, @@ -2042,19 +2050,22 @@ static int ip_vs_set_timeout(struct ip_vs_timeout_user *u) #ifdef CONFIG_IP_VS_PROTO_TCP if (u->tcp_timeout) { - ip_vs_protocol_tcp.timeout_table[IP_VS_TCP_S_ESTABLISHED] + pd = ip_vs_proto_data_get(net, IPPROTO_TCP); + pd->timeout_table[IP_VS_TCP_S_ESTABLISHED] = u->tcp_timeout * HZ; } if (u->tcp_fin_timeout) { - ip_vs_protocol_tcp.timeout_table[IP_VS_TCP_S_FIN_WAIT] + pd = ip_vs_proto_data_get(net, IPPROTO_TCP); + pd->timeout_table[IP_VS_TCP_S_FIN_WAIT] = u->tcp_fin_timeout * HZ; } #endif #ifdef CONFIG_IP_VS_PROTO_UDP if (u->udp_timeout) { - ip_vs_protocol_udp.timeout_table[IP_VS_UDP_S_NORMAL] + pd = ip_vs_proto_data_get(net, IPPROTO_UDP); + pd->timeout_table[IP_VS_UDP_S_NORMAL] = u->udp_timeout * HZ; } #endif @@ -2158,7 +2169,7 @@ do_ip_vs_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) goto out_unlock; } else if (cmd == IP_VS_SO_SET_TIMEOUT) { /* Set timeout values for (tcp tcpfin udp) */ - ret = ip_vs_set_timeout((struct ip_vs_timeout_user *)arg); + ret = ip_vs_set_timeout(net, (struct ip_vs_timeout_user *)arg); goto out_unlock; } else if (cmd == IP_VS_SO_SET_STARTDAEMON) { struct ip_vs_daemon_user *dm = (struct ip_vs_daemon_user *)arg; @@ -2370,17 +2381,19 @@ __ip_vs_get_dest_entries(struct net *net, const struct ip_vs_get_dests *get, } static inline void -__ip_vs_get_timeouts(struct ip_vs_timeout_user *u) +__ip_vs_get_timeouts(struct net *net, struct ip_vs_timeout_user *u) { + struct ip_vs_proto_data *pd; + #ifdef CONFIG_IP_VS_PROTO_TCP - u->tcp_timeout = - ip_vs_protocol_tcp.timeout_table[IP_VS_TCP_S_ESTABLISHED] / HZ; - u->tcp_fin_timeout = - ip_vs_protocol_tcp.timeout_table[IP_VS_TCP_S_FIN_WAIT] / HZ; + pd = ip_vs_proto_data_get(net, IPPROTO_TCP); + u->tcp_timeout = pd->timeout_table[IP_VS_TCP_S_ESTABLISHED] / HZ; + u->tcp_fin_timeout = pd->timeout_table[IP_VS_TCP_S_FIN_WAIT] / HZ; #endif #ifdef CONFIG_IP_VS_PROTO_UDP + pd = ip_vs_proto_data_get(net, IPPROTO_UDP); u->udp_timeout = - ip_vs_protocol_udp.timeout_table[IP_VS_UDP_S_NORMAL] / HZ; + pd->timeout_table[IP_VS_UDP_S_NORMAL] / HZ; #endif } @@ -2521,7 +2534,7 @@ do_ip_vs_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) { struct ip_vs_timeout_user t; - __ip_vs_get_timeouts(&t); + __ip_vs_get_timeouts(net, &t); if (copy_to_user(user, &t, sizeof(t)) != 0) ret = -EFAULT; } @@ -3092,11 +3105,11 @@ static int ip_vs_genl_del_daemon(struct nlattr **attrs) return stop_sync_thread(nla_get_u32(attrs[IPVS_DAEMON_ATTR_STATE])); } -static int ip_vs_genl_set_config(struct nlattr **attrs) +static int ip_vs_genl_set_config(struct net *net, struct nlattr **attrs) { struct ip_vs_timeout_user t; - __ip_vs_get_timeouts(&t); + __ip_vs_get_timeouts(net, &t); if (attrs[IPVS_CMD_ATTR_TIMEOUT_TCP]) t.tcp_timeout = nla_get_u32(attrs[IPVS_CMD_ATTR_TIMEOUT_TCP]); @@ -3108,7 +3121,7 @@ static int ip_vs_genl_set_config(struct nlattr **attrs) if (attrs[IPVS_CMD_ATTR_TIMEOUT_UDP]) t.udp_timeout = nla_get_u32(attrs[IPVS_CMD_ATTR_TIMEOUT_UDP]); - return ip_vs_set_timeout(&t); + return ip_vs_set_timeout(net, &t); } static int ip_vs_genl_set_cmd(struct sk_buff *skb, struct genl_info *info) @@ -3129,7 +3142,7 @@ static int ip_vs_genl_set_cmd(struct sk_buff *skb, struct genl_info *info) ret = ip_vs_flush(net); goto out; } else if (cmd == IPVS_CMD_SET_CONFIG) { - ret = ip_vs_genl_set_config(info->attrs); + ret = ip_vs_genl_set_config(net, info->attrs); goto out; } else if (cmd == IPVS_CMD_NEW_DAEMON || cmd == IPVS_CMD_DEL_DAEMON) { @@ -3281,7 +3294,7 @@ static int ip_vs_genl_get_cmd(struct sk_buff *skb, struct genl_info *info) { struct ip_vs_timeout_user t; - __ip_vs_get_timeouts(&t); + __ip_vs_get_timeouts(net, &t); #ifdef CONFIG_IP_VS_PROTO_TCP NLA_PUT_U32(msg, IPVS_CMD_ATTR_TIMEOUT_TCP, t.tcp_timeout); NLA_PUT_U32(msg, IPVS_CMD_ATTR_TIMEOUT_TCP_FIN, diff --git a/net/netfilter/ipvs/ip_vs_proto.c b/net/netfilter/ipvs/ip_vs_proto.c index 9f609d4d5d58..6ac986cdcff3 100644 --- a/net/netfilter/ipvs/ip_vs_proto.c +++ b/net/netfilter/ipvs/ip_vs_proto.c @@ -152,9 +152,8 @@ EXPORT_SYMBOL(ip_vs_proto_get); * get ip_vs_protocol object data by netns and proto */ struct ip_vs_proto_data * -ip_vs_proto_data_get(struct net *net, unsigned short proto) +__ipvs_proto_data_get(struct netns_ipvs *ipvs, unsigned short proto) { - struct netns_ipvs *ipvs = net_ipvs(net); struct ip_vs_proto_data *pd; unsigned hash = IP_VS_PROTO_HASH(proto); @@ -165,20 +164,28 @@ ip_vs_proto_data_get(struct net *net, unsigned short proto) return NULL; } + +struct ip_vs_proto_data * +ip_vs_proto_data_get(struct net *net, unsigned short proto) +{ + struct netns_ipvs *ipvs = net_ipvs(net); + + return __ipvs_proto_data_get(ipvs, proto); +} EXPORT_SYMBOL(ip_vs_proto_data_get); /* * Propagate event for state change to all protocols */ -void ip_vs_protocol_timeout_change(int flags) +void ip_vs_protocol_timeout_change(struct netns_ipvs *ipvs, int flags) { - struct ip_vs_protocol *pp; + struct ip_vs_proto_data *pd; int i; for (i = 0; i < IP_VS_PROTO_TAB_SIZE; i++) { - for (pp = ip_vs_proto_table[i]; pp; pp = pp->next) { - if (pp->timeout_change) - pp->timeout_change(pp, flags); + for (pd = ipvs->proto_data_table[i]; pd; pd = pd->next) { + if (pd->pp->timeout_change) + pd->pp->timeout_change(pd, flags); } } } diff --git a/net/netfilter/ipvs/ip_vs_proto_ah_esp.c b/net/netfilter/ipvs/ip_vs_proto_ah_esp.c index b8b37fafc988..28039cbfcff4 100644 --- a/net/netfilter/ipvs/ip_vs_proto_ah_esp.c +++ b/net/netfilter/ipvs/ip_vs_proto_ah_esp.c @@ -55,7 +55,7 @@ ah_esp_conn_fill_param_proto(int af, const struct ip_vs_iphdr *iph, } static struct ip_vs_conn * -ah_esp_conn_in_get(int af, const struct sk_buff *skb, struct ip_vs_protocol *pp, +ah_esp_conn_in_get(int af, const struct sk_buff *skb, const struct ip_vs_iphdr *iph, unsigned int proto_off, int inverse) { @@ -72,7 +72,7 @@ ah_esp_conn_in_get(int af, const struct sk_buff *skb, struct ip_vs_protocol *pp, IP_VS_DBG_BUF(12, "Unknown ISAKMP entry for outin packet " "%s%s %s->%s\n", inverse ? "ICMP+" : "", - pp->name, + ip_vs_proto_get(iph->protocol)->name, IP_VS_DBG_ADDR(af, &iph->saddr), IP_VS_DBG_ADDR(af, &iph->daddr)); } @@ -83,7 +83,6 @@ ah_esp_conn_in_get(int af, const struct sk_buff *skb, struct ip_vs_protocol *pp, static struct ip_vs_conn * ah_esp_conn_out_get(int af, const struct sk_buff *skb, - struct ip_vs_protocol *pp, const struct ip_vs_iphdr *iph, unsigned int proto_off, int inverse) @@ -97,7 +96,7 @@ ah_esp_conn_out_get(int af, const struct sk_buff *skb, IP_VS_DBG_BUF(12, "Unknown ISAKMP entry for inout packet " "%s%s %s->%s\n", inverse ? "ICMP+" : "", - pp->name, + ip_vs_proto_get(iph->protocol)->name, IP_VS_DBG_ADDR(af, &iph->saddr), IP_VS_DBG_ADDR(af, &iph->daddr)); } @@ -107,7 +106,7 @@ ah_esp_conn_out_get(int af, const struct sk_buff *skb, static int -ah_esp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, +ah_esp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, int *verdict, struct ip_vs_conn **cpp) { /* @@ -137,7 +136,6 @@ struct ip_vs_protocol ip_vs_protocol_ah = { .app_conn_bind = NULL, .debug_packet = ip_vs_tcpudp_debug_packet, .timeout_change = NULL, /* ISAKMP */ - .set_state_timeout = NULL, }; #endif diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c index f826dd1e4630..19bc37976ea7 100644 --- a/net/netfilter/ipvs/ip_vs_proto_sctp.c +++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c @@ -9,7 +9,7 @@ #include static int -sctp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, +sctp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, int *verdict, struct ip_vs_conn **cpp) { struct net *net; @@ -47,10 +47,10 @@ sctp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, * Let the virtual server select a real server for the * incoming connection, and create a connection entry. */ - *cpp = ip_vs_schedule(svc, skb, pp, &ignored); + *cpp = ip_vs_schedule(svc, skb, pd, &ignored); if (!*cpp && ignored <= 0) { if (!ignored) - *verdict = ip_vs_leave(svc, skb, pp); + *verdict = ip_vs_leave(svc, skb, pd); else { ip_vs_service_put(svc); *verdict = NF_DROP; @@ -907,14 +907,13 @@ static const char *sctp_state_name(int state) } static inline int -set_sctp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, +set_sctp_state(struct ip_vs_proto_data *pd, struct ip_vs_conn *cp, int direction, const struct sk_buff *skb) { sctp_chunkhdr_t _sctpch, *sch; unsigned char chunk_type; int event, next_state; int ihl; - struct ip_vs_proto_data *pd; #ifdef CONFIG_IP_VS_IPV6 ihl = cp->af == AF_INET ? ip_hdrlen(skb) : sizeof(struct ipv6hdr); @@ -966,7 +965,7 @@ set_sctp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, IP_VS_DBG_BUF(8, "%s %s %s:%d->" "%s:%d state: %s->%s conn->refcnt:%d\n", - pp->name, + pd->pp->name, ((direction == IP_VS_DIR_OUTPUT) ? "output " : "input "), IP_VS_DBG_ADDR(cp->af, &cp->daddr), @@ -990,7 +989,6 @@ set_sctp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, } } } - pd = ip_vs_proto_data_get(&init_net, pp->protocol); /* tmp fix */ if (likely(pd)) cp->timeout = pd->timeout_table[cp->state = next_state]; else /* What to do ? */ @@ -1001,12 +999,12 @@ set_sctp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, static int sctp_state_transition(struct ip_vs_conn *cp, int direction, - const struct sk_buff *skb, struct ip_vs_protocol *pp) + const struct sk_buff *skb, struct ip_vs_proto_data *pd) { int ret = 0; spin_lock(&cp->lock); - ret = set_sctp_state(pp, cp, direction, skb); + ret = set_sctp_state(pd, cp, direction, skb); spin_unlock(&cp->lock); return ret; diff --git a/net/netfilter/ipvs/ip_vs_proto_tcp.c b/net/netfilter/ipvs/ip_vs_proto_tcp.c index 9d9df3d61093..d7c245532798 100644 --- a/net/netfilter/ipvs/ip_vs_proto_tcp.c +++ b/net/netfilter/ipvs/ip_vs_proto_tcp.c @@ -32,7 +32,7 @@ #include static int -tcp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, +tcp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, int *verdict, struct ip_vs_conn **cpp) { struct net *net; @@ -68,10 +68,10 @@ tcp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, * Let the virtual server select a real server for the * incoming connection, and create a connection entry. */ - *cpp = ip_vs_schedule(svc, skb, pp, &ignored); + *cpp = ip_vs_schedule(svc, skb, pd, &ignored); if (!*cpp && ignored <= 0) { if (!ignored) - *verdict = ip_vs_leave(svc, skb, pp); + *verdict = ip_vs_leave(svc, skb, pd); else { ip_vs_service_put(svc); *verdict = NF_DROP; @@ -448,10 +448,7 @@ static struct tcp_states_t tcp_states_dos [] = { /*rst*/ {{sCL, sCL, sCL, sSR, sCL, sCL, sCL, sCL, sLA, sLI, sCL }}, }; -static struct tcp_states_t *tcp_state_table = tcp_states; - - -static void tcp_timeout_change(struct ip_vs_protocol *pp, int flags) +static void tcp_timeout_change(struct ip_vs_proto_data *pd, int flags) { int on = (flags & 1); /* secure_tcp */ @@ -461,7 +458,7 @@ static void tcp_timeout_change(struct ip_vs_protocol *pp, int flags) ** for most if not for all of the applications. Something ** like "capabilities" (flags) for each object. */ - tcp_state_table = (on? tcp_states_dos : tcp_states); + pd->tcp_state_table = (on ? tcp_states_dos : tcp_states); } static inline int tcp_state_idx(struct tcphdr *th) @@ -478,13 +475,12 @@ static inline int tcp_state_idx(struct tcphdr *th) } static inline void -set_tcp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, +set_tcp_state(struct ip_vs_proto_data *pd, struct ip_vs_conn *cp, int direction, struct tcphdr *th) { int state_idx; int new_state = IP_VS_TCP_S_CLOSE; int state_off = tcp_state_off[direction]; - struct ip_vs_proto_data *pd; /* Temp fix */ /* * Update state offset to INPUT_ONLY if necessary @@ -502,7 +498,8 @@ set_tcp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, goto tcp_state_out; } - new_state = tcp_state_table[state_off+state_idx].next_state[cp->state]; + new_state = + pd->tcp_state_table[state_off+state_idx].next_state[cp->state]; tcp_state_out: if (new_state != cp->state) { @@ -510,7 +507,7 @@ set_tcp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, IP_VS_DBG_BUF(8, "%s %s [%c%c%c%c] %s:%d->" "%s:%d state: %s->%s conn->refcnt:%d\n", - pp->name, + pd->pp->name, ((state_off == TCP_DIR_OUTPUT) ? "output " : "input "), th->syn ? 'S' : '.', @@ -540,7 +537,6 @@ set_tcp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, } } - pd = ip_vs_proto_data_get(&init_net, pp->protocol); if (likely(pd)) cp->timeout = pd->timeout_table[cp->state = new_state]; else /* What to do ? */ @@ -553,7 +549,7 @@ set_tcp_state(struct ip_vs_protocol *pp, struct ip_vs_conn *cp, static int tcp_state_transition(struct ip_vs_conn *cp, int direction, const struct sk_buff *skb, - struct ip_vs_protocol *pp) + struct ip_vs_proto_data *pd) { struct tcphdr _tcph, *th; @@ -568,7 +564,7 @@ tcp_state_transition(struct ip_vs_conn *cp, int direction, return 0; spin_lock(&cp->lock); - set_tcp_state(pp, cp, direction, th); + set_tcp_state(pd, cp, direction, th); spin_unlock(&cp->lock); return 1; @@ -691,6 +687,7 @@ static void __ip_vs_tcp_init(struct net *net, struct ip_vs_proto_data *pd) spin_lock_init(&ipvs->tcp_app_lock); pd->timeout_table = ip_vs_create_timeout_table((int *)tcp_timeouts, sizeof(tcp_timeouts)); + pd->tcp_state_table = tcp_states; } static void __ip_vs_tcp_exit(struct net *net, struct ip_vs_proto_data *pd) diff --git a/net/netfilter/ipvs/ip_vs_proto_udp.c b/net/netfilter/ipvs/ip_vs_proto_udp.c index 71a4721a8f8a..aa85df2f14a0 100644 --- a/net/netfilter/ipvs/ip_vs_proto_udp.c +++ b/net/netfilter/ipvs/ip_vs_proto_udp.c @@ -29,7 +29,7 @@ #include static int -udp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, +udp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, int *verdict, struct ip_vs_conn **cpp) { struct net *net; @@ -64,10 +64,10 @@ udp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, * Let the virtual server select a real server for the * incoming connection, and create a connection entry. */ - *cpp = ip_vs_schedule(svc, skb, pp, &ignored); + *cpp = ip_vs_schedule(svc, skb, pd, &ignored); if (!*cpp && ignored <= 0) { if (!ignored) - *verdict = ip_vs_leave(svc, skb, pp); + *verdict = ip_vs_leave(svc, skb, pd); else { ip_vs_service_put(svc); *verdict = NF_DROP; @@ -457,11 +457,8 @@ static const char * udp_state_name(int state) static int udp_state_transition(struct ip_vs_conn *cp, int direction, const struct sk_buff *skb, - struct ip_vs_protocol *pp) + struct ip_vs_proto_data *pd) { - struct ip_vs_proto_data *pd; /* Temp fix, pp will be replaced by pd */ - - pd = ip_vs_proto_data_get(&init_net, IPPROTO_UDP); if (unlikely(!pd)) { pr_err("UDP no ns data\n"); return 0; diff --git a/net/netfilter/xt_ipvs.c b/net/netfilter/xt_ipvs.c index 9127a3d8aa35..bb10b0717f1b 100644 --- a/net/netfilter/xt_ipvs.c +++ b/net/netfilter/xt_ipvs.c @@ -85,7 +85,7 @@ ipvs_mt(const struct sk_buff *skb, struct xt_action_param *par) /* * Check if the packet belongs to an existing entry */ - cp = pp->conn_out_get(family, skb, pp, &iph, iph.len, 1 /* inverse */); + cp = pp->conn_out_get(family, skb, &iph, iph.len, 1 /* inverse */); if (unlikely(cp == NULL)) { match = false; goto out; -- cgit v1.2.3 From 9bbac6a904d0816dae58b454692c54d6773cc20d Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:52 +0100 Subject: IPVS: netns, common protocol changes and use of appcnt. appcnt and timeout_table moved from struct ip_vs_protocol to ip_vs proto_data. struct net *net added as first param to - register_app() - unregister_app() - app_conn_bind() - ip_vs_conn_new() [horms@verge.net.au: removed cosmetic-change-only hunk] Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_conn.c | 6 ++-- net/netfilter/ipvs/ip_vs_proto_sctp.c | 4 +-- net/netfilter/ipvs/ip_vs_proto_tcp.c | 5 ++-- net/netfilter/ipvs/ip_vs_proto_udp.c | 4 +-- net/netfilter/ipvs/ip_vs_sync.c | 55 +++++++++++++++++++---------------- 5 files changed, 39 insertions(+), 35 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index a7aba6a4697e..b2024c942345 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -804,7 +804,7 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p, struct ip_vs_dest *dest, __u32 fwmark) { struct ip_vs_conn *cp; - struct ip_vs_protocol *pp = ip_vs_proto_get(p->protocol); + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, p->protocol); cp = kmem_cache_zalloc(ip_vs_conn_cachep, GFP_ATOMIC); if (cp == NULL) { @@ -863,8 +863,8 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p, #endif ip_vs_bind_xmit(cp); - if (unlikely(pp && atomic_read(&pp->appcnt))) - ip_vs_bind_app(cp, pp); + if (unlikely(pd && atomic_read(&pd->appcnt))) + ip_vs_bind_app(cp, pd->pp); /* * Allow conntrack to be preserved. By default, conntrack diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c index 19bc37976ea7..0f14f793318a 100644 --- a/net/netfilter/ipvs/ip_vs_proto_sctp.c +++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c @@ -1035,7 +1035,7 @@ static int sctp_register_app(struct ip_vs_app *inc) } } list_add(&inc->p_list, &ipvs->sctp_apps[hash]); - atomic_inc(&pd->pp->appcnt); + atomic_inc(&pd->appcnt); out: spin_unlock_bh(&ipvs->sctp_app_lock); @@ -1048,7 +1048,7 @@ static void sctp_unregister_app(struct ip_vs_app *inc) struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, IPPROTO_SCTP); spin_lock_bh(&ipvs->sctp_app_lock); - atomic_dec(&pd->pp->appcnt); + atomic_dec(&pd->appcnt); list_del(&inc->p_list); spin_unlock_bh(&ipvs->sctp_app_lock); } diff --git a/net/netfilter/ipvs/ip_vs_proto_tcp.c b/net/netfilter/ipvs/ip_vs_proto_tcp.c index d7c245532798..290b3803d8ce 100644 --- a/net/netfilter/ipvs/ip_vs_proto_tcp.c +++ b/net/netfilter/ipvs/ip_vs_proto_tcp.c @@ -596,7 +596,7 @@ static int tcp_register_app(struct ip_vs_app *inc) } } list_add(&inc->p_list, &ipvs->tcp_apps[hash]); - atomic_inc(&pd->pp->appcnt); + atomic_inc(&pd->appcnt); out: spin_unlock_bh(&ipvs->tcp_app_lock); @@ -611,7 +611,7 @@ tcp_unregister_app(struct ip_vs_app *inc) struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, IPPROTO_TCP); spin_lock_bh(&ipvs->tcp_app_lock); - atomic_dec(&pd->pp->appcnt); + atomic_dec(&pd->appcnt); list_del(&inc->p_list); spin_unlock_bh(&ipvs->tcp_app_lock); } @@ -701,7 +701,6 @@ struct ip_vs_protocol ip_vs_protocol_tcp = { .protocol = IPPROTO_TCP, .num_states = IP_VS_TCP_S_LAST, .dont_defrag = 0, - .appcnt = ATOMIC_INIT(0), .init = NULL, .exit = NULL, .init_netns = __ip_vs_tcp_init, diff --git a/net/netfilter/ipvs/ip_vs_proto_udp.c b/net/netfilter/ipvs/ip_vs_proto_udp.c index aa85df2f14a0..3719837a8fdc 100644 --- a/net/netfilter/ipvs/ip_vs_proto_udp.c +++ b/net/netfilter/ipvs/ip_vs_proto_udp.c @@ -373,7 +373,7 @@ static int udp_register_app(struct ip_vs_app *inc) } } list_add(&inc->p_list, &ipvs->udp_apps[hash]); - atomic_inc(&pd->pp->appcnt); + atomic_inc(&pd->appcnt); out: spin_unlock_bh(&ipvs->udp_app_lock); @@ -388,7 +388,7 @@ udp_unregister_app(struct ip_vs_app *inc) struct netns_ipvs *ipvs = net_ipvs(&init_net); spin_lock_bh(&ipvs->udp_app_lock); - atomic_dec(&pd->pp->appcnt); + atomic_dec(&pd->appcnt); list_del(&inc->p_list); spin_unlock_bh(&ipvs->udp_app_lock); } diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index 662aa2c22a05..6831e8fac8db 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -725,17 +725,16 @@ ip_vs_conn_fill_param_sync(int af, union ip_vs_sync_conn *sc, * Param: ... * timeout is in sec. */ -static void ip_vs_proc_conn(struct ip_vs_conn_param *param, unsigned flags, - unsigned state, unsigned protocol, unsigned type, +static void ip_vs_proc_conn(struct net *net, struct ip_vs_conn_param *param, + unsigned int flags, unsigned int state, + unsigned int protocol, unsigned int type, const union nf_inet_addr *daddr, __be16 dport, unsigned long timeout, __u32 fwmark, - struct ip_vs_sync_conn_options *opt, - struct ip_vs_protocol *pp) + struct ip_vs_sync_conn_options *opt) { struct ip_vs_dest *dest; struct ip_vs_conn *cp; - if (!(flags & IP_VS_CONN_F_TEMPLATE)) cp = ip_vs_conn_in_get(param); else @@ -821,17 +820,23 @@ static void ip_vs_proc_conn(struct ip_vs_conn_param *param, unsigned flags, if (timeout > MAX_SCHEDULE_TIMEOUT / HZ) timeout = MAX_SCHEDULE_TIMEOUT / HZ; cp->timeout = timeout*HZ; - } else if (!(flags & IP_VS_CONN_F_TEMPLATE) && pp->timeout_table) - cp->timeout = pp->timeout_table[state]; - else - cp->timeout = (3*60*HZ); + } else { + struct ip_vs_proto_data *pd; + + pd = ip_vs_proto_data_get(net, protocol); + if (!(flags & IP_VS_CONN_F_TEMPLATE) && pd && pd->timeout_table) + cp->timeout = pd->timeout_table[state]; + else + cp->timeout = (3*60*HZ); + } ip_vs_conn_put(cp); } /* * Process received multicast message for Version 0 */ -static void ip_vs_process_message_v0(const char *buffer, const size_t buflen) +static void ip_vs_process_message_v0(struct net *net, const char *buffer, + const size_t buflen) { struct ip_vs_sync_mesg_v0 *m = (struct ip_vs_sync_mesg_v0 *)buffer; struct ip_vs_sync_conn_v0 *s; @@ -879,7 +884,6 @@ static void ip_vs_process_message_v0(const char *buffer, const size_t buflen) } } else { /* protocol in templates is not used for state/timeout */ - pp = NULL; if (state > 0) { IP_VS_DBG(2, "BACKUP v0, Invalid template state %u\n", state); @@ -894,9 +898,9 @@ static void ip_vs_process_message_v0(const char *buffer, const size_t buflen) s->vport, ¶m); /* Send timeout as Zero */ - ip_vs_proc_conn(¶m, flags, state, s->protocol, AF_INET, + ip_vs_proc_conn(net, ¶m, flags, state, s->protocol, AF_INET, (union nf_inet_addr *)&s->daddr, s->dport, - 0, 0, opt, pp); + 0, 0, opt); } } @@ -945,7 +949,7 @@ static int ip_vs_proc_str(__u8 *p, unsigned int plen, unsigned int *data_len, /* * Process a Version 1 sync. connection */ -static inline int ip_vs_proc_sync_conn(__u8 *p, __u8 *msg_end) +static inline int ip_vs_proc_sync_conn(struct net *net, __u8 *p, __u8 *msg_end) { struct ip_vs_sync_conn_options opt; union ip_vs_sync_conn *s; @@ -1043,7 +1047,6 @@ static inline int ip_vs_proc_sync_conn(__u8 *p, __u8 *msg_end) } } else { /* protocol in templates is not used for state/timeout */ - pp = NULL; if (state > 0) { IP_VS_DBG(3, "BACKUP, Invalid template state %u\n", state); @@ -1058,18 +1061,18 @@ static inline int ip_vs_proc_sync_conn(__u8 *p, __u8 *msg_end) } /* If only IPv4, just silent skip IPv6 */ if (af == AF_INET) - ip_vs_proc_conn(¶m, flags, state, s->v4.protocol, af, + ip_vs_proc_conn(net, ¶m, flags, state, s->v4.protocol, af, (union nf_inet_addr *)&s->v4.daddr, s->v4.dport, ntohl(s->v4.timeout), ntohl(s->v4.fwmark), - (opt_flags & IPVS_OPT_F_SEQ_DATA ? &opt : NULL), - pp); + (opt_flags & IPVS_OPT_F_SEQ_DATA ? &opt : NULL) + ); #ifdef CONFIG_IP_VS_IPV6 else - ip_vs_proc_conn(¶m, flags, state, s->v6.protocol, af, + ip_vs_proc_conn(net, ¶m, flags, state, s->v6.protocol, af, (union nf_inet_addr *)&s->v6.daddr, s->v6.dport, ntohl(s->v6.timeout), ntohl(s->v6.fwmark), - (opt_flags & IPVS_OPT_F_SEQ_DATA ? &opt : NULL), - pp); + (opt_flags & IPVS_OPT_F_SEQ_DATA ? &opt : NULL) + ); #endif return 0; /* Error exit */ @@ -1083,7 +1086,8 @@ out: * ip_vs_conn entries. * Handles Version 0 & 1 */ -static void ip_vs_process_message(__u8 *buffer, const size_t buflen) +static void ip_vs_process_message(struct net *net, __u8 *buffer, + const size_t buflen) { struct ip_vs_sync_mesg *m2 = (struct ip_vs_sync_mesg *)buffer; __u8 *p, *msg_end; @@ -1136,7 +1140,8 @@ static void ip_vs_process_message(__u8 *buffer, const size_t buflen) return; } /* Process a single sync_conn */ - if ((retc=ip_vs_proc_sync_conn(p, msg_end)) < 0) { + retc = ip_vs_proc_sync_conn(net, p, msg_end); + if (retc < 0) { IP_VS_ERR_RL("BACKUP, Dropping buffer, Err: %d in decoding\n", retc); return; @@ -1146,7 +1151,7 @@ static void ip_vs_process_message(__u8 *buffer, const size_t buflen) } } else { /* Old type of message */ - ip_vs_process_message_v0(buffer, buflen); + ip_vs_process_message_v0(net, buffer, buflen); return; } } @@ -1500,7 +1505,7 @@ static int sync_thread_backup(void *data) /* disable bottom half, because it accesses the data shared by softirq while getting/creating conns */ local_bh_disable(); - ip_vs_process_message(tinfo->buf, len); + ip_vs_process_message(&init_net, tinfo->buf, len); local_bh_enable(); } } -- cgit v1.2.3 From ab8a5e8408c3df2d654611bffc3aaf04f418b266 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:53 +0100 Subject: IPVS: netns awareness to ip_vs_app All variables moved to struct ipvs, most external changes fixed (i.e. init_net removed) in ip_vs_protocol param struct net *net added to: - register_app() - unregister_app() This affected almost all proto_xxx.c files Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_app.c | 73 +++++++++++++++++++++-------------- net/netfilter/ipvs/ip_vs_ftp.c | 8 ++-- net/netfilter/ipvs/ip_vs_proto_sctp.c | 12 +++--- net/netfilter/ipvs/ip_vs_proto_tcp.c | 12 +++--- net/netfilter/ipvs/ip_vs_proto_udp.c | 12 +++--- 5 files changed, 65 insertions(+), 52 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_app.c b/net/netfilter/ipvs/ip_vs_app.c index 40b09ccc4896..286f46594e0e 100644 --- a/net/netfilter/ipvs/ip_vs_app.c +++ b/net/netfilter/ipvs/ip_vs_app.c @@ -43,11 +43,6 @@ EXPORT_SYMBOL(register_ip_vs_app); EXPORT_SYMBOL(unregister_ip_vs_app); EXPORT_SYMBOL(register_ip_vs_app_inc); -/* ipvs application list head */ -static LIST_HEAD(ip_vs_app_list); -static DEFINE_MUTEX(__ip_vs_app_mutex); - - /* * Get an ip_vs_app object */ @@ -67,7 +62,8 @@ static inline void ip_vs_app_put(struct ip_vs_app *app) * Allocate/initialize app incarnation and register it in proto apps. */ static int -ip_vs_app_inc_new(struct ip_vs_app *app, __u16 proto, __u16 port) +ip_vs_app_inc_new(struct net *net, struct ip_vs_app *app, __u16 proto, + __u16 port) { struct ip_vs_protocol *pp; struct ip_vs_app *inc; @@ -98,7 +94,7 @@ ip_vs_app_inc_new(struct ip_vs_app *app, __u16 proto, __u16 port) } } - ret = pp->register_app(inc); + ret = pp->register_app(net, inc); if (ret) goto out; @@ -119,7 +115,7 @@ ip_vs_app_inc_new(struct ip_vs_app *app, __u16 proto, __u16 port) * Release app incarnation */ static void -ip_vs_app_inc_release(struct ip_vs_app *inc) +ip_vs_app_inc_release(struct net *net, struct ip_vs_app *inc) { struct ip_vs_protocol *pp; @@ -127,7 +123,7 @@ ip_vs_app_inc_release(struct ip_vs_app *inc) return; if (pp->unregister_app) - pp->unregister_app(inc); + pp->unregister_app(net, inc); IP_VS_DBG(9, "%s App %s:%u unregistered\n", pp->name, inc->name, ntohs(inc->port)); @@ -168,15 +164,17 @@ void ip_vs_app_inc_put(struct ip_vs_app *inc) * Register an application incarnation in protocol applications */ int -register_ip_vs_app_inc(struct ip_vs_app *app, __u16 proto, __u16 port) +register_ip_vs_app_inc(struct net *net, struct ip_vs_app *app, __u16 proto, + __u16 port) { + struct netns_ipvs *ipvs = net_ipvs(net); int result; - mutex_lock(&__ip_vs_app_mutex); + mutex_lock(&ipvs->app_mutex); - result = ip_vs_app_inc_new(app, proto, port); + result = ip_vs_app_inc_new(net, app, proto, port); - mutex_unlock(&__ip_vs_app_mutex); + mutex_unlock(&ipvs->app_mutex); return result; } @@ -185,16 +183,17 @@ register_ip_vs_app_inc(struct ip_vs_app *app, __u16 proto, __u16 port) /* * ip_vs_app registration routine */ -int register_ip_vs_app(struct ip_vs_app *app) +int register_ip_vs_app(struct net *net, struct ip_vs_app *app) { + struct netns_ipvs *ipvs = net_ipvs(net); /* increase the module use count */ ip_vs_use_count_inc(); - mutex_lock(&__ip_vs_app_mutex); + mutex_lock(&ipvs->app_mutex); - list_add(&app->a_list, &ip_vs_app_list); + list_add(&app->a_list, &ipvs->app_list); - mutex_unlock(&__ip_vs_app_mutex); + mutex_unlock(&ipvs->app_mutex); return 0; } @@ -204,19 +203,20 @@ int register_ip_vs_app(struct ip_vs_app *app) * ip_vs_app unregistration routine * We are sure there are no app incarnations attached to services */ -void unregister_ip_vs_app(struct ip_vs_app *app) +void unregister_ip_vs_app(struct net *net, struct ip_vs_app *app) { + struct netns_ipvs *ipvs = net_ipvs(net); struct ip_vs_app *inc, *nxt; - mutex_lock(&__ip_vs_app_mutex); + mutex_lock(&ipvs->app_mutex); list_for_each_entry_safe(inc, nxt, &app->incs_list, a_list) { - ip_vs_app_inc_release(inc); + ip_vs_app_inc_release(net, inc); } list_del(&app->a_list); - mutex_unlock(&__ip_vs_app_mutex); + mutex_unlock(&ipvs->app_mutex); /* decrease the module use count */ ip_vs_use_count_dec(); @@ -226,7 +226,8 @@ void unregister_ip_vs_app(struct ip_vs_app *app) /* * Bind ip_vs_conn to its ip_vs_app (called by cp constructor) */ -int ip_vs_bind_app(struct ip_vs_conn *cp, struct ip_vs_protocol *pp) +int ip_vs_bind_app(struct ip_vs_conn *cp, + struct ip_vs_protocol *pp) { return pp->app_conn_bind(cp); } @@ -481,11 +482,11 @@ int ip_vs_app_pkt_in(struct ip_vs_conn *cp, struct sk_buff *skb) * /proc/net/ip_vs_app entry function */ -static struct ip_vs_app *ip_vs_app_idx(loff_t pos) +static struct ip_vs_app *ip_vs_app_idx(struct netns_ipvs *ipvs, loff_t pos) { struct ip_vs_app *app, *inc; - list_for_each_entry(app, &ip_vs_app_list, a_list) { + list_for_each_entry(app, &ipvs->app_list, a_list) { list_for_each_entry(inc, &app->incs_list, a_list) { if (pos-- == 0) return inc; @@ -497,19 +498,24 @@ static struct ip_vs_app *ip_vs_app_idx(loff_t pos) static void *ip_vs_app_seq_start(struct seq_file *seq, loff_t *pos) { - mutex_lock(&__ip_vs_app_mutex); + struct net *net = seq_file_net(seq); + struct netns_ipvs *ipvs = net_ipvs(net); + + mutex_lock(&ipvs->app_mutex); - return *pos ? ip_vs_app_idx(*pos - 1) : SEQ_START_TOKEN; + return *pos ? ip_vs_app_idx(ipvs, *pos - 1) : SEQ_START_TOKEN; } static void *ip_vs_app_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct ip_vs_app *inc, *app; struct list_head *e; + struct net *net = seq_file_net(seq); + struct netns_ipvs *ipvs = net_ipvs(net); ++*pos; if (v == SEQ_START_TOKEN) - return ip_vs_app_idx(0); + return ip_vs_app_idx(ipvs, 0); inc = v; app = inc->app; @@ -518,7 +524,7 @@ static void *ip_vs_app_seq_next(struct seq_file *seq, void *v, loff_t *pos) return list_entry(e, struct ip_vs_app, a_list); /* go on to next application */ - for (e = app->a_list.next; e != &ip_vs_app_list; e = e->next) { + for (e = app->a_list.next; e != &ipvs->app_list; e = e->next) { app = list_entry(e, struct ip_vs_app, a_list); list_for_each_entry(inc, &app->incs_list, a_list) { return inc; @@ -529,7 +535,9 @@ static void *ip_vs_app_seq_next(struct seq_file *seq, void *v, loff_t *pos) static void ip_vs_app_seq_stop(struct seq_file *seq, void *v) { - mutex_unlock(&__ip_vs_app_mutex); + struct netns_ipvs *ipvs = net_ipvs(seq_file_net(seq)); + + mutex_unlock(&ipvs->app_mutex); } static int ip_vs_app_seq_show(struct seq_file *seq, void *v) @@ -557,7 +565,8 @@ static const struct seq_operations ip_vs_app_seq_ops = { static int ip_vs_app_open(struct inode *inode, struct file *file) { - return seq_open(file, &ip_vs_app_seq_ops); + return seq_open_net(inode, file, &ip_vs_app_seq_ops, + sizeof(struct seq_net_private)); } static const struct file_operations ip_vs_app_fops = { @@ -571,9 +580,13 @@ static const struct file_operations ip_vs_app_fops = { static int __net_init __ip_vs_app_init(struct net *net) { + struct netns_ipvs *ipvs = net_ipvs(net); + if (!net_eq(net, &init_net)) /* netns not enabled yet */ return -EPERM; + INIT_LIST_HEAD(&ipvs->app_list); + __mutex_init(&ipvs->app_mutex, "ipvs->app_mutex", &ipvs->app_key); proc_net_fops_create(net, "ip_vs_app", 0, &ip_vs_app_fops); return 0; } diff --git a/net/netfilter/ipvs/ip_vs_ftp.c b/net/netfilter/ipvs/ip_vs_ftp.c index b38ae941f677..77b0036dcb73 100644 --- a/net/netfilter/ipvs/ip_vs_ftp.c +++ b/net/netfilter/ipvs/ip_vs_ftp.c @@ -414,14 +414,14 @@ static int __net_init __ip_vs_ftp_init(struct net *net) if (!net_eq(net, &init_net)) /* netns not enabled yet */ return -EPERM; - ret = register_ip_vs_app(app); + ret = register_ip_vs_app(net, app); if (ret) return ret; for (i=0; iprotocol, ports[i]); + ret = register_ip_vs_app_inc(net, app, app->protocol, ports[i]); if (ret) break; pr_info("%s: loaded support on port[%d] = %d\n", @@ -429,7 +429,7 @@ static int __net_init __ip_vs_ftp_init(struct net *net) } if (ret) - unregister_ip_vs_app(app); + unregister_ip_vs_app(net, app); return ret; } @@ -443,7 +443,7 @@ static void __ip_vs_ftp_exit(struct net *net) if (!net_eq(net, &init_net)) /* netns not enabled yet */ return; - unregister_ip_vs_app(app); + unregister_ip_vs_app(net, app); } static struct pernet_operations ip_vs_ftp_ops = { diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c index 0f14f793318a..569e77bf08c4 100644 --- a/net/netfilter/ipvs/ip_vs_proto_sctp.c +++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c @@ -1016,14 +1016,14 @@ static inline __u16 sctp_app_hashkey(__be16 port) & SCTP_APP_TAB_MASK; } -static int sctp_register_app(struct ip_vs_app *inc) +static int sctp_register_app(struct net *net, struct ip_vs_app *inc) { struct ip_vs_app *i; __u16 hash; __be16 port = inc->port; int ret = 0; - struct netns_ipvs *ipvs = net_ipvs(&init_net); - struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, IPPROTO_SCTP); + struct netns_ipvs *ipvs = net_ipvs(net); + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(net, IPPROTO_SCTP); hash = sctp_app_hashkey(port); @@ -1042,10 +1042,10 @@ out: return ret; } -static void sctp_unregister_app(struct ip_vs_app *inc) +static void sctp_unregister_app(struct net *net, struct ip_vs_app *inc) { - struct netns_ipvs *ipvs = net_ipvs(&init_net); - struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, IPPROTO_SCTP); + struct netns_ipvs *ipvs = net_ipvs(net); + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(net, IPPROTO_SCTP); spin_lock_bh(&ipvs->sctp_app_lock); atomic_dec(&pd->appcnt); diff --git a/net/netfilter/ipvs/ip_vs_proto_tcp.c b/net/netfilter/ipvs/ip_vs_proto_tcp.c index 290b3803d8ce..757aaaf083bb 100644 --- a/net/netfilter/ipvs/ip_vs_proto_tcp.c +++ b/net/netfilter/ipvs/ip_vs_proto_tcp.c @@ -577,14 +577,14 @@ static inline __u16 tcp_app_hashkey(__be16 port) } -static int tcp_register_app(struct ip_vs_app *inc) +static int tcp_register_app(struct net *net, struct ip_vs_app *inc) { struct ip_vs_app *i; __u16 hash; __be16 port = inc->port; int ret = 0; - struct netns_ipvs *ipvs = net_ipvs(&init_net); - struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, IPPROTO_TCP); + struct netns_ipvs *ipvs = net_ipvs(net); + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(net, IPPROTO_TCP); hash = tcp_app_hashkey(port); @@ -605,10 +605,10 @@ static int tcp_register_app(struct ip_vs_app *inc) static void -tcp_unregister_app(struct ip_vs_app *inc) +tcp_unregister_app(struct net *net, struct ip_vs_app *inc) { - struct netns_ipvs *ipvs = net_ipvs(&init_net); - struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, IPPROTO_TCP); + struct netns_ipvs *ipvs = net_ipvs(net); + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(net, IPPROTO_TCP); spin_lock_bh(&ipvs->tcp_app_lock); atomic_dec(&pd->appcnt); diff --git a/net/netfilter/ipvs/ip_vs_proto_udp.c b/net/netfilter/ipvs/ip_vs_proto_udp.c index 3719837a8fdc..1dc394100fa8 100644 --- a/net/netfilter/ipvs/ip_vs_proto_udp.c +++ b/net/netfilter/ipvs/ip_vs_proto_udp.c @@ -353,14 +353,14 @@ static inline __u16 udp_app_hashkey(__be16 port) } -static int udp_register_app(struct ip_vs_app *inc) +static int udp_register_app(struct net *net, struct ip_vs_app *inc) { struct ip_vs_app *i; __u16 hash; __be16 port = inc->port; int ret = 0; - struct netns_ipvs *ipvs = net_ipvs(&init_net); - struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, IPPROTO_UDP); + struct netns_ipvs *ipvs = net_ipvs(net); + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(net, IPPROTO_UDP); hash = udp_app_hashkey(port); @@ -382,10 +382,10 @@ static int udp_register_app(struct ip_vs_app *inc) static void -udp_unregister_app(struct ip_vs_app *inc) +udp_unregister_app(struct net *net, struct ip_vs_app *inc) { - struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, IPPROTO_UDP); - struct netns_ipvs *ipvs = net_ipvs(&init_net); + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(net, IPPROTO_UDP); + struct netns_ipvs *ipvs = net_ipvs(net); spin_lock_bh(&ipvs->udp_app_lock); atomic_dec(&pd->appcnt); -- cgit v1.2.3 From 29c2026fd4980c144d9c746dc1565060f08e5796 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:54 +0100 Subject: IPVS: netns awareness to ip_vs_est All variables moved to struct ipvs, most external changes fixed (i.e. init_net removed) *v3 timer per ns instead of a common timer in estimator. Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_ctl.c | 20 +++++----- net/netfilter/ipvs/ip_vs_est.c | 86 +++++++++++++++++++++++------------------- 2 files changed, 58 insertions(+), 48 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 88474f1e828a..c89beb8eafbb 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -816,7 +816,7 @@ __ip_vs_update_dest(struct ip_vs_service *svc, struct ip_vs_dest *dest, spin_unlock(&dest->dst_lock); if (add) - ip_vs_new_estimator(&dest->stats); + ip_vs_new_estimator(svc->net, &dest->stats); write_lock_bh(&__ip_vs_svc_lock); @@ -1009,9 +1009,9 @@ ip_vs_edit_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) /* * Delete a destination (must be already unlinked from the service) */ -static void __ip_vs_del_dest(struct ip_vs_dest *dest) +static void __ip_vs_del_dest(struct net *net, struct ip_vs_dest *dest) { - ip_vs_kill_estimator(&dest->stats); + ip_vs_kill_estimator(net, &dest->stats); /* * Remove it from the d-linked list with the real services. @@ -1080,6 +1080,7 @@ static int ip_vs_del_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) { struct ip_vs_dest *dest; + struct net *net = svc->net; __be16 dport = udest->port; EnterFunction(2); @@ -1108,7 +1109,7 @@ ip_vs_del_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) /* * Delete the destination */ - __ip_vs_del_dest(dest); + __ip_vs_del_dest(net, dest); LeaveFunction(2); @@ -1197,7 +1198,7 @@ ip_vs_add_service(struct net *net, struct ip_vs_service_user_kern *u, else if (svc->port == 0) atomic_inc(&ip_vs_nullsvc_counter); - ip_vs_new_estimator(&svc->stats); + ip_vs_new_estimator(net, &svc->stats); /* Count only IPv4 services for old get/setsockopt interface */ if (svc->af == AF_INET) @@ -1345,7 +1346,7 @@ static void __ip_vs_del_service(struct ip_vs_service *svc) if (svc->af == AF_INET) ip_vs_num_services--; - ip_vs_kill_estimator(&svc->stats); + ip_vs_kill_estimator(svc->net, &svc->stats); /* Unbind scheduler */ old_sched = svc->scheduler; @@ -1368,7 +1369,7 @@ static void __ip_vs_del_service(struct ip_vs_service *svc) */ list_for_each_entry_safe(dest, nxt, &svc->destinations, n_list) { __ip_vs_unlink_dest(svc, dest, 0); - __ip_vs_del_dest(dest); + __ip_vs_del_dest(svc->net, dest); } /* @@ -3460,7 +3461,7 @@ int __net_init __ip_vs_control_init(struct net *net) vs_vars); if (sysctl_header == NULL) goto err_reg; - ip_vs_new_estimator(&ip_vs_stats); + ip_vs_new_estimator(net, &ip_vs_stats); return 0; err_reg: @@ -3472,7 +3473,7 @@ static void __net_exit __ip_vs_control_cleanup(struct net *net) if (!net_eq(net, &init_net)) /* netns not enabled yet */ return; - ip_vs_kill_estimator(&ip_vs_stats); + ip_vs_kill_estimator(net, &ip_vs_stats); unregister_net_sysctl_table(sysctl_header); proc_net_remove(net, "ip_vs_stats"); proc_net_remove(net, "ip_vs"); @@ -3536,7 +3537,6 @@ void ip_vs_control_cleanup(void) ip_vs_trash_cleanup(); cancel_delayed_work_sync(&defense_work); cancel_work_sync(&defense_work.work); - ip_vs_kill_estimator(&ip_vs_stats); unregister_pernet_subsys(&ipvs_control_ops); ip_vs_genl_unregister(); nf_unregister_sockopt(&ip_vs_sockopts); diff --git a/net/netfilter/ipvs/ip_vs_est.c b/net/netfilter/ipvs/ip_vs_est.c index 7417a0c1408b..07d839bef537 100644 --- a/net/netfilter/ipvs/ip_vs_est.c +++ b/net/netfilter/ipvs/ip_vs_est.c @@ -8,8 +8,12 @@ * as published by the Free Software Foundation; either version * 2 of the License, or (at your option) any later version. * - * Changes: - * + * Changes: Hans Schillstrom + * Network name space (netns) aware. + * Global data moved to netns i.e struct netns_ipvs + * Affected data: est_list and est_lock. + * estimation_timer() runs with timer per netns. + * get_stats()) do the per cpu summing. */ #define KMSG_COMPONENT "IPVS" @@ -48,12 +52,6 @@ */ -static void estimation_timer(unsigned long arg); - -static LIST_HEAD(est_list); -static DEFINE_SPINLOCK(est_lock); -static DEFINE_TIMER(est_timer, estimation_timer, 0, 0); - static void estimation_timer(unsigned long arg) { struct ip_vs_estimator *e; @@ -62,9 +60,12 @@ static void estimation_timer(unsigned long arg) u32 n_inpkts, n_outpkts; u64 n_inbytes, n_outbytes; u32 rate; + struct net *net = (struct net *)arg; + struct netns_ipvs *ipvs; - spin_lock(&est_lock); - list_for_each_entry(e, &est_list, list) { + ipvs = net_ipvs(net); + spin_lock(&ipvs->est_lock); + list_for_each_entry(e, &ipvs->est_list, list) { s = container_of(e, struct ip_vs_stats, est); spin_lock(&s->lock); @@ -75,38 +76,39 @@ static void estimation_timer(unsigned long arg) n_outbytes = s->ustats.outbytes; /* scaled by 2^10, but divided 2 seconds */ - rate = (n_conns - e->last_conns)<<9; + rate = (n_conns - e->last_conns) << 9; e->last_conns = n_conns; - e->cps += ((long)rate - (long)e->cps)>>2; - s->ustats.cps = (e->cps+0x1FF)>>10; + e->cps += ((long)rate - (long)e->cps) >> 2; + s->ustats.cps = (e->cps + 0x1FF) >> 10; - rate = (n_inpkts - e->last_inpkts)<<9; + rate = (n_inpkts - e->last_inpkts) << 9; e->last_inpkts = n_inpkts; - e->inpps += ((long)rate - (long)e->inpps)>>2; - s->ustats.inpps = (e->inpps+0x1FF)>>10; + e->inpps += ((long)rate - (long)e->inpps) >> 2; + s->ustats.inpps = (e->inpps + 0x1FF) >> 10; - rate = (n_outpkts - e->last_outpkts)<<9; + rate = (n_outpkts - e->last_outpkts) << 9; e->last_outpkts = n_outpkts; - e->outpps += ((long)rate - (long)e->outpps)>>2; - s->ustats.outpps = (e->outpps+0x1FF)>>10; + e->outpps += ((long)rate - (long)e->outpps) >> 2; + s->ustats.outpps = (e->outpps + 0x1FF) >> 10; - rate = (n_inbytes - e->last_inbytes)<<4; + rate = (n_inbytes - e->last_inbytes) << 4; e->last_inbytes = n_inbytes; - e->inbps += ((long)rate - (long)e->inbps)>>2; - s->ustats.inbps = (e->inbps+0xF)>>5; + e->inbps += ((long)rate - (long)e->inbps) >> 2; + s->ustats.inbps = (e->inbps + 0xF) >> 5; - rate = (n_outbytes - e->last_outbytes)<<4; + rate = (n_outbytes - e->last_outbytes) << 4; e->last_outbytes = n_outbytes; - e->outbps += ((long)rate - (long)e->outbps)>>2; - s->ustats.outbps = (e->outbps+0xF)>>5; + e->outbps += ((long)rate - (long)e->outbps) >> 2; + s->ustats.outbps = (e->outbps + 0xF) >> 5; spin_unlock(&s->lock); } - spin_unlock(&est_lock); - mod_timer(&est_timer, jiffies + 2*HZ); + spin_unlock(&ipvs->est_lock); + mod_timer(&ipvs->est_timer, jiffies + 2*HZ); } -void ip_vs_new_estimator(struct ip_vs_stats *stats) +void ip_vs_new_estimator(struct net *net, struct ip_vs_stats *stats) { + struct netns_ipvs *ipvs = net_ipvs(net); struct ip_vs_estimator *est = &stats->est; INIT_LIST_HEAD(&est->list); @@ -126,18 +128,19 @@ void ip_vs_new_estimator(struct ip_vs_stats *stats) est->last_outbytes = stats->ustats.outbytes; est->outbps = stats->ustats.outbps<<5; - spin_lock_bh(&est_lock); - list_add(&est->list, &est_list); - spin_unlock_bh(&est_lock); + spin_lock_bh(&ipvs->est_lock); + list_add(&est->list, &ipvs->est_list); + spin_unlock_bh(&ipvs->est_lock); } -void ip_vs_kill_estimator(struct ip_vs_stats *stats) +void ip_vs_kill_estimator(struct net *net, struct ip_vs_stats *stats) { + struct netns_ipvs *ipvs = net_ipvs(net); struct ip_vs_estimator *est = &stats->est; - spin_lock_bh(&est_lock); + spin_lock_bh(&ipvs->est_lock); list_del(&est->list); - spin_unlock_bh(&est_lock); + spin_unlock_bh(&ipvs->est_lock); } void ip_vs_zero_estimator(struct ip_vs_stats *stats) @@ -159,14 +162,25 @@ void ip_vs_zero_estimator(struct ip_vs_stats *stats) static int __net_init __ip_vs_estimator_init(struct net *net) { + struct netns_ipvs *ipvs = net_ipvs(net); + if (!net_eq(net, &init_net)) /* netns not enabled yet */ return -EPERM; + INIT_LIST_HEAD(&ipvs->est_list); + spin_lock_init(&ipvs->est_lock); + setup_timer(&ipvs->est_timer, estimation_timer, (unsigned long)net); + mod_timer(&ipvs->est_timer, jiffies + 2 * HZ); return 0; } +static void __net_exit __ip_vs_estimator_exit(struct net *net) +{ + del_timer_sync(&net_ipvs(net)->est_timer); +} static struct pernet_operations ip_vs_app_ops = { .init = __ip_vs_estimator_init, + .exit = __ip_vs_estimator_exit, }; int __init ip_vs_estimator_init(void) @@ -174,14 +188,10 @@ int __init ip_vs_estimator_init(void) int rv; rv = register_pernet_subsys(&ip_vs_app_ops); - if (rv < 0) - return rv; - mod_timer(&est_timer, jiffies + 2 * HZ); return rv; } void ip_vs_estimator_cleanup(void) { - del_timer_sync(&est_timer); unregister_pernet_subsys(&ip_vs_app_ops); } -- cgit v1.2.3 From f131315fa272d337dfca7dad2f033ff5296dad65 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:55 +0100 Subject: IPVS: netns awareness to ip_vs_sync All global variables moved to struct ipvs, most external changes fixed (i.e. init_net removed) in sync_buf create + 4 replaced by sizeof(struct..) Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 15 +- net/netfilter/ipvs/ip_vs_ctl.c | 52 ++++--- net/netfilter/ipvs/ip_vs_sync.c | 334 +++++++++++++++++++++------------------- 3 files changed, 219 insertions(+), 182 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 9317affc5ea1..5531d569aa5e 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -1471,12 +1471,13 @@ ip_vs_in_icmp_v6(struct sk_buff *skb, int *related, unsigned int hooknum) static unsigned int ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) { - struct net *net = NULL; + struct net *net; struct ip_vs_iphdr iph; struct ip_vs_protocol *pp; struct ip_vs_proto_data *pd; struct ip_vs_conn *cp; int ret, restart, pkts; + struct netns_ipvs *ipvs; /* Already marked as IPVS request or reply? */ if (skb->ipvs_property) @@ -1556,7 +1557,8 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) } IP_VS_DBG_PKT(11, af, pp, skb, 0, "Incoming packet"); - + net = skb_net(skb); + ipvs = net_ipvs(net); /* Check the server status */ if (cp->dest && !(cp->dest->flags & IP_VS_DEST_F_AVAILABLE)) { /* the destination server is not available */ @@ -1589,12 +1591,13 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) * * For ONE_PKT let ip_vs_sync_conn() do the filter work. */ + if (cp->flags & IP_VS_CONN_F_ONE_PACKET) pkts = sysctl_ip_vs_sync_threshold[0]; else pkts = atomic_add_return(1, &cp->in_pkts); - if ((ip_vs_sync_state & IP_VS_STATE_MASTER) && + if ((ipvs->sync_state & IP_VS_STATE_MASTER) && cp->protocol == IPPROTO_SCTP) { if ((cp->state == IP_VS_SCTP_S_ESTABLISHED && (pkts % sysctl_ip_vs_sync_threshold[1] @@ -1603,13 +1606,13 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) ((cp->state == IP_VS_SCTP_S_CLOSED) || (cp->state == IP_VS_SCTP_S_SHUT_ACK_CLI) || (cp->state == IP_VS_SCTP_S_SHUT_ACK_SER)))) { - ip_vs_sync_conn(cp); + ip_vs_sync_conn(net, cp); goto out; } } /* Keep this block last: TCP and others with pp->num_states <= 1 */ - else if ((ip_vs_sync_state & IP_VS_STATE_MASTER) && + else if ((ipvs->sync_state & IP_VS_STATE_MASTER) && (((cp->protocol != IPPROTO_TCP || cp->state == IP_VS_TCP_S_ESTABLISHED) && (pkts % sysctl_ip_vs_sync_threshold[1] @@ -1619,7 +1622,7 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) (cp->state == IP_VS_TCP_S_CLOSE) || (cp->state == IP_VS_TCP_S_CLOSE_WAIT) || (cp->state == IP_VS_TCP_S_TIME_WAIT))))) - ip_vs_sync_conn(cp); + ip_vs_sync_conn(net, cp); out: cp->old_state = cp->state; diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index c89beb8eafbb..03f86312b4bb 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -1559,7 +1559,8 @@ proc_do_sync_mode(ctl_table *table, int write, /* Restore the correct value */ *valp = val; } else { - ip_vs_sync_switch_mode(val); + struct net *net = current->nsproxy->net_ns; + ip_vs_sync_switch_mode(net, val); } } return rc; @@ -2174,11 +2175,12 @@ do_ip_vs_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) goto out_unlock; } else if (cmd == IP_VS_SO_SET_STARTDAEMON) { struct ip_vs_daemon_user *dm = (struct ip_vs_daemon_user *)arg; - ret = start_sync_thread(dm->state, dm->mcast_ifn, dm->syncid); + ret = start_sync_thread(net, dm->state, dm->mcast_ifn, + dm->syncid); goto out_unlock; } else if (cmd == IP_VS_SO_SET_STOPDAEMON) { struct ip_vs_daemon_user *dm = (struct ip_vs_daemon_user *)arg; - ret = stop_sync_thread(dm->state); + ret = stop_sync_thread(net, dm->state); goto out_unlock; } @@ -2424,6 +2426,7 @@ do_ip_vs_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) int ret = 0; unsigned int copylen; struct net *net = sock_net(sk); + struct netns_ipvs *ipvs = net_ipvs(net); BUG_ON(!net); if (!capable(CAP_NET_ADMIN)) @@ -2546,15 +2549,17 @@ do_ip_vs_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) struct ip_vs_daemon_user d[2]; memset(&d, 0, sizeof(d)); - if (ip_vs_sync_state & IP_VS_STATE_MASTER) { + if (ipvs->sync_state & IP_VS_STATE_MASTER) { d[0].state = IP_VS_STATE_MASTER; - strlcpy(d[0].mcast_ifn, ip_vs_master_mcast_ifn, sizeof(d[0].mcast_ifn)); - d[0].syncid = ip_vs_master_syncid; + strlcpy(d[0].mcast_ifn, ipvs->master_mcast_ifn, + sizeof(d[0].mcast_ifn)); + d[0].syncid = ipvs->master_syncid; } - if (ip_vs_sync_state & IP_VS_STATE_BACKUP) { + if (ipvs->sync_state & IP_VS_STATE_BACKUP) { d[1].state = IP_VS_STATE_BACKUP; - strlcpy(d[1].mcast_ifn, ip_vs_backup_mcast_ifn, sizeof(d[1].mcast_ifn)); - d[1].syncid = ip_vs_backup_syncid; + strlcpy(d[1].mcast_ifn, ipvs->backup_mcast_ifn, + sizeof(d[1].mcast_ifn)); + d[1].syncid = ipvs->backup_syncid; } if (copy_to_user(user, &d, sizeof(d)) != 0) ret = -EFAULT; @@ -3061,20 +3066,23 @@ nla_put_failure: static int ip_vs_genl_dump_daemons(struct sk_buff *skb, struct netlink_callback *cb) { + struct net *net = skb_net(skb); + struct netns_ipvs *ipvs = net_ipvs(net); + mutex_lock(&__ip_vs_mutex); - if ((ip_vs_sync_state & IP_VS_STATE_MASTER) && !cb->args[0]) { + if ((ipvs->sync_state & IP_VS_STATE_MASTER) && !cb->args[0]) { if (ip_vs_genl_dump_daemon(skb, IP_VS_STATE_MASTER, - ip_vs_master_mcast_ifn, - ip_vs_master_syncid, cb) < 0) + ipvs->master_mcast_ifn, + ipvs->master_syncid, cb) < 0) goto nla_put_failure; cb->args[0] = 1; } - if ((ip_vs_sync_state & IP_VS_STATE_BACKUP) && !cb->args[1]) { + if ((ipvs->sync_state & IP_VS_STATE_BACKUP) && !cb->args[1]) { if (ip_vs_genl_dump_daemon(skb, IP_VS_STATE_BACKUP, - ip_vs_backup_mcast_ifn, - ip_vs_backup_syncid, cb) < 0) + ipvs->backup_mcast_ifn, + ipvs->backup_syncid, cb) < 0) goto nla_put_failure; cb->args[1] = 1; @@ -3086,24 +3094,26 @@ nla_put_failure: return skb->len; } -static int ip_vs_genl_new_daemon(struct nlattr **attrs) +static int ip_vs_genl_new_daemon(struct net *net, struct nlattr **attrs) { if (!(attrs[IPVS_DAEMON_ATTR_STATE] && attrs[IPVS_DAEMON_ATTR_MCAST_IFN] && attrs[IPVS_DAEMON_ATTR_SYNC_ID])) return -EINVAL; - return start_sync_thread(nla_get_u32(attrs[IPVS_DAEMON_ATTR_STATE]), + return start_sync_thread(net, + nla_get_u32(attrs[IPVS_DAEMON_ATTR_STATE]), nla_data(attrs[IPVS_DAEMON_ATTR_MCAST_IFN]), nla_get_u32(attrs[IPVS_DAEMON_ATTR_SYNC_ID])); } -static int ip_vs_genl_del_daemon(struct nlattr **attrs) +static int ip_vs_genl_del_daemon(struct net *net, struct nlattr **attrs) { if (!attrs[IPVS_DAEMON_ATTR_STATE]) return -EINVAL; - return stop_sync_thread(nla_get_u32(attrs[IPVS_DAEMON_ATTR_STATE])); + return stop_sync_thread(net, + nla_get_u32(attrs[IPVS_DAEMON_ATTR_STATE])); } static int ip_vs_genl_set_config(struct net *net, struct nlattr **attrs) @@ -3159,9 +3169,9 @@ static int ip_vs_genl_set_cmd(struct sk_buff *skb, struct genl_info *info) } if (cmd == IPVS_CMD_NEW_DAEMON) - ret = ip_vs_genl_new_daemon(daemon_attrs); + ret = ip_vs_genl_new_daemon(net, daemon_attrs); else - ret = ip_vs_genl_del_daemon(daemon_attrs); + ret = ip_vs_genl_del_daemon(net, daemon_attrs); goto out; } else if (cmd == IPVS_CMD_ZERO && !info->attrs[IPVS_CMD_ATTR_SERVICE]) { diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index 6831e8fac8db..c29e73d686fb 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -192,6 +192,7 @@ union ip_vs_sync_conn { #define IPVS_OPT_F_PARAM (1 << (IPVS_OPT_PARAM-1)) struct ip_vs_sync_thread_data { + struct net *net; struct socket *sock; char *buf; }; @@ -259,10 +260,6 @@ struct ip_vs_sync_mesg { /* ip_vs_sync_conn entries start here */ }; -/* the maximum length of sync (sending/receiving) message */ -static int sync_send_mesg_maxlen; -static int sync_recv_mesg_maxlen; - struct ip_vs_sync_buff { struct list_head list; unsigned long firstuse; @@ -273,28 +270,6 @@ struct ip_vs_sync_buff { unsigned char *end; }; - -/* the sync_buff list head and the lock */ -static LIST_HEAD(ip_vs_sync_queue); -static DEFINE_SPINLOCK(ip_vs_sync_lock); - -/* current sync_buff for accepting new conn entries */ -static struct ip_vs_sync_buff *curr_sb = NULL; -static DEFINE_SPINLOCK(curr_sb_lock); - -/* ipvs sync daemon state */ -volatile int ip_vs_sync_state = IP_VS_STATE_NONE; -volatile int ip_vs_master_syncid = 0; -volatile int ip_vs_backup_syncid = 0; - -/* multicast interface name */ -char ip_vs_master_mcast_ifn[IP_VS_IFNAME_MAXLEN]; -char ip_vs_backup_mcast_ifn[IP_VS_IFNAME_MAXLEN]; - -/* sync daemon tasks */ -static struct task_struct *sync_master_thread; -static struct task_struct *sync_backup_thread; - /* multicast addr */ static struct sockaddr_in mcast_addr = { .sin_family = AF_INET, @@ -324,20 +299,20 @@ static void hton_seq(struct ip_vs_seq *ho, struct ip_vs_seq *no) put_unaligned_be32(ho->previous_delta, &no->previous_delta); } -static inline struct ip_vs_sync_buff *sb_dequeue(void) +static inline struct ip_vs_sync_buff *sb_dequeue(struct netns_ipvs *ipvs) { struct ip_vs_sync_buff *sb; - spin_lock_bh(&ip_vs_sync_lock); - if (list_empty(&ip_vs_sync_queue)) { + spin_lock_bh(&ipvs->sync_lock); + if (list_empty(&ipvs->sync_queue)) { sb = NULL; } else { - sb = list_entry(ip_vs_sync_queue.next, + sb = list_entry(ipvs->sync_queue.next, struct ip_vs_sync_buff, list); list_del(&sb->list); } - spin_unlock_bh(&ip_vs_sync_lock); + spin_unlock_bh(&ipvs->sync_lock); return sb; } @@ -345,25 +320,27 @@ static inline struct ip_vs_sync_buff *sb_dequeue(void) /* * Create a new sync buffer for Version 1 proto. */ -static inline struct ip_vs_sync_buff * ip_vs_sync_buff_create(void) +static inline struct ip_vs_sync_buff * +ip_vs_sync_buff_create(struct netns_ipvs *ipvs) { struct ip_vs_sync_buff *sb; if (!(sb=kmalloc(sizeof(struct ip_vs_sync_buff), GFP_ATOMIC))) return NULL; - if (!(sb->mesg=kmalloc(sync_send_mesg_maxlen, GFP_ATOMIC))) { + sb->mesg = kmalloc(ipvs->send_mesg_maxlen, GFP_ATOMIC); + if (!sb->mesg) { kfree(sb); return NULL; } sb->mesg->reserved = 0; /* old nr_conns i.e. must be zeo now */ sb->mesg->version = SYNC_PROTO_VER; - sb->mesg->syncid = ip_vs_master_syncid; + sb->mesg->syncid = ipvs->master_syncid; sb->mesg->size = sizeof(struct ip_vs_sync_mesg); sb->mesg->nr_conns = 0; sb->mesg->spare = 0; sb->head = (unsigned char *)sb->mesg + sizeof(struct ip_vs_sync_mesg); - sb->end = (unsigned char *)sb->mesg + sync_send_mesg_maxlen; + sb->end = (unsigned char *)sb->mesg + ipvs->send_mesg_maxlen; sb->firstuse = jiffies; return sb; @@ -375,14 +352,16 @@ static inline void ip_vs_sync_buff_release(struct ip_vs_sync_buff *sb) kfree(sb); } -static inline void sb_queue_tail(struct ip_vs_sync_buff *sb) +static inline void sb_queue_tail(struct netns_ipvs *ipvs) { - spin_lock(&ip_vs_sync_lock); - if (ip_vs_sync_state & IP_VS_STATE_MASTER) - list_add_tail(&sb->list, &ip_vs_sync_queue); + struct ip_vs_sync_buff *sb = ipvs->sync_buff; + + spin_lock(&ipvs->sync_lock); + if (ipvs->sync_state & IP_VS_STATE_MASTER) + list_add_tail(&sb->list, &ipvs->sync_queue); else ip_vs_sync_buff_release(sb); - spin_unlock(&ip_vs_sync_lock); + spin_unlock(&ipvs->sync_lock); } /* @@ -390,18 +369,18 @@ static inline void sb_queue_tail(struct ip_vs_sync_buff *sb) * than the specified time or the specified time is zero. */ static inline struct ip_vs_sync_buff * -get_curr_sync_buff(unsigned long time) +get_curr_sync_buff(struct netns_ipvs *ipvs, unsigned long time) { struct ip_vs_sync_buff *sb; - spin_lock_bh(&curr_sb_lock); - if (curr_sb && (time == 0 || - time_before(jiffies - curr_sb->firstuse, time))) { - sb = curr_sb; - curr_sb = NULL; + spin_lock_bh(&ipvs->sync_buff_lock); + if (ipvs->sync_buff && (time == 0 || + time_before(jiffies - ipvs->sync_buff->firstuse, time))) { + sb = ipvs->sync_buff; + ipvs->sync_buff = NULL; } else sb = NULL; - spin_unlock_bh(&curr_sb_lock); + spin_unlock_bh(&ipvs->sync_buff_lock); return sb; } @@ -409,33 +388,37 @@ get_curr_sync_buff(unsigned long time) * Switch mode from sending version 0 or 1 * - must handle sync_buf */ -void ip_vs_sync_switch_mode(int mode) { +void ip_vs_sync_switch_mode(struct net *net, int mode) +{ + struct netns_ipvs *ipvs = net_ipvs(net); - if (!ip_vs_sync_state & IP_VS_STATE_MASTER) + if (!ipvs->sync_state & IP_VS_STATE_MASTER) return; - if (mode == sysctl_ip_vs_sync_ver || !curr_sb) + if (mode == sysctl_ip_vs_sync_ver || !ipvs->sync_buff) return; - spin_lock_bh(&curr_sb_lock); + spin_lock_bh(&ipvs->sync_buff_lock); /* Buffer empty ? then let buf_create do the job */ - if ( curr_sb->mesg->size <= sizeof(struct ip_vs_sync_mesg)) { - kfree(curr_sb); - curr_sb = NULL; + if (ipvs->sync_buff->mesg->size <= sizeof(struct ip_vs_sync_mesg)) { + kfree(ipvs->sync_buff); + ipvs->sync_buff = NULL; } else { - spin_lock_bh(&ip_vs_sync_lock); - if (ip_vs_sync_state & IP_VS_STATE_MASTER) - list_add_tail(&curr_sb->list, &ip_vs_sync_queue); + spin_lock_bh(&ipvs->sync_lock); + if (ipvs->sync_state & IP_VS_STATE_MASTER) + list_add_tail(&ipvs->sync_buff->list, + &ipvs->sync_queue); else - ip_vs_sync_buff_release(curr_sb); - spin_unlock_bh(&ip_vs_sync_lock); + ip_vs_sync_buff_release(ipvs->sync_buff); + spin_unlock_bh(&ipvs->sync_lock); } - spin_unlock_bh(&curr_sb_lock); + spin_unlock_bh(&ipvs->sync_buff_lock); } /* * Create a new sync buffer for Version 0 proto. */ -static inline struct ip_vs_sync_buff * ip_vs_sync_buff_create_v0(void) +static inline struct ip_vs_sync_buff * +ip_vs_sync_buff_create_v0(struct netns_ipvs *ipvs) { struct ip_vs_sync_buff *sb; struct ip_vs_sync_mesg_v0 *mesg; @@ -443,16 +426,17 @@ static inline struct ip_vs_sync_buff * ip_vs_sync_buff_create_v0(void) if (!(sb=kmalloc(sizeof(struct ip_vs_sync_buff), GFP_ATOMIC))) return NULL; - if (!(sb->mesg=kmalloc(sync_send_mesg_maxlen, GFP_ATOMIC))) { + sb->mesg = kmalloc(ipvs->send_mesg_maxlen, GFP_ATOMIC); + if (!sb->mesg) { kfree(sb); return NULL; } mesg = (struct ip_vs_sync_mesg_v0 *)sb->mesg; mesg->nr_conns = 0; - mesg->syncid = ip_vs_master_syncid; - mesg->size = 4; - sb->head = (unsigned char *)mesg + 4; - sb->end = (unsigned char *)mesg + sync_send_mesg_maxlen; + mesg->syncid = ipvs->master_syncid; + mesg->size = sizeof(struct ip_vs_sync_mesg_v0); + sb->head = (unsigned char *)mesg + sizeof(struct ip_vs_sync_mesg_v0); + sb->end = (unsigned char *)mesg + ipvs->send_mesg_maxlen; sb->firstuse = jiffies; return sb; } @@ -461,8 +445,9 @@ static inline struct ip_vs_sync_buff * ip_vs_sync_buff_create_v0(void) * Version 0 , could be switched in by sys_ctl. * Add an ip_vs_conn information into the current sync_buff. */ -void ip_vs_sync_conn_v0(struct ip_vs_conn *cp) +void ip_vs_sync_conn_v0(struct net *net, struct ip_vs_conn *cp) { + struct netns_ipvs *ipvs = net_ipvs(net); struct ip_vs_sync_mesg_v0 *m; struct ip_vs_sync_conn_v0 *s; int len; @@ -473,10 +458,12 @@ void ip_vs_sync_conn_v0(struct ip_vs_conn *cp) if (cp->flags & IP_VS_CONN_F_ONE_PACKET) return; - spin_lock(&curr_sb_lock); - if (!curr_sb) { - if (!(curr_sb=ip_vs_sync_buff_create_v0())) { - spin_unlock(&curr_sb_lock); + spin_lock(&ipvs->sync_buff_lock); + if (!ipvs->sync_buff) { + ipvs->sync_buff = + ip_vs_sync_buff_create_v0(ipvs); + if (!ipvs->sync_buff) { + spin_unlock(&ipvs->sync_buff_lock); pr_err("ip_vs_sync_buff_create failed.\n"); return; } @@ -484,8 +471,8 @@ void ip_vs_sync_conn_v0(struct ip_vs_conn *cp) len = (cp->flags & IP_VS_CONN_F_SEQ_MASK) ? FULL_CONN_SIZE : SIMPLE_CONN_SIZE; - m = (struct ip_vs_sync_mesg_v0 *)curr_sb->mesg; - s = (struct ip_vs_sync_conn_v0 *)curr_sb->head; + m = (struct ip_vs_sync_mesg_v0 *)ipvs->sync_buff->mesg; + s = (struct ip_vs_sync_conn_v0 *)ipvs->sync_buff->head; /* copy members */ s->reserved = 0; @@ -506,18 +493,18 @@ void ip_vs_sync_conn_v0(struct ip_vs_conn *cp) m->nr_conns++; m->size += len; - curr_sb->head += len; + ipvs->sync_buff->head += len; /* check if there is a space for next one */ - if (curr_sb->head + FULL_CONN_SIZE > curr_sb->end) { - sb_queue_tail(curr_sb); - curr_sb = NULL; + if (ipvs->sync_buff->head + FULL_CONN_SIZE > ipvs->sync_buff->end) { + sb_queue_tail(ipvs); + ipvs->sync_buff = NULL; } - spin_unlock(&curr_sb_lock); + spin_unlock(&ipvs->sync_buff_lock); /* synchronize its controller if it has */ if (cp->control) - ip_vs_sync_conn(cp->control); + ip_vs_sync_conn(net, cp->control); } /* @@ -525,8 +512,9 @@ void ip_vs_sync_conn_v0(struct ip_vs_conn *cp) * Called by ip_vs_in. * Sending Version 1 messages */ -void ip_vs_sync_conn(struct ip_vs_conn *cp) +void ip_vs_sync_conn(struct net *net, struct ip_vs_conn *cp) { + struct netns_ipvs *ipvs = net_ipvs(net); struct ip_vs_sync_mesg *m; union ip_vs_sync_conn *s; __u8 *p; @@ -534,7 +522,7 @@ void ip_vs_sync_conn(struct ip_vs_conn *cp) /* Handle old version of the protocol */ if (sysctl_ip_vs_sync_ver == 0) { - ip_vs_sync_conn_v0(cp); + ip_vs_sync_conn_v0(net, cp); return; } /* Do not sync ONE PACKET */ @@ -551,7 +539,7 @@ sloop: pe_name_len = strnlen(cp->pe->name, IP_VS_PENAME_MAXLEN); } - spin_lock(&curr_sb_lock); + spin_lock(&ipvs->sync_buff_lock); #ifdef CONFIG_IP_VS_IPV6 if (cp->af == AF_INET6) @@ -570,26 +558,27 @@ sloop: /* check if there is a space for this one */ pad = 0; - if (curr_sb) { - pad = (4 - (size_t)curr_sb->head) & 3; - if (curr_sb->head + len + pad > curr_sb->end) { - sb_queue_tail(curr_sb); - curr_sb = NULL; + if (ipvs->sync_buff) { + pad = (4 - (size_t)ipvs->sync_buff->head) & 3; + if (ipvs->sync_buff->head + len + pad > ipvs->sync_buff->end) { + sb_queue_tail(ipvs); + ipvs->sync_buff = NULL; pad = 0; } } - if (!curr_sb) { - if (!(curr_sb=ip_vs_sync_buff_create())) { - spin_unlock(&curr_sb_lock); + if (!ipvs->sync_buff) { + ipvs->sync_buff = ip_vs_sync_buff_create(ipvs); + if (!ipvs->sync_buff) { + spin_unlock(&ipvs->sync_buff_lock); pr_err("ip_vs_sync_buff_create failed.\n"); return; } } - m = curr_sb->mesg; - p = curr_sb->head; - curr_sb->head += pad + len; + m = ipvs->sync_buff->mesg; + p = ipvs->sync_buff->head; + ipvs->sync_buff->head += pad + len; m->size += pad + len; /* Add ev. padding from prev. sync_conn */ while (pad--) @@ -647,7 +636,7 @@ sloop: } } - spin_unlock(&curr_sb_lock); + spin_unlock(&ipvs->sync_buff_lock); control: /* synchronize its controller if it has */ @@ -699,7 +688,8 @@ ip_vs_conn_fill_param_sync(int af, union ip_vs_sync_conn *sc, buff[pe_name_len]=0; p->pe = __ip_vs_pe_getbyname(buff); if (!p->pe) { - IP_VS_DBG(3, "BACKUP, no %s engine found/loaded\n", buff); + IP_VS_DBG(3, "BACKUP, no %s engine found/loaded\n", + buff); return 1; } } else { @@ -748,7 +738,7 @@ static void ip_vs_proc_conn(struct net *net, struct ip_vs_conn_param *param, * If it is not found the connection will remain unbound * but still handled. */ - dest = ip_vs_find_dest(&init_net, type, daddr, dport, param->vaddr, + dest = ip_vs_find_dest(net, type, daddr, dport, param->vaddr, param->vport, protocol, fwmark); /* Set the approprite ativity flag */ @@ -1089,6 +1079,7 @@ out: static void ip_vs_process_message(struct net *net, __u8 *buffer, const size_t buflen) { + struct netns_ipvs *ipvs = net_ipvs(net); struct ip_vs_sync_mesg *m2 = (struct ip_vs_sync_mesg *)buffer; __u8 *p, *msg_end; int i, nr_conns; @@ -1105,7 +1096,7 @@ static void ip_vs_process_message(struct net *net, __u8 *buffer, return; } /* SyncID sanity check */ - if (ip_vs_backup_syncid != 0 && m2->syncid != ip_vs_backup_syncid) { + if (ipvs->backup_syncid != 0 && m2->syncid != ipvs->backup_syncid) { IP_VS_DBG(7, "BACKUP, Ignoring syncid = %d\n", m2->syncid); return; } @@ -1190,8 +1181,10 @@ static int set_mcast_if(struct sock *sk, char *ifname) { struct net_device *dev; struct inet_sock *inet = inet_sk(sk); + struct net *net = sock_net(sk); - if ((dev = __dev_get_by_name(&init_net, ifname)) == NULL) + dev = __dev_get_by_name(net, ifname); + if (!dev) return -ENODEV; if (sk->sk_bound_dev_if && dev->ifindex != sk->sk_bound_dev_if) @@ -1210,30 +1203,33 @@ static int set_mcast_if(struct sock *sk, char *ifname) * Set the maximum length of sync message according to the * specified interface's MTU. */ -static int set_sync_mesg_maxlen(int sync_state) +static int set_sync_mesg_maxlen(struct net *net, int sync_state) { + struct netns_ipvs *ipvs = net_ipvs(net); struct net_device *dev; int num; if (sync_state == IP_VS_STATE_MASTER) { - if ((dev = __dev_get_by_name(&init_net, ip_vs_master_mcast_ifn)) == NULL) + dev = __dev_get_by_name(net, ipvs->master_mcast_ifn); + if (!dev) return -ENODEV; num = (dev->mtu - sizeof(struct iphdr) - sizeof(struct udphdr) - SYNC_MESG_HEADER_LEN - 20) / SIMPLE_CONN_SIZE; - sync_send_mesg_maxlen = SYNC_MESG_HEADER_LEN + + ipvs->send_mesg_maxlen = SYNC_MESG_HEADER_LEN + SIMPLE_CONN_SIZE * min(num, MAX_CONNS_PER_SYNCBUFF); IP_VS_DBG(7, "setting the maximum length of sync sending " - "message %d.\n", sync_send_mesg_maxlen); + "message %d.\n", ipvs->send_mesg_maxlen); } else if (sync_state == IP_VS_STATE_BACKUP) { - if ((dev = __dev_get_by_name(&init_net, ip_vs_backup_mcast_ifn)) == NULL) + dev = __dev_get_by_name(net, ipvs->backup_mcast_ifn); + if (!dev) return -ENODEV; - sync_recv_mesg_maxlen = dev->mtu - + ipvs->recv_mesg_maxlen = dev->mtu - sizeof(struct iphdr) - sizeof(struct udphdr); IP_VS_DBG(7, "setting the maximum length of sync receiving " - "message %d.\n", sync_recv_mesg_maxlen); + "message %d.\n", ipvs->recv_mesg_maxlen); } return 0; @@ -1248,6 +1244,7 @@ static int set_sync_mesg_maxlen(int sync_state) static int join_mcast_group(struct sock *sk, struct in_addr *addr, char *ifname) { + struct net *net = sock_net(sk); struct ip_mreqn mreq; struct net_device *dev; int ret; @@ -1255,7 +1252,8 @@ join_mcast_group(struct sock *sk, struct in_addr *addr, char *ifname) memset(&mreq, 0, sizeof(mreq)); memcpy(&mreq.imr_multiaddr, addr, sizeof(struct in_addr)); - if ((dev = __dev_get_by_name(&init_net, ifname)) == NULL) + dev = __dev_get_by_name(net, ifname); + if (!dev) return -ENODEV; if (sk->sk_bound_dev_if && dev->ifindex != sk->sk_bound_dev_if) return -EINVAL; @@ -1272,11 +1270,13 @@ join_mcast_group(struct sock *sk, struct in_addr *addr, char *ifname) static int bind_mcastif_addr(struct socket *sock, char *ifname) { + struct net *net = sock_net(sock->sk); struct net_device *dev; __be32 addr; struct sockaddr_in sin; - if ((dev = __dev_get_by_name(&init_net, ifname)) == NULL) + dev = __dev_get_by_name(net, ifname); + if (!dev) return -ENODEV; addr = inet_select_addr(dev, 0, RT_SCOPE_UNIVERSE); @@ -1298,8 +1298,9 @@ static int bind_mcastif_addr(struct socket *sock, char *ifname) /* * Set up sending multicast socket over UDP */ -static struct socket * make_send_sock(void) +static struct socket *make_send_sock(struct net *net) { + struct netns_ipvs *ipvs = net_ipvs(net); struct socket *sock; int result; @@ -1310,7 +1311,7 @@ static struct socket * make_send_sock(void) return ERR_PTR(result); } - result = set_mcast_if(sock->sk, ip_vs_master_mcast_ifn); + result = set_mcast_if(sock->sk, ipvs->master_mcast_ifn); if (result < 0) { pr_err("Error setting outbound mcast interface\n"); goto error; @@ -1319,7 +1320,7 @@ static struct socket * make_send_sock(void) set_mcast_loop(sock->sk, 0); set_mcast_ttl(sock->sk, 1); - result = bind_mcastif_addr(sock, ip_vs_master_mcast_ifn); + result = bind_mcastif_addr(sock, ipvs->master_mcast_ifn); if (result < 0) { pr_err("Error binding address of the mcast interface\n"); goto error; @@ -1343,8 +1344,9 @@ static struct socket * make_send_sock(void) /* * Set up receiving multicast socket over UDP */ -static struct socket * make_receive_sock(void) +static struct socket *make_receive_sock(struct net *net) { + struct netns_ipvs *ipvs = net_ipvs(net); struct socket *sock; int result; @@ -1368,7 +1370,7 @@ static struct socket * make_receive_sock(void) /* join the multicast group */ result = join_mcast_group(sock->sk, (struct in_addr *) &mcast_addr.sin_addr, - ip_vs_backup_mcast_ifn); + ipvs->backup_mcast_ifn); if (result < 0) { pr_err("Error joining to the multicast group\n"); goto error; @@ -1439,20 +1441,21 @@ ip_vs_receive(struct socket *sock, char *buffer, const size_t buflen) static int sync_thread_master(void *data) { struct ip_vs_sync_thread_data *tinfo = data; + struct netns_ipvs *ipvs = net_ipvs(tinfo->net); struct ip_vs_sync_buff *sb; pr_info("sync thread started: state = MASTER, mcast_ifn = %s, " "syncid = %d\n", - ip_vs_master_mcast_ifn, ip_vs_master_syncid); + ipvs->master_mcast_ifn, ipvs->master_syncid); while (!kthread_should_stop()) { - while ((sb = sb_dequeue())) { + while ((sb = sb_dequeue(ipvs))) { ip_vs_send_sync_msg(tinfo->sock, sb->mesg); ip_vs_sync_buff_release(sb); } - /* check if entries stay in curr_sb for 2 seconds */ - sb = get_curr_sync_buff(2 * HZ); + /* check if entries stay in ipvs->sync_buff for 2 seconds */ + sb = get_curr_sync_buff(ipvs, 2 * HZ); if (sb) { ip_vs_send_sync_msg(tinfo->sock, sb->mesg); ip_vs_sync_buff_release(sb); @@ -1462,14 +1465,13 @@ static int sync_thread_master(void *data) } /* clean up the sync_buff queue */ - while ((sb=sb_dequeue())) { + while ((sb = sb_dequeue(ipvs))) ip_vs_sync_buff_release(sb); - } /* clean up the current sync_buff */ - if ((sb = get_curr_sync_buff(0))) { + sb = get_curr_sync_buff(ipvs, 0); + if (sb) ip_vs_sync_buff_release(sb); - } /* release the sending multicast socket */ sock_release(tinfo->sock); @@ -1482,11 +1484,12 @@ static int sync_thread_master(void *data) static int sync_thread_backup(void *data) { struct ip_vs_sync_thread_data *tinfo = data; + struct netns_ipvs *ipvs = net_ipvs(tinfo->net); int len; pr_info("sync thread started: state = BACKUP, mcast_ifn = %s, " "syncid = %d\n", - ip_vs_backup_mcast_ifn, ip_vs_backup_syncid); + ipvs->backup_mcast_ifn, ipvs->backup_syncid); while (!kthread_should_stop()) { wait_event_interruptible(*sk_sleep(tinfo->sock->sk), @@ -1496,7 +1499,7 @@ static int sync_thread_backup(void *data) /* do we have data now? */ while (!skb_queue_empty(&(tinfo->sock->sk->sk_receive_queue))) { len = ip_vs_receive(tinfo->sock, tinfo->buf, - sync_recv_mesg_maxlen); + ipvs->recv_mesg_maxlen); if (len <= 0) { pr_err("receiving message error\n"); break; @@ -1505,7 +1508,7 @@ static int sync_thread_backup(void *data) /* disable bottom half, because it accesses the data shared by softirq while getting/creating conns */ local_bh_disable(); - ip_vs_process_message(&init_net, tinfo->buf, len); + ip_vs_process_message(tinfo->net, tinfo->buf, len); local_bh_enable(); } } @@ -1519,11 +1522,12 @@ static int sync_thread_backup(void *data) } -int start_sync_thread(int state, char *mcast_ifn, __u8 syncid) +int start_sync_thread(struct net *net, int state, char *mcast_ifn, __u8 syncid) { struct ip_vs_sync_thread_data *tinfo; struct task_struct **realtask, *task; struct socket *sock; + struct netns_ipvs *ipvs = net_ipvs(net); char *name, *buf = NULL; int (*threadfn)(void *data); int result = -ENOMEM; @@ -1533,27 +1537,27 @@ int start_sync_thread(int state, char *mcast_ifn, __u8 syncid) sizeof(struct ip_vs_sync_conn_v0)); if (state == IP_VS_STATE_MASTER) { - if (sync_master_thread) + if (ipvs->master_thread) return -EEXIST; - strlcpy(ip_vs_master_mcast_ifn, mcast_ifn, - sizeof(ip_vs_master_mcast_ifn)); - ip_vs_master_syncid = syncid; - realtask = &sync_master_thread; - name = "ipvs_syncmaster"; + strlcpy(ipvs->master_mcast_ifn, mcast_ifn, + sizeof(ipvs->master_mcast_ifn)); + ipvs->master_syncid = syncid; + realtask = &ipvs->master_thread; + name = "ipvs_master:%d"; threadfn = sync_thread_master; - sock = make_send_sock(); + sock = make_send_sock(net); } else if (state == IP_VS_STATE_BACKUP) { - if (sync_backup_thread) + if (ipvs->backup_thread) return -EEXIST; - strlcpy(ip_vs_backup_mcast_ifn, mcast_ifn, - sizeof(ip_vs_backup_mcast_ifn)); - ip_vs_backup_syncid = syncid; - realtask = &sync_backup_thread; - name = "ipvs_syncbackup"; + strlcpy(ipvs->backup_mcast_ifn, mcast_ifn, + sizeof(ipvs->backup_mcast_ifn)); + ipvs->backup_syncid = syncid; + realtask = &ipvs->backup_thread; + name = "ipvs_backup:%d"; threadfn = sync_thread_backup; - sock = make_receive_sock(); + sock = make_receive_sock(net); } else { return -EINVAL; } @@ -1563,9 +1567,9 @@ int start_sync_thread(int state, char *mcast_ifn, __u8 syncid) goto out; } - set_sync_mesg_maxlen(state); + set_sync_mesg_maxlen(net, state); if (state == IP_VS_STATE_BACKUP) { - buf = kmalloc(sync_recv_mesg_maxlen, GFP_KERNEL); + buf = kmalloc(ipvs->recv_mesg_maxlen, GFP_KERNEL); if (!buf) goto outsocket; } @@ -1574,10 +1578,11 @@ int start_sync_thread(int state, char *mcast_ifn, __u8 syncid) if (!tinfo) goto outbuf; + tinfo->net = net; tinfo->sock = sock; tinfo->buf = buf; - task = kthread_run(threadfn, tinfo, name); + task = kthread_run(threadfn, tinfo, name, ipvs->gen); if (IS_ERR(task)) { result = PTR_ERR(task); goto outtinfo; @@ -1585,7 +1590,7 @@ int start_sync_thread(int state, char *mcast_ifn, __u8 syncid) /* mark as active */ *realtask = task; - ip_vs_sync_state |= state; + ipvs->sync_state |= state; /* increase the module use count */ ip_vs_use_count_inc(); @@ -1603,16 +1608,18 @@ out: } -int stop_sync_thread(int state) +int stop_sync_thread(struct net *net, int state) { + struct netns_ipvs *ipvs = net_ipvs(net); + IP_VS_DBG(7, "%s(): pid %d\n", __func__, task_pid_nr(current)); if (state == IP_VS_STATE_MASTER) { - if (!sync_master_thread) + if (!ipvs->master_thread) return -ESRCH; pr_info("stopping master sync thread %d ...\n", - task_pid_nr(sync_master_thread)); + task_pid_nr(ipvs->master_thread)); /* * The lock synchronizes with sb_queue_tail(), so that we don't @@ -1620,21 +1627,21 @@ int stop_sync_thread(int state) * progress of stopping the master sync daemon. */ - spin_lock_bh(&ip_vs_sync_lock); - ip_vs_sync_state &= ~IP_VS_STATE_MASTER; - spin_unlock_bh(&ip_vs_sync_lock); - kthread_stop(sync_master_thread); - sync_master_thread = NULL; + spin_lock_bh(&ipvs->sync_lock); + ipvs->sync_state &= ~IP_VS_STATE_MASTER; + spin_unlock_bh(&ipvs->sync_lock); + kthread_stop(ipvs->master_thread); + ipvs->master_thread = NULL; } else if (state == IP_VS_STATE_BACKUP) { - if (!sync_backup_thread) + if (!ipvs->backup_thread) return -ESRCH; pr_info("stopping backup sync thread %d ...\n", - task_pid_nr(sync_backup_thread)); + task_pid_nr(ipvs->backup_thread)); - ip_vs_sync_state &= ~IP_VS_STATE_BACKUP; - kthread_stop(sync_backup_thread); - sync_backup_thread = NULL; + ipvs->sync_state &= ~IP_VS_STATE_BACKUP; + kthread_stop(ipvs->backup_thread); + ipvs->backup_thread = NULL; } else { return -EINVAL; } @@ -1650,12 +1657,29 @@ int stop_sync_thread(int state) */ static int __net_init __ip_vs_sync_init(struct net *net) { + struct netns_ipvs *ipvs = net_ipvs(net); + + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + + INIT_LIST_HEAD(&ipvs->sync_queue); + spin_lock_init(&ipvs->sync_lock); + spin_lock_init(&ipvs->sync_buff_lock); + + ipvs->sync_mcast_addr.sin_family = AF_INET; + ipvs->sync_mcast_addr.sin_port = cpu_to_be16(IP_VS_SYNC_PORT); + ipvs->sync_mcast_addr.sin_addr.s_addr = cpu_to_be32(IP_VS_SYNC_GROUP); return 0; } static void __ip_vs_sync_cleanup(struct net *net) { + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return; + stop_sync_thread(net, IP_VS_STATE_MASTER); + stop_sync_thread(net, IP_VS_STATE_BACKUP); } + static struct pernet_operations ipvs_sync_ops = { .init = __ip_vs_sync_init, .exit = __ip_vs_sync_cleanup, -- cgit v1.2.3 From b17fc9963f837ef1acfe36e193108fb16ed58647 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:56 +0100 Subject: IPVS: netns, ip_vs_stats and its procfs The statistic counter locks for every packet are now removed, and that statistic is now per CPU, i.e. no locks needed. However summing is made in ip_vs_est into ip_vs_stats struct which is moved to ipvs struc. procfs, ip_vs_stats now have a "per cpu" count and a grand total. A new function seq_file_single_net() in ip_vs.h created for handling of single_open_net() since it does not place net ptr in a struct, like others. /var/lib/lxc # cat /proc/net/ip_vs_stats_percpu Total Incoming Outgoing Incoming Outgoing CPU Conns Packets Packets Bytes Bytes 0 0 3 1 9D 34 1 0 1 2 49 70 2 0 1 2 34 76 3 1 2 2 70 74 ~ 1 7 7 18A 18E Conns/s Pkts/s Pkts/s Bytes/s Bytes/s 0 0 0 0 0 *v3 ip_vs_stats reamains as before, instead ip_vs_stats_percpu is added. u64 seq lock added *v4 Bug correction inbytes and outbytes as own vars.. per_cpu counter for all stats now as suggested by Julian. [horms@verge.net.au: removed whitespace-change-only hunk] Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 89 ++++++++++++++------------ net/netfilter/ipvs/ip_vs_ctl.c | 134 ++++++++++++++++++++++++++++++++++------ net/netfilter/ipvs/ip_vs_est.c | 39 ++++++++++++ 3 files changed, 204 insertions(+), 58 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 5531d569aa5e..7e6a2a046bf5 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -115,21 +115,28 @@ static inline void ip_vs_in_stats(struct ip_vs_conn *cp, struct sk_buff *skb) { struct ip_vs_dest *dest = cp->dest; + struct netns_ipvs *ipvs = net_ipvs(skb_net(skb)); + if (dest && (dest->flags & IP_VS_DEST_F_AVAILABLE)) { - spin_lock(&dest->stats.lock); - dest->stats.ustats.inpkts++; - dest->stats.ustats.inbytes += skb->len; - spin_unlock(&dest->stats.lock); - - spin_lock(&dest->svc->stats.lock); - dest->svc->stats.ustats.inpkts++; - dest->svc->stats.ustats.inbytes += skb->len; - spin_unlock(&dest->svc->stats.lock); - - spin_lock(&ip_vs_stats.lock); - ip_vs_stats.ustats.inpkts++; - ip_vs_stats.ustats.inbytes += skb->len; - spin_unlock(&ip_vs_stats.lock); + struct ip_vs_cpu_stats *s; + + s = this_cpu_ptr(dest->stats.cpustats); + s->ustats.inpkts++; + u64_stats_update_begin(&s->syncp); + s->ustats.inbytes += skb->len; + u64_stats_update_end(&s->syncp); + + s = this_cpu_ptr(dest->svc->stats.cpustats); + s->ustats.inpkts++; + u64_stats_update_begin(&s->syncp); + s->ustats.inbytes += skb->len; + u64_stats_update_end(&s->syncp); + + s = this_cpu_ptr(ipvs->cpustats); + s->ustats.inpkts++; + u64_stats_update_begin(&s->syncp); + s->ustats.inbytes += skb->len; + u64_stats_update_end(&s->syncp); } } @@ -138,21 +145,28 @@ static inline void ip_vs_out_stats(struct ip_vs_conn *cp, struct sk_buff *skb) { struct ip_vs_dest *dest = cp->dest; + struct netns_ipvs *ipvs = net_ipvs(skb_net(skb)); + if (dest && (dest->flags & IP_VS_DEST_F_AVAILABLE)) { - spin_lock(&dest->stats.lock); - dest->stats.ustats.outpkts++; - dest->stats.ustats.outbytes += skb->len; - spin_unlock(&dest->stats.lock); - - spin_lock(&dest->svc->stats.lock); - dest->svc->stats.ustats.outpkts++; - dest->svc->stats.ustats.outbytes += skb->len; - spin_unlock(&dest->svc->stats.lock); - - spin_lock(&ip_vs_stats.lock); - ip_vs_stats.ustats.outpkts++; - ip_vs_stats.ustats.outbytes += skb->len; - spin_unlock(&ip_vs_stats.lock); + struct ip_vs_cpu_stats *s; + + s = this_cpu_ptr(dest->stats.cpustats); + s->ustats.outpkts++; + u64_stats_update_begin(&s->syncp); + s->ustats.outbytes += skb->len; + u64_stats_update_end(&s->syncp); + + s = this_cpu_ptr(dest->svc->stats.cpustats); + s->ustats.outpkts++; + u64_stats_update_begin(&s->syncp); + s->ustats.outbytes += skb->len; + u64_stats_update_end(&s->syncp); + + s = this_cpu_ptr(ipvs->cpustats); + s->ustats.outpkts++; + u64_stats_update_begin(&s->syncp); + s->ustats.outbytes += skb->len; + u64_stats_update_end(&s->syncp); } } @@ -160,17 +174,17 @@ ip_vs_out_stats(struct ip_vs_conn *cp, struct sk_buff *skb) static inline void ip_vs_conn_stats(struct ip_vs_conn *cp, struct ip_vs_service *svc) { - spin_lock(&cp->dest->stats.lock); - cp->dest->stats.ustats.conns++; - spin_unlock(&cp->dest->stats.lock); + struct netns_ipvs *ipvs = net_ipvs(svc->net); + struct ip_vs_cpu_stats *s; - spin_lock(&svc->stats.lock); - svc->stats.ustats.conns++; - spin_unlock(&svc->stats.lock); + s = this_cpu_ptr(cp->dest->stats.cpustats); + s->ustats.conns++; - spin_lock(&ip_vs_stats.lock); - ip_vs_stats.ustats.conns++; - spin_unlock(&ip_vs_stats.lock); + s = this_cpu_ptr(svc->stats.cpustats); + s->ustats.conns++; + + s = this_cpu_ptr(ipvs->cpustats); + s->ustats.conns++; } @@ -1841,7 +1855,6 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = { }, #endif }; - /* * Initialize IP Virtual Server netns mem. */ diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 03f86312b4bb..cbd58c60e1bf 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -257,8 +257,7 @@ static DECLARE_DELAYED_WORK(defense_work, defense_work_handler); static void defense_work_handler(struct work_struct *work) { - struct net *net = &init_net; - struct netns_ipvs *ipvs = net_ipvs(net); + struct netns_ipvs *ipvs = net_ipvs(&init_net); update_defense_level(ipvs); if (atomic_read(&ip_vs_dropentry)) @@ -519,6 +518,7 @@ __ip_vs_unbind_svc(struct ip_vs_dest *dest) svc->fwmark, IP_VS_DBG_ADDR(svc->af, &svc->addr), ntohs(svc->port), atomic_read(&svc->usecnt)); + free_percpu(svc->stats.cpustats); kfree(svc); } } @@ -722,6 +722,7 @@ ip_vs_trash_get_dest(struct ip_vs_service *svc, const union nf_inet_addr *daddr, list_del(&dest->n_list); ip_vs_dst_reset(dest); __ip_vs_unbind_svc(dest); + free_percpu(dest->stats.cpustats); kfree(dest); } } @@ -747,6 +748,7 @@ static void ip_vs_trash_cleanup(void) list_del(&dest->n_list); ip_vs_dst_reset(dest); __ip_vs_unbind_svc(dest); + free_percpu(dest->stats.cpustats); kfree(dest); } } @@ -868,6 +870,11 @@ ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest, pr_err("%s(): no memory.\n", __func__); return -ENOMEM; } + dest->stats.cpustats = alloc_percpu(struct ip_vs_cpu_stats); + if (!dest->stats.cpustats) { + pr_err("%s() alloc_percpu failed\n", __func__); + goto err_alloc; + } dest->af = svc->af; dest->protocol = svc->protocol; @@ -891,6 +898,10 @@ ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest, LeaveFunction(2); return 0; + +err_alloc: + kfree(dest); + return -ENOMEM; } @@ -1037,6 +1048,7 @@ static void __ip_vs_del_dest(struct net *net, struct ip_vs_dest *dest) and only one user context can update virtual service at a time, so the operation here is OK */ atomic_dec(&dest->svc->refcnt); + free_percpu(dest->stats.cpustats); kfree(dest); } else { IP_VS_DBG_BUF(3, "Moving dest %s:%u into trash, " @@ -1163,6 +1175,11 @@ ip_vs_add_service(struct net *net, struct ip_vs_service_user_kern *u, ret = -ENOMEM; goto out_err; } + svc->stats.cpustats = alloc_percpu(struct ip_vs_cpu_stats); + if (!svc->stats.cpustats) { + pr_err("%s() alloc_percpu failed\n", __func__); + goto out_err; + } /* I'm the first user of the service */ atomic_set(&svc->usecnt, 0); @@ -1212,6 +1229,7 @@ ip_vs_add_service(struct net *net, struct ip_vs_service_user_kern *u, *svc_p = svc; return 0; + out_err: if (svc != NULL) { ip_vs_unbind_scheduler(svc); @@ -1220,6 +1238,8 @@ ip_vs_add_service(struct net *net, struct ip_vs_service_user_kern *u, ip_vs_app_inc_put(svc->inc); local_bh_enable(); } + if (svc->stats.cpustats) + free_percpu(svc->stats.cpustats); kfree(svc); } ip_vs_scheduler_put(sched); @@ -1388,6 +1408,7 @@ static void __ip_vs_del_service(struct ip_vs_service *svc) svc->fwmark, IP_VS_DBG_ADDR(svc->af, &svc->addr), ntohs(svc->port), atomic_read(&svc->usecnt)); + free_percpu(svc->stats.cpustats); kfree(svc); } @@ -1499,7 +1520,7 @@ static int ip_vs_zero_all(struct net *net) } } - ip_vs_zero_stats(&ip_vs_stats); + ip_vs_zero_stats(net_ipvs(net)->tot_stats); return 0; } @@ -1989,13 +2010,11 @@ static const struct file_operations ip_vs_info_fops = { #endif -struct ip_vs_stats ip_vs_stats = { - .lock = __SPIN_LOCK_UNLOCKED(ip_vs_stats.lock), -}; - #ifdef CONFIG_PROC_FS static int ip_vs_stats_show(struct seq_file *seq, void *v) { + struct net *net = seq_file_single_net(seq); + struct ip_vs_stats *tot_stats = net_ipvs(net)->tot_stats; /* 01234567 01234567 01234567 0123456701234567 0123456701234567 */ seq_puts(seq, @@ -2003,22 +2022,22 @@ static int ip_vs_stats_show(struct seq_file *seq, void *v) seq_printf(seq, " Conns Packets Packets Bytes Bytes\n"); - spin_lock_bh(&ip_vs_stats.lock); - seq_printf(seq, "%8X %8X %8X %16LX %16LX\n\n", ip_vs_stats.ustats.conns, - ip_vs_stats.ustats.inpkts, ip_vs_stats.ustats.outpkts, - (unsigned long long) ip_vs_stats.ustats.inbytes, - (unsigned long long) ip_vs_stats.ustats.outbytes); + spin_lock_bh(&tot_stats->lock); + seq_printf(seq, "%8X %8X %8X %16LX %16LX\n\n", tot_stats->ustats.conns, + tot_stats->ustats.inpkts, tot_stats->ustats.outpkts, + (unsigned long long) tot_stats->ustats.inbytes, + (unsigned long long) tot_stats->ustats.outbytes); /* 01234567 01234567 01234567 0123456701234567 0123456701234567 */ seq_puts(seq, " Conns/s Pkts/s Pkts/s Bytes/s Bytes/s\n"); seq_printf(seq,"%8X %8X %8X %16X %16X\n", - ip_vs_stats.ustats.cps, - ip_vs_stats.ustats.inpps, - ip_vs_stats.ustats.outpps, - ip_vs_stats.ustats.inbps, - ip_vs_stats.ustats.outbps); - spin_unlock_bh(&ip_vs_stats.lock); + tot_stats->ustats.cps, + tot_stats->ustats.inpps, + tot_stats->ustats.outpps, + tot_stats->ustats.inbps, + tot_stats->ustats.outbps); + spin_unlock_bh(&tot_stats->lock); return 0; } @@ -2036,6 +2055,59 @@ static const struct file_operations ip_vs_stats_fops = { .release = single_release, }; +static int ip_vs_stats_percpu_show(struct seq_file *seq, void *v) +{ + struct net *net = seq_file_single_net(seq); + struct ip_vs_stats *tot_stats = net_ipvs(net)->tot_stats; + int i; + +/* 01234567 01234567 01234567 0123456701234567 0123456701234567 */ + seq_puts(seq, + " Total Incoming Outgoing Incoming Outgoing\n"); + seq_printf(seq, + "CPU Conns Packets Packets Bytes Bytes\n"); + + for_each_possible_cpu(i) { + struct ip_vs_cpu_stats *u = per_cpu_ptr(net->ipvs->cpustats, i); + seq_printf(seq, "%3X %8X %8X %8X %16LX %16LX\n", + i, u->ustats.conns, u->ustats.inpkts, + u->ustats.outpkts, (__u64)u->ustats.inbytes, + (__u64)u->ustats.outbytes); + } + + spin_lock_bh(&tot_stats->lock); + seq_printf(seq, " ~ %8X %8X %8X %16LX %16LX\n\n", + tot_stats->ustats.conns, tot_stats->ustats.inpkts, + tot_stats->ustats.outpkts, + (unsigned long long) tot_stats->ustats.inbytes, + (unsigned long long) tot_stats->ustats.outbytes); + +/* 01234567 01234567 01234567 0123456701234567 0123456701234567 */ + seq_puts(seq, + " Conns/s Pkts/s Pkts/s Bytes/s Bytes/s\n"); + seq_printf(seq, " %8X %8X %8X %16X %16X\n", + tot_stats->ustats.cps, + tot_stats->ustats.inpps, + tot_stats->ustats.outpps, + tot_stats->ustats.inbps, + tot_stats->ustats.outbps); + spin_unlock_bh(&tot_stats->lock); + + return 0; +} + +static int ip_vs_stats_percpu_seq_open(struct inode *inode, struct file *file) +{ + return single_open_net(inode, file, ip_vs_stats_percpu_show); +} + +static const struct file_operations ip_vs_stats_percpu_fops = { + .owner = THIS_MODULE, + .open = ip_vs_stats_percpu_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; #endif /* @@ -3461,32 +3533,54 @@ int __net_init __ip_vs_control_init(struct net *net) if (!net_eq(net, &init_net)) /* netns not enabled yet */ return -EPERM; + /* procfs stats */ + ipvs->tot_stats = kzalloc(sizeof(struct ip_vs_stats), GFP_KERNEL); + if (ipvs->tot_stats == NULL) { + pr_err("%s(): no memory.\n", __func__); + return -ENOMEM; + } + ipvs->cpustats = alloc_percpu(struct ip_vs_cpu_stats); + if (!ipvs->cpustats) { + pr_err("%s() alloc_percpu failed\n", __func__); + goto err_alloc; + } + spin_lock_init(&ipvs->tot_stats->lock); for (idx = 0; idx < IP_VS_RTAB_SIZE; idx++) INIT_LIST_HEAD(&ipvs->rs_table[idx]); proc_net_fops_create(net, "ip_vs", 0, &ip_vs_info_fops); proc_net_fops_create(net, "ip_vs_stats", 0, &ip_vs_stats_fops); + proc_net_fops_create(net, "ip_vs_stats_percpu", 0, + &ip_vs_stats_percpu_fops); sysctl_header = register_net_sysctl_table(net, net_vs_ctl_path, vs_vars); if (sysctl_header == NULL) goto err_reg; - ip_vs_new_estimator(net, &ip_vs_stats); + ip_vs_new_estimator(net, ipvs->tot_stats); return 0; err_reg: + free_percpu(ipvs->cpustats); +err_alloc: + kfree(ipvs->tot_stats); return -ENOMEM; } static void __net_exit __ip_vs_control_cleanup(struct net *net) { + struct netns_ipvs *ipvs = net_ipvs(net); + if (!net_eq(net, &init_net)) /* netns not enabled yet */ return; - ip_vs_kill_estimator(net, &ip_vs_stats); + ip_vs_kill_estimator(net, ipvs->tot_stats); unregister_net_sysctl_table(sysctl_header); + proc_net_remove(net, "ip_vs_stats_percpu"); proc_net_remove(net, "ip_vs_stats"); proc_net_remove(net, "ip_vs"); + free_percpu(ipvs->cpustats); + kfree(ipvs->tot_stats); } static struct pernet_operations ipvs_control_ops = { diff --git a/net/netfilter/ipvs/ip_vs_est.c b/net/netfilter/ipvs/ip_vs_est.c index 07d839bef537..d13616b138cd 100644 --- a/net/netfilter/ipvs/ip_vs_est.c +++ b/net/netfilter/ipvs/ip_vs_est.c @@ -52,6 +52,43 @@ */ +/* + * Make a summary from each cpu + */ +static void ip_vs_read_cpu_stats(struct ip_vs_stats_user *sum, + struct ip_vs_cpu_stats *stats) +{ + int i; + + for_each_possible_cpu(i) { + struct ip_vs_cpu_stats *s = per_cpu_ptr(stats, i); + unsigned int start; + __u64 inbytes, outbytes; + if (i) { + sum->conns += s->ustats.conns; + sum->inpkts += s->ustats.inpkts; + sum->outpkts += s->ustats.outpkts; + do { + start = u64_stats_fetch_begin_bh(&s->syncp); + inbytes = s->ustats.inbytes; + outbytes = s->ustats.outbytes; + } while (u64_stats_fetch_retry_bh(&s->syncp, start)); + sum->inbytes += inbytes; + sum->outbytes += outbytes; + } else { + sum->conns = s->ustats.conns; + sum->inpkts = s->ustats.inpkts; + sum->outpkts = s->ustats.outpkts; + do { + start = u64_stats_fetch_begin_bh(&s->syncp); + sum->inbytes = s->ustats.inbytes; + sum->outbytes = s->ustats.outbytes; + } while (u64_stats_fetch_retry_bh(&s->syncp, start)); + } + } +} + + static void estimation_timer(unsigned long arg) { struct ip_vs_estimator *e; @@ -64,10 +101,12 @@ static void estimation_timer(unsigned long arg) struct netns_ipvs *ipvs; ipvs = net_ipvs(net); + ip_vs_read_cpu_stats(&ipvs->tot_stats->ustats, ipvs->cpustats); spin_lock(&ipvs->est_lock); list_for_each_entry(e, &ipvs->est_list, list) { s = container_of(e, struct ip_vs_stats, est); + ip_vs_read_cpu_stats(&s->ustats, s->cpustats); spin_lock(&s->lock); n_conns = s->ustats.conns; n_inpkts = s->ustats.inpkts; -- cgit v1.2.3 From 6e67e586e7289c144d5a189d6e0fa7141d025746 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:57 +0100 Subject: IPVS: netns, connection hash got net as param. Connection hash table is now name space aware. i.e. net ptr >> 8 is xor:ed to the hash, and this is the first param to be compared. The net struct is 0xa40 in size ( a little bit smaller for 32 bit arch:s) and cache-line aligned, so a ptr >> 5 might be a more clever solution ? All lookups where net is compared uses net_eq() which returns 1 when netns is disabled, and the compiler seems to do something clever in that case. ip_vs_conn_fill_param() have *net as first param now. Three new inlines added to keep conn struct smaller when names space is disabled. - ip_vs_conn_net() - ip_vs_conn_net_set() - ip_vs_conn_net_eq() *v3 moved net compare to the end in "fast path" Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_conn.c | 112 ++++++++++++++++++++------------ net/netfilter/ipvs/ip_vs_core.c | 15 +++-- net/netfilter/ipvs/ip_vs_ftp.c | 14 ++-- net/netfilter/ipvs/ip_vs_nfct.c | 6 +- net/netfilter/ipvs/ip_vs_proto_ah_esp.c | 15 +++-- net/netfilter/ipvs/ip_vs_proto_sctp.c | 2 +- net/netfilter/ipvs/ip_vs_proto_tcp.c | 2 +- net/netfilter/ipvs/ip_vs_proto_udp.c | 2 +- net/netfilter/ipvs/ip_vs_sync.c | 13 ++-- 9 files changed, 109 insertions(+), 72 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index b2024c942345..0d5e4feabc1b 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -64,9 +64,6 @@ static struct list_head *ip_vs_conn_tab __read_mostly; /* SLAB cache for IPVS connections */ static struct kmem_cache *ip_vs_conn_cachep __read_mostly; -/* counter for current IPVS connections */ -static atomic_t ip_vs_conn_count = ATOMIC_INIT(0); - /* counter for no client port connections */ static atomic_t ip_vs_conn_no_cport_cnt = ATOMIC_INIT(0); @@ -76,7 +73,7 @@ static unsigned int ip_vs_conn_rnd __read_mostly; /* * Fine locking granularity for big connection hash table */ -#define CT_LOCKARRAY_BITS 4 +#define CT_LOCKARRAY_BITS 5 #define CT_LOCKARRAY_SIZE (1<>8)) & ip_vs_conn_tab_mask; #endif - return jhash_3words((__force u32)addr->ip, (__force u32)port, proto, - ip_vs_conn_rnd) - & ip_vs_conn_tab_mask; + return (jhash_3words((__force u32)addr->ip, (__force u32)port, proto, + ip_vs_conn_rnd) ^ + ((size_t)net>>8)) & ip_vs_conn_tab_mask; } static unsigned int ip_vs_conn_hashkey_param(const struct ip_vs_conn_param *p, @@ -166,15 +163,15 @@ static unsigned int ip_vs_conn_hashkey_param(const struct ip_vs_conn_param *p, port = p->vport; } - return ip_vs_conn_hashkey(p->af, p->protocol, addr, port); + return ip_vs_conn_hashkey(p->net, p->af, p->protocol, addr, port); } static unsigned int ip_vs_conn_hashkey_conn(const struct ip_vs_conn *cp) { struct ip_vs_conn_param p; - ip_vs_conn_fill_param(cp->af, cp->protocol, &cp->caddr, cp->cport, - NULL, 0, &p); + ip_vs_conn_fill_param(ip_vs_conn_net(cp), cp->af, cp->protocol, + &cp->caddr, cp->cport, NULL, 0, &p); if (cp->pe) { p.pe = cp->pe; @@ -186,7 +183,7 @@ static unsigned int ip_vs_conn_hashkey_conn(const struct ip_vs_conn *cp) } /* - * Hashes ip_vs_conn in ip_vs_conn_tab by proto,addr,port. + * Hashes ip_vs_conn in ip_vs_conn_tab by netns,proto,addr,port. * returns bool success. */ static inline int ip_vs_conn_hash(struct ip_vs_conn *cp) @@ -269,11 +266,12 @@ __ip_vs_conn_in_get(const struct ip_vs_conn_param *p) list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { if (cp->af == p->af && + p->cport == cp->cport && p->vport == cp->vport && ip_vs_addr_equal(p->af, p->caddr, &cp->caddr) && ip_vs_addr_equal(p->af, p->vaddr, &cp->vaddr) && - p->cport == cp->cport && p->vport == cp->vport && ((!p->cport) ^ (!(cp->flags & IP_VS_CONN_F_NO_CPORT))) && - p->protocol == cp->protocol) { + p->protocol == cp->protocol && + ip_vs_conn_net_eq(cp, p->net)) { /* HIT */ atomic_inc(&cp->refcnt); ct_read_unlock(hash); @@ -313,17 +311,18 @@ ip_vs_conn_fill_param_proto(int af, const struct sk_buff *skb, struct ip_vs_conn_param *p) { __be16 _ports[2], *pptr; + struct net *net = skb_net(skb); pptr = skb_header_pointer(skb, proto_off, sizeof(_ports), _ports); if (pptr == NULL) return 1; if (likely(!inverse)) - ip_vs_conn_fill_param(af, iph->protocol, &iph->saddr, pptr[0], - &iph->daddr, pptr[1], p); + ip_vs_conn_fill_param(net, af, iph->protocol, &iph->saddr, + pptr[0], &iph->daddr, pptr[1], p); else - ip_vs_conn_fill_param(af, iph->protocol, &iph->daddr, pptr[1], - &iph->saddr, pptr[0], p); + ip_vs_conn_fill_param(net, af, iph->protocol, &iph->daddr, + pptr[1], &iph->saddr, pptr[0], p); return 0; } @@ -352,6 +351,8 @@ struct ip_vs_conn *ip_vs_ct_in_get(const struct ip_vs_conn_param *p) ct_read_lock(hash); list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { + if (!ip_vs_conn_net_eq(cp, p->net)) + continue; if (p->pe_data && p->pe->ct_match) { if (p->pe == cp->pe && p->pe->ct_match(p, cp)) goto out; @@ -403,10 +404,11 @@ struct ip_vs_conn *ip_vs_conn_out_get(const struct ip_vs_conn_param *p) list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { if (cp->af == p->af && + p->vport == cp->cport && p->cport == cp->dport && ip_vs_addr_equal(p->af, p->vaddr, &cp->caddr) && ip_vs_addr_equal(p->af, p->caddr, &cp->daddr) && - p->vport == cp->cport && p->cport == cp->dport && - p->protocol == cp->protocol) { + p->protocol == cp->protocol && + ip_vs_conn_net_eq(cp, p->net)) { /* HIT */ atomic_inc(&cp->refcnt); ret = cp; @@ -609,8 +611,8 @@ struct ip_vs_dest *ip_vs_try_bind_dest(struct ip_vs_conn *cp) struct ip_vs_dest *dest; if ((cp) && (!cp->dest)) { - dest = ip_vs_find_dest(&init_net, cp->af, &cp->daddr, cp->dport, - &cp->vaddr, cp->vport, + dest = ip_vs_find_dest(ip_vs_conn_net(cp), cp->af, &cp->daddr, + cp->dport, &cp->vaddr, cp->vport, cp->protocol, cp->fwmark); ip_vs_bind_dest(cp, dest); return dest; @@ -728,6 +730,7 @@ int ip_vs_check_template(struct ip_vs_conn *ct) static void ip_vs_conn_expire(unsigned long data) { struct ip_vs_conn *cp = (struct ip_vs_conn *)data; + struct netns_ipvs *ipvs = net_ipvs(ip_vs_conn_net(cp)); cp->timeout = 60*HZ; @@ -770,7 +773,7 @@ static void ip_vs_conn_expire(unsigned long data) ip_vs_unbind_dest(cp); if (cp->flags & IP_VS_CONN_F_NO_CPORT) atomic_dec(&ip_vs_conn_no_cport_cnt); - atomic_dec(&ip_vs_conn_count); + atomic_dec(&ipvs->conn_count); kmem_cache_free(ip_vs_conn_cachep, cp); return; @@ -804,7 +807,9 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p, struct ip_vs_dest *dest, __u32 fwmark) { struct ip_vs_conn *cp; - struct ip_vs_proto_data *pd = ip_vs_proto_data_get(&init_net, p->protocol); + struct netns_ipvs *ipvs = net_ipvs(p->net); + struct ip_vs_proto_data *pd = ip_vs_proto_data_get(p->net, + p->protocol); cp = kmem_cache_zalloc(ip_vs_conn_cachep, GFP_ATOMIC); if (cp == NULL) { @@ -814,6 +819,7 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p, INIT_LIST_HEAD(&cp->c_list); setup_timer(&cp->timer, ip_vs_conn_expire, (unsigned long)cp); + ip_vs_conn_net_set(cp, p->net); cp->af = p->af; cp->protocol = p->protocol; ip_vs_addr_copy(p->af, &cp->caddr, p->caddr); @@ -844,7 +850,7 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p, atomic_set(&cp->n_control, 0); atomic_set(&cp->in_pkts, 0); - atomic_inc(&ip_vs_conn_count); + atomic_inc(&ipvs->conn_count); if (flags & IP_VS_CONN_F_NO_CPORT) atomic_inc(&ip_vs_conn_no_cport_cnt); @@ -886,17 +892,22 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p, * /proc/net/ip_vs_conn entries */ #ifdef CONFIG_PROC_FS +struct ip_vs_iter_state { + struct seq_net_private p; + struct list_head *l; +}; static void *ip_vs_conn_array(struct seq_file *seq, loff_t pos) { int idx; struct ip_vs_conn *cp; + struct ip_vs_iter_state *iter = seq->private; for (idx = 0; idx < ip_vs_conn_tab_size; idx++) { ct_read_lock_bh(idx); list_for_each_entry(cp, &ip_vs_conn_tab[idx], c_list) { if (pos-- == 0) { - seq->private = &ip_vs_conn_tab[idx]; + iter->l = &ip_vs_conn_tab[idx]; return cp; } } @@ -908,14 +919,17 @@ static void *ip_vs_conn_array(struct seq_file *seq, loff_t pos) static void *ip_vs_conn_seq_start(struct seq_file *seq, loff_t *pos) { - seq->private = NULL; + struct ip_vs_iter_state *iter = seq->private; + + iter->l = NULL; return *pos ? ip_vs_conn_array(seq, *pos - 1) :SEQ_START_TOKEN; } static void *ip_vs_conn_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct ip_vs_conn *cp = v; - struct list_head *e, *l = seq->private; + struct ip_vs_iter_state *iter = seq->private; + struct list_head *e, *l = iter->l; int idx; ++*pos; @@ -932,18 +946,19 @@ static void *ip_vs_conn_seq_next(struct seq_file *seq, void *v, loff_t *pos) while (++idx < ip_vs_conn_tab_size) { ct_read_lock_bh(idx); list_for_each_entry(cp, &ip_vs_conn_tab[idx], c_list) { - seq->private = &ip_vs_conn_tab[idx]; + iter->l = &ip_vs_conn_tab[idx]; return cp; } ct_read_unlock_bh(idx); } - seq->private = NULL; + iter->l = NULL; return NULL; } static void ip_vs_conn_seq_stop(struct seq_file *seq, void *v) { - struct list_head *l = seq->private; + struct ip_vs_iter_state *iter = seq->private; + struct list_head *l = iter->l; if (l) ct_read_unlock_bh(l - ip_vs_conn_tab); @@ -957,9 +972,12 @@ static int ip_vs_conn_seq_show(struct seq_file *seq, void *v) "Pro FromIP FPrt ToIP TPrt DestIP DPrt State Expires PEName PEData\n"); else { const struct ip_vs_conn *cp = v; + struct net *net = seq_file_net(seq); char pe_data[IP_VS_PENAME_MAXLEN + IP_VS_PEDATA_MAXLEN + 3]; size_t len = 0; + if (!ip_vs_conn_net_eq(cp, net)) + return 0; if (cp->pe_data) { pe_data[0] = ' '; len = strlen(cp->pe->name); @@ -1004,7 +1022,8 @@ static const struct seq_operations ip_vs_conn_seq_ops = { static int ip_vs_conn_open(struct inode *inode, struct file *file) { - return seq_open(file, &ip_vs_conn_seq_ops); + return seq_open_net(inode, file, &ip_vs_conn_seq_ops, + sizeof(struct ip_vs_iter_state)); } static const struct file_operations ip_vs_conn_fops = { @@ -1031,6 +1050,10 @@ static int ip_vs_conn_sync_seq_show(struct seq_file *seq, void *v) "Pro FromIP FPrt ToIP TPrt DestIP DPrt State Origin Expires\n"); else { const struct ip_vs_conn *cp = v; + struct net *net = seq_file_net(seq); + + if (!ip_vs_conn_net_eq(cp, net)) + return 0; #ifdef CONFIG_IP_VS_IPV6 if (cp->af == AF_INET6) @@ -1067,7 +1090,8 @@ static const struct seq_operations ip_vs_conn_sync_seq_ops = { static int ip_vs_conn_sync_open(struct inode *inode, struct file *file) { - return seq_open(file, &ip_vs_conn_sync_seq_ops); + return seq_open_net(inode, file, &ip_vs_conn_sync_seq_ops, + sizeof(struct ip_vs_iter_state)); } static const struct file_operations ip_vs_conn_sync_fops = { @@ -1168,10 +1192,11 @@ void ip_vs_random_dropentry(void) /* * Flush all the connection entries in the ip_vs_conn_tab */ -static void ip_vs_conn_flush(void) +static void ip_vs_conn_flush(struct net *net) { int idx; struct ip_vs_conn *cp; + struct netns_ipvs *ipvs = net_ipvs(net); flush_again: for (idx = 0; idx < ip_vs_conn_tab_size; idx++) { @@ -1181,7 +1206,8 @@ static void ip_vs_conn_flush(void) ct_write_lock_bh(idx); list_for_each_entry(cp, &ip_vs_conn_tab[idx], c_list) { - + if (!ip_vs_conn_net_eq(cp, net)) + continue; IP_VS_DBG(4, "del connection\n"); ip_vs_conn_expire_now(cp); if (cp->control) { @@ -1194,7 +1220,7 @@ static void ip_vs_conn_flush(void) /* the counter may be not NULL, because maybe some conn entries are run by slow timer handler or unhashed but still referred */ - if (atomic_read(&ip_vs_conn_count) != 0) { + if (atomic_read(&ipvs->conn_count) != 0) { schedule(); goto flush_again; } @@ -1204,8 +1230,11 @@ static void ip_vs_conn_flush(void) */ int __net_init __ip_vs_conn_init(struct net *net) { + struct netns_ipvs *ipvs = net_ipvs(net); + if (!net_eq(net, &init_net)) /* netns not enabled yet */ return -EPERM; + atomic_set(&ipvs->conn_count, 0); proc_net_fops_create(net, "ip_vs_conn", 0, &ip_vs_conn_fops); proc_net_fops_create(net, "ip_vs_conn_sync", 0, &ip_vs_conn_sync_fops); @@ -1217,6 +1246,8 @@ static void __net_exit __ip_vs_conn_cleanup(struct net *net) if (!net_eq(net, &init_net)) /* netns not enabled yet */ return; + /* flush all the connection entries first */ + ip_vs_conn_flush(net); proc_net_remove(net, "ip_vs_conn"); proc_net_remove(net, "ip_vs_conn_sync"); } @@ -1277,9 +1308,6 @@ int __init ip_vs_conn_init(void) void ip_vs_conn_cleanup(void) { unregister_pernet_subsys(&ipvs_conn_ops); - /* flush all the connection entries first */ - ip_vs_conn_flush(); - /* Release the empty cache */ kmem_cache_destroy(ip_vs_conn_cachep); vfree(ip_vs_conn_tab); diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 7e6a2a046bf5..7205b49c56c1 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -205,7 +205,8 @@ ip_vs_conn_fill_param_persist(const struct ip_vs_service *svc, const union nf_inet_addr *vaddr, __be16 vport, struct ip_vs_conn_param *p) { - ip_vs_conn_fill_param(svc->af, protocol, caddr, cport, vaddr, vport, p); + ip_vs_conn_fill_param(svc->net, svc->af, protocol, caddr, cport, vaddr, + vport, p); p->pe = svc->pe; if (p->pe && p->pe->fill_param) return p->pe->fill_param(p, skb); @@ -348,8 +349,8 @@ ip_vs_sched_persist(struct ip_vs_service *svc, /* * Create a new connection according to the template */ - ip_vs_conn_fill_param(svc->af, iph.protocol, &iph.saddr, src_port, - &iph.daddr, dst_port, ¶m); + ip_vs_conn_fill_param(svc->net, svc->af, iph.protocol, &iph.saddr, + src_port, &iph.daddr, dst_port, ¶m); cp = ip_vs_conn_new(¶m, &dest->addr, dport, flags, dest, skb->mark); if (cp == NULL) { @@ -464,8 +465,10 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, */ { struct ip_vs_conn_param p; - ip_vs_conn_fill_param(svc->af, iph.protocol, &iph.saddr, - pptr[0], &iph.daddr, pptr[1], &p); + + ip_vs_conn_fill_param(svc->net, svc->af, iph.protocol, + &iph.saddr, pptr[0], &iph.daddr, pptr[1], + &p); cp = ip_vs_conn_new(&p, &dest->addr, dest->port ? dest->port : pptr[1], flags, dest, skb->mark); @@ -532,7 +535,7 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, IP_VS_DBG(6, "%s(): create a cache_bypass entry\n", __func__); { struct ip_vs_conn_param p; - ip_vs_conn_fill_param(svc->af, iph.protocol, + ip_vs_conn_fill_param(svc->net, svc->af, iph.protocol, &iph.saddr, pptr[0], &iph.daddr, pptr[1], &p); cp = ip_vs_conn_new(&p, &daddr, 0, diff --git a/net/netfilter/ipvs/ip_vs_ftp.c b/net/netfilter/ipvs/ip_vs_ftp.c index 77b0036dcb73..6a04f9ab9d0d 100644 --- a/net/netfilter/ipvs/ip_vs_ftp.c +++ b/net/netfilter/ipvs/ip_vs_ftp.c @@ -198,13 +198,15 @@ static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp, */ { struct ip_vs_conn_param p; - ip_vs_conn_fill_param(AF_INET, iph->protocol, - &from, port, &cp->caddr, 0, &p); + ip_vs_conn_fill_param(ip_vs_conn_net(cp), AF_INET, + iph->protocol, &from, port, + &cp->caddr, 0, &p); n_cp = ip_vs_conn_out_get(&p); } if (!n_cp) { struct ip_vs_conn_param p; - ip_vs_conn_fill_param(AF_INET, IPPROTO_TCP, &cp->caddr, + ip_vs_conn_fill_param(ip_vs_conn_net(cp), + AF_INET, IPPROTO_TCP, &cp->caddr, 0, &cp->vaddr, port, &p); n_cp = ip_vs_conn_new(&p, &from, port, IP_VS_CONN_F_NO_CPORT | @@ -361,9 +363,9 @@ static int ip_vs_ftp_in(struct ip_vs_app *app, struct ip_vs_conn *cp, { struct ip_vs_conn_param p; - ip_vs_conn_fill_param(AF_INET, iph->protocol, &to, port, - &cp->vaddr, htons(ntohs(cp->vport)-1), - &p); + ip_vs_conn_fill_param(ip_vs_conn_net(cp), AF_INET, + iph->protocol, &to, port, &cp->vaddr, + htons(ntohs(cp->vport)-1), &p); n_cp = ip_vs_conn_in_get(&p); if (!n_cp) { n_cp = ip_vs_conn_new(&p, &cp->daddr, diff --git a/net/netfilter/ipvs/ip_vs_nfct.c b/net/netfilter/ipvs/ip_vs_nfct.c index 4680647cd450..f454c80df0a7 100644 --- a/net/netfilter/ipvs/ip_vs_nfct.c +++ b/net/netfilter/ipvs/ip_vs_nfct.c @@ -141,6 +141,7 @@ static void ip_vs_nfct_expect_callback(struct nf_conn *ct, struct nf_conntrack_tuple *orig, new_reply; struct ip_vs_conn *cp; struct ip_vs_conn_param p; + struct net *net = nf_ct_net(ct); if (exp->tuple.src.l3num != PF_INET) return; @@ -155,7 +156,7 @@ static void ip_vs_nfct_expect_callback(struct nf_conn *ct, /* RS->CLIENT */ orig = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; - ip_vs_conn_fill_param(exp->tuple.src.l3num, orig->dst.protonum, + ip_vs_conn_fill_param(net, exp->tuple.src.l3num, orig->dst.protonum, &orig->src.u3, orig->src.u.tcp.port, &orig->dst.u3, orig->dst.u.tcp.port, &p); cp = ip_vs_conn_out_get(&p); @@ -268,7 +269,8 @@ void ip_vs_conn_drop_conntrack(struct ip_vs_conn *cp) " for conn " FMT_CONN "\n", __func__, ARG_TUPLE(&tuple), ARG_CONN(cp)); - h = nf_conntrack_find_get(&init_net, NF_CT_DEFAULT_ZONE, &tuple); + h = nf_conntrack_find_get(ip_vs_conn_net(cp), NF_CT_DEFAULT_ZONE, + &tuple); if (h) { ct = nf_ct_tuplehash_to_ctrack(h); /* Show what happens instead of calling nf_ct_kill() */ diff --git a/net/netfilter/ipvs/ip_vs_proto_ah_esp.c b/net/netfilter/ipvs/ip_vs_proto_ah_esp.c index 28039cbfcff4..5b8eb8b12c3e 100644 --- a/net/netfilter/ipvs/ip_vs_proto_ah_esp.c +++ b/net/netfilter/ipvs/ip_vs_proto_ah_esp.c @@ -41,15 +41,16 @@ struct isakmp_hdr { #define PORT_ISAKMP 500 static void -ah_esp_conn_fill_param_proto(int af, const struct ip_vs_iphdr *iph, - int inverse, struct ip_vs_conn_param *p) +ah_esp_conn_fill_param_proto(struct net *net, int af, + const struct ip_vs_iphdr *iph, int inverse, + struct ip_vs_conn_param *p) { if (likely(!inverse)) - ip_vs_conn_fill_param(af, IPPROTO_UDP, + ip_vs_conn_fill_param(net, af, IPPROTO_UDP, &iph->saddr, htons(PORT_ISAKMP), &iph->daddr, htons(PORT_ISAKMP), p); else - ip_vs_conn_fill_param(af, IPPROTO_UDP, + ip_vs_conn_fill_param(net, af, IPPROTO_UDP, &iph->daddr, htons(PORT_ISAKMP), &iph->saddr, htons(PORT_ISAKMP), p); } @@ -61,8 +62,9 @@ ah_esp_conn_in_get(int af, const struct sk_buff *skb, { struct ip_vs_conn *cp; struct ip_vs_conn_param p; + struct net *net = skb_net(skb); - ah_esp_conn_fill_param_proto(af, iph, inverse, &p); + ah_esp_conn_fill_param_proto(net, af, iph, inverse, &p); cp = ip_vs_conn_in_get(&p); if (!cp) { /* @@ -89,8 +91,9 @@ ah_esp_conn_out_get(int af, const struct sk_buff *skb, { struct ip_vs_conn *cp; struct ip_vs_conn_param p; + struct net *net = skb_net(skb); - ah_esp_conn_fill_param_proto(af, iph, inverse, &p); + ah_esp_conn_fill_param_proto(net, af, iph, inverse, &p); cp = ip_vs_conn_out_get(&p); if (!cp) { IP_VS_DBG_BUF(12, "Unknown ISAKMP entry for inout packet " diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c index 569e77bf08c4..550365a690c7 100644 --- a/net/netfilter/ipvs/ip_vs_proto_sctp.c +++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c @@ -1055,7 +1055,7 @@ static void sctp_unregister_app(struct net *net, struct ip_vs_app *inc) static int sctp_app_conn_bind(struct ip_vs_conn *cp) { - struct netns_ipvs *ipvs = net_ipvs(&init_net); + struct netns_ipvs *ipvs = net_ipvs(ip_vs_conn_net(cp)); int hash; struct ip_vs_app *inc; int result = 0; diff --git a/net/netfilter/ipvs/ip_vs_proto_tcp.c b/net/netfilter/ipvs/ip_vs_proto_tcp.c index 757aaaf083bb..d8b3f9f15826 100644 --- a/net/netfilter/ipvs/ip_vs_proto_tcp.c +++ b/net/netfilter/ipvs/ip_vs_proto_tcp.c @@ -620,7 +620,7 @@ tcp_unregister_app(struct net *net, struct ip_vs_app *inc) static int tcp_app_conn_bind(struct ip_vs_conn *cp) { - struct netns_ipvs *ipvs = net_ipvs(&init_net); + struct netns_ipvs *ipvs = net_ipvs(ip_vs_conn_net(cp)); int hash; struct ip_vs_app *inc; int result = 0; diff --git a/net/netfilter/ipvs/ip_vs_proto_udp.c b/net/netfilter/ipvs/ip_vs_proto_udp.c index 1dc394100fa8..581157bbded5 100644 --- a/net/netfilter/ipvs/ip_vs_proto_udp.c +++ b/net/netfilter/ipvs/ip_vs_proto_udp.c @@ -396,7 +396,7 @@ udp_unregister_app(struct net *net, struct ip_vs_app *inc) static int udp_app_conn_bind(struct ip_vs_conn *cp) { - struct netns_ipvs *ipvs = net_ipvs(&init_net); + struct netns_ipvs *ipvs = net_ipvs(ip_vs_conn_net(cp)); int hash; struct ip_vs_app *inc; int result = 0; diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index c29e73d686fb..f85e47daecc3 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -660,21 +660,21 @@ control: * fill_param used by version 1 */ static inline int -ip_vs_conn_fill_param_sync(int af, union ip_vs_sync_conn *sc, +ip_vs_conn_fill_param_sync(struct net *net, int af, union ip_vs_sync_conn *sc, struct ip_vs_conn_param *p, __u8 *pe_data, unsigned int pe_data_len, __u8 *pe_name, unsigned int pe_name_len) { #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) - ip_vs_conn_fill_param(af, sc->v6.protocol, + ip_vs_conn_fill_param(net, af, sc->v6.protocol, (const union nf_inet_addr *)&sc->v6.caddr, sc->v6.cport, (const union nf_inet_addr *)&sc->v6.vaddr, sc->v6.vport, p); else #endif - ip_vs_conn_fill_param(af, sc->v4.protocol, + ip_vs_conn_fill_param(net, af, sc->v4.protocol, (const union nf_inet_addr *)&sc->v4.caddr, sc->v4.cport, (const union nf_inet_addr *)&sc->v4.vaddr, @@ -881,7 +881,7 @@ static void ip_vs_process_message_v0(struct net *net, const char *buffer, } } - ip_vs_conn_fill_param(AF_INET, s->protocol, + ip_vs_conn_fill_param(net, AF_INET, s->protocol, (const union nf_inet_addr *)&s->caddr, s->cport, (const union nf_inet_addr *)&s->vaddr, @@ -1043,9 +1043,8 @@ static inline int ip_vs_proc_sync_conn(struct net *net, __u8 *p, __u8 *msg_end) state = 0; } } - if (ip_vs_conn_fill_param_sync(af, s, ¶m, - pe_data, pe_data_len, - pe_name, pe_name_len)) { + if (ip_vs_conn_fill_param_sync(net, af, s, ¶m, pe_data, + pe_data_len, pe_name, pe_name_len)) { retc = 50; goto out; } -- cgit v1.2.3 From a0840e2e165a370ca24a59545e564e9881a55891 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:58 +0100 Subject: IPVS: netns, ip_vs_ctl local vars moved to ipvs struct. Moving global vars to ipvs struct, except for svc table lock. Next patch for ctl will be drop-rate handling. *v3 __ip_vs_mutex remains global ip_vs_conntrack_enabled(struct netns_ipvs *ipvs) Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_conn.c | 7 +- net/netfilter/ipvs/ip_vs_core.c | 34 ++-- net/netfilter/ipvs/ip_vs_ctl.c | 291 ++++++++++++++++++---------------- net/netfilter/ipvs/ip_vs_proto_sctp.c | 2 +- net/netfilter/ipvs/ip_vs_proto_tcp.c | 2 +- net/netfilter/ipvs/ip_vs_proto_udp.c | 2 +- net/netfilter/ipvs/ip_vs_sync.c | 9 +- 7 files changed, 184 insertions(+), 163 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 0d5e4feabc1b..5ba205a4d79c 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -686,13 +686,14 @@ static inline void ip_vs_unbind_dest(struct ip_vs_conn *cp) int ip_vs_check_template(struct ip_vs_conn *ct) { struct ip_vs_dest *dest = ct->dest; + struct netns_ipvs *ipvs = net_ipvs(ip_vs_conn_net(ct)); /* * Checking the dest server status. */ if ((dest == NULL) || !(dest->flags & IP_VS_DEST_F_AVAILABLE) || - (sysctl_ip_vs_expire_quiescent_template && + (ipvs->sysctl_expire_quiescent_template && (atomic_read(&dest->weight) == 0))) { IP_VS_DBG_BUF(9, "check_template: dest not available for " "protocol %s s:%s:%d v:%s:%d " @@ -879,7 +880,7 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p, * IP_VS_CONN_F_ONE_PACKET too. */ - if (ip_vs_conntrack_enabled()) + if (ip_vs_conntrack_enabled(ipvs)) cp->flags |= IP_VS_CONN_F_NFCT; /* Hash it in the ip_vs_conn_tab finally */ @@ -1198,7 +1199,7 @@ static void ip_vs_conn_flush(struct net *net) struct ip_vs_conn *cp; struct netns_ipvs *ipvs = net_ipvs(net); - flush_again: +flush_again: for (idx = 0; idx < ip_vs_conn_tab_size; idx++) { /* * Lock is actually needed in this loop. diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 7205b49c56c1..a7c59a722af3 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -499,6 +499,7 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, struct ip_vs_proto_data *pd) { + struct netns_ipvs *ipvs; __be16 _ports[2], *pptr; struct ip_vs_iphdr iph; int unicast; @@ -521,7 +522,8 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, /* if it is fwmark-based service, the cache_bypass sysctl is up and the destination is a non-local unicast, then create a cache_bypass connection entry */ - if (sysctl_ip_vs_cache_bypass && svc->fwmark && unicast) { + ipvs = net_ipvs(skb_net(skb)); + if (ipvs->sysctl_cache_bypass && svc->fwmark && unicast) { int ret, cs; struct ip_vs_conn *cp; unsigned int flags = (svc->flags & IP_VS_SVC_F_ONEPACKET && @@ -733,6 +735,7 @@ static int handle_response_icmp(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, unsigned int offset, unsigned int ihl) { + struct netns_ipvs *ipvs; unsigned int verdict = NF_DROP; if (IP_VS_FWD_METHOD(cp) != 0) { @@ -754,6 +757,8 @@ static int handle_response_icmp(int af, struct sk_buff *skb, if (!skb_make_writable(skb, offset)) goto out; + ipvs = net_ipvs(skb_net(skb)); + #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) ip_vs_nat_icmp_v6(skb, pp, cp, 1); @@ -763,11 +768,11 @@ static int handle_response_icmp(int af, struct sk_buff *skb, #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) { - if (sysctl_ip_vs_snat_reroute && ip6_route_me_harder(skb) != 0) + if (ipvs->sysctl_snat_reroute && ip6_route_me_harder(skb) != 0) goto out; } else #endif - if ((sysctl_ip_vs_snat_reroute || + if ((ipvs->sysctl_snat_reroute || skb_rtable(skb)->rt_flags & RTCF_LOCAL) && ip_route_me_harder(skb, RTN_LOCAL) != 0) goto out; @@ -979,6 +984,7 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, struct ip_vs_conn *cp, int ihl) { struct ip_vs_protocol *pp = pd->pp; + struct netns_ipvs *ipvs; IP_VS_DBG_PKT(11, af, pp, skb, 0, "Outgoing packet"); @@ -1014,13 +1020,15 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, * if it came from this machine itself. So re-compute * the routing information. */ + ipvs = net_ipvs(skb_net(skb)); + #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) { - if (sysctl_ip_vs_snat_reroute && ip6_route_me_harder(skb) != 0) + if (ipvs->sysctl_snat_reroute && ip6_route_me_harder(skb) != 0) goto drop; } else #endif - if ((sysctl_ip_vs_snat_reroute || + if ((ipvs->sysctl_snat_reroute || skb_rtable(skb)->rt_flags & RTCF_LOCAL) && ip_route_me_harder(skb, RTN_LOCAL) != 0) goto drop; @@ -1057,6 +1065,7 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af) struct ip_vs_protocol *pp; struct ip_vs_proto_data *pd; struct ip_vs_conn *cp; + struct netns_ipvs *ipvs; EnterFunction(11); @@ -1131,10 +1140,11 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af) * Check if the packet belongs to an existing entry */ cp = pp->conn_out_get(af, skb, &iph, iph.len, 0); + ipvs = net_ipvs(net); if (likely(cp)) return handle_response(af, skb, pd, cp, iph.len); - if (sysctl_ip_vs_nat_icmp_send && + if (ipvs->sysctl_nat_icmp_send && (pp->protocol == IPPROTO_TCP || pp->protocol == IPPROTO_UDP || pp->protocol == IPPROTO_SCTP)) { @@ -1580,7 +1590,7 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) if (cp->dest && !(cp->dest->flags & IP_VS_DEST_F_AVAILABLE)) { /* the destination server is not available */ - if (sysctl_ip_vs_expire_nodest_conn) { + if (ipvs->sysctl_expire_nodest_conn) { /* try to expire the connection immediately */ ip_vs_conn_expire_now(cp); } @@ -1610,15 +1620,15 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) */ if (cp->flags & IP_VS_CONN_F_ONE_PACKET) - pkts = sysctl_ip_vs_sync_threshold[0]; + pkts = ipvs->sysctl_sync_threshold[0]; else pkts = atomic_add_return(1, &cp->in_pkts); if ((ipvs->sync_state & IP_VS_STATE_MASTER) && cp->protocol == IPPROTO_SCTP) { if ((cp->state == IP_VS_SCTP_S_ESTABLISHED && - (pkts % sysctl_ip_vs_sync_threshold[1] - == sysctl_ip_vs_sync_threshold[0])) || + (pkts % ipvs->sysctl_sync_threshold[1] + == ipvs->sysctl_sync_threshold[0])) || (cp->old_state != cp->state && ((cp->state == IP_VS_SCTP_S_CLOSED) || (cp->state == IP_VS_SCTP_S_SHUT_ACK_CLI) || @@ -1632,8 +1642,8 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) else if ((ipvs->sync_state & IP_VS_STATE_MASTER) && (((cp->protocol != IPPROTO_TCP || cp->state == IP_VS_TCP_S_ESTABLISHED) && - (pkts % sysctl_ip_vs_sync_threshold[1] - == sysctl_ip_vs_sync_threshold[0])) || + (pkts % ipvs->sysctl_sync_threshold[1] + == ipvs->sysctl_sync_threshold[0])) || ((cp->protocol == IPPROTO_TCP) && (cp->old_state != cp->state) && ((cp->state == IP_VS_TCP_S_FIN_WAIT) || (cp->state == IP_VS_TCP_S_CLOSE) || diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index cbd58c60e1bf..183ac18bded5 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -58,42 +58,7 @@ static DEFINE_MUTEX(__ip_vs_mutex); /* lock for service table */ static DEFINE_RWLOCK(__ip_vs_svc_lock); -/* lock for table with the real services */ -static DEFINE_RWLOCK(__ip_vs_rs_lock); - -/* lock for state and timeout tables */ -static DEFINE_SPINLOCK(ip_vs_securetcp_lock); - -/* lock for drop entry handling */ -static DEFINE_SPINLOCK(__ip_vs_dropentry_lock); - -/* lock for drop packet handling */ -static DEFINE_SPINLOCK(__ip_vs_droppacket_lock); - -/* 1/rate drop and drop-entry variables */ -int ip_vs_drop_rate = 0; -int ip_vs_drop_counter = 0; -static atomic_t ip_vs_dropentry = ATOMIC_INIT(0); - -/* number of virtual services */ -static int ip_vs_num_services = 0; - /* sysctl variables */ -static int sysctl_ip_vs_drop_entry = 0; -static int sysctl_ip_vs_drop_packet = 0; -static int sysctl_ip_vs_secure_tcp = 0; -static int sysctl_ip_vs_amemthresh = 1024; -static int sysctl_ip_vs_am_droprate = 10; -int sysctl_ip_vs_cache_bypass = 0; -int sysctl_ip_vs_expire_nodest_conn = 0; -int sysctl_ip_vs_expire_quiescent_template = 0; -int sysctl_ip_vs_sync_threshold[2] = { 3, 50 }; -int sysctl_ip_vs_nat_icmp_send = 0; -#ifdef CONFIG_IP_VS_NFCT -int sysctl_ip_vs_conntrack; -#endif -int sysctl_ip_vs_snat_reroute = 1; -int sysctl_ip_vs_sync_ver = 1; /* Default version of sync proto */ #ifdef CONFIG_IP_VS_DEBUG static int sysctl_ip_vs_debug_level = 0; @@ -142,73 +107,73 @@ static void update_defense_level(struct netns_ipvs *ipvs) /* si_swapinfo(&i); */ /* availmem = availmem - (i.totalswap - i.freeswap); */ - nomem = (availmem < sysctl_ip_vs_amemthresh); + nomem = (availmem < ipvs->sysctl_amemthresh); local_bh_disable(); /* drop_entry */ - spin_lock(&__ip_vs_dropentry_lock); - switch (sysctl_ip_vs_drop_entry) { + spin_lock(&ipvs->dropentry_lock); + switch (ipvs->sysctl_drop_entry) { case 0: - atomic_set(&ip_vs_dropentry, 0); + atomic_set(&ipvs->dropentry, 0); break; case 1: if (nomem) { - atomic_set(&ip_vs_dropentry, 1); - sysctl_ip_vs_drop_entry = 2; + atomic_set(&ipvs->dropentry, 1); + ipvs->sysctl_drop_entry = 2; } else { - atomic_set(&ip_vs_dropentry, 0); + atomic_set(&ipvs->dropentry, 0); } break; case 2: if (nomem) { - atomic_set(&ip_vs_dropentry, 1); + atomic_set(&ipvs->dropentry, 1); } else { - atomic_set(&ip_vs_dropentry, 0); - sysctl_ip_vs_drop_entry = 1; + atomic_set(&ipvs->dropentry, 0); + ipvs->sysctl_drop_entry = 1; }; break; case 3: - atomic_set(&ip_vs_dropentry, 1); + atomic_set(&ipvs->dropentry, 1); break; } - spin_unlock(&__ip_vs_dropentry_lock); + spin_unlock(&ipvs->dropentry_lock); /* drop_packet */ - spin_lock(&__ip_vs_droppacket_lock); - switch (sysctl_ip_vs_drop_packet) { + spin_lock(&ipvs->droppacket_lock); + switch (ipvs->sysctl_drop_packet) { case 0: - ip_vs_drop_rate = 0; + ipvs->drop_rate = 0; break; case 1: if (nomem) { - ip_vs_drop_rate = ip_vs_drop_counter - = sysctl_ip_vs_amemthresh / - (sysctl_ip_vs_amemthresh-availmem); - sysctl_ip_vs_drop_packet = 2; + ipvs->drop_rate = ipvs->drop_counter + = ipvs->sysctl_amemthresh / + (ipvs->sysctl_amemthresh-availmem); + ipvs->sysctl_drop_packet = 2; } else { - ip_vs_drop_rate = 0; + ipvs->drop_rate = 0; } break; case 2: if (nomem) { - ip_vs_drop_rate = ip_vs_drop_counter - = sysctl_ip_vs_amemthresh / - (sysctl_ip_vs_amemthresh-availmem); + ipvs->drop_rate = ipvs->drop_counter + = ipvs->sysctl_amemthresh / + (ipvs->sysctl_amemthresh-availmem); } else { - ip_vs_drop_rate = 0; - sysctl_ip_vs_drop_packet = 1; + ipvs->drop_rate = 0; + ipvs->sysctl_drop_packet = 1; } break; case 3: - ip_vs_drop_rate = sysctl_ip_vs_am_droprate; + ipvs->drop_rate = ipvs->sysctl_am_droprate; break; } - spin_unlock(&__ip_vs_droppacket_lock); + spin_unlock(&ipvs->droppacket_lock); /* secure_tcp */ - spin_lock(&ip_vs_securetcp_lock); - switch (sysctl_ip_vs_secure_tcp) { + spin_lock(&ipvs->securetcp_lock); + switch (ipvs->sysctl_secure_tcp) { case 0: if (old_secure_tcp >= 2) to_change = 0; @@ -217,7 +182,7 @@ static void update_defense_level(struct netns_ipvs *ipvs) if (nomem) { if (old_secure_tcp < 2) to_change = 1; - sysctl_ip_vs_secure_tcp = 2; + ipvs->sysctl_secure_tcp = 2; } else { if (old_secure_tcp >= 2) to_change = 0; @@ -230,7 +195,7 @@ static void update_defense_level(struct netns_ipvs *ipvs) } else { if (old_secure_tcp >= 2) to_change = 0; - sysctl_ip_vs_secure_tcp = 1; + ipvs->sysctl_secure_tcp = 1; } break; case 3: @@ -238,11 +203,11 @@ static void update_defense_level(struct netns_ipvs *ipvs) to_change = 1; break; } - old_secure_tcp = sysctl_ip_vs_secure_tcp; + old_secure_tcp = ipvs->sysctl_secure_tcp; if (to_change >= 0) ip_vs_protocol_timeout_change(ipvs, - sysctl_ip_vs_secure_tcp > 1); - spin_unlock(&ip_vs_securetcp_lock); + ipvs->sysctl_secure_tcp > 1); + spin_unlock(&ipvs->securetcp_lock); local_bh_enable(); } @@ -260,7 +225,7 @@ static void defense_work_handler(struct work_struct *work) struct netns_ipvs *ipvs = net_ipvs(&init_net); update_defense_level(ipvs); - if (atomic_read(&ip_vs_dropentry)) + if (atomic_read(&ipvs->dropentry)) ip_vs_random_dropentry(); schedule_delayed_work(&defense_work, DEFENSE_TIMER_PERIOD); @@ -602,7 +567,7 @@ ip_vs_lookup_real_service(struct net *net, int af, __u16 protocol, */ hash = ip_vs_rs_hashkey(af, daddr, dport); - read_lock(&__ip_vs_rs_lock); + read_lock(&ipvs->rs_lock); list_for_each_entry(dest, &ipvs->rs_table[hash], d_list) { if ((dest->af == af) && ip_vs_addr_equal(af, &dest->addr, daddr) @@ -610,11 +575,11 @@ ip_vs_lookup_real_service(struct net *net, int af, __u16 protocol, && ((dest->protocol == protocol) || dest->vfwmark)) { /* HIT */ - read_unlock(&__ip_vs_rs_lock); + read_unlock(&ipvs->rs_lock); return dest; } } - read_unlock(&__ip_vs_rs_lock); + read_unlock(&ipvs->rs_lock); return NULL; } @@ -788,9 +753,9 @@ __ip_vs_update_dest(struct ip_vs_service *svc, struct ip_vs_dest *dest, * Put the real service in rs_table if not present. * For now only for NAT! */ - write_lock_bh(&__ip_vs_rs_lock); + write_lock_bh(&ipvs->rs_lock); ip_vs_rs_hash(ipvs, dest); - write_unlock_bh(&__ip_vs_rs_lock); + write_unlock_bh(&ipvs->rs_lock); } atomic_set(&dest->conn_flags, conn_flags); @@ -1022,14 +987,16 @@ ip_vs_edit_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) */ static void __ip_vs_del_dest(struct net *net, struct ip_vs_dest *dest) { + struct netns_ipvs *ipvs = net_ipvs(net); + ip_vs_kill_estimator(net, &dest->stats); /* * Remove it from the d-linked list with the real services. */ - write_lock_bh(&__ip_vs_rs_lock); + write_lock_bh(&ipvs->rs_lock); ip_vs_rs_unhash(dest); - write_unlock_bh(&__ip_vs_rs_lock); + write_unlock_bh(&ipvs->rs_lock); /* * Decrease the refcnt of the dest, and free the dest @@ -1092,7 +1059,6 @@ static int ip_vs_del_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) { struct ip_vs_dest *dest; - struct net *net = svc->net; __be16 dport = udest->port; EnterFunction(2); @@ -1121,7 +1087,7 @@ ip_vs_del_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest) /* * Delete the destination */ - __ip_vs_del_dest(net, dest); + __ip_vs_del_dest(svc->net, dest); LeaveFunction(2); @@ -1140,6 +1106,7 @@ ip_vs_add_service(struct net *net, struct ip_vs_service_user_kern *u, struct ip_vs_scheduler *sched = NULL; struct ip_vs_pe *pe = NULL; struct ip_vs_service *svc = NULL; + struct netns_ipvs *ipvs = net_ipvs(net); /* increase the module use count */ ip_vs_use_count_inc(); @@ -1219,7 +1186,7 @@ ip_vs_add_service(struct net *net, struct ip_vs_service_user_kern *u, /* Count only IPv4 services for old get/setsockopt interface */ if (svc->af == AF_INET) - ip_vs_num_services++; + ipvs->num_services++; /* Hash the service into the service table */ write_lock_bh(&__ip_vs_svc_lock); @@ -1359,12 +1326,13 @@ static void __ip_vs_del_service(struct ip_vs_service *svc) struct ip_vs_dest *dest, *nxt; struct ip_vs_scheduler *old_sched; struct ip_vs_pe *old_pe; + struct netns_ipvs *ipvs = net_ipvs(svc->net); pr_info("%s: enter\n", __func__); /* Count only IPv4 services for old get/setsockopt interface */ if (svc->af == AF_INET) - ip_vs_num_services--; + ipvs->num_services--; ip_vs_kill_estimator(svc->net, &svc->stats); @@ -1589,42 +1557,31 @@ proc_do_sync_mode(ctl_table *table, int write, /* * IPVS sysctl table (under the /proc/sys/net/ipv4/vs/) + * Do not change order or insert new entries without + * align with netns init in __ip_vs_control_init() */ static struct ctl_table vs_vars[] = { { .procname = "amemthresh", - .data = &sysctl_ip_vs_amemthresh, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec, - }, -#ifdef CONFIG_IP_VS_DEBUG - { - .procname = "debug_level", - .data = &sysctl_ip_vs_debug_level, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec, }, -#endif { .procname = "am_droprate", - .data = &sysctl_ip_vs_am_droprate, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec, }, { .procname = "drop_entry", - .data = &sysctl_ip_vs_drop_entry, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_do_defense_mode, }, { .procname = "drop_packet", - .data = &sysctl_ip_vs_drop_packet, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_do_defense_mode, @@ -1632,7 +1589,6 @@ static struct ctl_table vs_vars[] = { #ifdef CONFIG_IP_VS_NFCT { .procname = "conntrack", - .data = &sysctl_ip_vs_conntrack, .maxlen = sizeof(int), .mode = 0644, .proc_handler = &proc_dointvec, @@ -1640,25 +1596,62 @@ static struct ctl_table vs_vars[] = { #endif { .procname = "secure_tcp", - .data = &sysctl_ip_vs_secure_tcp, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_do_defense_mode, }, { .procname = "snat_reroute", - .data = &sysctl_ip_vs_snat_reroute, .maxlen = sizeof(int), .mode = 0644, .proc_handler = &proc_dointvec, }, { .procname = "sync_version", - .data = &sysctl_ip_vs_sync_ver, .maxlen = sizeof(int), .mode = 0644, .proc_handler = &proc_do_sync_mode, }, + { + .procname = "cache_bypass", + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .procname = "expire_nodest_conn", + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .procname = "expire_quiescent_template", + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .procname = "sync_threshold", + .maxlen = + sizeof(((struct netns_ipvs *)0)->sysctl_sync_threshold), + .mode = 0644, + .proc_handler = proc_do_sync_threshold, + }, + { + .procname = "nat_icmp_send", + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, +#ifdef CONFIG_IP_VS_DEBUG + { + .procname = "debug_level", + .data = &sysctl_ip_vs_debug_level, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, +#endif #if 0 { .procname = "timeout_established", @@ -1745,41 +1738,6 @@ static struct ctl_table vs_vars[] = { .proc_handler = proc_dointvec_jiffies, }, #endif - { - .procname = "cache_bypass", - .data = &sysctl_ip_vs_cache_bypass, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec, - }, - { - .procname = "expire_nodest_conn", - .data = &sysctl_ip_vs_expire_nodest_conn, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec, - }, - { - .procname = "expire_quiescent_template", - .data = &sysctl_ip_vs_expire_quiescent_template, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec, - }, - { - .procname = "sync_threshold", - .data = &sysctl_ip_vs_sync_threshold, - .maxlen = sizeof(sysctl_ip_vs_sync_threshold), - .mode = 0644, - .proc_handler = proc_do_sync_threshold, - }, - { - .procname = "nat_icmp_send", - .data = &sysctl_ip_vs_nat_icmp_send, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec, - }, { } }; @@ -1791,8 +1749,6 @@ const struct ctl_path net_vs_ctl_path[] = { }; EXPORT_SYMBOL_GPL(net_vs_ctl_path); -static struct ctl_table_header * sysctl_header; - #ifdef CONFIG_PROC_FS struct ip_vs_iter { @@ -2543,7 +2499,7 @@ do_ip_vs_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) struct ip_vs_getinfo info; info.version = IP_VS_VERSION_CODE; info.size = ip_vs_conn_tab_size; - info.num_services = ip_vs_num_services; + info.num_services = ipvs->num_services; if (copy_to_user(user, &info, sizeof(info)) != 0) ret = -EFAULT; } @@ -3014,7 +2970,7 @@ static int ip_vs_genl_dump_dests(struct sk_buff *skb, struct ip_vs_service *svc; struct ip_vs_dest *dest; struct nlattr *attrs[IPVS_CMD_ATTR_MAX + 1]; - struct net *net; + struct net *net = skb_sknet(skb); mutex_lock(&__ip_vs_mutex); @@ -3023,7 +2979,7 @@ static int ip_vs_genl_dump_dests(struct sk_buff *skb, IPVS_CMD_ATTR_MAX, ip_vs_cmd_policy)) goto out_err; - net = skb_sknet(skb); + svc = ip_vs_genl_find_service(net, attrs[IPVS_CMD_ATTR_SERVICE]); if (IS_ERR(svc) || svc == NULL) goto out_err; @@ -3215,8 +3171,10 @@ static int ip_vs_genl_set_cmd(struct sk_buff *skb, struct genl_info *info) int ret = 0, cmd; int need_full_svc = 0, need_full_dest = 0; struct net *net; + struct netns_ipvs *ipvs; net = skb_sknet(skb); + ipvs = net_ipvs(net); cmd = info->genlhdr->cmd; mutex_lock(&__ip_vs_mutex); @@ -3326,8 +3284,10 @@ static int ip_vs_genl_get_cmd(struct sk_buff *skb, struct genl_info *info) void *reply; int ret, cmd, reply_cmd; struct net *net; + struct netns_ipvs *ipvs; net = skb_sknet(skb); + ipvs = net_ipvs(net); cmd = info->genlhdr->cmd; if (cmd == IPVS_CMD_GET_SERVICE) @@ -3530,9 +3490,21 @@ int __net_init __ip_vs_control_init(struct net *net) { int idx; struct netns_ipvs *ipvs = net_ipvs(net); + struct ctl_table *tbl; if (!net_eq(net, &init_net)) /* netns not enabled yet */ return -EPERM; + + atomic_set(&ipvs->dropentry, 0); + spin_lock_init(&ipvs->dropentry_lock); + spin_lock_init(&ipvs->droppacket_lock); + spin_lock_init(&ipvs->securetcp_lock); + ipvs->rs_lock = __RW_LOCK_UNLOCKED(ipvs->rs_lock); + + /* Initialize rs_table */ + for (idx = 0; idx < IP_VS_RTAB_SIZE; idx++) + INIT_LIST_HEAD(&ipvs->rs_table[idx]); + /* procfs stats */ ipvs->tot_stats = kzalloc(sizeof(struct ip_vs_stats), GFP_KERNEL); if (ipvs->tot_stats == NULL) { @@ -3553,14 +3525,51 @@ int __net_init __ip_vs_control_init(struct net *net) proc_net_fops_create(net, "ip_vs_stats", 0, &ip_vs_stats_fops); proc_net_fops_create(net, "ip_vs_stats_percpu", 0, &ip_vs_stats_percpu_fops); - sysctl_header = register_net_sysctl_table(net, net_vs_ctl_path, + + if (!net_eq(net, &init_net)) { + tbl = kmemdup(vs_vars, sizeof(vs_vars), GFP_KERNEL); + if (tbl == NULL) + goto err_dup; + } else + tbl = vs_vars; + /* Initialize sysctl defaults */ + idx = 0; + ipvs->sysctl_amemthresh = 1024; + tbl[idx++].data = &ipvs->sysctl_amemthresh; + ipvs->sysctl_am_droprate = 10; + tbl[idx++].data = &ipvs->sysctl_am_droprate; + tbl[idx++].data = &ipvs->sysctl_drop_entry; + tbl[idx++].data = &ipvs->sysctl_drop_packet; +#ifdef CONFIG_IP_VS_NFCT + tbl[idx++].data = &ipvs->sysctl_conntrack; +#endif + tbl[idx++].data = &ipvs->sysctl_secure_tcp; + ipvs->sysctl_snat_reroute = 1; + tbl[idx++].data = &ipvs->sysctl_snat_reroute; + ipvs->sysctl_sync_ver = 1; + tbl[idx++].data = &ipvs->sysctl_sync_ver; + tbl[idx++].data = &ipvs->sysctl_cache_bypass; + tbl[idx++].data = &ipvs->sysctl_expire_nodest_conn; + tbl[idx++].data = &ipvs->sysctl_expire_quiescent_template; + ipvs->sysctl_sync_threshold[0] = 3; + ipvs->sysctl_sync_threshold[1] = 50; + tbl[idx].data = &ipvs->sysctl_sync_threshold; + tbl[idx++].maxlen = sizeof(ipvs->sysctl_sync_threshold); + tbl[idx++].data = &ipvs->sysctl_nat_icmp_send; + + + ipvs->sysctl_hdr = register_net_sysctl_table(net, net_vs_ctl_path, vs_vars); - if (sysctl_header == NULL) + if (ipvs->sysctl_hdr == NULL) goto err_reg; ip_vs_new_estimator(net, ipvs->tot_stats); + ipvs->sysctl_tbl = tbl; return 0; err_reg: + if (!net_eq(net, &init_net)) + kfree(tbl); +err_dup: free_percpu(ipvs->cpustats); err_alloc: kfree(ipvs->tot_stats); @@ -3575,7 +3584,7 @@ static void __net_exit __ip_vs_control_cleanup(struct net *net) return; ip_vs_kill_estimator(net, ipvs->tot_stats); - unregister_net_sysctl_table(sysctl_header); + unregister_net_sysctl_table(ipvs->sysctl_hdr); proc_net_remove(net, "ip_vs_stats_percpu"); proc_net_remove(net, "ip_vs_stats"); proc_net_remove(net, "ip_vs"); diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c index 550365a690c7..fb2d04ac5d4e 100644 --- a/net/netfilter/ipvs/ip_vs_proto_sctp.c +++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c @@ -34,7 +34,7 @@ sctp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, &iph.daddr, sh->dest))) { int ignored; - if (ip_vs_todrop()) { + if (ip_vs_todrop(net_ipvs(net))) { /* * It seems that we are very loaded. * We have to drop this packet :( diff --git a/net/netfilter/ipvs/ip_vs_proto_tcp.c b/net/netfilter/ipvs/ip_vs_proto_tcp.c index d8b3f9f15826..c0cc341b840d 100644 --- a/net/netfilter/ipvs/ip_vs_proto_tcp.c +++ b/net/netfilter/ipvs/ip_vs_proto_tcp.c @@ -54,7 +54,7 @@ tcp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, &iph.daddr, th->dest))) { int ignored; - if (ip_vs_todrop()) { + if (ip_vs_todrop(net_ipvs(net))) { /* * It seems that we are very loaded. * We have to drop this packet :( diff --git a/net/netfilter/ipvs/ip_vs_proto_udp.c b/net/netfilter/ipvs/ip_vs_proto_udp.c index 581157bbded5..f1282cbe6fe3 100644 --- a/net/netfilter/ipvs/ip_vs_proto_udp.c +++ b/net/netfilter/ipvs/ip_vs_proto_udp.c @@ -50,7 +50,7 @@ udp_conn_schedule(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, if (svc) { int ignored; - if (ip_vs_todrop()) { + if (ip_vs_todrop(net_ipvs(net))) { /* * It seems that we are very loaded. * We have to drop this packet :( diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index f85e47daecc3..b1780562c42b 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -394,7 +394,7 @@ void ip_vs_sync_switch_mode(struct net *net, int mode) if (!ipvs->sync_state & IP_VS_STATE_MASTER) return; - if (mode == sysctl_ip_vs_sync_ver || !ipvs->sync_buff) + if (mode == ipvs->sysctl_sync_ver || !ipvs->sync_buff) return; spin_lock_bh(&ipvs->sync_buff_lock); @@ -521,7 +521,7 @@ void ip_vs_sync_conn(struct net *net, struct ip_vs_conn *cp) unsigned int len, pe_name_len, pad; /* Handle old version of the protocol */ - if (sysctl_ip_vs_sync_ver == 0) { + if (ipvs->sysctl_sync_ver == 0) { ip_vs_sync_conn_v0(net, cp); return; } @@ -650,7 +650,7 @@ control: if (cp->flags & IP_VS_CONN_F_TEMPLATE) { int pkts = atomic_add_return(1, &cp->in_pkts); - if (pkts % sysctl_ip_vs_sync_threshold[1] != 1) + if (pkts % ipvs->sysctl_sync_threshold[1] != 1) return; } goto sloop; @@ -724,6 +724,7 @@ static void ip_vs_proc_conn(struct net *net, struct ip_vs_conn_param *param, { struct ip_vs_dest *dest; struct ip_vs_conn *cp; + struct netns_ipvs *ipvs = net_ipvs(net); if (!(flags & IP_VS_CONN_F_TEMPLATE)) cp = ip_vs_conn_in_get(param); @@ -794,7 +795,7 @@ static void ip_vs_proc_conn(struct net *net, struct ip_vs_conn_param *param, if (opt) memcpy(&cp->in_seq, opt, sizeof(*opt)); - atomic_set(&cp->in_pkts, sysctl_ip_vs_sync_threshold[0]); + atomic_set(&cp->in_pkts, ipvs->sysctl_sync_threshold[0]); cp->state = state; cp->old_state = cp->state; /* -- cgit v1.2.3 From f6340ee0c6b9498ec918a7bb2f44e20abb8b2833 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:59 +0100 Subject: IPVS: netns, defense work timer. This patch makes defense work timer per name-space, A net ptr had to be added to the ipvs struct, since it's needed by defense_work_handler. [ horms@verge.net.au: Use cancel_delayed_work_sync() instead of cancel_rearming_delayed_work(). Found during merge conflict resoliution ] Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_conn.c | 5 +++-- net/netfilter/ipvs/ip_vs_core.c | 1 + net/netfilter/ipvs/ip_vs_ctl.c | 20 +++++++++----------- 3 files changed, 13 insertions(+), 13 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 5ba205a4d79c..28bdaf7c02f4 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -1138,7 +1138,7 @@ static inline int todrop_entry(struct ip_vs_conn *cp) } /* Called from keventd and must protect itself from softirqs */ -void ip_vs_random_dropentry(void) +void ip_vs_random_dropentry(struct net *net) { int idx; struct ip_vs_conn *cp; @@ -1158,7 +1158,8 @@ void ip_vs_random_dropentry(void) if (cp->flags & IP_VS_CONN_F_TEMPLATE) /* connection template */ continue; - + if (!ip_vs_conn_net_eq(cp, net)) + continue; if (cp->protocol == IPPROTO_TCP) { switch(cp->state) { case IP_VS_TCP_S_SYN_RECV: diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index a7c59a722af3..bdda346a4f30 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -1884,6 +1884,7 @@ static int __net_init __ip_vs_init(struct net *net) pr_err("%s(): no memory.\n", __func__); return -ENOMEM; } + ipvs->net = net; /* Counters used for creating unique names */ ipvs->gen = atomic_read(&ipvs_netns_cnt); atomic_inc(&ipvs_netns_cnt); diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 183ac18bded5..6a963d44df48 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -217,18 +217,16 @@ static void update_defense_level(struct netns_ipvs *ipvs) * Timer for checking the defense */ #define DEFENSE_TIMER_PERIOD 1*HZ -static void defense_work_handler(struct work_struct *work); -static DECLARE_DELAYED_WORK(defense_work, defense_work_handler); static void defense_work_handler(struct work_struct *work) { - struct netns_ipvs *ipvs = net_ipvs(&init_net); + struct netns_ipvs *ipvs = + container_of(work, struct netns_ipvs, defense_work.work); update_defense_level(ipvs); if (atomic_read(&ipvs->dropentry)) - ip_vs_random_dropentry(); - - schedule_delayed_work(&defense_work, DEFENSE_TIMER_PERIOD); + ip_vs_random_dropentry(ipvs->net); + schedule_delayed_work(&ipvs->defense_work, DEFENSE_TIMER_PERIOD); } int @@ -3564,6 +3562,9 @@ int __net_init __ip_vs_control_init(struct net *net) goto err_reg; ip_vs_new_estimator(net, ipvs->tot_stats); ipvs->sysctl_tbl = tbl; + /* Schedule defense work */ + INIT_DELAYED_WORK(&ipvs->defense_work, defense_work_handler); + schedule_delayed_work(&ipvs->defense_work, DEFENSE_TIMER_PERIOD); return 0; err_reg: @@ -3588,6 +3589,8 @@ static void __net_exit __ip_vs_control_cleanup(struct net *net) proc_net_remove(net, "ip_vs_stats_percpu"); proc_net_remove(net, "ip_vs_stats"); proc_net_remove(net, "ip_vs"); + cancel_delayed_work_sync(&ipvs->defense_work); + cancel_work_sync(&ipvs->defense_work.work); free_percpu(ipvs->cpustats); kfree(ipvs->tot_stats); } @@ -3631,9 +3634,6 @@ int __init ip_vs_control_init(void) goto err_net; } - /* Hook the defense timer */ - schedule_delayed_work(&defense_work, DEFENSE_TIMER_PERIOD); - LeaveFunction(2); return 0; @@ -3648,8 +3648,6 @@ void ip_vs_control_cleanup(void) { EnterFunction(2); ip_vs_trash_cleanup(); - cancel_delayed_work_sync(&defense_work); - cancel_work_sync(&defense_work.work); unregister_pernet_subsys(&ipvs_control_ops); ip_vs_genl_unregister(); nf_unregister_sockopt(&ip_vs_sockopts); -- cgit v1.2.3 From f2431e6e9255461eb1476340a89ad32ad4b38b03 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:45:00 +0100 Subject: IPVS: netns, trash handling trash list per namspace, and reordering of some params in dst struct. [ horms@verge.net.au: Use cancel_delayed_work_sync() instead of cancel_rearming_delayed_work(). Found during merge conflict resoliution ] Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_ctl.c | 23 +++++++++++------------ 1 file changed, 11 insertions(+), 12 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 6a963d44df48..442edf4be644 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -254,11 +254,6 @@ static struct list_head ip_vs_svc_table[IP_VS_SVC_TAB_SIZE]; /* the service table hashed by fwmark */ static struct list_head ip_vs_svc_fwm_table[IP_VS_SVC_TAB_SIZE]; -/* - * Trash for destinations - */ -static LIST_HEAD(ip_vs_dest_trash); - /* * FTP & NULL virtual service counters */ @@ -650,11 +645,12 @@ ip_vs_trash_get_dest(struct ip_vs_service *svc, const union nf_inet_addr *daddr, __be16 dport) { struct ip_vs_dest *dest, *nxt; + struct netns_ipvs *ipvs = net_ipvs(svc->net); /* * Find the destination in trash */ - list_for_each_entry_safe(dest, nxt, &ip_vs_dest_trash, n_list) { + list_for_each_entry_safe(dest, nxt, &ipvs->dest_trash, n_list) { IP_VS_DBG_BUF(3, "Destination %u/%s:%u still in trash, " "dest->refcnt=%d\n", dest->vfwmark, @@ -703,11 +699,12 @@ ip_vs_trash_get_dest(struct ip_vs_service *svc, const union nf_inet_addr *daddr, * are expired, and the refcnt of each destination in the trash must * be 1, so we simply release them here. */ -static void ip_vs_trash_cleanup(void) +static void ip_vs_trash_cleanup(struct net *net) { struct ip_vs_dest *dest, *nxt; + struct netns_ipvs *ipvs = net_ipvs(net); - list_for_each_entry_safe(dest, nxt, &ip_vs_dest_trash, n_list) { + list_for_each_entry_safe(dest, nxt, &ipvs->dest_trash, n_list) { list_del(&dest->n_list); ip_vs_dst_reset(dest); __ip_vs_unbind_svc(dest); @@ -1021,7 +1018,7 @@ static void __ip_vs_del_dest(struct net *net, struct ip_vs_dest *dest) IP_VS_DBG_ADDR(dest->af, &dest->addr), ntohs(dest->port), atomic_read(&dest->refcnt)); - list_add(&dest->n_list, &ip_vs_dest_trash); + list_add(&dest->n_list, &ipvs->dest_trash); atomic_inc(&dest->refcnt); } } @@ -3503,6 +3500,8 @@ int __net_init __ip_vs_control_init(struct net *net) for (idx = 0; idx < IP_VS_RTAB_SIZE; idx++) INIT_LIST_HEAD(&ipvs->rs_table[idx]); + INIT_LIST_HEAD(&ipvs->dest_trash); + /* procfs stats */ ipvs->tot_stats = kzalloc(sizeof(struct ip_vs_stats), GFP_KERNEL); if (ipvs->tot_stats == NULL) { @@ -3584,13 +3583,14 @@ static void __net_exit __ip_vs_control_cleanup(struct net *net) if (!net_eq(net, &init_net)) /* netns not enabled yet */ return; + ip_vs_trash_cleanup(net); ip_vs_kill_estimator(net, ipvs->tot_stats); + cancel_delayed_work_sync(&ipvs->defense_work); + cancel_work_sync(&ipvs->defense_work.work); unregister_net_sysctl_table(ipvs->sysctl_hdr); proc_net_remove(net, "ip_vs_stats_percpu"); proc_net_remove(net, "ip_vs_stats"); proc_net_remove(net, "ip_vs"); - cancel_delayed_work_sync(&ipvs->defense_work); - cancel_work_sync(&ipvs->defense_work.work); free_percpu(ipvs->cpustats); kfree(ipvs->tot_stats); } @@ -3647,7 +3647,6 @@ err: void ip_vs_control_cleanup(void) { EnterFunction(2); - ip_vs_trash_cleanup(); unregister_pernet_subsys(&ipvs_control_ops); ip_vs_genl_unregister(); nf_unregister_sockopt(&ip_vs_sockopts); -- cgit v1.2.3 From 763f8d0ed4f1ce38b35cc0e05482b7799b82789b Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:45:01 +0100 Subject: IPVS: netns, svc counters moved in ip_vs_ctl,c Last two global vars to be moved, ip_vs_ftpsvc_counter and ip_vs_nullsvc_counter. [horms@verge.net.au: removed whitespace-change-only hunk] Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_ctl.c | 21 +++++++++------------ 1 file changed, 9 insertions(+), 12 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 442edf4be644..65f5de405ad2 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -254,12 +254,6 @@ static struct list_head ip_vs_svc_table[IP_VS_SVC_TAB_SIZE]; /* the service table hashed by fwmark */ static struct list_head ip_vs_svc_fwm_table[IP_VS_SVC_TAB_SIZE]; -/* - * FTP & NULL virtual service counters - */ -static atomic_t ip_vs_ftpsvc_counter = ATOMIC_INIT(0); -static atomic_t ip_vs_nullsvc_counter = ATOMIC_INIT(0); - /* * Returns hash value for virtual service @@ -409,6 +403,7 @@ ip_vs_service_get(struct net *net, int af, __u32 fwmark, __u16 protocol, const union nf_inet_addr *vaddr, __be16 vport) { struct ip_vs_service *svc; + struct netns_ipvs *ipvs = net_ipvs(net); read_lock(&__ip_vs_svc_lock); @@ -427,7 +422,7 @@ ip_vs_service_get(struct net *net, int af, __u32 fwmark, __u16 protocol, if (svc == NULL && protocol == IPPROTO_TCP - && atomic_read(&ip_vs_ftpsvc_counter) + && atomic_read(&ipvs->ftpsvc_counter) && (vport == FTPDATA || ntohs(vport) >= PROT_SOCK)) { /* * Check if ftp service entry exists, the packet @@ -437,7 +432,7 @@ ip_vs_service_get(struct net *net, int af, __u32 fwmark, __u16 protocol, } if (svc == NULL - && atomic_read(&ip_vs_nullsvc_counter)) { + && atomic_read(&ipvs->nullsvc_counter)) { /* * Check if the catch-all port (port zero) exists */ @@ -1173,9 +1168,9 @@ ip_vs_add_service(struct net *net, struct ip_vs_service_user_kern *u, /* Update the virtual service counters */ if (svc->port == FTPPORT) - atomic_inc(&ip_vs_ftpsvc_counter); + atomic_inc(&ipvs->ftpsvc_counter); else if (svc->port == 0) - atomic_inc(&ip_vs_nullsvc_counter); + atomic_inc(&ipvs->nullsvc_counter); ip_vs_new_estimator(net, &svc->stats); @@ -1359,9 +1354,9 @@ static void __ip_vs_del_service(struct ip_vs_service *svc) * Update the virtual service counters */ if (svc->port == FTPPORT) - atomic_dec(&ip_vs_ftpsvc_counter); + atomic_dec(&ipvs->ftpsvc_counter); else if (svc->port == 0) - atomic_dec(&ip_vs_nullsvc_counter); + atomic_dec(&ipvs->nullsvc_counter); /* * Free the service if nobody refers to it @@ -3501,6 +3496,8 @@ int __net_init __ip_vs_control_init(struct net *net) INIT_LIST_HEAD(&ipvs->rs_table[idx]); INIT_LIST_HEAD(&ipvs->dest_trash); + atomic_set(&ipvs->ftpsvc_counter, 0); + atomic_set(&ipvs->nullsvc_counter, 0); /* procfs stats */ ipvs->tot_stats = kzalloc(sizeof(struct ip_vs_stats), GFP_KERNEL); -- cgit v1.2.3 From 4a98480bccc2f5998c5564d254392635b9aa04c2 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:45:02 +0100 Subject: IPVS: netns, misc init_net removal in core. init_net removed in __ip_vs_addr_is_local_v6, and got net as param. Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 6 ++++-- net/netfilter/ipvs/ip_vs_ctl.c | 9 +++++---- 2 files changed, 9 insertions(+), 6 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index bdda346a4f30..9e10c7a0720a 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -499,6 +499,7 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, struct ip_vs_proto_data *pd) { + struct net *net; struct netns_ipvs *ipvs; __be16 _ports[2], *pptr; struct ip_vs_iphdr iph; @@ -511,18 +512,19 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, ip_vs_service_put(svc); return NF_DROP; } + net = skb_net(skb); #ifdef CONFIG_IP_VS_IPV6 if (svc->af == AF_INET6) unicast = ipv6_addr_type(&iph.daddr.in6) & IPV6_ADDR_UNICAST; else #endif - unicast = (inet_addr_type(&init_net, iph.daddr.ip) == RTN_UNICAST); + unicast = (inet_addr_type(net, iph.daddr.ip) == RTN_UNICAST); /* if it is fwmark-based service, the cache_bypass sysctl is up and the destination is a non-local unicast, then create a cache_bypass connection entry */ - ipvs = net_ipvs(skb_net(skb)); + ipvs = net_ipvs(net); if (ipvs->sysctl_cache_bypass && svc->fwmark && unicast) { int ret, cs; struct ip_vs_conn *cp; diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 65f5de405ad2..edf2b6dee972 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -71,7 +71,8 @@ int ip_vs_get_debug_level(void) #ifdef CONFIG_IP_VS_IPV6 /* Taken from rt6_fill_node() in net/ipv6/route.c, is there a better way? */ -static int __ip_vs_addr_is_local_v6(const struct in6_addr *addr) +static int __ip_vs_addr_is_local_v6(struct net *net, + const struct in6_addr *addr) { struct rt6_info *rt; struct flowi fl = { @@ -80,7 +81,7 @@ static int __ip_vs_addr_is_local_v6(const struct in6_addr *addr) .fl6_src = { .s6_addr32 = {0, 0, 0, 0} }, }; - rt = (struct rt6_info *)ip6_route_output(&init_net, NULL, &fl); + rt = (struct rt6_info *)ip6_route_output(net, NULL, &fl); if (rt && rt->rt6i_dev && (rt->rt6i_dev->flags & IFF_LOOPBACK)) return 1; @@ -810,12 +811,12 @@ ip_vs_new_dest(struct ip_vs_service *svc, struct ip_vs_dest_user_kern *udest, atype = ipv6_addr_type(&udest->addr.in6); if ((!(atype & IPV6_ADDR_UNICAST) || atype & IPV6_ADDR_LINKLOCAL) && - !__ip_vs_addr_is_local_v6(&udest->addr.in6)) + !__ip_vs_addr_is_local_v6(svc->net, &udest->addr.in6)) return -EINVAL; } else #endif { - atype = inet_addr_type(&init_net, udest->addr.ip); + atype = inet_addr_type(svc->net, udest->addr.ip); if (atype != RTN_LOCAL && atype != RTN_UNICAST) return -EINVAL; } -- cgit v1.2.3 From c6d2d445d8dee04cde47eb4021636399a4239e9f Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:45:03 +0100 Subject: IPVS: netns, final patch enabling network name space. all init_net removed, (except for some alloc related that needs to be there) Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_app.c | 3 --- net/netfilter/ipvs/ip_vs_conn.c | 5 ----- net/netfilter/ipvs/ip_vs_core.c | 4 ---- net/netfilter/ipvs/ip_vs_ctl.c | 7 +------ net/netfilter/ipvs/ip_vs_est.c | 3 --- net/netfilter/ipvs/ip_vs_ftp.c | 6 ------ net/netfilter/ipvs/ip_vs_sync.c | 5 ----- 7 files changed, 1 insertion(+), 32 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_app.c b/net/netfilter/ipvs/ip_vs_app.c index 286f46594e0e..5c48ffb60c28 100644 --- a/net/netfilter/ipvs/ip_vs_app.c +++ b/net/netfilter/ipvs/ip_vs_app.c @@ -582,9 +582,6 @@ static int __net_init __ip_vs_app_init(struct net *net) { struct netns_ipvs *ipvs = net_ipvs(net); - if (!net_eq(net, &init_net)) /* netns not enabled yet */ - return -EPERM; - INIT_LIST_HEAD(&ipvs->app_list); __mutex_init(&ipvs->app_mutex, "ipvs->app_mutex", &ipvs->app_key); proc_net_fops_create(net, "ip_vs_app", 0, &ip_vs_app_fops); diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 28bdaf7c02f4..83233fe24a08 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -1234,8 +1234,6 @@ int __net_init __ip_vs_conn_init(struct net *net) { struct netns_ipvs *ipvs = net_ipvs(net); - if (!net_eq(net, &init_net)) /* netns not enabled yet */ - return -EPERM; atomic_set(&ipvs->conn_count, 0); proc_net_fops_create(net, "ip_vs_conn", 0, &ip_vs_conn_fops); @@ -1245,9 +1243,6 @@ int __net_init __ip_vs_conn_init(struct net *net) static void __net_exit __ip_vs_conn_cleanup(struct net *net) { - if (!net_eq(net, &init_net)) /* netns not enabled yet */ - return; - /* flush all the connection entries first */ ip_vs_conn_flush(net); proc_net_remove(net, "ip_vs_conn"); diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 9e10c7a0720a..f36a84f33efb 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -1877,10 +1877,6 @@ static int __net_init __ip_vs_init(struct net *net) { struct netns_ipvs *ipvs; - if (!net_eq(net, &init_net)) { - pr_err("The final patch for enabling netns is missing\n"); - return -EPERM; - } ipvs = net_generic(net, ip_vs_net_id); if (ipvs == NULL) { pr_err("%s(): no memory.\n", __func__); diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index edf2b6dee972..09ca2ce2f2b7 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -2617,6 +2617,7 @@ static struct genl_family ip_vs_genl_family = { .name = IPVS_GENL_NAME, .version = IPVS_GENL_VERSION, .maxattr = IPVS_CMD_MAX, + .netnsok = true, /* Make ipvsadm to work on netns */ }; /* Policy used for first-level command attributes */ @@ -3483,9 +3484,6 @@ int __net_init __ip_vs_control_init(struct net *net) struct netns_ipvs *ipvs = net_ipvs(net); struct ctl_table *tbl; - if (!net_eq(net, &init_net)) /* netns not enabled yet */ - return -EPERM; - atomic_set(&ipvs->dropentry, 0); spin_lock_init(&ipvs->dropentry_lock); spin_lock_init(&ipvs->droppacket_lock); @@ -3578,9 +3576,6 @@ static void __net_exit __ip_vs_control_cleanup(struct net *net) { struct netns_ipvs *ipvs = net_ipvs(net); - if (!net_eq(net, &init_net)) /* netns not enabled yet */ - return; - ip_vs_trash_cleanup(net); ip_vs_kill_estimator(net, ipvs->tot_stats); cancel_delayed_work_sync(&ipvs->defense_work); diff --git a/net/netfilter/ipvs/ip_vs_est.c b/net/netfilter/ipvs/ip_vs_est.c index d13616b138cd..f560a05c965a 100644 --- a/net/netfilter/ipvs/ip_vs_est.c +++ b/net/netfilter/ipvs/ip_vs_est.c @@ -203,9 +203,6 @@ static int __net_init __ip_vs_estimator_init(struct net *net) { struct netns_ipvs *ipvs = net_ipvs(net); - if (!net_eq(net, &init_net)) /* netns not enabled yet */ - return -EPERM; - INIT_LIST_HEAD(&ipvs->est_list); spin_lock_init(&ipvs->est_lock); setup_timer(&ipvs->est_timer, estimation_timer, (unsigned long)net); diff --git a/net/netfilter/ipvs/ip_vs_ftp.c b/net/netfilter/ipvs/ip_vs_ftp.c index 6a04f9ab9d0d..6b5dd6ddaae9 100644 --- a/net/netfilter/ipvs/ip_vs_ftp.c +++ b/net/netfilter/ipvs/ip_vs_ftp.c @@ -413,9 +413,6 @@ static int __net_init __ip_vs_ftp_init(struct net *net) int i, ret; struct ip_vs_app *app = &ip_vs_ftp; - if (!net_eq(net, &init_net)) /* netns not enabled yet */ - return -EPERM; - ret = register_ip_vs_app(net, app); if (ret) return ret; @@ -442,9 +439,6 @@ static void __ip_vs_ftp_exit(struct net *net) { struct ip_vs_app *app = &ip_vs_ftp; - if (!net_eq(net, &init_net)) /* netns not enabled yet */ - return; - unregister_ip_vs_app(net, app); } diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index b1780562c42b..d1adf988eb08 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -1659,9 +1659,6 @@ static int __net_init __ip_vs_sync_init(struct net *net) { struct netns_ipvs *ipvs = net_ipvs(net); - if (!net_eq(net, &init_net)) /* netns not enabled yet */ - return -EPERM; - INIT_LIST_HEAD(&ipvs->sync_queue); spin_lock_init(&ipvs->sync_lock); spin_lock_init(&ipvs->sync_buff_lock); @@ -1674,8 +1671,6 @@ static int __net_init __ip_vs_sync_init(struct net *net) static void __ip_vs_sync_cleanup(struct net *net) { - if (!net_eq(net, &init_net)) /* netns not enabled yet */ - return; stop_sync_thread(net, IP_VS_STATE_MASTER); stop_sync_thread(net, IP_VS_STATE_BACKUP); } -- cgit v1.2.3 From b017900aac4a158b9bf7ffdcb8a369a91115b3e4 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 15 Dec 2010 09:46:26 +0100 Subject: netfilter: xt_conntrack: support matching on port ranges Add a new revision 3 that contains port ranges for all of origsrc, origdst, replsrc and repldst. The high ports are appended to the original v2 data structure to allow sharing most of the code with v1 and v2. Use of the revision specific port matching function is made dependant on par->match->revision. Signed-off-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- net/netfilter/xt_conntrack.c | 75 ++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 73 insertions(+), 2 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_conntrack.c b/net/netfilter/xt_conntrack.c index e536710ad916..4ef1b63ad73f 100644 --- a/net/netfilter/xt_conntrack.c +++ b/net/netfilter/xt_conntrack.c @@ -112,6 +112,54 @@ ct_proto_port_check(const struct xt_conntrack_mtinfo2 *info, return true; } +static inline bool +port_match(u16 min, u16 max, u16 port, bool invert) +{ + return (port >= min && port <= max) ^ invert; +} + +static inline bool +ct_proto_port_check_v3(const struct xt_conntrack_mtinfo3 *info, + const struct nf_conn *ct) +{ + const struct nf_conntrack_tuple *tuple; + + tuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; + if ((info->match_flags & XT_CONNTRACK_PROTO) && + (nf_ct_protonum(ct) == info->l4proto) ^ + !(info->invert_flags & XT_CONNTRACK_PROTO)) + return false; + + /* Shortcut to match all recognized protocols by using ->src.all. */ + if ((info->match_flags & XT_CONNTRACK_ORIGSRC_PORT) && + !port_match(info->origsrc_port, info->origsrc_port_high, + ntohs(tuple->src.u.all), + info->invert_flags & XT_CONNTRACK_ORIGSRC_PORT)) + return false; + + if ((info->match_flags & XT_CONNTRACK_ORIGDST_PORT) && + !port_match(info->origdst_port, info->origdst_port_high, + ntohs(tuple->dst.u.all), + info->invert_flags & XT_CONNTRACK_ORIGDST_PORT)) + return false; + + tuple = &ct->tuplehash[IP_CT_DIR_REPLY].tuple; + + if ((info->match_flags & XT_CONNTRACK_REPLSRC_PORT) && + !port_match(info->replsrc_port, info->replsrc_port_high, + ntohs(tuple->src.u.all), + info->invert_flags & XT_CONNTRACK_REPLSRC_PORT)) + return false; + + if ((info->match_flags & XT_CONNTRACK_REPLDST_PORT) && + !port_match(info->repldst_port, info->repldst_port_high, + ntohs(tuple->dst.u.all), + info->invert_flags & XT_CONNTRACK_REPLDST_PORT)) + return false; + + return true; +} + static bool conntrack_mt(const struct sk_buff *skb, struct xt_action_param *par, u16 state_mask, u16 status_mask) @@ -170,8 +218,13 @@ conntrack_mt(const struct sk_buff *skb, struct xt_action_param *par, !(info->invert_flags & XT_CONNTRACK_REPLDST)) return false; - if (!ct_proto_port_check(info, ct)) - return false; + if (par->match->revision != 3) { + if (!ct_proto_port_check(info, ct)) + return false; + } else { + if (!ct_proto_port_check_v3(par->matchinfo, ct)) + return false; + } if ((info->match_flags & XT_CONNTRACK_STATUS) && (!!(status_mask & ct->status) ^ @@ -207,6 +260,14 @@ conntrack_mt_v2(const struct sk_buff *skb, struct xt_action_param *par) return conntrack_mt(skb, par, info->state_mask, info->status_mask); } +static bool +conntrack_mt_v3(const struct sk_buff *skb, struct xt_action_param *par) +{ + const struct xt_conntrack_mtinfo3 *info = par->matchinfo; + + return conntrack_mt(skb, par, info->state_mask, info->status_mask); +} + static int conntrack_mt_check(const struct xt_mtchk_param *par) { int ret; @@ -244,6 +305,16 @@ static struct xt_match conntrack_mt_reg[] __read_mostly = { .destroy = conntrack_mt_destroy, .me = THIS_MODULE, }, + { + .name = "conntrack", + .revision = 3, + .family = NFPROTO_UNSPEC, + .matchsize = sizeof(struct xt_conntrack_mtinfo3), + .match = conntrack_mt_v3, + .checkentry = conntrack_mt_check, + .destroy = conntrack_mt_destroy, + .me = THIS_MODULE, + }, }; static int __init conntrack_mt_init(void) -- cgit v1.2.3 From 255d0dc34068a976550ce555e153c0bfcfec7cc6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 18 Dec 2010 18:35:15 +0100 Subject: netfilter: x_table: speedup compat operations One iptables invocation with 135000 rules takes 35 seconds of cpu time on a recent server, using a 32bit distro and a 64bit kernel. We eventually trigger NMI/RCU watchdog. INFO: rcu_sched_state detected stall on CPU 3 (t=6000 jiffies) COMPAT mode has quadratic behavior and consume 16 bytes of memory per rule. Switch the xt_compat algos to use an array instead of list, and use a binary search to locate an offset in the sorted array. This halves memory need (8 bytes per rule), and removes quadratic behavior [ O(N*N) -> O(N*log2(N)) ] Time of iptables goes from 35 s to 150 ms. Signed-off-by: Eric Dumazet Signed-off-by: Pablo Neira Ayuso --- net/netfilter/x_tables.c | 82 ++++++++++++++++++++++++++++-------------------- 1 file changed, 48 insertions(+), 34 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 80463507420e..ee5de3af510a 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -38,9 +38,8 @@ MODULE_DESCRIPTION("{ip,ip6,arp,eb}_tables backend module"); #define SMP_ALIGN(x) (((x) + SMP_CACHE_BYTES-1) & ~(SMP_CACHE_BYTES-1)) struct compat_delta { - struct compat_delta *next; - unsigned int offset; - int delta; + unsigned int offset; /* offset in kernel */ + int delta; /* delta in 32bit user land */ }; struct xt_af { @@ -49,7 +48,9 @@ struct xt_af { struct list_head target; #ifdef CONFIG_COMPAT struct mutex compat_mutex; - struct compat_delta *compat_offsets; + struct compat_delta *compat_tab; + unsigned int number; /* number of slots in compat_tab[] */ + unsigned int cur; /* number of used slots in compat_tab[] */ #endif }; @@ -414,54 +415,67 @@ int xt_check_match(struct xt_mtchk_param *par, EXPORT_SYMBOL_GPL(xt_check_match); #ifdef CONFIG_COMPAT -int xt_compat_add_offset(u_int8_t af, unsigned int offset, short delta) +int xt_compat_add_offset(u_int8_t af, unsigned int offset, int delta) { - struct compat_delta *tmp; + struct xt_af *xp = &xt[af]; - tmp = kmalloc(sizeof(struct compat_delta), GFP_KERNEL); - if (!tmp) - return -ENOMEM; + if (!xp->compat_tab) { + if (!xp->number) + return -EINVAL; + xp->compat_tab = vmalloc(sizeof(struct compat_delta) * xp->number); + if (!xp->compat_tab) + return -ENOMEM; + xp->cur = 0; + } - tmp->offset = offset; - tmp->delta = delta; + if (xp->cur >= xp->number) + return -EINVAL; - if (xt[af].compat_offsets) { - tmp->next = xt[af].compat_offsets->next; - xt[af].compat_offsets->next = tmp; - } else { - xt[af].compat_offsets = tmp; - tmp->next = NULL; - } + if (xp->cur) + delta += xp->compat_tab[xp->cur - 1].delta; + xp->compat_tab[xp->cur].offset = offset; + xp->compat_tab[xp->cur].delta = delta; + xp->cur++; return 0; } EXPORT_SYMBOL_GPL(xt_compat_add_offset); void xt_compat_flush_offsets(u_int8_t af) { - struct compat_delta *tmp, *next; - - if (xt[af].compat_offsets) { - for (tmp = xt[af].compat_offsets; tmp; tmp = next) { - next = tmp->next; - kfree(tmp); - } - xt[af].compat_offsets = NULL; + if (xt[af].compat_tab) { + vfree(xt[af].compat_tab); + xt[af].compat_tab = NULL; + xt[af].number = 0; } } EXPORT_SYMBOL_GPL(xt_compat_flush_offsets); int xt_compat_calc_jump(u_int8_t af, unsigned int offset) { - struct compat_delta *tmp; - int delta; - - for (tmp = xt[af].compat_offsets, delta = 0; tmp; tmp = tmp->next) - if (tmp->offset < offset) - delta += tmp->delta; - return delta; + struct compat_delta *tmp = xt[af].compat_tab; + int mid, left = 0, right = xt[af].cur - 1; + + while (left <= right) { + mid = (left + right) >> 1; + if (offset > tmp[mid].offset) + left = mid + 1; + else if (offset < tmp[mid].offset) + right = mid - 1; + else + return mid ? tmp[mid - 1].delta : 0; + } + WARN_ON_ONCE(1); + return 0; } EXPORT_SYMBOL_GPL(xt_compat_calc_jump); +void xt_compat_init_offsets(u_int8_t af, unsigned int number) +{ + xt[af].number = number; + xt[af].cur = 0; +} +EXPORT_SYMBOL(xt_compat_init_offsets); + int xt_compat_match_offset(const struct xt_match *match) { u_int16_t csize = match->compatsize ? : match->matchsize; @@ -1337,7 +1351,7 @@ static int __init xt_init(void) mutex_init(&xt[i].mutex); #ifdef CONFIG_COMPAT mutex_init(&xt[i].compat_mutex); - xt[i].compat_offsets = NULL; + xt[i].compat_tab = NULL; #endif INIT_LIST_HEAD(&xt[i].target); INIT_LIST_HEAD(&xt[i].match); -- cgit v1.2.3 From c7066f70d9610df0b9406cc635fc09e86136e714 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 14 Jan 2011 13:36:42 +0100 Subject: netfilter: fix Kconfig dependencies Fix dependencies of netfilter realm match: it depends on NET_CLS_ROUTE, which itself depends on NET_SCHED; this dependency is missing from netfilter. Since matching on realms is also useful without having NET_SCHED enabled and the option really only controls whether the tclassid member is included in route and dst entries, rename the config option to IP_ROUTE_CLASSID and move it outside of traffic scheduling context to get rid of the NET_SCHED dependeny. Reported-by: Vladis Kletnieks Signed-off-by: Patrick McHardy --- net/netfilter/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 1534f2b44caf..1b79353a5d29 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -886,7 +886,7 @@ config NETFILTER_XT_MATCH_RATEEST config NETFILTER_XT_MATCH_REALM tristate '"realm" match support' depends on NETFILTER_ADVANCED - select NET_CLS_ROUTE + select IP_ROUTE_CLASSID help This option adds a `realm' match, which allows you to use the realm key from the routing subsystem inside iptables. -- cgit v1.2.3 From d862a6622e9db508d4b28cc7c5bc28bd548cc24e Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 14 Jan 2011 15:45:56 +0100 Subject: netfilter: nf_conntrack: use is_vmalloc_addr() Use is_vmalloc_addr() in nf_ct_free_hashtable() and get rid of the vmalloc flags to indicate that a hash table has been allocated using vmalloc(). Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_core.c | 26 +++++++++----------------- net/netfilter/nf_conntrack_expect.c | 9 +++------ net/netfilter/nf_conntrack_helper.c | 10 +++------- 3 files changed, 15 insertions(+), 30 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index e95ac42ef673..dc2ff2cd0a7e 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -1202,9 +1202,9 @@ static int kill_all(struct nf_conn *i, void *data) return 1; } -void nf_ct_free_hashtable(void *hash, int vmalloced, unsigned int size) +void nf_ct_free_hashtable(void *hash, unsigned int size) { - if (vmalloced) + if (is_vmalloc_addr(hash)) vfree(hash); else free_pages((unsigned long)hash, @@ -1271,8 +1271,7 @@ static void nf_conntrack_cleanup_net(struct net *net) goto i_see_dead_people; } - nf_ct_free_hashtable(net->ct.hash, net->ct.hash_vmalloc, - net->ct.htable_size); + nf_ct_free_hashtable(net->ct.hash, net->ct.htable_size); nf_conntrack_ecache_fini(net); nf_conntrack_acct_fini(net); nf_conntrack_expect_fini(net); @@ -1301,21 +1300,18 @@ void nf_conntrack_cleanup(struct net *net) } } -void *nf_ct_alloc_hashtable(unsigned int *sizep, int *vmalloced, int nulls) +void *nf_ct_alloc_hashtable(unsigned int *sizep, int nulls) { struct hlist_nulls_head *hash; unsigned int nr_slots, i; size_t sz; - *vmalloced = 0; - BUILD_BUG_ON(sizeof(struct hlist_nulls_head) != sizeof(struct hlist_head)); nr_slots = *sizep = roundup(*sizep, PAGE_SIZE / sizeof(struct hlist_nulls_head)); sz = nr_slots * sizeof(struct hlist_nulls_head); hash = (void *)__get_free_pages(GFP_KERNEL | __GFP_NOWARN | __GFP_ZERO, get_order(sz)); if (!hash) { - *vmalloced = 1; printk(KERN_WARNING "nf_conntrack: falling back to vmalloc.\n"); hash = __vmalloc(sz, GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO, PAGE_KERNEL); @@ -1331,7 +1327,7 @@ EXPORT_SYMBOL_GPL(nf_ct_alloc_hashtable); int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) { - int i, bucket, vmalloced, old_vmalloced; + int i, bucket; unsigned int hashsize, old_size; struct hlist_nulls_head *hash, *old_hash; struct nf_conntrack_tuple_hash *h; @@ -1348,7 +1344,7 @@ int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) if (!hashsize) return -EINVAL; - hash = nf_ct_alloc_hashtable(&hashsize, &vmalloced, 1); + hash = nf_ct_alloc_hashtable(&hashsize, 1); if (!hash) return -ENOMEM; @@ -1370,15 +1366,13 @@ int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp) } } old_size = init_net.ct.htable_size; - old_vmalloced = init_net.ct.hash_vmalloc; old_hash = init_net.ct.hash; init_net.ct.htable_size = nf_conntrack_htable_size = hashsize; - init_net.ct.hash_vmalloc = vmalloced; init_net.ct.hash = hash; spin_unlock_bh(&nf_conntrack_lock); - nf_ct_free_hashtable(old_hash, old_vmalloced, old_size); + nf_ct_free_hashtable(old_hash, old_size); return 0; } EXPORT_SYMBOL_GPL(nf_conntrack_set_hashsize); @@ -1491,8 +1485,7 @@ static int nf_conntrack_init_net(struct net *net) } net->ct.htable_size = nf_conntrack_htable_size; - net->ct.hash = nf_ct_alloc_hashtable(&net->ct.htable_size, - &net->ct.hash_vmalloc, 1); + net->ct.hash = nf_ct_alloc_hashtable(&net->ct.htable_size, 1); if (!net->ct.hash) { ret = -ENOMEM; printk(KERN_ERR "Unable to create nf_conntrack_hash\n"); @@ -1515,8 +1508,7 @@ err_ecache: err_acct: nf_conntrack_expect_fini(net); err_expect: - nf_ct_free_hashtable(net->ct.hash, net->ct.hash_vmalloc, - net->ct.htable_size); + nf_ct_free_hashtable(net->ct.hash, net->ct.htable_size); err_hash: kmem_cache_destroy(net->ct.nf_conntrack_cachep); err_cache: diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 4a9ed23180df..cd1e8e0970f2 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -639,8 +639,7 @@ int nf_conntrack_expect_init(struct net *net) } net->ct.expect_count = 0; - net->ct.expect_hash = nf_ct_alloc_hashtable(&nf_ct_expect_hsize, - &net->ct.expect_vmalloc, 0); + net->ct.expect_hash = nf_ct_alloc_hashtable(&nf_ct_expect_hsize, 0); if (net->ct.expect_hash == NULL) goto err1; @@ -662,8 +661,7 @@ err3: if (net_eq(net, &init_net)) kmem_cache_destroy(nf_ct_expect_cachep); err2: - nf_ct_free_hashtable(net->ct.expect_hash, net->ct.expect_vmalloc, - nf_ct_expect_hsize); + nf_ct_free_hashtable(net->ct.expect_hash, nf_ct_expect_hsize); err1: return err; } @@ -675,6 +673,5 @@ void nf_conntrack_expect_fini(struct net *net) rcu_barrier(); /* Wait for call_rcu() before destroy */ kmem_cache_destroy(nf_ct_expect_cachep); } - nf_ct_free_hashtable(net->ct.expect_hash, net->ct.expect_vmalloc, - nf_ct_expect_hsize); + nf_ct_free_hashtable(net->ct.expect_hash, nf_ct_expect_hsize); } diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index 767bbe98a0f0..1bdfea357955 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -33,7 +33,6 @@ static DEFINE_MUTEX(nf_ct_helper_mutex); static struct hlist_head *nf_ct_helper_hash __read_mostly; static unsigned int nf_ct_helper_hsize __read_mostly; static unsigned int nf_ct_helper_count __read_mostly; -static int nf_ct_helper_vmalloc; /* Stupid hash, but collision free for the default registrations of the @@ -267,8 +266,7 @@ int nf_conntrack_helper_init(void) int err; nf_ct_helper_hsize = 1; /* gets rounded up to use one page */ - nf_ct_helper_hash = nf_ct_alloc_hashtable(&nf_ct_helper_hsize, - &nf_ct_helper_vmalloc, 0); + nf_ct_helper_hash = nf_ct_alloc_hashtable(&nf_ct_helper_hsize, 0); if (!nf_ct_helper_hash) return -ENOMEM; @@ -279,14 +277,12 @@ int nf_conntrack_helper_init(void) return 0; err1: - nf_ct_free_hashtable(nf_ct_helper_hash, nf_ct_helper_vmalloc, - nf_ct_helper_hsize); + nf_ct_free_hashtable(nf_ct_helper_hash, nf_ct_helper_hsize); return err; } void nf_conntrack_helper_fini(void) { nf_ct_extend_unregister(&helper_extend); - nf_ct_free_hashtable(nf_ct_helper_hash, nf_ct_helper_vmalloc, - nf_ct_helper_hsize); + nf_ct_free_hashtable(nf_ct_helper_hash, nf_ct_helper_hsize); } -- cgit v1.2.3 From 43f393caec0362abe03c72799d3f342af3973070 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sun, 16 Jan 2011 18:10:28 +0100 Subject: netfilter: audit target to record accepted/dropped packets This patch adds a new netfilter target which creates audit records for packets traversing a certain chain. It can be used to record packets which are rejected administraively as follows: -N AUDIT_DROP -A AUDIT_DROP -j AUDIT --type DROP -A AUDIT_DROP -j DROP a rule which would typically drop or reject a packet would then invoke the new chain to record packets before dropping them. -j AUDIT_DROP The module is protocol independant and works for iptables, ip6tables and ebtables. The following information is logged: - netfilter hook - packet length - incomming/outgoing interface - MAC src/dst/proto for ethernet packets - src/dst/protocol address for IPv4/IPv6 - src/dst port for TCP/UDP/UDPLITE - icmp type/code Cc: Patrick McHardy Cc: Eric Paris Cc: Al Viro Signed-off-by: Thomas Graf Signed-off-by: Patrick McHardy --- net/netfilter/Kconfig | 10 +++ net/netfilter/Makefile | 1 + net/netfilter/xt_AUDIT.c | 204 +++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 215 insertions(+) create mode 100644 net/netfilter/xt_AUDIT.c (limited to 'net/netfilter') diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 1b79353a5d29..93918f022555 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -326,6 +326,16 @@ config NETFILTER_XT_CONNMARK comment "Xtables targets" +config NETFILTER_XT_TARGET_AUDIT + tristate "AUDIT target support" + depends on AUDIT + depends on NETFILTER_ADVANCED + ---help--- + This option adds a 'AUDIT' target, which can be used to create + audit records for packets dropped/accepted. + + To compileit as a module, choose M here. If unsure, say N. + config NETFILTER_XT_TARGET_CHECKSUM tristate "CHECKSUM target support" depends on IP_NF_MANGLE || IP6_NF_MANGLE diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 441050f31111..401d574bdfd2 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -45,6 +45,7 @@ obj-$(CONFIG_NETFILTER_XT_MARK) += xt_mark.o obj-$(CONFIG_NETFILTER_XT_CONNMARK) += xt_connmark.o # targets +obj-$(CONFIG_NETFILTER_XT_TARGET_AUDIT) += xt_AUDIT.o obj-$(CONFIG_NETFILTER_XT_TARGET_CHECKSUM) += xt_CHECKSUM.o obj-$(CONFIG_NETFILTER_XT_TARGET_CLASSIFY) += xt_CLASSIFY.o obj-$(CONFIG_NETFILTER_XT_TARGET_CONNSECMARK) += xt_CONNSECMARK.o diff --git a/net/netfilter/xt_AUDIT.c b/net/netfilter/xt_AUDIT.c new file mode 100644 index 000000000000..81802d27346e --- /dev/null +++ b/net/netfilter/xt_AUDIT.c @@ -0,0 +1,204 @@ +/* + * Creates audit record for dropped/accepted packets + * + * (C) 2010-2011 Thomas Graf + * (C) 2010-2011 Red Hat, Inc. + * + * 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. +*/ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Thomas Graf "); +MODULE_DESCRIPTION("Xtables: creates audit records for dropped/accepted packets"); +MODULE_ALIAS("ipt_AUDIT"); +MODULE_ALIAS("ip6t_AUDIT"); +MODULE_ALIAS("ebt_AUDIT"); +MODULE_ALIAS("arpt_AUDIT"); + +static void audit_proto(struct audit_buffer *ab, struct sk_buff *skb, + unsigned int proto, unsigned int offset) +{ + switch (proto) { + case IPPROTO_TCP: + case IPPROTO_UDP: + case IPPROTO_UDPLITE: { + const __be16 *pptr; + __be16 _ports[2]; + + pptr = skb_header_pointer(skb, offset, sizeof(_ports), _ports); + if (pptr == NULL) { + audit_log_format(ab, " truncated=1"); + return; + } + + audit_log_format(ab, " sport=%hu dport=%hu", + ntohs(pptr[0]), ntohs(pptr[1])); + } + break; + + case IPPROTO_ICMP: + case IPPROTO_ICMPV6: { + const u8 *iptr; + u8 _ih[2]; + + iptr = skb_header_pointer(skb, offset, sizeof(_ih), &_ih); + if (iptr == NULL) { + audit_log_format(ab, " truncated=1"); + return; + } + + audit_log_format(ab, " icmptype=%hhu icmpcode=%hhu", + iptr[0], iptr[1]); + + } + break; + } +} + +static void audit_ip4(struct audit_buffer *ab, struct sk_buff *skb) +{ + struct iphdr _iph; + const struct iphdr *ih; + + ih = skb_header_pointer(skb, 0, sizeof(_iph), &_iph); + if (!ih) { + audit_log_format(ab, " truncated=1"); + return; + } + + audit_log_format(ab, " saddr=%pI4 daddr=%pI4 ipid=%hu proto=%hhu", + &ih->saddr, &ih->daddr, ntohs(ih->id), ih->protocol); + + if (ntohs(ih->frag_off) & IP_OFFSET) { + audit_log_format(ab, " frag=1"); + return; + } + + audit_proto(ab, skb, ih->protocol, ih->ihl * 4); +} + +static void audit_ip6(struct audit_buffer *ab, struct sk_buff *skb) +{ + struct ipv6hdr _ip6h; + const struct ipv6hdr *ih; + u8 nexthdr; + int offset; + + ih = skb_header_pointer(skb, skb_network_offset(skb), sizeof(_ip6h), &_ip6h); + if (!ih) { + audit_log_format(ab, " truncated=1"); + return; + } + + nexthdr = ih->nexthdr; + offset = ipv6_skip_exthdr(skb, skb_network_offset(skb) + sizeof(_ip6h), + &nexthdr); + + audit_log_format(ab, " saddr=%pI6c daddr=%pI6c proto=%hhu", + &ih->saddr, &ih->daddr, nexthdr); + + if (offset) + audit_proto(ab, skb, nexthdr, offset); +} + +static unsigned int +audit_tg(struct sk_buff *skb, const struct xt_action_param *par) +{ + const struct xt_audit_info *info = par->targinfo; + struct audit_buffer *ab; + + ab = audit_log_start(NULL, GFP_ATOMIC, AUDIT_NETFILTER_PKT); + if (ab == NULL) + goto errout; + + audit_log_format(ab, "action=%hhu hook=%u len=%u inif=%s outif=%s", + info->type, par->hooknum, skb->len, + par->in ? par->in->name : "?", + par->out ? par->out->name : "?"); + + if (skb->mark) + audit_log_format(ab, " mark=%#x", skb->mark); + + if (skb->dev && skb->dev->type == ARPHRD_ETHER) { + audit_log_format(ab, " smac=%pM dmac=%pM macproto=0x%04x", + eth_hdr(skb)->h_source, eth_hdr(skb)->h_dest, + ntohs(eth_hdr(skb)->h_proto)); + + if (par->family == NFPROTO_BRIDGE) { + switch (eth_hdr(skb)->h_proto) { + case __constant_htons(ETH_P_IP): + audit_ip4(ab, skb); + break; + + case __constant_htons(ETH_P_IPV6): + audit_ip6(ab, skb); + break; + } + } + } + + switch (par->family) { + case NFPROTO_IPV4: + audit_ip4(ab, skb); + break; + + case NFPROTO_IPV6: + audit_ip6(ab, skb); + break; + } + + audit_log_end(ab); + +errout: + return XT_CONTINUE; +} + +static int audit_tg_check(const struct xt_tgchk_param *par) +{ + const struct xt_audit_info *info = par->targinfo; + + if (info->type > XT_AUDIT_TYPE_MAX) { + pr_info("Audit type out of range (valid range: 0..%hhu)\n", + XT_AUDIT_TYPE_MAX); + return -ERANGE; + } + + return 0; +} + +static struct xt_target audit_tg_reg __read_mostly = { + .name = "AUDIT", + .family = NFPROTO_UNSPEC, + .target = audit_tg, + .targetsize = sizeof(struct xt_audit_info), + .checkentry = audit_tg_check, + .me = THIS_MODULE, +}; + +static int __init audit_tg_init(void) +{ + return xt_register_target(&audit_tg_reg); +} + +static void __exit audit_tg_exit(void) +{ + xt_unregister_target(&audit_tg_reg); +} + +module_init(audit_tg_init); +module_exit(audit_tg_exit); -- cgit v1.2.3 From fbabf31e4d482149b5e2704eb0287cf9117bdcf3 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sun, 16 Jan 2011 18:12:59 +0100 Subject: netfilter: create audit records for x_tables replaces The setsockopt() syscall to replace tables is already recorded in the audit logs. This patch stores additional information such as table name and netfilter protocol. Cc: Patrick McHardy Cc: Eric Paris Cc: Al Viro Signed-off-by: Thomas Graf Signed-off-by: Patrick McHardy --- net/netfilter/x_tables.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) (limited to 'net/netfilter') diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index ee5de3af510a..fbc2b72091e0 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include @@ -834,6 +835,21 @@ xt_replace_table(struct xt_table *table, */ local_bh_enable(); +#ifdef CONFIG_AUDIT + if (audit_enabled) { + struct audit_buffer *ab; + + ab = audit_log_start(current->audit_context, GFP_KERNEL, + AUDIT_NETFILTER_CFG); + if (ab) { + audit_log_format(ab, "table=%s family=%u entries=%u", + table->name, table->af, + private->number); + audit_log_end(ab); + } + } +#endif + return private; } EXPORT_SYMBOL_GPL(xt_replace_table); -- cgit v1.2.3 From f1e231a356f90a67f8547c2881a62c92084683c6 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Tue, 18 Jan 2011 06:30:13 +0100 Subject: netfilter: xtables: add missing aliases for autoloading via iptables Signed-off-by: Jan Engelhardt --- net/netfilter/xt_IDLETIMER.c | 2 ++ net/netfilter/xt_LED.c | 2 ++ net/netfilter/xt_cpu.c | 2 ++ 3 files changed, 6 insertions(+) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_IDLETIMER.c b/net/netfilter/xt_IDLETIMER.c index be1f22e13545..3bdd443aaf15 100644 --- a/net/netfilter/xt_IDLETIMER.c +++ b/net/netfilter/xt_IDLETIMER.c @@ -313,3 +313,5 @@ MODULE_AUTHOR("Timo Teras "); MODULE_AUTHOR("Luciano Coelho "); MODULE_DESCRIPTION("Xtables: idle time monitor"); MODULE_LICENSE("GPL v2"); +MODULE_ALIAS("ipt_IDLETIMER"); +MODULE_ALIAS("ip6t_IDLETIMER"); diff --git a/net/netfilter/xt_LED.c b/net/netfilter/xt_LED.c index a4140509eea1..993de2ba89d3 100644 --- a/net/netfilter/xt_LED.c +++ b/net/netfilter/xt_LED.c @@ -31,6 +31,8 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Adam Nielsen "); MODULE_DESCRIPTION("Xtables: trigger LED devices on packet match"); +MODULE_ALIAS("ipt_LED"); +MODULE_ALIAS("ip6t_LED"); static LIST_HEAD(xt_led_triggers); static DEFINE_MUTEX(xt_led_mutex); diff --git a/net/netfilter/xt_cpu.c b/net/netfilter/xt_cpu.c index b39db8a5cbae..c7a2e5466bc4 100644 --- a/net/netfilter/xt_cpu.c +++ b/net/netfilter/xt_cpu.c @@ -22,6 +22,8 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Eric Dumazet "); MODULE_DESCRIPTION("Xtables: CPU match"); +MODULE_ALIAS("ipt_cpu"); +MODULE_ALIAS("ip6t_cpu"); static int cpu_mt_check(const struct xt_mtchk_param *par) { -- cgit v1.2.3 From 1cc34c30be0e27d4ba8c1ce04a8a4f46c927d121 Mon Sep 17 00:00:00 2001 From: Richard Weinberger Date: Tue, 18 Jan 2011 01:36:57 +0100 Subject: netfilter: xt_connlimit: use hotdrop jump mark Signed-off-by: Richard Weinberger Signed-off-by: Jan Engelhardt --- net/netfilter/xt_connlimit.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c index 5c5b6b921b84..452bc16af56c 100644 --- a/net/netfilter/xt_connlimit.c +++ b/net/netfilter/xt_connlimit.c @@ -204,11 +204,9 @@ connlimit_mt(const struct sk_buff *skb, struct xt_action_param *par) &info->mask, par->family); spin_unlock_bh(&info->data->lock); - if (connections < 0) { + if (connections < 0) /* kmalloc failed, drop it entirely */ - par->hotdrop = true; - return false; - } + goto hotdrop; return (connections > info->limit) ^ info->inverse; -- cgit v1.2.3 From 45eec34195853e918518231dcefaca1ea4ebacfc Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Tue, 18 Jan 2011 15:08:13 +0100 Subject: netfilter: nf_conntrack: remove an atomic bit operation As this ct won't be seen by the others, we don't need to set the IPS_CONFIRMED_BIT in atomic way. Signed-off-by: Changli Gao Cc: Tim Gardner Cc: Eric Dumazet Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index dc2ff2cd0a7e..f47ac67e1bfe 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -486,7 +486,7 @@ __nf_conntrack_confirm(struct sk_buff *skb) ct->timeout.expires += jiffies; add_timer(&ct->timeout); atomic_inc(&ct->ct_general.use); - set_bit(IPS_CONFIRMED_BIT, &ct->status); + ct->status |= IPS_CONFIRMED; /* Since the lookup is lockless, hash insertion must be done after * starting the timer and setting the CONFIRMED bit. The RCU barriers -- cgit v1.2.3 From 5f2cafe73671d865af88494159f3e8c1b322e1c5 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 18 Jan 2011 15:18:08 +0100 Subject: netfilter: Kconfig: NFQUEUE is useless without NETFILTER_NETLINK_QUEUE NFLOG already does the same thing for NETFILTER_NETLINK_LOG. Signed-off-by: Florian Westphal Signed-off-by: Patrick McHardy --- net/netfilter/Kconfig | 1 + 1 file changed, 1 insertion(+) (limited to 'net/netfilter') diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 93918f022555..e2480bddbfd5 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -487,6 +487,7 @@ config NETFILTER_XT_TARGET_NFLOG config NETFILTER_XT_TARGET_NFQUEUE tristate '"NFQUEUE" target Support' depends on NETFILTER_ADVANCED + select NETFILTER_NETLINK_QUEUE help This target replaced the old obsolete QUEUE target. -- cgit v1.2.3 From f15850861860636c905b33a9a5be3dcbc2b0d56a Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 18 Jan 2011 15:27:28 +0100 Subject: netfilter: nfnetlink_queue: return error number to caller instead of returning -1 on error, return an error number to allow the caller to handle some errors differently. ECANCELED is used to indicate that the hook is going away and should be ignored. A followup patch will introduce more 'ignore this hook' conditions, (depending on queue settings) and will move kfree_skb responsibility to the caller. Signed-off-by: Florian Westphal Signed-off-by: Patrick McHardy --- net/netfilter/core.c | 6 ++++-- net/netfilter/nf_queue.c | 44 +++++++++++++++++++++++++++++------------ net/netfilter/nfnetlink_queue.c | 22 +++++++++++++-------- 3 files changed, 49 insertions(+), 23 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/core.c b/net/netfilter/core.c index e69d537362c7..91d66d2f8cd9 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -179,9 +179,11 @@ next_hook: if (ret == 0) ret = -EPERM; } else if ((verdict & NF_VERDICT_MASK) == NF_QUEUE) { - if (!nf_queue(skb, elem, pf, hook, indev, outdev, okfn, - verdict >> NF_VERDICT_BITS)) + ret = nf_queue(skb, elem, pf, hook, indev, outdev, okfn, + verdict >> NF_VERDICT_BITS); + if (ret == -ECANCELED) goto next_hook; + ret = 0; } rcu_read_unlock(); return ret; diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index 1876f7411561..ad25c7e726bc 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -125,7 +125,7 @@ static int __nf_queue(struct sk_buff *skb, int (*okfn)(struct sk_buff *), unsigned int queuenum) { - int status; + int status = -ENOENT; struct nf_queue_entry *entry = NULL; #ifdef CONFIG_BRIDGE_NETFILTER struct net_device *physindev; @@ -146,8 +146,10 @@ static int __nf_queue(struct sk_buff *skb, goto err_unlock; entry = kmalloc(sizeof(*entry) + afinfo->route_key_size, GFP_ATOMIC); - if (!entry) + if (!entry) { + status = -ENOMEM; goto err_unlock; + } *entry = (struct nf_queue_entry) { .skb = skb, @@ -163,9 +165,8 @@ static int __nf_queue(struct sk_buff *skb, if (!try_module_get(entry->elem->owner)) { rcu_read_unlock(); kfree(entry); - return 0; + return -ECANCELED; } - /* Bump dev refs so they don't vanish while packet is out */ if (indev) dev_hold(indev); @@ -192,14 +193,14 @@ static int __nf_queue(struct sk_buff *skb, goto err; } - return 1; + return 0; err_unlock: rcu_read_unlock(); err: kfree_skb(skb); kfree(entry); - return 1; + return status; } int nf_queue(struct sk_buff *skb, @@ -211,6 +212,8 @@ int nf_queue(struct sk_buff *skb, unsigned int queuenum) { struct sk_buff *segs; + int err; + unsigned int queued; if (!skb_is_gso(skb)) return __nf_queue(skb, elem, pf, hook, indev, outdev, okfn, @@ -227,19 +230,32 @@ int nf_queue(struct sk_buff *skb, segs = skb_gso_segment(skb, 0); kfree_skb(skb); + /* Does not use PTR_ERR to limit the number of error codes that can be + * returned by nf_queue. For instance, callers rely on -ECANCELED to mean + * 'ignore this hook'. + */ if (IS_ERR(segs)) - return 1; + return -EINVAL; + queued = 0; + err = 0; do { struct sk_buff *nskb = segs->next; segs->next = NULL; - if (!__nf_queue(segs, elem, pf, hook, indev, outdev, okfn, - queuenum)) + if (err == 0) + err = __nf_queue(segs, elem, pf, hook, indev, + outdev, okfn, queuenum); + if (err == 0) + queued++; + else kfree_skb(segs); segs = nskb; } while (segs); - return 1; + + if (unlikely(err && queued)) + err = 0; + return err; } void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict) @@ -247,6 +263,7 @@ void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict) struct sk_buff *skb = entry->skb; struct list_head *elem = &entry->elem->list; const struct nf_afinfo *afinfo; + int err; rcu_read_lock(); @@ -280,9 +297,10 @@ void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict) local_bh_enable(); break; case NF_QUEUE: - if (!__nf_queue(skb, elem, entry->pf, entry->hook, - entry->indev, entry->outdev, entry->okfn, - verdict >> NF_VERDICT_BITS)) + err = __nf_queue(skb, elem, entry->pf, entry->hook, + entry->indev, entry->outdev, entry->okfn, + verdict >> NF_VERDICT_BITS); + if (err == -ECANCELED) goto next_hook; break; case NF_STOLEN: diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 68e67d19724d..b83123f12b42 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -387,25 +387,31 @@ nfqnl_enqueue_packet(struct nf_queue_entry *entry, unsigned int queuenum) { struct sk_buff *nskb; struct nfqnl_instance *queue; - int err; + int err = -ENOBUFS; /* rcu_read_lock()ed by nf_hook_slow() */ queue = instance_lookup(queuenum); - if (!queue) + if (!queue) { + err = -ESRCH; goto err_out; + } - if (queue->copy_mode == NFQNL_COPY_NONE) + if (queue->copy_mode == NFQNL_COPY_NONE) { + err = -EINVAL; goto err_out; + } nskb = nfqnl_build_packet_message(queue, entry); - if (nskb == NULL) + if (nskb == NULL) { + err = -ENOMEM; goto err_out; - + } spin_lock_bh(&queue->lock); - if (!queue->peer_pid) + if (!queue->peer_pid) { + err = -EINVAL; goto err_out_free_nskb; - + } if (queue->queue_total >= queue->queue_maxlen) { queue->queue_dropped++; if (net_ratelimit()) @@ -432,7 +438,7 @@ err_out_free_nskb: err_out_unlock: spin_unlock_bh(&queue->lock); err_out: - return -1; + return err; } static int -- cgit v1.2.3 From 06cdb6349c1f3fd439398dbc04ce4c696f0a41ab Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 18 Jan 2011 15:28:38 +0100 Subject: netfilter: nfnetlink_queue: do not free skb on error Move free responsibility from nf_queue to caller. This enables more flexible error handling; we can now accept the skb instead of freeing it. Signed-off-by: Florian Westphal Signed-off-by: Patrick McHardy --- net/netfilter/core.c | 7 +++++-- net/netfilter/nf_queue.c | 17 ++++++++++------- 2 files changed, 15 insertions(+), 9 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/core.c b/net/netfilter/core.c index 91d66d2f8cd9..0c5b796ef527 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -181,8 +181,11 @@ next_hook: } else if ((verdict & NF_VERDICT_MASK) == NF_QUEUE) { ret = nf_queue(skb, elem, pf, hook, indev, outdev, okfn, verdict >> NF_VERDICT_BITS); - if (ret == -ECANCELED) - goto next_hook; + if (ret < 0) { + if (ret == -ECANCELED) + goto next_hook; + kfree_skb(skb); + } ret = 0; } rcu_read_unlock(); diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index ad25c7e726bc..5c4b730a2e68 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -163,9 +163,8 @@ static int __nf_queue(struct sk_buff *skb, /* If it's going away, ignore hook. */ if (!try_module_get(entry->elem->owner)) { - rcu_read_unlock(); - kfree(entry); - return -ECANCELED; + status = -ECANCELED; + goto err_unlock; } /* Bump dev refs so they don't vanish while packet is out */ if (indev) @@ -198,7 +197,6 @@ static int __nf_queue(struct sk_buff *skb, err_unlock: rcu_read_unlock(); err: - kfree_skb(skb); kfree(entry); return status; } @@ -229,7 +227,6 @@ int nf_queue(struct sk_buff *skb, } segs = skb_gso_segment(skb, 0); - kfree_skb(skb); /* Does not use PTR_ERR to limit the number of error codes that can be * returned by nf_queue. For instance, callers rely on -ECANCELED to mean * 'ignore this hook'. @@ -253,8 +250,11 @@ int nf_queue(struct sk_buff *skb, segs = nskb; } while (segs); + /* also free orig skb if only some segments were queued */ if (unlikely(err && queued)) err = 0; + if (err == 0) + kfree_skb(skb); return err; } @@ -300,8 +300,11 @@ void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict) err = __nf_queue(skb, elem, entry->pf, entry->hook, entry->indev, entry->outdev, entry->okfn, verdict >> NF_VERDICT_BITS); - if (err == -ECANCELED) - goto next_hook; + if (err < 0) { + if (err == -ECANCELED) + goto next_hook; + kfree_skb(skb); + } break; case NF_STOLEN: default: -- cgit v1.2.3 From f615df76ed862b7d3927ec5f55b805ca19be29d9 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 18 Jan 2011 15:52:14 +0100 Subject: netfilter: reduce NF_VERDICT_MASK to 0xff NF_VERDICT_MASK is currently 0xffff. This is because the upper 16 bits are used to store errno (for NF_DROP) or the queue number (NF_QUEUE verdict). As there are up to 0xffff different queues available, there is no more room to store additional flags. At the moment there are only 6 different verdicts, i.e. we can reduce NF_VERDICT_MASK to 0xff to allow storing additional flags in the 0xff00 space. NF_VERDICT_BITS would then be reduced to 8, but because the value is exported to userspace, this might cause breakage; e.g.: e.g. 'queuenr = (1 << NF_VERDICT_BITS) | NF_QUEUE' would now break. Thus, remove NF_VERDICT_BITS usage in the kernel and move the old value to the 'userspace compat' section. Signed-off-by: Florian Westphal Signed-off-by: Patrick McHardy --- net/netfilter/core.c | 4 ++-- net/netfilter/nf_queue.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/core.c b/net/netfilter/core.c index 0c5b796ef527..4d88e45b978e 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -175,12 +175,12 @@ next_hook: ret = 1; } else if ((verdict & NF_VERDICT_MASK) == NF_DROP) { kfree_skb(skb); - ret = -(verdict >> NF_VERDICT_BITS); + ret = NF_DROP_GETERR(verdict); if (ret == 0) ret = -EPERM; } else if ((verdict & NF_VERDICT_MASK) == NF_QUEUE) { ret = nf_queue(skb, elem, pf, hook, indev, outdev, okfn, - verdict >> NF_VERDICT_BITS); + verdict >> NF_VERDICT_QBITS); if (ret < 0) { if (ret == -ECANCELED) goto next_hook; diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index 5c4b730a2e68..ce1150d4a3f2 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -299,7 +299,7 @@ void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict) case NF_QUEUE: err = __nf_queue(skb, elem, entry->pf, entry->hook, entry->indev, entry->outdev, entry->okfn, - verdict >> NF_VERDICT_BITS); + verdict >> NF_VERDICT_QBITS); if (err < 0) { if (err == -ECANCELED) goto next_hook; -- cgit v1.2.3 From 94b27cc36123069966616670c3653cd6873babe9 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 18 Jan 2011 16:08:30 +0100 Subject: netfilter: allow NFQUEUE bypass if no listener is available If an skb is to be NF_QUEUE'd, but no program has opened the queue, the packet is dropped. This adds a v2 target revision of xt_NFQUEUE that allows packets to continue through the ruleset instead. Because the actual queueing happens outside of the target context, the 'bypass' flag has to be communicated back to the netfilter core. Unfortunately the only choice to do this without adding a new function argument is to use the target function return value (i.e. the verdict). In the NF_QUEUE case, the upper 16bit already contain the queue number to use. The previous patch reduced NF_VERDICT_MASK to 0xff, i.e. we now have extra room for a new flag. If a hook issued a NF_QUEUE verdict, then the netfilter core will continue packet processing if the queueing hook returns -ESRCH (== "this queue does not exist") and the new NF_VERDICT_FLAG_QUEUE_BYPASS flag is set in the verdict value. Note: If the queue exists, but userspace does not consume packets fast enough, the skb will still be dropped. Signed-off-by: Florian Westphal Signed-off-by: Patrick McHardy --- net/netfilter/core.c | 3 +++ net/netfilter/nf_queue.c | 7 ++++++- net/netfilter/xt_NFQUEUE.c | 28 +++++++++++++++++++++++++--- 3 files changed, 34 insertions(+), 4 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/core.c b/net/netfilter/core.c index 4d88e45b978e..1e00bf7d27c5 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -184,6 +184,9 @@ next_hook: if (ret < 0) { if (ret == -ECANCELED) goto next_hook; + if (ret == -ESRCH && + (verdict & NF_VERDICT_FLAG_QUEUE_BYPASS)) + goto next_hook; kfree_skb(skb); } ret = 0; diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c index ce1150d4a3f2..5ab22e2bbd7d 100644 --- a/net/netfilter/nf_queue.c +++ b/net/netfilter/nf_queue.c @@ -138,8 +138,10 @@ static int __nf_queue(struct sk_buff *skb, rcu_read_lock(); qh = rcu_dereference(queue_handler[pf]); - if (!qh) + if (!qh) { + status = -ESRCH; goto err_unlock; + } afinfo = nf_get_afinfo(pf); if (!afinfo) @@ -303,6 +305,9 @@ void nf_reinject(struct nf_queue_entry *entry, unsigned int verdict) if (err < 0) { if (err == -ECANCELED) goto next_hook; + if (err == -ESRCH && + (verdict & NF_VERDICT_FLAG_QUEUE_BYPASS)) + goto next_hook; kfree_skb(skb); } break; diff --git a/net/netfilter/xt_NFQUEUE.c b/net/netfilter/xt_NFQUEUE.c index 39627706aac6..d4f4b5d66b20 100644 --- a/net/netfilter/xt_NFQUEUE.c +++ b/net/netfilter/xt_NFQUEUE.c @@ -83,9 +83,20 @@ nfqueue_tg_v1(struct sk_buff *skb, const struct xt_action_param *par) return NF_QUEUE_NR(queue); } -static int nfqueue_tg_v1_check(const struct xt_tgchk_param *par) +static unsigned int +nfqueue_tg_v2(struct sk_buff *skb, const struct xt_action_param *par) { - const struct xt_NFQ_info_v1 *info = par->targinfo; + const struct xt_NFQ_info_v2 *info = par->targinfo; + unsigned int ret = nfqueue_tg_v1(skb, par); + + if (info->bypass) + ret |= NF_VERDICT_FLAG_QUEUE_BYPASS; + return ret; +} + +static int nfqueue_tg_check(const struct xt_tgchk_param *par) +{ + const struct xt_NFQ_info_v2 *info = par->targinfo; u32 maxid; if (unlikely(!rnd_inited)) { @@ -102,6 +113,8 @@ static int nfqueue_tg_v1_check(const struct xt_tgchk_param *par) info->queues_total, maxid); return -ERANGE; } + if (par->target->revision == 2 && info->bypass > 1) + return -EINVAL; return 0; } @@ -117,11 +130,20 @@ static struct xt_target nfqueue_tg_reg[] __read_mostly = { .name = "NFQUEUE", .revision = 1, .family = NFPROTO_UNSPEC, - .checkentry = nfqueue_tg_v1_check, + .checkentry = nfqueue_tg_check, .target = nfqueue_tg_v1, .targetsize = sizeof(struct xt_NFQ_info_v1), .me = THIS_MODULE, }, + { + .name = "NFQUEUE", + .revision = 2, + .family = NFPROTO_UNSPEC, + .checkentry = nfqueue_tg_check, + .target = nfqueue_tg_v2, + .targetsize = sizeof(struct xt_NFQ_info_v2), + .me = THIS_MODULE, + }, }; static int __init nfqueue_tg_init(void) -- cgit v1.2.3 From 93557f53e1fbd9e2b6574ab0a9b5852628fde9e3 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Tue, 18 Jan 2011 18:12:24 +0100 Subject: netfilter: nf_conntrack: nf_conntrack snmp helper Adding support for SNMP broadcast connection tracking. The SNMP broadcast requests are now paired with the SNMP responses. Thus allowing using SNMP broadcasts with firewall enabled. Please refer to the following conversation: http://marc.info/?l=netfilter-devel&m=125992205006600&w=2 Patrick McHardy wrote: > > The best solution would be to add generic broadcast tracking, the > > use of expectations for this is a bit of abuse. > > The second best choice I guess would be to move the help() function > > to a shared module and generalize it so it can be used for both. This patch implements the "second best choice". Since the netbios-ns conntrack module uses the same helper functionality as the snmp, only one helper function is added for both snmp and netbios-ns modules into the new object - nf_conntrack_broadcast. Signed-off-by: Jiri Olsa Signed-off-by: Patrick McHardy --- net/netfilter/Kconfig | 19 ++++++++ net/netfilter/Makefile | 2 + net/netfilter/nf_conntrack_broadcast.c | 82 +++++++++++++++++++++++++++++++++ net/netfilter/nf_conntrack_netbios_ns.c | 74 ++++------------------------- net/netfilter/nf_conntrack_snmp.c | 77 +++++++++++++++++++++++++++++++ 5 files changed, 189 insertions(+), 65 deletions(-) create mode 100644 net/netfilter/nf_conntrack_broadcast.c create mode 100644 net/netfilter/nf_conntrack_snmp.c (limited to 'net/netfilter') diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index e2480bddbfd5..939b504604c2 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -185,9 +185,13 @@ config NF_CONNTRACK_IRC To compile it as a module, choose M here. If unsure, say N. +config NF_CONNTRACK_BROADCAST + tristate + config NF_CONNTRACK_NETBIOS_NS tristate "NetBIOS name service protocol support" depends on NETFILTER_ADVANCED + select NF_CONNTRACK_BROADCAST help NetBIOS name service requests are sent as broadcast messages from an unprivileged port and responded to with unicast messages to the @@ -204,6 +208,21 @@ config NF_CONNTRACK_NETBIOS_NS To compile it as a module, choose M here. If unsure, say N. +config NF_CONNTRACK_SNMP + tristate "SNMP service protocol support" + depends on NETFILTER_ADVANCED + select NF_CONNTRACK_BROADCAST + help + SNMP service requests are sent as broadcast messages from an + unprivileged port and responded to with unicast messages to the + same port. This make them hard to firewall properly because connection + tracking doesn't deal with broadcasts. This helper tracks locally + originating SNMP service requests and the corresponding + responses. It relies on correct IP address configuration, specifically + netmask and broadcast address. + + To compile it as a module, choose M here. If unsure, say N. + config NF_CONNTRACK_PPTP tristate "PPtP protocol support" depends on NETFILTER_ADVANCED diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 401d574bdfd2..2c2628de9d3f 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -28,7 +28,9 @@ obj-$(CONFIG_NF_CONNTRACK_AMANDA) += nf_conntrack_amanda.o obj-$(CONFIG_NF_CONNTRACK_FTP) += nf_conntrack_ftp.o obj-$(CONFIG_NF_CONNTRACK_H323) += nf_conntrack_h323.o obj-$(CONFIG_NF_CONNTRACK_IRC) += nf_conntrack_irc.o +obj-$(CONFIG_NF_CONNTRACK_BROADCAST) += nf_conntrack_broadcast.o obj-$(CONFIG_NF_CONNTRACK_NETBIOS_NS) += nf_conntrack_netbios_ns.o +obj-$(CONFIG_NF_CONNTRACK_SNMP) += nf_conntrack_snmp.o obj-$(CONFIG_NF_CONNTRACK_PPTP) += nf_conntrack_pptp.o obj-$(CONFIG_NF_CONNTRACK_SANE) += nf_conntrack_sane.o obj-$(CONFIG_NF_CONNTRACK_SIP) += nf_conntrack_sip.o diff --git a/net/netfilter/nf_conntrack_broadcast.c b/net/netfilter/nf_conntrack_broadcast.c new file mode 100644 index 000000000000..4e99cca61612 --- /dev/null +++ b/net/netfilter/nf_conntrack_broadcast.c @@ -0,0 +1,82 @@ +/* + * broadcast connection tracking helper + * + * (c) 2005 Patrick McHardy + * + * 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 + +int nf_conntrack_broadcast_help(struct sk_buff *skb, + unsigned int protoff, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo, + unsigned int timeout) +{ + struct nf_conntrack_expect *exp; + struct iphdr *iph = ip_hdr(skb); + struct rtable *rt = skb_rtable(skb); + struct in_device *in_dev; + struct nf_conn_help *help = nfct_help(ct); + __be32 mask = 0; + + /* we're only interested in locally generated packets */ + if (skb->sk == NULL) + goto out; + if (rt == NULL || !(rt->rt_flags & RTCF_BROADCAST)) + goto out; + if (CTINFO2DIR(ctinfo) != IP_CT_DIR_ORIGINAL) + goto out; + + rcu_read_lock(); + in_dev = __in_dev_get_rcu(rt->dst.dev); + if (in_dev != NULL) { + for_primary_ifa(in_dev) { + if (ifa->ifa_broadcast == iph->daddr) { + mask = ifa->ifa_mask; + break; + } + } endfor_ifa(in_dev); + } + rcu_read_unlock(); + + if (mask == 0) + goto out; + + exp = nf_ct_expect_alloc(ct); + if (exp == NULL) + goto out; + + exp->tuple = ct->tuplehash[IP_CT_DIR_REPLY].tuple; + exp->tuple.src.u.udp.port = help->helper->tuple.src.u.udp.port; + + exp->mask.src.u3.ip = mask; + exp->mask.src.u.udp.port = htons(0xFFFF); + + exp->expectfn = NULL; + exp->flags = NF_CT_EXPECT_PERMANENT; + exp->class = NF_CT_EXPECT_CLASS_DEFAULT; + exp->helper = NULL; + + nf_ct_expect_related(exp); + nf_ct_expect_put(exp); + + nf_ct_refresh(ct, skb, timeout * HZ); +out: + return NF_ACCEPT; +} +EXPORT_SYMBOL_GPL(nf_conntrack_broadcast_help); + +MODULE_LICENSE("GPL"); diff --git a/net/netfilter/nf_conntrack_netbios_ns.c b/net/netfilter/nf_conntrack_netbios_ns.c index aadde018a072..4c8f30a3d6d2 100644 --- a/net/netfilter/nf_conntrack_netbios_ns.c +++ b/net/netfilter/nf_conntrack_netbios_ns.c @@ -18,14 +18,7 @@ #include #include #include -#include -#include -#include -#include #include -#include -#include -#include #include #include @@ -40,75 +33,26 @@ MODULE_ALIAS("ip_conntrack_netbios_ns"); MODULE_ALIAS_NFCT_HELPER("netbios_ns"); static unsigned int timeout __read_mostly = 3; -module_param(timeout, uint, 0400); +module_param(timeout, uint, S_IRUSR); MODULE_PARM_DESC(timeout, "timeout for master connection/replies in seconds"); -static int help(struct sk_buff *skb, unsigned int protoff, - struct nf_conn *ct, enum ip_conntrack_info ctinfo) -{ - struct nf_conntrack_expect *exp; - struct iphdr *iph = ip_hdr(skb); - struct rtable *rt = skb_rtable(skb); - struct in_device *in_dev; - __be32 mask = 0; - - /* we're only interested in locally generated packets */ - if (skb->sk == NULL) - goto out; - if (rt == NULL || !(rt->rt_flags & RTCF_BROADCAST)) - goto out; - if (CTINFO2DIR(ctinfo) != IP_CT_DIR_ORIGINAL) - goto out; - - rcu_read_lock(); - in_dev = __in_dev_get_rcu(rt->dst.dev); - if (in_dev != NULL) { - for_primary_ifa(in_dev) { - if (ifa->ifa_broadcast == iph->daddr) { - mask = ifa->ifa_mask; - break; - } - } endfor_ifa(in_dev); - } - rcu_read_unlock(); - - if (mask == 0) - goto out; - - exp = nf_ct_expect_alloc(ct); - if (exp == NULL) - goto out; - - exp->tuple = ct->tuplehash[IP_CT_DIR_REPLY].tuple; - exp->tuple.src.u.udp.port = htons(NMBD_PORT); - - exp->mask.src.u3.ip = mask; - exp->mask.src.u.udp.port = htons(0xFFFF); - - exp->expectfn = NULL; - exp->flags = NF_CT_EXPECT_PERMANENT; - exp->class = NF_CT_EXPECT_CLASS_DEFAULT; - exp->helper = NULL; - - nf_ct_expect_related(exp); - nf_ct_expect_put(exp); - - nf_ct_refresh(ct, skb, timeout * HZ); -out: - return NF_ACCEPT; -} - static struct nf_conntrack_expect_policy exp_policy = { .max_expected = 1, }; +static int netbios_ns_help(struct sk_buff *skb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo) +{ + return nf_conntrack_broadcast_help(skb, protoff, ct, ctinfo, timeout); +} + static struct nf_conntrack_helper helper __read_mostly = { .name = "netbios-ns", - .tuple.src.l3num = AF_INET, + .tuple.src.l3num = NFPROTO_IPV4, .tuple.src.u.udp.port = cpu_to_be16(NMBD_PORT), .tuple.dst.protonum = IPPROTO_UDP, .me = THIS_MODULE, - .help = help, + .help = netbios_ns_help, .expect_policy = &exp_policy, }; diff --git a/net/netfilter/nf_conntrack_snmp.c b/net/netfilter/nf_conntrack_snmp.c new file mode 100644 index 000000000000..6e545e26289e --- /dev/null +++ b/net/netfilter/nf_conntrack_snmp.c @@ -0,0 +1,77 @@ +/* + * SNMP service broadcast connection tracking helper + * + * (c) 2011 Jiri Olsa + * + * 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 + +#define SNMP_PORT 161 + +MODULE_AUTHOR("Jiri Olsa "); +MODULE_DESCRIPTION("SNMP service broadcast connection tracking helper"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS_NFCT_HELPER("snmp"); + +static unsigned int timeout __read_mostly = 30; +module_param(timeout, uint, S_IRUSR); +MODULE_PARM_DESC(timeout, "timeout for master connection/replies in seconds"); + +int (*nf_nat_snmp_hook)(struct sk_buff *skb, + unsigned int protoff, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo); +EXPORT_SYMBOL_GPL(nf_nat_snmp_hook); + +static int snmp_conntrack_help(struct sk_buff *skb, unsigned int protoff, + struct nf_conn *ct, enum ip_conntrack_info ctinfo) +{ + typeof(nf_nat_snmp_hook) nf_nat_snmp; + + nf_conntrack_broadcast_help(skb, protoff, ct, ctinfo, timeout); + + nf_nat_snmp = rcu_dereference(nf_nat_snmp_hook); + if (nf_nat_snmp && ct->status & IPS_NAT_MASK) + return nf_nat_snmp(skb, protoff, ct, ctinfo); + + return NF_ACCEPT; +} + +static struct nf_conntrack_expect_policy exp_policy = { + .max_expected = 1, +}; + +static struct nf_conntrack_helper helper __read_mostly = { + .name = "snmp", + .tuple.src.l3num = NFPROTO_IPV4, + .tuple.src.u.udp.port = cpu_to_be16(SNMP_PORT), + .tuple.dst.protonum = IPPROTO_UDP, + .me = THIS_MODULE, + .help = snmp_conntrack_help, + .expect_policy = &exp_policy, +}; + +static int __init nf_conntrack_snmp_init(void) +{ + exp_policy.timeout = timeout; + return nf_conntrack_helper_register(&helper); +} + +static void __exit nf_conntrack_snmp_fini(void) +{ + nf_conntrack_helper_unregister(&helper); +} + +module_init(nf_conntrack_snmp_init); +module_exit(nf_conntrack_snmp_fini); -- cgit v1.2.3 From a992ca2a0498edd22a88ac8c41570f536de29c9e Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 19 Jan 2011 16:00:07 +0100 Subject: netfilter: nf_conntrack_tstamp: add flow-based timestamp extension This patch adds flow-based timestamping for conntracks. This conntrack extension is disabled by default. Basically, we use two 64-bits variables to store the creation timestamp once the conntrack has been confirmed and the other to store the deletion time. This extension is disabled by default, to enable it, you have to: echo 1 > /proc/sys/net/netfilter/nf_conntrack_timestamp This patch allows to save memory for user-space flow-based loogers such as ulogd2. In short, ulogd2 does not need to keep a hashtable with the conntrack in user-space to know when they were created and destroyed, instead we use the kernel timestamp. If we want to have a sane IPFIX implementation in user-space, this nanosecs resolution timestamps are also useful. Other custom user-space applications can benefit from this via libnetfilter_conntrack. This patch modifies the /proc output to display the delta time in seconds since the flow start. You can also obtain the flow-start date by means of the conntrack-tools. Signed-off-by: Pablo Neira Ayuso Signed-off-by: Patrick McHardy --- net/netfilter/Kconfig | 11 +++ net/netfilter/Makefile | 1 + net/netfilter/nf_conntrack_core.c | 26 +++++++ net/netfilter/nf_conntrack_netlink.c | 46 +++++++++++- net/netfilter/nf_conntrack_standalone.c | 41 +++++++++++ net/netfilter/nf_conntrack_timestamp.c | 120 ++++++++++++++++++++++++++++++++ 6 files changed, 244 insertions(+), 1 deletion(-) create mode 100644 net/netfilter/nf_conntrack_timestamp.c (limited to 'net/netfilter') diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 939b504604c2..faf7412ea453 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -85,6 +85,17 @@ config NF_CONNTRACK_EVENTS If unsure, say `N'. +config NF_CONNTRACK_TIMESTAMP + bool 'Connection tracking timestamping' + depends on NETFILTER_ADVANCED + help + This option enables support for connection tracking timestamping. + This allows you to store the flow start-time and to obtain + the flow-stop time (once it has been destroyed) via Connection + tracking events. + + If unsure, say `N'. + config NF_CT_PROTO_DCCP tristate 'DCCP protocol connection tracking support (EXPERIMENTAL)' depends on EXPERIMENTAL diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 2c2628de9d3f..9ae6878a85b1 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -1,6 +1,7 @@ netfilter-objs := core.o nf_log.o nf_queue.o nf_sockopt.o nf_conntrack-y := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_expect.o nf_conntrack_helper.o nf_conntrack_proto.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o nf_conntrack_extend.o nf_conntrack_acct.o +nf_conntrack-$(CONFIG_NF_CONNTRACK_TIMESTAMP) += nf_conntrack_timestamp.o nf_conntrack-$(CONFIG_NF_CONNTRACK_EVENTS) += nf_conntrack_ecache.o obj-$(CONFIG_NETFILTER) = netfilter.o diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index f47ac67e1bfe..1909311c392a 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -43,6 +43,7 @@ #include #include #include +#include #include #include @@ -282,6 +283,11 @@ EXPORT_SYMBOL_GPL(nf_ct_insert_dying_list); static void death_by_timeout(unsigned long ul_conntrack) { struct nf_conn *ct = (void *)ul_conntrack; + struct nf_conn_tstamp *tstamp; + + tstamp = nf_conn_tstamp_find(ct); + if (tstamp && tstamp->stop == 0) + tstamp->stop = ktime_to_ns(ktime_get_real()); if (!test_bit(IPS_DYING_BIT, &ct->status) && unlikely(nf_conntrack_event(IPCT_DESTROY, ct) < 0)) { @@ -419,6 +425,7 @@ __nf_conntrack_confirm(struct sk_buff *skb) struct nf_conntrack_tuple_hash *h; struct nf_conn *ct; struct nf_conn_help *help; + struct nf_conn_tstamp *tstamp; struct hlist_nulls_node *n; enum ip_conntrack_info ctinfo; struct net *net; @@ -488,6 +495,14 @@ __nf_conntrack_confirm(struct sk_buff *skb) atomic_inc(&ct->ct_general.use); ct->status |= IPS_CONFIRMED; + /* set conntrack timestamp, if enabled. */ + tstamp = nf_conn_tstamp_find(ct); + if (tstamp) { + if (skb->tstamp.tv64 == 0) + __net_timestamp((struct sk_buff *)skb); + + tstamp->start = ktime_to_ns(skb->tstamp); + } /* Since the lookup is lockless, hash insertion must be done after * starting the timer and setting the CONFIRMED bit. The RCU barriers * guarantee that no other CPU can find the conntrack before the above @@ -746,6 +761,7 @@ init_conntrack(struct net *net, struct nf_conn *tmpl, } nf_ct_acct_ext_add(ct, GFP_ATOMIC); + nf_ct_tstamp_ext_add(ct, GFP_ATOMIC); ecache = tmpl ? nf_ct_ecache_find(tmpl) : NULL; nf_ct_ecache_ext_add(ct, ecache ? ecache->ctmask : 0, @@ -1186,6 +1202,11 @@ struct __nf_ct_flush_report { static int kill_report(struct nf_conn *i, void *data) { struct __nf_ct_flush_report *fr = (struct __nf_ct_flush_report *)data; + struct nf_conn_tstamp *tstamp; + + tstamp = nf_conn_tstamp_find(i); + if (tstamp && tstamp->stop == 0) + tstamp->stop = ktime_to_ns(ktime_get_real()); /* If we fail to deliver the event, death_by_timeout() will retry */ if (nf_conntrack_event_report(IPCT_DESTROY, i, @@ -1497,6 +1518,9 @@ static int nf_conntrack_init_net(struct net *net) ret = nf_conntrack_acct_init(net); if (ret < 0) goto err_acct; + ret = nf_conntrack_tstamp_init(net); + if (ret < 0) + goto err_tstamp; ret = nf_conntrack_ecache_init(net); if (ret < 0) goto err_ecache; @@ -1504,6 +1528,8 @@ static int nf_conntrack_init_net(struct net *net) return 0; err_ecache: + nf_conntrack_tstamp_fini(net); +err_tstamp: nf_conntrack_acct_fini(net); err_acct: nf_conntrack_expect_fini(net); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 9eabaa6f28a8..715d56c85475 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -42,6 +42,7 @@ #include #include #include +#include #ifdef CONFIG_NF_NAT_NEEDED #include #include @@ -230,6 +231,33 @@ nla_put_failure: return -1; } +static int +ctnetlink_dump_timestamp(struct sk_buff *skb, const struct nf_conn *ct) +{ + struct nlattr *nest_count; + const struct nf_conn_tstamp *tstamp; + + tstamp = nf_conn_tstamp_find(ct); + if (!tstamp) + return 0; + + nest_count = nla_nest_start(skb, CTA_TIMESTAMP | NLA_F_NESTED); + if (!nest_count) + goto nla_put_failure; + + NLA_PUT_BE64(skb, CTA_TIMESTAMP_START, cpu_to_be64(tstamp->start)); + if (tstamp->stop != 0) { + NLA_PUT_BE64(skb, CTA_TIMESTAMP_STOP, + cpu_to_be64(tstamp->stop)); + } + nla_nest_end(skb, nest_count); + + return 0; + +nla_put_failure: + return -1; +} + #ifdef CONFIG_NF_CONNTRACK_MARK static inline int ctnetlink_dump_mark(struct sk_buff *skb, const struct nf_conn *ct) @@ -404,6 +432,7 @@ ctnetlink_fill_info(struct sk_buff *skb, u32 pid, u32 seq, ctnetlink_dump_timeout(skb, ct) < 0 || ctnetlink_dump_counters(skb, ct, IP_CT_DIR_ORIGINAL) < 0 || ctnetlink_dump_counters(skb, ct, IP_CT_DIR_REPLY) < 0 || + ctnetlink_dump_timestamp(skb, ct) < 0 || ctnetlink_dump_protoinfo(skb, ct) < 0 || ctnetlink_dump_helpinfo(skb, ct) < 0 || ctnetlink_dump_mark(skb, ct) < 0 || @@ -470,6 +499,18 @@ ctnetlink_secctx_size(const struct nf_conn *ct) #endif } +static inline size_t +ctnetlink_timestamp_size(const struct nf_conn *ct) +{ +#ifdef CONFIG_NF_CONNTRACK_TIMESTAMP + if (!nf_ct_ext_exist(ct, NF_CT_EXT_TSTAMP)) + return 0; + return nla_total_size(0) + 2 * nla_total_size(sizeof(uint64_t)); +#else + return 0; +#endif +} + static inline size_t ctnetlink_nlmsg_size(const struct nf_conn *ct) { @@ -481,6 +522,7 @@ ctnetlink_nlmsg_size(const struct nf_conn *ct) + nla_total_size(sizeof(u_int32_t)) /* CTA_ID */ + nla_total_size(sizeof(u_int32_t)) /* CTA_STATUS */ + ctnetlink_counters_size(ct) + + ctnetlink_timestamp_size(ct) + nla_total_size(sizeof(u_int32_t)) /* CTA_TIMEOUT */ + nla_total_size(0) /* CTA_PROTOINFO */ + nla_total_size(0) /* CTA_HELP */ @@ -571,7 +613,8 @@ ctnetlink_conntrack_event(unsigned int events, struct nf_ct_event *item) if (events & (1 << IPCT_DESTROY)) { if (ctnetlink_dump_counters(skb, ct, IP_CT_DIR_ORIGINAL) < 0 || - ctnetlink_dump_counters(skb, ct, IP_CT_DIR_REPLY) < 0) + ctnetlink_dump_counters(skb, ct, IP_CT_DIR_REPLY) < 0 || + ctnetlink_dump_timestamp(skb, ct) < 0) goto nla_put_failure; } else { if (ctnetlink_dump_timeout(skb, ct) < 0) @@ -1360,6 +1403,7 @@ ctnetlink_create_conntrack(struct net *net, u16 zone, } nf_ct_acct_ext_add(ct, GFP_ATOMIC); + nf_ct_tstamp_ext_add(ct, GFP_ATOMIC); nf_ct_ecache_ext_add(ct, 0, 0, GFP_ATOMIC); /* we must add conntrack extensions before confirmation. */ ct->status |= IPS_CONFIRMED; diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 8257bf643593..69107fd78d3e 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -29,6 +29,7 @@ #include #include #include +#include #include MODULE_LICENSE("GPL"); @@ -46,6 +47,7 @@ EXPORT_SYMBOL_GPL(print_tuple); struct ct_iter_state { struct seq_net_private p; unsigned int bucket; + u_int64_t time_now; }; static struct hlist_nulls_node *ct_get_first(struct seq_file *seq) @@ -96,6 +98,9 @@ static struct hlist_nulls_node *ct_get_idx(struct seq_file *seq, loff_t pos) static void *ct_seq_start(struct seq_file *seq, loff_t *pos) __acquires(RCU) { + struct ct_iter_state *st = seq->private; + + st->time_now = ktime_to_ns(ktime_get_real()); rcu_read_lock(); return ct_get_idx(seq, *pos); } @@ -135,6 +140,39 @@ static inline int ct_show_secctx(struct seq_file *s, const struct nf_conn *ct) } #endif +#ifdef CONFIG_NF_CONNTRACK_TIMESTAMP +static u_int64_t ct_delta_time(u_int64_t time_now, const struct nf_conn *ct) +{ + struct nf_conn_tstamp *tstamp; + + tstamp = nf_conn_tstamp_find(ct); + if (tstamp) { + u_int64_t delta_time = time_now - tstamp->start; + return delta_time > 0 ? div_s64(delta_time, NSEC_PER_SEC) : 0; + } + return -1; +} + +static int ct_show_delta_time(struct seq_file *s, const struct nf_conn *ct) +{ + struct ct_iter_state *st = s->private; + u_int64_t delta_time; + + delta_time = ct_delta_time(st->time_now, ct); + if (delta_time < 0) + return 0; + + return seq_printf(s, "delta-time=%llu ", + (unsigned long long)delta_time); +} +#else +static inline int +ct_show_delta_time(struct seq_file *s, const struct nf_conn *ct) +{ + return 0; +} +#endif + /* return 0 on success, 1 in case of error */ static int ct_seq_show(struct seq_file *s, void *v) { @@ -203,6 +241,9 @@ static int ct_seq_show(struct seq_file *s, void *v) goto release; #endif + if (ct_show_delta_time(s, ct)) + goto release; + if (seq_printf(s, "use=%u\n", atomic_read(&ct->ct_general.use))) goto release; diff --git a/net/netfilter/nf_conntrack_timestamp.c b/net/netfilter/nf_conntrack_timestamp.c new file mode 100644 index 000000000000..af7dd31af0a1 --- /dev/null +++ b/net/netfilter/nf_conntrack_timestamp.c @@ -0,0 +1,120 @@ +/* + * (C) 2010 Pablo Neira Ayuso + * + * 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 (or any later at your option). + */ + +#include +#include +#include +#include + +#include +#include +#include + +static int nf_ct_tstamp __read_mostly; + +module_param_named(tstamp, nf_ct_tstamp, bool, 0644); +MODULE_PARM_DESC(tstamp, "Enable connection tracking flow timestamping."); + +#ifdef CONFIG_SYSCTL +static struct ctl_table tstamp_sysctl_table[] = { + { + .procname = "nf_conntrack_timestamp", + .data = &init_net.ct.sysctl_tstamp, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + {} +}; +#endif /* CONFIG_SYSCTL */ + +static struct nf_ct_ext_type tstamp_extend __read_mostly = { + .len = sizeof(struct nf_conn_tstamp), + .align = __alignof__(struct nf_conn_tstamp), + .id = NF_CT_EXT_TSTAMP, +}; + +#ifdef CONFIG_SYSCTL +static int nf_conntrack_tstamp_init_sysctl(struct net *net) +{ + struct ctl_table *table; + + table = kmemdup(tstamp_sysctl_table, sizeof(tstamp_sysctl_table), + GFP_KERNEL); + if (!table) + goto out; + + table[0].data = &net->ct.sysctl_tstamp; + + net->ct.tstamp_sysctl_header = register_net_sysctl_table(net, + nf_net_netfilter_sysctl_path, table); + if (!net->ct.tstamp_sysctl_header) { + printk(KERN_ERR "nf_ct_tstamp: can't register to sysctl.\n"); + goto out_register; + } + return 0; + +out_register: + kfree(table); +out: + return -ENOMEM; +} + +static void nf_conntrack_tstamp_fini_sysctl(struct net *net) +{ + struct ctl_table *table; + + table = net->ct.tstamp_sysctl_header->ctl_table_arg; + unregister_net_sysctl_table(net->ct.tstamp_sysctl_header); + kfree(table); +} +#else +static int nf_conntrack_tstamp_init_sysctl(struct net *net) +{ + return 0; +} + +static void nf_conntrack_tstamp_fini_sysctl(struct net *net) +{ +} +#endif + +int nf_conntrack_tstamp_init(struct net *net) +{ + int ret; + + net->ct.sysctl_tstamp = nf_ct_tstamp; + + if (net_eq(net, &init_net)) { + ret = nf_ct_extend_register(&tstamp_extend); + if (ret < 0) { + printk(KERN_ERR "nf_ct_tstamp: Unable to register " + "extension\n"); + goto out_extend_register; + } + } + + ret = nf_conntrack_tstamp_init_sysctl(net); + if (ret < 0) + goto out_sysctl; + + return 0; + +out_sysctl: + if (net_eq(net, &init_net)) + nf_ct_extend_unregister(&tstamp_extend); +out_extend_register: + return ret; +} + +void nf_conntrack_tstamp_fini(struct net *net) +{ + nf_conntrack_tstamp_fini_sysctl(net); + if (net_eq(net, &init_net)) + nf_ct_extend_unregister(&tstamp_extend); +} -- cgit v1.2.3 From cc4fc022571376412986e27e08b0765e9cb2aafb Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Tue, 18 Jan 2011 17:32:40 +0100 Subject: netfilter: xtables: connlimit revision 1 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This adds destination address-based selection. The old "inverse" member is overloaded (memory-wise) with a new "flags" variable, similar to how J.Park did it with xt_string rev 1. Since revision 0 userspace only sets flag 0x1, no great changes are made to explicitly test for different revisions. Signed-off-by: Jan Engelhardt --- net/netfilter/xt_connlimit.c | 44 ++++++++++++++++++++++++++++++-------------- 1 file changed, 30 insertions(+), 14 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c index 452bc16af56c..7fd3fd51f274 100644 --- a/net/netfilter/xt_connlimit.c +++ b/net/netfilter/xt_connlimit.c @@ -193,10 +193,12 @@ connlimit_mt(const struct sk_buff *skb, struct xt_action_param *par) if (par->family == NFPROTO_IPV6) { const struct ipv6hdr *iph = ipv6_hdr(skb); - memcpy(&addr.ip6, &iph->saddr, sizeof(iph->saddr)); + memcpy(&addr.ip6, (info->flags & XT_CONNLIMIT_DADDR) ? + &iph->daddr : &iph->saddr, sizeof(addr.ip6)); } else { const struct iphdr *iph = ip_hdr(skb); - addr.ip = iph->saddr; + addr.ip = (info->flags & XT_CONNLIMIT_DADDR) ? + iph->daddr : iph->saddr; } spin_lock_bh(&info->data->lock); @@ -208,7 +210,8 @@ connlimit_mt(const struct sk_buff *skb, struct xt_action_param *par) /* kmalloc failed, drop it entirely */ goto hotdrop; - return (connections > info->limit) ^ info->inverse; + return (connections > info->limit) ^ + !!(info->flags & XT_CONNLIMIT_INVERT); hotdrop: par->hotdrop = true; @@ -266,25 +269,38 @@ static void connlimit_mt_destroy(const struct xt_mtdtor_param *par) kfree(info->data); } -static struct xt_match connlimit_mt_reg __read_mostly = { - .name = "connlimit", - .revision = 0, - .family = NFPROTO_UNSPEC, - .checkentry = connlimit_mt_check, - .match = connlimit_mt, - .matchsize = sizeof(struct xt_connlimit_info), - .destroy = connlimit_mt_destroy, - .me = THIS_MODULE, +static struct xt_match connlimit_mt_reg[] __read_mostly = { + { + .name = "connlimit", + .revision = 0, + .family = NFPROTO_UNSPEC, + .checkentry = connlimit_mt_check, + .match = connlimit_mt, + .matchsize = sizeof(struct xt_connlimit_info), + .destroy = connlimit_mt_destroy, + .me = THIS_MODULE, + }, + { + .name = "connlimit", + .revision = 1, + .family = NFPROTO_UNSPEC, + .checkentry = connlimit_mt_check, + .match = connlimit_mt, + .matchsize = sizeof(struct xt_connlimit_info), + .destroy = connlimit_mt_destroy, + .me = THIS_MODULE, + }, }; static int __init connlimit_mt_init(void) { - return xt_register_match(&connlimit_mt_reg); + return xt_register_matches(connlimit_mt_reg, + ARRAY_SIZE(connlimit_mt_reg)); } static void __exit connlimit_mt_exit(void) { - xt_unregister_match(&connlimit_mt_reg); + xt_unregister_matches(connlimit_mt_reg, ARRAY_SIZE(connlimit_mt_reg)); } module_init(connlimit_mt_init); -- cgit v1.2.3 From f5c88f56b35599ab9ff2d3398e0153e4cd4a4c82 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 19 Jan 2011 19:10:49 +0100 Subject: netfilter: nf_conntrack: fix lifetime display for disabled connections When no tstamp extension exists, ct_delta_time() returns -1, which is then assigned to an u64 and tested for negative values to decide whether to display the lifetime. This obviously doesn't work, use a s64 and merge the two minor functions into one. Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_standalone.c | 29 ++++++++++++----------------- 1 file changed, 12 insertions(+), 17 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 69107fd78d3e..0ae142825881 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -141,29 +141,24 @@ static inline int ct_show_secctx(struct seq_file *s, const struct nf_conn *ct) #endif #ifdef CONFIG_NF_CONNTRACK_TIMESTAMP -static u_int64_t ct_delta_time(u_int64_t time_now, const struct nf_conn *ct) +static int ct_show_delta_time(struct seq_file *s, const struct nf_conn *ct) { + struct ct_iter_state *st = s->private; struct nf_conn_tstamp *tstamp; + s64 delta_time; tstamp = nf_conn_tstamp_find(ct); if (tstamp) { - u_int64_t delta_time = time_now - tstamp->start; - return delta_time > 0 ? div_s64(delta_time, NSEC_PER_SEC) : 0; + delta_time = st->time_now - tstamp->start; + if (delta_time > 0) + delta_time = div_s64(delta_time, NSEC_PER_SEC); + else + delta_time = 0; + + return seq_printf(s, "delta-time=%llu ", + (unsigned long long)delta_time); } - return -1; -} - -static int ct_show_delta_time(struct seq_file *s, const struct nf_conn *ct) -{ - struct ct_iter_state *st = s->private; - u_int64_t delta_time; - - delta_time = ct_delta_time(st->time_now, ct); - if (delta_time < 0) - return 0; - - return seq_printf(s, "delta-time=%llu ", - (unsigned long long)delta_time); + return 0; } #else static inline int -- cgit v1.2.3 From 091bb34c143674d37a59b2d4857534f7106c5d7d Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Fri, 21 Jan 2011 18:02:13 +0800 Subject: netfilter: ipvs: fix compiler warnings Fix compiler warnings when no transport protocol load balancing support is configured. [horms@verge.net.au: removed suprious __ip_vs_cleanup() clean-up hunk] Signed-off-by: Changli Gao Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_ctl.c | 4 ++++ net/netfilter/ipvs/ip_vs_proto.c | 4 ++++ 2 files changed, 8 insertions(+) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 09ca2ce2f2b7..68b8033a96f7 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -2062,7 +2062,9 @@ static const struct file_operations ip_vs_stats_percpu_fops = { */ static int ip_vs_set_timeout(struct net *net, struct ip_vs_timeout_user *u) { +#if defined(CONFIG_IP_VS_PROTO_TCP) || defined(CONFIG_IP_VS_PROTO_UDP) struct ip_vs_proto_data *pd; +#endif IP_VS_DBG(2, "Setting timeout tcp:%d tcpfin:%d udp:%d\n", u->tcp_timeout, @@ -2405,7 +2407,9 @@ __ip_vs_get_dest_entries(struct net *net, const struct ip_vs_get_dests *get, static inline void __ip_vs_get_timeouts(struct net *net, struct ip_vs_timeout_user *u) { +#if defined(CONFIG_IP_VS_PROTO_TCP) || defined(CONFIG_IP_VS_PROTO_UDP) struct ip_vs_proto_data *pd; +#endif #ifdef CONFIG_IP_VS_PROTO_TCP pd = ip_vs_proto_data_get(net, IPPROTO_TCP); diff --git a/net/netfilter/ipvs/ip_vs_proto.c b/net/netfilter/ipvs/ip_vs_proto.c index 6ac986cdcff3..17484a4416ef 100644 --- a/net/netfilter/ipvs/ip_vs_proto.c +++ b/net/netfilter/ipvs/ip_vs_proto.c @@ -60,6 +60,9 @@ static int __used __init register_ip_vs_protocol(struct ip_vs_protocol *pp) return 0; } +#if defined(CONFIG_IP_VS_PROTO_TCP) || defined(CONFIG_IP_VS_PROTO_UDP) || \ + defined(CONFIG_IP_VS_PROTO_SCTP) || defined(CONFIG_IP_VS_PROTO_AH) || \ + defined(CONFIG_IP_VS_PROTO_ESP) /* * register an ipvs protocols netns related data */ @@ -85,6 +88,7 @@ register_ip_vs_proto_netns(struct net *net, struct ip_vs_protocol *pp) return 0; } +#endif /* * unregister an ipvs protocol -- cgit v1.2.3 From 4b3fd57138c969dd940651fadf90db627254edbf Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Sat, 22 Jan 2011 13:48:01 +1100 Subject: IPVS: Change sock_create_kernel() to __sock_create() The recent netns changes omitted to change sock_create_kernel() to __sock_create() in ip_vs_sync.c The effect of this is that the interface will be selected in the root-namespace, from my point of view it's a major bug. Reported-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_sync.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index d1adf988eb08..d5a6e640ea45 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -1305,7 +1305,7 @@ static struct socket *make_send_sock(struct net *net) int result; /* First create a socket */ - result = sock_create_kern(PF_INET, SOCK_DGRAM, IPPROTO_UDP, &sock); + result = __sock_create(net, PF_INET, SOCK_DGRAM, IPPROTO_UDP, &sock, 1); if (result < 0) { pr_err("Error during creation of socket; terminating\n"); return ERR_PTR(result); @@ -1351,7 +1351,7 @@ static struct socket *make_receive_sock(struct net *net) int result; /* First create a socket */ - result = sock_create_kern(PF_INET, SOCK_DGRAM, IPPROTO_UDP, &sock); + result = __sock_create(net, PF_INET, SOCK_DGRAM, IPPROTO_UDP, &sock, 1); if (result < 0) { pr_err("Error during creation of socket; terminating\n"); return ERR_PTR(result); -- cgit v1.2.3 From 07924709f68b3f4f701d4efd6acd18ca4ee14de3 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 24 Jan 2011 15:14:41 +0100 Subject: IPVS netns BUG, register sysctl for root ns The newly created table was not used when register sysctl for a new namespace. I.e. sysctl doesn't work for other than root namespace (init_net) Signed-off-by: Hans Schillstrom Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_ctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 68b8033a96f7..98df59a12453 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -3556,7 +3556,7 @@ int __net_init __ip_vs_control_init(struct net *net) ipvs->sysctl_hdr = register_net_sysctl_table(net, net_vs_ctl_path, - vs_vars); + tbl); if (ipvs->sysctl_hdr == NULL) goto err_reg; ip_vs_new_estimator(net, ipvs->tot_stats); -- cgit v1.2.3 From 9f4e1ccd80530609bbceec68ae3831697b5c6a68 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Tue, 25 Jan 2011 12:40:18 +0800 Subject: netfilter: ipvs: fix compiler warnings Fix compiler warnings when IP_VS_DBG() isn't defined. Signed-off-by: Changli Gao Acked-by: Hans Schillstrom Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index f36a84f33efb..d889f4f6be99 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -1894,9 +1894,7 @@ static int __net_init __ip_vs_init(struct net *net) static void __net_exit __ip_vs_cleanup(struct net *net) { - struct netns_ipvs *ipvs = net_ipvs(net); - - IP_VS_DBG(10, "ipvs netns %d released\n", ipvs->gen); + IP_VS_DBG(10, "ipvs netns %d released\n", net_ipvs(net)->gen); } static struct pernet_operations ipvs_core_ops = { -- cgit v1.2.3 From ad86e1f27a9a97a9e50810b10bca678407b1d6fd Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Wed, 26 Jan 2011 11:50:03 +0100 Subject: netfilter: xt_connlimit: pick right dstaddr in NAT scenario MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit xt_connlimit normally records the "original" tuples in a hashlist (such as "1.2.3.4 -> 5.6.7.8"), and looks in this list for iph->daddr when counting. When the user however uses DNAT in PREROUTING, looking for iph->daddr -- which is now 192.168.9.10 -- will not match. Thus in daddr mode, we need to record the reverse direction tuple ("192.168.9.10 -> 1.2.3.4") instead. In the reverse tuple, the dst addr is on the src side, which is convenient, as count_them still uses &conn->tuple.src.u3. Signed-off-by: Jan Engelhardt --- net/netfilter/xt_connlimit.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c index 7fd3fd51f274..e029c4807404 100644 --- a/net/netfilter/xt_connlimit.c +++ b/net/netfilter/xt_connlimit.c @@ -185,11 +185,15 @@ connlimit_mt(const struct sk_buff *skb, struct xt_action_param *par) int connections; ct = nf_ct_get(skb, &ctinfo); - if (ct != NULL) - tuple_ptr = &ct->tuplehash[0].tuple; - else if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), - par->family, &tuple)) + if (ct != NULL) { + if (info->flags & XT_CONNLIMIT_DADDR) + tuple_ptr = &ct->tuplehash[IP_CT_DIR_REPLY].tuple; + else + tuple_ptr = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; + } else if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), + par->family, &tuple)) { goto hotdrop; + } if (par->family == NFPROTO_IPV6) { const struct ipv6hdr *iph = ipv6_hdr(skb); -- cgit v1.2.3 From 705ca147176090203afd7503392e6e770637499b Mon Sep 17 00:00:00 2001 From: Thomas Jacob Date: Thu, 27 Jan 2011 10:56:32 +0100 Subject: netfilter: xt_iprange: typo in IPv4 match debug print code Signed-off-by: Thomas Jacob Signed-off-by: Patrick McHardy --- net/netfilter/xt_iprange.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_iprange.c b/net/netfilter/xt_iprange.c index 88f7c3511c72..77b9ebcc90b6 100644 --- a/net/netfilter/xt_iprange.c +++ b/net/netfilter/xt_iprange.c @@ -31,7 +31,7 @@ iprange_mt4(const struct sk_buff *skb, struct xt_action_param *par) pr_debug("src IP %pI4 NOT in range %s%pI4-%pI4\n", &iph->saddr, (info->flags & IPRANGE_SRC_INV) ? "(INV) " : "", - &info->src_max.ip, + &info->src_min.ip, &info->src_max.ip); return false; } -- cgit v1.2.3 From 6a4ddef2a3805d5b0664a94579b7a651bc202266 Mon Sep 17 00:00:00 2001 From: Thomas Jacob Date: Fri, 28 Jan 2011 19:33:13 +0100 Subject: netfilter: xt_iprange: add IPv6 match debug print code Signed-off-by: Thomas Jacob Signed-off-by: Patrick McHardy --- net/netfilter/xt_iprange.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_iprange.c b/net/netfilter/xt_iprange.c index 77b9ebcc90b6..d3eb5ed1892f 100644 --- a/net/netfilter/xt_iprange.c +++ b/net/netfilter/xt_iprange.c @@ -78,15 +78,27 @@ iprange_mt6(const struct sk_buff *skb, struct xt_action_param *par) m = iprange_ipv6_sub(&iph->saddr, &info->src_min.in6) < 0; m |= iprange_ipv6_sub(&iph->saddr, &info->src_max.in6) > 0; m ^= !!(info->flags & IPRANGE_SRC_INV); - if (m) + if (m) { + pr_debug("src IP %pI6 NOT in range %s%pI6-%pI6\n", + &iph->saddr, + (info->flags & IPRANGE_SRC_INV) ? "(INV) " : "", + &info->src_min.in6, + &info->src_max.in6); return false; + } } if (info->flags & IPRANGE_DST) { m = iprange_ipv6_sub(&iph->daddr, &info->dst_min.in6) < 0; m |= iprange_ipv6_sub(&iph->daddr, &info->dst_max.in6) > 0; m ^= !!(info->flags & IPRANGE_DST_INV); - if (m) + if (m) { + pr_debug("dst IP %pI6 NOT in range %s%pI6-%pI6\n", + &iph->daddr, + (info->flags & IPRANGE_DST_INV) ? "(INV) " : "", + &info->dst_min.in6, + &info->dst_max.in6); return false; + } } return true; } -- cgit v1.2.3 From a7b4f989a629493bb4ec4a354def784d440b32c4 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 1 Feb 2011 15:28:35 +0100 Subject: netfilter: ipset: IP set core support The patch adds the IP set core support to the kernel. The IP set core implements a netlink (nfnetlink) based protocol by which one can create, destroy, flush, rename, swap, list, save, restore sets, and add, delete, test elements from userspace. For simplicity (and backward compatibilty and for not to force ip(6)tables to be linked with a netlink library) reasons a small getsockopt-based protocol is also kept in order to communicate with the ip(6)tables match and target. The netlink protocol passes all u16, etc values in network order with NLA_F_NET_BYTEORDER flag. The protocol enforces the proper use of the NLA_F_NESTED and NLA_F_NET_BYTEORDER flags. For other kernel subsystems (netfilter match and target) the API contains the functions to add, delete and test elements in sets and the required calls to get/put refereces to the sets before those operations can be performed. The set types (which are implemented in independent modules) are stored in a simple RCU protected list. A set type may have variants: for example without timeout or with timeout support, for IPv4 or for IPv6. The sets (i.e. the pointers to the sets) are stored in an array. The sets are identified by their index in the array, which makes possible easy and fast swapping of sets. The array is protected indirectly by the nfnl mutex from nfnetlink. The content of the sets are protected by the rwlock of the set. There are functional differences between the add/del/test functions for the kernel and userspace: - kernel add/del/test: works on the current packet (i.e. one element) - kernel test: may trigger an "add" operation in order to fill out unspecified parts of the element from the packet (like MAC address) - userspace add/del: works on the netlink message and thus possibly on multiple elements from the IPSET_ATTR_ADT container attribute. - userspace add: may trigger resizing of a set Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/Kconfig | 2 + net/netfilter/Makefile | 3 + net/netfilter/ipset/Kconfig | 26 + net/netfilter/ipset/Makefile | 8 + net/netfilter/ipset/ip_set_core.c | 1662 ++++++++++++++++++++++++++++++++++ net/netfilter/ipset/ip_set_getport.c | 136 +++ net/netfilter/ipset/pfxlen.c | 291 ++++++ 7 files changed, 2128 insertions(+) create mode 100644 net/netfilter/ipset/Kconfig create mode 100644 net/netfilter/ipset/Makefile create mode 100644 net/netfilter/ipset/ip_set_core.c create mode 100644 net/netfilter/ipset/ip_set_getport.c create mode 100644 net/netfilter/ipset/pfxlen.c (limited to 'net/netfilter') diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index faf7412ea453..351abf8ace13 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -1052,4 +1052,6 @@ endif # NETFILTER_XTABLES endmenu +source "net/netfilter/ipset/Kconfig" + source "net/netfilter/ipvs/Kconfig" diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 9ae6878a85b1..510b586ccb7f 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -105,5 +105,8 @@ obj-$(CONFIG_NETFILTER_XT_MATCH_TCPMSS) += xt_tcpmss.o obj-$(CONFIG_NETFILTER_XT_MATCH_TIME) += xt_time.o obj-$(CONFIG_NETFILTER_XT_MATCH_U32) += xt_u32.o +# ipset +obj-$(CONFIG_IP_SET) += ipset/ + # IPVS obj-$(CONFIG_IP_VS) += ipvs/ diff --git a/net/netfilter/ipset/Kconfig b/net/netfilter/ipset/Kconfig new file mode 100644 index 000000000000..5ade156dd470 --- /dev/null +++ b/net/netfilter/ipset/Kconfig @@ -0,0 +1,26 @@ +menuconfig IP_SET + tristate "IP set support" + depends on INET && NETFILTER + help + This option adds IP set support to the kernel. + In order to define and use the sets, you need the userspace utility + ipset(8). You can use the sets in netfilter via the "set" match + and "SET" target. + + To compile it as a module, choose M here. If unsure, say N. + +if IP_SET + +config IP_SET_MAX + int "Maximum number of IP sets" + default 256 + range 2 65534 + depends on IP_SET + help + You can define here default value of the maximum number + of IP sets for the kernel. + + The value can be overriden by the 'max_sets' module + parameter of the 'ip_set' module. + +endif # IP_SET diff --git a/net/netfilter/ipset/Makefile b/net/netfilter/ipset/Makefile new file mode 100644 index 000000000000..910cd425c067 --- /dev/null +++ b/net/netfilter/ipset/Makefile @@ -0,0 +1,8 @@ +# +# Makefile for the ipset modules +# + +ip_set-y := ip_set_core.o ip_set_getport.o pfxlen.o + +# ipset core +obj-$(CONFIG_IP_SET) += ip_set.o diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c new file mode 100644 index 000000000000..8a736247e85f --- /dev/null +++ b/net/netfilter/ipset/ip_set_core.c @@ -0,0 +1,1662 @@ +/* Copyright (C) 2000-2002 Joakim Axelsson + * Patrick Schaaf + * Copyright (C) 2003-2011 Jozsef Kadlecsik + * + * 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. + */ + +/* Kernel module for IP set management */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +static LIST_HEAD(ip_set_type_list); /* all registered set types */ +static DEFINE_MUTEX(ip_set_type_mutex); /* protects ip_set_type_list */ + +static struct ip_set **ip_set_list; /* all individual sets */ +static ip_set_id_t ip_set_max = CONFIG_IP_SET_MAX; /* max number of sets */ + +#define STREQ(a, b) (strncmp(a, b, IPSET_MAXNAMELEN) == 0) + +static unsigned int max_sets; + +module_param(max_sets, int, 0600); +MODULE_PARM_DESC(max_sets, "maximal number of sets"); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Jozsef Kadlecsik "); +MODULE_DESCRIPTION("core IP set support"); +MODULE_ALIAS_NFNL_SUBSYS(NFNL_SUBSYS_IPSET); + +/* + * The set types are implemented in modules and registered set types + * can be found in ip_set_type_list. Adding/deleting types is + * serialized by ip_set_type_mutex. + */ + +static inline void +ip_set_type_lock(void) +{ + mutex_lock(&ip_set_type_mutex); +} + +static inline void +ip_set_type_unlock(void) +{ + mutex_unlock(&ip_set_type_mutex); +} + +/* Register and deregister settype */ + +static struct ip_set_type * +find_set_type(const char *name, u8 family, u8 revision) +{ + struct ip_set_type *type; + + list_for_each_entry_rcu(type, &ip_set_type_list, list) + if (STREQ(type->name, name) && + (type->family == family || type->family == AF_UNSPEC) && + type->revision == revision) + return type; + return NULL; +} + +/* Unlock, try to load a set type module and lock again */ +static int +try_to_load_type(const char *name) +{ + nfnl_unlock(); + pr_debug("try to load ip_set_%s\n", name); + if (request_module("ip_set_%s", name) < 0) { + pr_warning("Can't find ip_set type %s\n", name); + nfnl_lock(); + return -IPSET_ERR_FIND_TYPE; + } + nfnl_lock(); + return -EAGAIN; +} + +/* Find a set type and reference it */ +static int +find_set_type_get(const char *name, u8 family, u8 revision, + struct ip_set_type **found) +{ + rcu_read_lock(); + *found = find_set_type(name, family, revision); + if (*found) { + int err = !try_module_get((*found)->me); + rcu_read_unlock(); + return err ? -EFAULT : 0; + } + rcu_read_unlock(); + + return try_to_load_type(name); +} + +/* Find a given set type by name and family. + * If we succeeded, the supported minimal and maximum revisions are + * filled out. + */ +static int +find_set_type_minmax(const char *name, u8 family, u8 *min, u8 *max) +{ + struct ip_set_type *type; + bool found = false; + + *min = *max = 0; + rcu_read_lock(); + list_for_each_entry_rcu(type, &ip_set_type_list, list) + if (STREQ(type->name, name) && + (type->family == family || type->family == AF_UNSPEC)) { + found = true; + if (type->revision < *min) + *min = type->revision; + else if (type->revision > *max) + *max = type->revision; + } + rcu_read_unlock(); + if (found) + return 0; + + return try_to_load_type(name); +} + +#define family_name(f) ((f) == AF_INET ? "inet" : \ + (f) == AF_INET6 ? "inet6" : "any") + +/* Register a set type structure. The type is identified by + * the unique triple of name, family and revision. + */ +int +ip_set_type_register(struct ip_set_type *type) +{ + int ret = 0; + + if (type->protocol != IPSET_PROTOCOL) { + pr_warning("ip_set type %s, family %s, revision %u uses " + "wrong protocol version %u (want %u)\n", + type->name, family_name(type->family), + type->revision, type->protocol, IPSET_PROTOCOL); + return -EINVAL; + } + + ip_set_type_lock(); + if (find_set_type(type->name, type->family, type->revision)) { + /* Duplicate! */ + pr_warning("ip_set type %s, family %s, revision %u " + "already registered!\n", type->name, + family_name(type->family), type->revision); + ret = -EINVAL; + goto unlock; + } + list_add_rcu(&type->list, &ip_set_type_list); + pr_debug("type %s, family %s, revision %u registered.\n", + type->name, family_name(type->family), type->revision); +unlock: + ip_set_type_unlock(); + return ret; +} +EXPORT_SYMBOL_GPL(ip_set_type_register); + +/* Unregister a set type. There's a small race with ip_set_create */ +void +ip_set_type_unregister(struct ip_set_type *type) +{ + ip_set_type_lock(); + if (!find_set_type(type->name, type->family, type->revision)) { + pr_warning("ip_set type %s, family %s, revision %u " + "not registered\n", type->name, + family_name(type->family), type->revision); + goto unlock; + } + list_del_rcu(&type->list); + pr_debug("type %s, family %s, revision %u unregistered.\n", + type->name, family_name(type->family), type->revision); +unlock: + ip_set_type_unlock(); + + synchronize_rcu(); +} +EXPORT_SYMBOL_GPL(ip_set_type_unregister); + +/* Utility functions */ +void * +ip_set_alloc(size_t size) +{ + void *members = NULL; + + if (size < KMALLOC_MAX_SIZE) + members = kzalloc(size, GFP_KERNEL | __GFP_NOWARN); + + if (members) { + pr_debug("%p: allocated with kmalloc\n", members); + return members; + } + + members = vzalloc(size); + if (!members) + return NULL; + pr_debug("%p: allocated with vmalloc\n", members); + + return members; +} +EXPORT_SYMBOL_GPL(ip_set_alloc); + +void +ip_set_free(void *members) +{ + pr_debug("%p: free with %s\n", members, + is_vmalloc_addr(members) ? "vfree" : "kfree"); + if (is_vmalloc_addr(members)) + vfree(members); + else + kfree(members); +} +EXPORT_SYMBOL_GPL(ip_set_free); + +static inline bool +flag_nested(const struct nlattr *nla) +{ + return nla->nla_type & NLA_F_NESTED; +} + +static const struct nla_policy ipaddr_policy[IPSET_ATTR_IPADDR_MAX + 1] = { + [IPSET_ATTR_IPADDR_IPV4] = { .type = NLA_U32 }, + [IPSET_ATTR_IPADDR_IPV6] = { .type = NLA_BINARY, + .len = sizeof(struct in6_addr) }, +}; + +int +ip_set_get_ipaddr4(struct nlattr *nla, __be32 *ipaddr) +{ + struct nlattr *tb[IPSET_ATTR_IPADDR_MAX+1]; + + if (unlikely(!flag_nested(nla))) + return -IPSET_ERR_PROTOCOL; + if (nla_parse(tb, IPSET_ATTR_IPADDR_MAX, nla_data(nla), nla_len(nla), + ipaddr_policy)) + return -IPSET_ERR_PROTOCOL; + if (unlikely(!ip_set_attr_netorder(tb, IPSET_ATTR_IPADDR_IPV4))) + return -IPSET_ERR_PROTOCOL; + + *ipaddr = nla_get_be32(tb[IPSET_ATTR_IPADDR_IPV4]); + return 0; +} +EXPORT_SYMBOL_GPL(ip_set_get_ipaddr4); + +int +ip_set_get_ipaddr6(struct nlattr *nla, union nf_inet_addr *ipaddr) +{ + struct nlattr *tb[IPSET_ATTR_IPADDR_MAX+1]; + + if (unlikely(!flag_nested(nla))) + return -IPSET_ERR_PROTOCOL; + + if (nla_parse(tb, IPSET_ATTR_IPADDR_MAX, nla_data(nla), nla_len(nla), + ipaddr_policy)) + return -IPSET_ERR_PROTOCOL; + if (unlikely(!ip_set_attr_netorder(tb, IPSET_ATTR_IPADDR_IPV6))) + return -IPSET_ERR_PROTOCOL; + + memcpy(ipaddr, nla_data(tb[IPSET_ATTR_IPADDR_IPV6]), + sizeof(struct in6_addr)); + return 0; +} +EXPORT_SYMBOL_GPL(ip_set_get_ipaddr6); + +/* + * Creating/destroying/renaming/swapping affect the existence and + * the properties of a set. All of these can be executed from userspace + * only and serialized by the nfnl mutex indirectly from nfnetlink. + * + * Sets are identified by their index in ip_set_list and the index + * is used by the external references (set/SET netfilter modules). + * + * The set behind an index may change by swapping only, from userspace. + */ + +static inline void +__ip_set_get(ip_set_id_t index) +{ + atomic_inc(&ip_set_list[index]->ref); +} + +static inline void +__ip_set_put(ip_set_id_t index) +{ + atomic_dec(&ip_set_list[index]->ref); +} + +/* + * Add, del and test set entries from kernel. + * + * The set behind the index must exist and must be referenced + * so it can't be destroyed (or changed) under our foot. + */ + +int +ip_set_test(ip_set_id_t index, const struct sk_buff *skb, + u8 family, u8 dim, u8 flags) +{ + struct ip_set *set = ip_set_list[index]; + int ret = 0; + + BUG_ON(set == NULL || atomic_read(&set->ref) == 0); + pr_debug("set %s, index %u\n", set->name, index); + + if (dim < set->type->dimension || + !(family == set->family || set->family == AF_UNSPEC)) + return 0; + + read_lock_bh(&set->lock); + ret = set->variant->kadt(set, skb, IPSET_TEST, family, dim, flags); + read_unlock_bh(&set->lock); + + if (ret == -EAGAIN) { + /* Type requests element to be completed */ + pr_debug("element must be competed, ADD is triggered\n"); + write_lock_bh(&set->lock); + set->variant->kadt(set, skb, IPSET_ADD, family, dim, flags); + write_unlock_bh(&set->lock); + ret = 1; + } + + /* Convert error codes to nomatch */ + return (ret < 0 ? 0 : ret); +} +EXPORT_SYMBOL_GPL(ip_set_test); + +int +ip_set_add(ip_set_id_t index, const struct sk_buff *skb, + u8 family, u8 dim, u8 flags) +{ + struct ip_set *set = ip_set_list[index]; + int ret; + + BUG_ON(set == NULL || atomic_read(&set->ref) == 0); + pr_debug("set %s, index %u\n", set->name, index); + + if (dim < set->type->dimension || + !(family == set->family || set->family == AF_UNSPEC)) + return 0; + + write_lock_bh(&set->lock); + ret = set->variant->kadt(set, skb, IPSET_ADD, family, dim, flags); + write_unlock_bh(&set->lock); + + return ret; +} +EXPORT_SYMBOL_GPL(ip_set_add); + +int +ip_set_del(ip_set_id_t index, const struct sk_buff *skb, + u8 family, u8 dim, u8 flags) +{ + struct ip_set *set = ip_set_list[index]; + int ret = 0; + + BUG_ON(set == NULL || atomic_read(&set->ref) == 0); + pr_debug("set %s, index %u\n", set->name, index); + + if (dim < set->type->dimension || + !(family == set->family || set->family == AF_UNSPEC)) + return 0; + + write_lock_bh(&set->lock); + ret = set->variant->kadt(set, skb, IPSET_DEL, family, dim, flags); + write_unlock_bh(&set->lock); + + return ret; +} +EXPORT_SYMBOL_GPL(ip_set_del); + +/* + * Find set by name, reference it once. The reference makes sure the + * thing pointed to, does not go away under our feet. + * + * The nfnl mutex must already be activated. + */ +ip_set_id_t +ip_set_get_byname(const char *name, struct ip_set **set) +{ + ip_set_id_t i, index = IPSET_INVALID_ID; + struct ip_set *s; + + for (i = 0; i < ip_set_max; i++) { + s = ip_set_list[i]; + if (s != NULL && STREQ(s->name, name)) { + __ip_set_get(i); + index = i; + *set = s; + } + } + + return index; +} +EXPORT_SYMBOL_GPL(ip_set_get_byname); + +/* + * If the given set pointer points to a valid set, decrement + * reference count by 1. The caller shall not assume the index + * to be valid, after calling this function. + * + * The nfnl mutex must already be activated. + */ +void +ip_set_put_byindex(ip_set_id_t index) +{ + if (ip_set_list[index] != NULL) { + BUG_ON(atomic_read(&ip_set_list[index]->ref) == 0); + __ip_set_put(index); + } +} +EXPORT_SYMBOL_GPL(ip_set_put_byindex); + +/* + * Get the name of a set behind a set index. + * We assume the set is referenced, so it does exist and + * can't be destroyed. The set cannot be renamed due to + * the referencing either. + * + * The nfnl mutex must already be activated. + */ +const char * +ip_set_name_byindex(ip_set_id_t index) +{ + const struct ip_set *set = ip_set_list[index]; + + BUG_ON(set == NULL); + BUG_ON(atomic_read(&set->ref) == 0); + + /* Referenced, so it's safe */ + return set->name; +} +EXPORT_SYMBOL_GPL(ip_set_name_byindex); + +/* + * Routines to call by external subsystems, which do not + * call nfnl_lock for us. + */ + +/* + * Find set by name, reference it once. The reference makes sure the + * thing pointed to, does not go away under our feet. + * + * The nfnl mutex is used in the function. + */ +ip_set_id_t +ip_set_nfnl_get(const char *name) +{ + struct ip_set *s; + ip_set_id_t index; + + nfnl_lock(); + index = ip_set_get_byname(name, &s); + nfnl_unlock(); + + return index; +} +EXPORT_SYMBOL_GPL(ip_set_nfnl_get); + +/* + * Find set by index, reference it once. The reference makes sure the + * thing pointed to, does not go away under our feet. + * + * The nfnl mutex is used in the function. + */ +ip_set_id_t +ip_set_nfnl_get_byindex(ip_set_id_t index) +{ + if (index > ip_set_max) + return IPSET_INVALID_ID; + + nfnl_lock(); + if (ip_set_list[index]) + __ip_set_get(index); + else + index = IPSET_INVALID_ID; + nfnl_unlock(); + + return index; +} +EXPORT_SYMBOL_GPL(ip_set_nfnl_get_byindex); + +/* + * If the given set pointer points to a valid set, decrement + * reference count by 1. The caller shall not assume the index + * to be valid, after calling this function. + * + * The nfnl mutex is used in the function. + */ +void +ip_set_nfnl_put(ip_set_id_t index) +{ + nfnl_lock(); + if (ip_set_list[index] != NULL) { + BUG_ON(atomic_read(&ip_set_list[index]->ref) == 0); + __ip_set_put(index); + } + nfnl_unlock(); +} +EXPORT_SYMBOL_GPL(ip_set_nfnl_put); + +/* + * Communication protocol with userspace over netlink. + * + * We already locked by nfnl_lock. + */ + +static inline bool +protocol_failed(const struct nlattr * const tb[]) +{ + return !tb[IPSET_ATTR_PROTOCOL] || + nla_get_u8(tb[IPSET_ATTR_PROTOCOL]) != IPSET_PROTOCOL; +} + +static inline u32 +flag_exist(const struct nlmsghdr *nlh) +{ + return nlh->nlmsg_flags & NLM_F_EXCL ? 0 : IPSET_FLAG_EXIST; +} + +static struct nlmsghdr * +start_msg(struct sk_buff *skb, u32 pid, u32 seq, unsigned int flags, + enum ipset_cmd cmd) +{ + struct nlmsghdr *nlh; + struct nfgenmsg *nfmsg; + + nlh = nlmsg_put(skb, pid, seq, cmd | (NFNL_SUBSYS_IPSET << 8), + sizeof(*nfmsg), flags); + if (nlh == NULL) + return NULL; + + nfmsg = nlmsg_data(nlh); + nfmsg->nfgen_family = AF_INET; + nfmsg->version = NFNETLINK_V0; + nfmsg->res_id = 0; + + return nlh; +} + +/* Create a set */ + +static const struct nla_policy ip_set_create_policy[IPSET_ATTR_CMD_MAX + 1] = { + [IPSET_ATTR_PROTOCOL] = { .type = NLA_U8 }, + [IPSET_ATTR_SETNAME] = { .type = NLA_NUL_STRING, + .len = IPSET_MAXNAMELEN - 1 }, + [IPSET_ATTR_TYPENAME] = { .type = NLA_NUL_STRING, + .len = IPSET_MAXNAMELEN - 1}, + [IPSET_ATTR_REVISION] = { .type = NLA_U8 }, + [IPSET_ATTR_FAMILY] = { .type = NLA_U8 }, + [IPSET_ATTR_DATA] = { .type = NLA_NESTED }, +}; + +static ip_set_id_t +find_set_id(const char *name) +{ + ip_set_id_t i, index = IPSET_INVALID_ID; + const struct ip_set *set; + + for (i = 0; index == IPSET_INVALID_ID && i < ip_set_max; i++) { + set = ip_set_list[i]; + if (set != NULL && STREQ(set->name, name)) + index = i; + } + return index; +} + +static inline struct ip_set * +find_set(const char *name) +{ + ip_set_id_t index = find_set_id(name); + + return index == IPSET_INVALID_ID ? NULL : ip_set_list[index]; +} + +static int +find_free_id(const char *name, ip_set_id_t *index, struct ip_set **set) +{ + ip_set_id_t i; + + *index = IPSET_INVALID_ID; + for (i = 0; i < ip_set_max; i++) { + if (ip_set_list[i] == NULL) { + if (*index == IPSET_INVALID_ID) + *index = i; + } else if (STREQ(name, ip_set_list[i]->name)) { + /* Name clash */ + *set = ip_set_list[i]; + return -EEXIST; + } + } + if (*index == IPSET_INVALID_ID) + /* No free slot remained */ + return -IPSET_ERR_MAX_SETS; + return 0; +} + +static int +ip_set_create(struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) +{ + struct ip_set *set, *clash; + ip_set_id_t index = IPSET_INVALID_ID; + struct nlattr *tb[IPSET_ATTR_CREATE_MAX+1] = {}; + const char *name, *typename; + u8 family, revision; + u32 flags = flag_exist(nlh); + int ret = 0; + + if (unlikely(protocol_failed(attr) || + attr[IPSET_ATTR_SETNAME] == NULL || + attr[IPSET_ATTR_TYPENAME] == NULL || + attr[IPSET_ATTR_REVISION] == NULL || + attr[IPSET_ATTR_FAMILY] == NULL || + (attr[IPSET_ATTR_DATA] != NULL && + !flag_nested(attr[IPSET_ATTR_DATA])))) + return -IPSET_ERR_PROTOCOL; + + name = nla_data(attr[IPSET_ATTR_SETNAME]); + typename = nla_data(attr[IPSET_ATTR_TYPENAME]); + family = nla_get_u8(attr[IPSET_ATTR_FAMILY]); + revision = nla_get_u8(attr[IPSET_ATTR_REVISION]); + pr_debug("setname: %s, typename: %s, family: %s, revision: %u\n", + name, typename, family_name(family), revision); + + /* + * First, and without any locks, allocate and initialize + * a normal base set structure. + */ + set = kzalloc(sizeof(struct ip_set), GFP_KERNEL); + if (!set) + return -ENOMEM; + rwlock_init(&set->lock); + strlcpy(set->name, name, IPSET_MAXNAMELEN); + atomic_set(&set->ref, 0); + set->family = family; + + /* + * Next, check that we know the type, and take + * a reference on the type, to make sure it stays available + * while constructing our new set. + * + * After referencing the type, we try to create the type + * specific part of the set without holding any locks. + */ + ret = find_set_type_get(typename, family, revision, &(set->type)); + if (ret) + goto out; + + /* + * Without holding any locks, create private part. + */ + if (attr[IPSET_ATTR_DATA] && + nla_parse(tb, IPSET_ATTR_CREATE_MAX, + nla_data(attr[IPSET_ATTR_DATA]), + nla_len(attr[IPSET_ATTR_DATA]), + set->type->create_policy)) { + ret = -IPSET_ERR_PROTOCOL; + goto put_out; + } + + ret = set->type->create(set, tb, flags); + if (ret != 0) + goto put_out; + + /* BTW, ret==0 here. */ + + /* + * Here, we have a valid, constructed set and we are protected + * by nfnl_lock. Find the first free index in ip_set_list and + * check clashing. + */ + if ((ret = find_free_id(set->name, &index, &clash)) != 0) { + /* If this is the same set and requested, ignore error */ + if (ret == -EEXIST && + (flags & IPSET_FLAG_EXIST) && + STREQ(set->type->name, clash->type->name) && + set->type->family == clash->type->family && + set->type->revision == clash->type->revision && + set->variant->same_set(set, clash)) + ret = 0; + goto cleanup; + } + + /* + * Finally! Add our shiny new set to the list, and be done. + */ + pr_debug("create: '%s' created with index %u!\n", set->name, index); + ip_set_list[index] = set; + + return ret; + +cleanup: + set->variant->destroy(set); +put_out: + module_put(set->type->me); +out: + kfree(set); + return ret; +} + +/* Destroy sets */ + +static const struct nla_policy +ip_set_setname_policy[IPSET_ATTR_CMD_MAX + 1] = { + [IPSET_ATTR_PROTOCOL] = { .type = NLA_U8 }, + [IPSET_ATTR_SETNAME] = { .type = NLA_NUL_STRING, + .len = IPSET_MAXNAMELEN - 1 }, +}; + +static void +ip_set_destroy_set(ip_set_id_t index) +{ + struct ip_set *set = ip_set_list[index]; + + pr_debug("set: %s\n", set->name); + ip_set_list[index] = NULL; + + /* Must call it without holding any lock */ + set->variant->destroy(set); + module_put(set->type->me); + kfree(set); +} + +static int +ip_set_destroy(struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) +{ + ip_set_id_t i; + + if (unlikely(protocol_failed(attr))) + return -IPSET_ERR_PROTOCOL; + + /* References are protected by the nfnl mutex */ + if (!attr[IPSET_ATTR_SETNAME]) { + for (i = 0; i < ip_set_max; i++) { + if (ip_set_list[i] != NULL && + (atomic_read(&ip_set_list[i]->ref))) + return -IPSET_ERR_BUSY; + } + for (i = 0; i < ip_set_max; i++) { + if (ip_set_list[i] != NULL) + ip_set_destroy_set(i); + } + } else { + i = find_set_id(nla_data(attr[IPSET_ATTR_SETNAME])); + if (i == IPSET_INVALID_ID) + return -ENOENT; + else if (atomic_read(&ip_set_list[i]->ref)) + return -IPSET_ERR_BUSY; + + ip_set_destroy_set(i); + } + return 0; +} + +/* Flush sets */ + +static void +ip_set_flush_set(struct ip_set *set) +{ + pr_debug("set: %s\n", set->name); + + write_lock_bh(&set->lock); + set->variant->flush(set); + write_unlock_bh(&set->lock); +} + +static int +ip_set_flush(struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) +{ + ip_set_id_t i; + + if (unlikely(protocol_failed(attr))) + return -EPROTO; + + if (!attr[IPSET_ATTR_SETNAME]) { + for (i = 0; i < ip_set_max; i++) + if (ip_set_list[i] != NULL) + ip_set_flush_set(ip_set_list[i]); + } else { + i = find_set_id(nla_data(attr[IPSET_ATTR_SETNAME])); + if (i == IPSET_INVALID_ID) + return -ENOENT; + + ip_set_flush_set(ip_set_list[i]); + } + + return 0; +} + +/* Rename a set */ + +static const struct nla_policy +ip_set_setname2_policy[IPSET_ATTR_CMD_MAX + 1] = { + [IPSET_ATTR_PROTOCOL] = { .type = NLA_U8 }, + [IPSET_ATTR_SETNAME] = { .type = NLA_NUL_STRING, + .len = IPSET_MAXNAMELEN - 1 }, + [IPSET_ATTR_SETNAME2] = { .type = NLA_NUL_STRING, + .len = IPSET_MAXNAMELEN - 1 }, +}; + +static int +ip_set_rename(struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) +{ + struct ip_set *set; + const char *name2; + ip_set_id_t i; + + if (unlikely(protocol_failed(attr) || + attr[IPSET_ATTR_SETNAME] == NULL || + attr[IPSET_ATTR_SETNAME2] == NULL)) + return -IPSET_ERR_PROTOCOL; + + set = find_set(nla_data(attr[IPSET_ATTR_SETNAME])); + if (set == NULL) + return -ENOENT; + if (atomic_read(&set->ref) != 0) + return -IPSET_ERR_REFERENCED; + + name2 = nla_data(attr[IPSET_ATTR_SETNAME2]); + for (i = 0; i < ip_set_max; i++) { + if (ip_set_list[i] != NULL && + STREQ(ip_set_list[i]->name, name2)) + return -IPSET_ERR_EXIST_SETNAME2; + } + strncpy(set->name, name2, IPSET_MAXNAMELEN); + + return 0; +} + +/* Swap two sets so that name/index points to the other. + * References and set names are also swapped. + * + * We are protected by the nfnl mutex and references are + * manipulated only by holding the mutex. The kernel interfaces + * do not hold the mutex but the pointer settings are atomic + * so the ip_set_list always contains valid pointers to the sets. + */ + +static int +ip_set_swap(struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) +{ + struct ip_set *from, *to; + ip_set_id_t from_id, to_id; + char from_name[IPSET_MAXNAMELEN]; + u32 from_ref; + + if (unlikely(protocol_failed(attr) || + attr[IPSET_ATTR_SETNAME] == NULL || + attr[IPSET_ATTR_SETNAME2] == NULL)) + return -IPSET_ERR_PROTOCOL; + + from_id = find_set_id(nla_data(attr[IPSET_ATTR_SETNAME])); + if (from_id == IPSET_INVALID_ID) + return -ENOENT; + + to_id = find_set_id(nla_data(attr[IPSET_ATTR_SETNAME2])); + if (to_id == IPSET_INVALID_ID) + return -IPSET_ERR_EXIST_SETNAME2; + + from = ip_set_list[from_id]; + to = ip_set_list[to_id]; + + /* Features must not change. + * Not an artifical restriction anymore, as we must prevent + * possible loops created by swapping in setlist type of sets. */ + if (!(from->type->features == to->type->features && + from->type->family == to->type->family)) + return -IPSET_ERR_TYPE_MISMATCH; + + /* No magic here: ref munging protected by the nfnl_lock */ + strncpy(from_name, from->name, IPSET_MAXNAMELEN); + from_ref = atomic_read(&from->ref); + + strncpy(from->name, to->name, IPSET_MAXNAMELEN); + atomic_set(&from->ref, atomic_read(&to->ref)); + strncpy(to->name, from_name, IPSET_MAXNAMELEN); + atomic_set(&to->ref, from_ref); + + ip_set_list[from_id] = to; + ip_set_list[to_id] = from; + + return 0; +} + +/* List/save set data */ + +#define DUMP_INIT 0L +#define DUMP_ALL 1L +#define DUMP_ONE 2L +#define DUMP_LAST 3L + +static int +ip_set_dump_done(struct netlink_callback *cb) +{ + if (cb->args[2]) { + pr_debug("release set %s\n", ip_set_list[cb->args[1]]->name); + __ip_set_put((ip_set_id_t) cb->args[1]); + } + return 0; +} + +static inline void +dump_attrs(struct nlmsghdr *nlh) +{ + const struct nlattr *attr; + int rem; + + pr_debug("dump nlmsg\n"); + nlmsg_for_each_attr(attr, nlh, sizeof(struct nfgenmsg), rem) { + pr_debug("type: %u, len %u\n", nla_type(attr), attr->nla_len); + } +} + +static int +dump_init(struct netlink_callback *cb) +{ + struct nlmsghdr *nlh = nlmsg_hdr(cb->skb); + int min_len = NLMSG_SPACE(sizeof(struct nfgenmsg)); + struct nlattr *cda[IPSET_ATTR_CMD_MAX+1]; + struct nlattr *attr = (void *)nlh + min_len; + ip_set_id_t index; + + /* Second pass, so parser can't fail */ + nla_parse(cda, IPSET_ATTR_CMD_MAX, + attr, nlh->nlmsg_len - min_len, ip_set_setname_policy); + + /* cb->args[0] : dump single set/all sets + * [1] : set index + * [..]: type specific + */ + + if (!cda[IPSET_ATTR_SETNAME]) { + cb->args[0] = DUMP_ALL; + return 0; + } + + index = find_set_id(nla_data(cda[IPSET_ATTR_SETNAME])); + if (index == IPSET_INVALID_ID) + return -ENOENT; + + cb->args[0] = DUMP_ONE; + cb->args[1] = index; + return 0; +} + +static int +ip_set_dump_start(struct sk_buff *skb, struct netlink_callback *cb) +{ + ip_set_id_t index = IPSET_INVALID_ID, max; + struct ip_set *set = NULL; + struct nlmsghdr *nlh = NULL; + unsigned int flags = NETLINK_CB(cb->skb).pid ? NLM_F_MULTI : 0; + int ret = 0; + + if (cb->args[0] == DUMP_INIT) { + ret = dump_init(cb); + if (ret < 0) { + nlh = nlmsg_hdr(cb->skb); + /* We have to create and send the error message + * manually :-( */ + if (nlh->nlmsg_flags & NLM_F_ACK) + netlink_ack(cb->skb, nlh, ret); + return ret; + } + } + + if (cb->args[1] >= ip_set_max) + goto out; + + pr_debug("args[0]: %ld args[1]: %ld\n", cb->args[0], cb->args[1]); + max = cb->args[0] == DUMP_ONE ? cb->args[1] + 1 : ip_set_max; + for (; cb->args[1] < max; cb->args[1]++) { + index = (ip_set_id_t) cb->args[1]; + set = ip_set_list[index]; + if (set == NULL) { + if (cb->args[0] == DUMP_ONE) { + ret = -ENOENT; + goto out; + } + continue; + } + /* When dumping all sets, we must dump "sorted" + * so that lists (unions of sets) are dumped last. + */ + if (cb->args[0] != DUMP_ONE && + !((cb->args[0] == DUMP_ALL) ^ + (set->type->features & IPSET_DUMP_LAST))) + continue; + pr_debug("List set: %s\n", set->name); + if (!cb->args[2]) { + /* Start listing: make sure set won't be destroyed */ + pr_debug("reference set\n"); + __ip_set_get(index); + } + nlh = start_msg(skb, NETLINK_CB(cb->skb).pid, + cb->nlh->nlmsg_seq, flags, + IPSET_CMD_LIST); + if (!nlh) { + ret = -EMSGSIZE; + goto release_refcount; + } + NLA_PUT_U8(skb, IPSET_ATTR_PROTOCOL, IPSET_PROTOCOL); + NLA_PUT_STRING(skb, IPSET_ATTR_SETNAME, set->name); + switch (cb->args[2]) { + case 0: + /* Core header data */ + NLA_PUT_STRING(skb, IPSET_ATTR_TYPENAME, + set->type->name); + NLA_PUT_U8(skb, IPSET_ATTR_FAMILY, + set->family); + NLA_PUT_U8(skb, IPSET_ATTR_REVISION, + set->type->revision); + ret = set->variant->head(set, skb); + if (ret < 0) + goto release_refcount; + /* Fall through and add elements */ + default: + read_lock_bh(&set->lock); + ret = set->variant->list(set, skb, cb); + read_unlock_bh(&set->lock); + if (!cb->args[2]) { + /* Set is done, proceed with next one */ + if (cb->args[0] == DUMP_ONE) + cb->args[1] = IPSET_INVALID_ID; + else + cb->args[1]++; + } + goto release_refcount; + } + } + goto out; + +nla_put_failure: + ret = -EFAULT; +release_refcount: + /* If there was an error or set is done, release set */ + if (ret || !cb->args[2]) { + pr_debug("release set %s\n", ip_set_list[index]->name); + __ip_set_put(index); + } + + /* If we dump all sets, continue with dumping last ones */ + if (cb->args[0] == DUMP_ALL && cb->args[1] >= max && !cb->args[2]) + cb->args[0] = DUMP_LAST; + +out: + if (nlh) { + nlmsg_end(skb, nlh); + pr_debug("nlmsg_len: %u\n", nlh->nlmsg_len); + dump_attrs(nlh); + } + + return ret < 0 ? ret : skb->len; +} + +static int +ip_set_dump(struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) +{ + if (unlikely(protocol_failed(attr))) + return -IPSET_ERR_PROTOCOL; + + return netlink_dump_start(ctnl, skb, nlh, + ip_set_dump_start, + ip_set_dump_done); +} + +/* Add, del and test */ + +static const struct nla_policy ip_set_adt_policy[IPSET_ATTR_CMD_MAX + 1] = { + [IPSET_ATTR_PROTOCOL] = { .type = NLA_U8 }, + [IPSET_ATTR_SETNAME] = { .type = NLA_NUL_STRING, + .len = IPSET_MAXNAMELEN - 1 }, + [IPSET_ATTR_LINENO] = { .type = NLA_U32 }, + [IPSET_ATTR_DATA] = { .type = NLA_NESTED }, + [IPSET_ATTR_ADT] = { .type = NLA_NESTED }, +}; + +static int +call_ad(struct sk_buff *skb, struct ip_set *set, + struct nlattr *tb[], enum ipset_adt adt, + u32 flags, bool use_lineno) +{ + int ret, retried = 0; + u32 lineno = 0; + bool eexist = flags & IPSET_FLAG_EXIST; + + do { + write_lock_bh(&set->lock); + ret = set->variant->uadt(set, tb, adt, &lineno, flags); + write_unlock_bh(&set->lock); + } while (ret == -EAGAIN && + set->variant->resize && + (ret = set->variant->resize(set, retried++)) == 0); + + if (!ret || (ret == -IPSET_ERR_EXIST && eexist)) + return 0; + if (lineno && use_lineno) { + /* Error in restore/batch mode: send back lineno */ + struct nlmsghdr *nlh = nlmsg_hdr(skb); + int min_len = NLMSG_SPACE(sizeof(struct nfgenmsg)); + struct nlattr *cda[IPSET_ATTR_CMD_MAX+1]; + struct nlattr *cmdattr = (void *)nlh + min_len; + u32 *errline; + + nla_parse(cda, IPSET_ATTR_CMD_MAX, + cmdattr, nlh->nlmsg_len - min_len, + ip_set_adt_policy); + + errline = nla_data(cda[IPSET_ATTR_LINENO]); + + *errline = lineno; + } + + return ret; +} + +static int +ip_set_uadd(struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) +{ + struct ip_set *set; + struct nlattr *tb[IPSET_ATTR_ADT_MAX+1] = {}; + const struct nlattr *nla; + u32 flags = flag_exist(nlh); + bool use_lineno; + int ret = 0; + + if (unlikely(protocol_failed(attr) || + attr[IPSET_ATTR_SETNAME] == NULL || + !((attr[IPSET_ATTR_DATA] != NULL) ^ + (attr[IPSET_ATTR_ADT] != NULL)) || + (attr[IPSET_ATTR_DATA] != NULL && + !flag_nested(attr[IPSET_ATTR_DATA])) || + (attr[IPSET_ATTR_ADT] != NULL && + (!flag_nested(attr[IPSET_ATTR_ADT]) || + attr[IPSET_ATTR_LINENO] == NULL)))) + return -IPSET_ERR_PROTOCOL; + + set = find_set(nla_data(attr[IPSET_ATTR_SETNAME])); + if (set == NULL) + return -ENOENT; + + use_lineno = !!attr[IPSET_ATTR_LINENO]; + if (attr[IPSET_ATTR_DATA]) { + if (nla_parse(tb, IPSET_ATTR_ADT_MAX, + nla_data(attr[IPSET_ATTR_DATA]), + nla_len(attr[IPSET_ATTR_DATA]), + set->type->adt_policy)) + return -IPSET_ERR_PROTOCOL; + ret = call_ad(skb, set, tb, IPSET_ADD, flags, use_lineno); + } else { + int nla_rem; + + nla_for_each_nested(nla, attr[IPSET_ATTR_ADT], nla_rem) { + memset(tb, 0, sizeof(tb)); + if (nla_type(nla) != IPSET_ATTR_DATA || + !flag_nested(nla) || + nla_parse(tb, IPSET_ATTR_ADT_MAX, + nla_data(nla), nla_len(nla), + set->type->adt_policy)) + return -IPSET_ERR_PROTOCOL; + ret = call_ad(skb, set, tb, IPSET_ADD, + flags, use_lineno); + if (ret < 0) + return ret; + } + } + return ret; +} + +static int +ip_set_udel(struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) +{ + struct ip_set *set; + struct nlattr *tb[IPSET_ATTR_ADT_MAX+1] = {}; + const struct nlattr *nla; + u32 flags = flag_exist(nlh); + bool use_lineno; + int ret = 0; + + if (unlikely(protocol_failed(attr) || + attr[IPSET_ATTR_SETNAME] == NULL || + !((attr[IPSET_ATTR_DATA] != NULL) ^ + (attr[IPSET_ATTR_ADT] != NULL)) || + (attr[IPSET_ATTR_DATA] != NULL && + !flag_nested(attr[IPSET_ATTR_DATA])) || + (attr[IPSET_ATTR_ADT] != NULL && + (!flag_nested(attr[IPSET_ATTR_ADT]) || + attr[IPSET_ATTR_LINENO] == NULL)))) + return -IPSET_ERR_PROTOCOL; + + set = find_set(nla_data(attr[IPSET_ATTR_SETNAME])); + if (set == NULL) + return -ENOENT; + + use_lineno = !!attr[IPSET_ATTR_LINENO]; + if (attr[IPSET_ATTR_DATA]) { + if (nla_parse(tb, IPSET_ATTR_ADT_MAX, + nla_data(attr[IPSET_ATTR_DATA]), + nla_len(attr[IPSET_ATTR_DATA]), + set->type->adt_policy)) + return -IPSET_ERR_PROTOCOL; + ret = call_ad(skb, set, tb, IPSET_DEL, flags, use_lineno); + } else { + int nla_rem; + + nla_for_each_nested(nla, attr[IPSET_ATTR_ADT], nla_rem) { + memset(tb, 0, sizeof(*tb)); + if (nla_type(nla) != IPSET_ATTR_DATA || + !flag_nested(nla) || + nla_parse(tb, IPSET_ATTR_ADT_MAX, + nla_data(nla), nla_len(nla), + set->type->adt_policy)) + return -IPSET_ERR_PROTOCOL; + ret = call_ad(skb, set, tb, IPSET_DEL, + flags, use_lineno); + if (ret < 0) + return ret; + } + } + return ret; +} + +static int +ip_set_utest(struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) +{ + struct ip_set *set; + struct nlattr *tb[IPSET_ATTR_ADT_MAX+1] = {}; + int ret = 0; + + if (unlikely(protocol_failed(attr) || + attr[IPSET_ATTR_SETNAME] == NULL || + attr[IPSET_ATTR_DATA] == NULL || + !flag_nested(attr[IPSET_ATTR_DATA]))) + return -IPSET_ERR_PROTOCOL; + + set = find_set(nla_data(attr[IPSET_ATTR_SETNAME])); + if (set == NULL) + return -ENOENT; + + if (nla_parse(tb, IPSET_ATTR_ADT_MAX, + nla_data(attr[IPSET_ATTR_DATA]), + nla_len(attr[IPSET_ATTR_DATA]), + set->type->adt_policy)) + return -IPSET_ERR_PROTOCOL; + + read_lock_bh(&set->lock); + ret = set->variant->uadt(set, tb, IPSET_TEST, NULL, 0); + read_unlock_bh(&set->lock); + /* Userspace can't trigger element to be re-added */ + if (ret == -EAGAIN) + ret = 1; + + return ret < 0 ? ret : ret > 0 ? 0 : -IPSET_ERR_EXIST; +} + +/* Get headed data of a set */ + +static int +ip_set_header(struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) +{ + const struct ip_set *set; + struct sk_buff *skb2; + struct nlmsghdr *nlh2; + ip_set_id_t index; + int ret = 0; + + if (unlikely(protocol_failed(attr) || + attr[IPSET_ATTR_SETNAME] == NULL)) + return -IPSET_ERR_PROTOCOL; + + index = find_set_id(nla_data(attr[IPSET_ATTR_SETNAME])); + if (index == IPSET_INVALID_ID) + return -ENOENT; + set = ip_set_list[index]; + + skb2 = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (skb2 == NULL) + return -ENOMEM; + + nlh2 = start_msg(skb2, NETLINK_CB(skb).pid, nlh->nlmsg_seq, 0, + IPSET_CMD_HEADER); + if (!nlh2) + goto nlmsg_failure; + NLA_PUT_U8(skb2, IPSET_ATTR_PROTOCOL, IPSET_PROTOCOL); + NLA_PUT_STRING(skb2, IPSET_ATTR_SETNAME, set->name); + NLA_PUT_STRING(skb2, IPSET_ATTR_TYPENAME, set->type->name); + NLA_PUT_U8(skb2, IPSET_ATTR_FAMILY, set->family); + NLA_PUT_U8(skb2, IPSET_ATTR_REVISION, set->type->revision); + nlmsg_end(skb2, nlh2); + + ret = netlink_unicast(ctnl, skb2, NETLINK_CB(skb).pid, MSG_DONTWAIT); + if (ret < 0) + return ret; + + return 0; + +nla_put_failure: + nlmsg_cancel(skb2, nlh2); +nlmsg_failure: + kfree_skb(skb2); + return -EMSGSIZE; +} + +/* Get type data */ + +static const struct nla_policy ip_set_type_policy[IPSET_ATTR_CMD_MAX + 1] = { + [IPSET_ATTR_PROTOCOL] = { .type = NLA_U8 }, + [IPSET_ATTR_TYPENAME] = { .type = NLA_NUL_STRING, + .len = IPSET_MAXNAMELEN - 1 }, + [IPSET_ATTR_FAMILY] = { .type = NLA_U8 }, +}; + +static int +ip_set_type(struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) +{ + struct sk_buff *skb2; + struct nlmsghdr *nlh2; + u8 family, min, max; + const char *typename; + int ret = 0; + + if (unlikely(protocol_failed(attr) || + attr[IPSET_ATTR_TYPENAME] == NULL || + attr[IPSET_ATTR_FAMILY] == NULL)) + return -IPSET_ERR_PROTOCOL; + + family = nla_get_u8(attr[IPSET_ATTR_FAMILY]); + typename = nla_data(attr[IPSET_ATTR_TYPENAME]); + ret = find_set_type_minmax(typename, family, &min, &max); + if (ret) + return ret; + + skb2 = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (skb2 == NULL) + return -ENOMEM; + + nlh2 = start_msg(skb2, NETLINK_CB(skb).pid, nlh->nlmsg_seq, 0, + IPSET_CMD_TYPE); + if (!nlh2) + goto nlmsg_failure; + NLA_PUT_U8(skb2, IPSET_ATTR_PROTOCOL, IPSET_PROTOCOL); + NLA_PUT_STRING(skb2, IPSET_ATTR_TYPENAME, typename); + NLA_PUT_U8(skb2, IPSET_ATTR_FAMILY, family); + NLA_PUT_U8(skb2, IPSET_ATTR_REVISION, max); + NLA_PUT_U8(skb2, IPSET_ATTR_REVISION_MIN, min); + nlmsg_end(skb2, nlh2); + + pr_debug("Send TYPE, nlmsg_len: %u\n", nlh2->nlmsg_len); + ret = netlink_unicast(ctnl, skb2, NETLINK_CB(skb).pid, MSG_DONTWAIT); + if (ret < 0) + return ret; + + return 0; + +nla_put_failure: + nlmsg_cancel(skb2, nlh2); +nlmsg_failure: + kfree_skb(skb2); + return -EMSGSIZE; +} + +/* Get protocol version */ + +static const struct nla_policy +ip_set_protocol_policy[IPSET_ATTR_CMD_MAX + 1] = { + [IPSET_ATTR_PROTOCOL] = { .type = NLA_U8 }, +}; + +static int +ip_set_protocol(struct sock *ctnl, struct sk_buff *skb, + const struct nlmsghdr *nlh, + const struct nlattr * const attr[]) +{ + struct sk_buff *skb2; + struct nlmsghdr *nlh2; + int ret = 0; + + if (unlikely(attr[IPSET_ATTR_PROTOCOL] == NULL)) + return -IPSET_ERR_PROTOCOL; + + skb2 = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (skb2 == NULL) + return -ENOMEM; + + nlh2 = start_msg(skb2, NETLINK_CB(skb).pid, nlh->nlmsg_seq, 0, + IPSET_CMD_PROTOCOL); + if (!nlh2) + goto nlmsg_failure; + NLA_PUT_U8(skb2, IPSET_ATTR_PROTOCOL, IPSET_PROTOCOL); + nlmsg_end(skb2, nlh2); + + ret = netlink_unicast(ctnl, skb2, NETLINK_CB(skb).pid, MSG_DONTWAIT); + if (ret < 0) + return ret; + + return 0; + +nla_put_failure: + nlmsg_cancel(skb2, nlh2); +nlmsg_failure: + kfree_skb(skb2); + return -EMSGSIZE; +} + +static const struct nfnl_callback ip_set_netlink_subsys_cb[IPSET_MSG_MAX] = { + [IPSET_CMD_CREATE] = { + .call = ip_set_create, + .attr_count = IPSET_ATTR_CMD_MAX, + .policy = ip_set_create_policy, + }, + [IPSET_CMD_DESTROY] = { + .call = ip_set_destroy, + .attr_count = IPSET_ATTR_CMD_MAX, + .policy = ip_set_setname_policy, + }, + [IPSET_CMD_FLUSH] = { + .call = ip_set_flush, + .attr_count = IPSET_ATTR_CMD_MAX, + .policy = ip_set_setname_policy, + }, + [IPSET_CMD_RENAME] = { + .call = ip_set_rename, + .attr_count = IPSET_ATTR_CMD_MAX, + .policy = ip_set_setname2_policy, + }, + [IPSET_CMD_SWAP] = { + .call = ip_set_swap, + .attr_count = IPSET_ATTR_CMD_MAX, + .policy = ip_set_setname2_policy, + }, + [IPSET_CMD_LIST] = { + .call = ip_set_dump, + .attr_count = IPSET_ATTR_CMD_MAX, + .policy = ip_set_setname_policy, + }, + [IPSET_CMD_SAVE] = { + .call = ip_set_dump, + .attr_count = IPSET_ATTR_CMD_MAX, + .policy = ip_set_setname_policy, + }, + [IPSET_CMD_ADD] = { + .call = ip_set_uadd, + .attr_count = IPSET_ATTR_CMD_MAX, + .policy = ip_set_adt_policy, + }, + [IPSET_CMD_DEL] = { + .call = ip_set_udel, + .attr_count = IPSET_ATTR_CMD_MAX, + .policy = ip_set_adt_policy, + }, + [IPSET_CMD_TEST] = { + .call = ip_set_utest, + .attr_count = IPSET_ATTR_CMD_MAX, + .policy = ip_set_adt_policy, + }, + [IPSET_CMD_HEADER] = { + .call = ip_set_header, + .attr_count = IPSET_ATTR_CMD_MAX, + .policy = ip_set_setname_policy, + }, + [IPSET_CMD_TYPE] = { + .call = ip_set_type, + .attr_count = IPSET_ATTR_CMD_MAX, + .policy = ip_set_type_policy, + }, + [IPSET_CMD_PROTOCOL] = { + .call = ip_set_protocol, + .attr_count = IPSET_ATTR_CMD_MAX, + .policy = ip_set_protocol_policy, + }, +}; + +static struct nfnetlink_subsystem ip_set_netlink_subsys __read_mostly = { + .name = "ip_set", + .subsys_id = NFNL_SUBSYS_IPSET, + .cb_count = IPSET_MSG_MAX, + .cb = ip_set_netlink_subsys_cb, +}; + +/* Interface to iptables/ip6tables */ + +static int +ip_set_sockfn_get(struct sock *sk, int optval, void __user *user, int *len) +{ + unsigned *op; + void *data; + int copylen = *len, ret = 0; + + if (!capable(CAP_NET_ADMIN)) + return -EPERM; + if (optval != SO_IP_SET) + return -EBADF; + if (*len < sizeof(unsigned)) + return -EINVAL; + + data = vmalloc(*len); + if (!data) + return -ENOMEM; + if (copy_from_user(data, user, *len) != 0) { + ret = -EFAULT; + goto done; + } + op = (unsigned *) data; + + if (*op < IP_SET_OP_VERSION) { + /* Check the version at the beginning of operations */ + struct ip_set_req_version *req_version = data; + if (req_version->version != IPSET_PROTOCOL) { + ret = -EPROTO; + goto done; + } + } + + switch (*op) { + case IP_SET_OP_VERSION: { + struct ip_set_req_version *req_version = data; + + if (*len != sizeof(struct ip_set_req_version)) { + ret = -EINVAL; + goto done; + } + + req_version->version = IPSET_PROTOCOL; + ret = copy_to_user(user, req_version, + sizeof(struct ip_set_req_version)); + goto done; + } + case IP_SET_OP_GET_BYNAME: { + struct ip_set_req_get_set *req_get = data; + + if (*len != sizeof(struct ip_set_req_get_set)) { + ret = -EINVAL; + goto done; + } + req_get->set.name[IPSET_MAXNAMELEN - 1] = '\0'; + nfnl_lock(); + req_get->set.index = find_set_id(req_get->set.name); + nfnl_unlock(); + goto copy; + } + case IP_SET_OP_GET_BYINDEX: { + struct ip_set_req_get_set *req_get = data; + + if (*len != sizeof(struct ip_set_req_get_set) || + req_get->set.index >= ip_set_max) { + ret = -EINVAL; + goto done; + } + nfnl_lock(); + strncpy(req_get->set.name, + ip_set_list[req_get->set.index] + ? ip_set_list[req_get->set.index]->name : "", + IPSET_MAXNAMELEN); + nfnl_unlock(); + goto copy; + } + default: + ret = -EBADMSG; + goto done; + } /* end of switch(op) */ + +copy: + ret = copy_to_user(user, data, copylen); + +done: + vfree(data); + if (ret > 0) + ret = 0; + return ret; +} + +static struct nf_sockopt_ops so_set __read_mostly = { + .pf = PF_INET, + .get_optmin = SO_IP_SET, + .get_optmax = SO_IP_SET + 1, + .get = &ip_set_sockfn_get, + .owner = THIS_MODULE, +}; + +static int __init +ip_set_init(void) +{ + int ret; + + if (max_sets) + ip_set_max = max_sets; + if (ip_set_max >= IPSET_INVALID_ID) + ip_set_max = IPSET_INVALID_ID - 1; + + ip_set_list = kzalloc(sizeof(struct ip_set *) * ip_set_max, + GFP_KERNEL); + if (!ip_set_list) { + pr_err("ip_set: Unable to create ip_set_list\n"); + return -ENOMEM; + } + + ret = nfnetlink_subsys_register(&ip_set_netlink_subsys); + if (ret != 0) { + pr_err("ip_set: cannot register with nfnetlink.\n"); + kfree(ip_set_list); + return ret; + } + ret = nf_register_sockopt(&so_set); + if (ret != 0) { + pr_err("SO_SET registry failed: %d\n", ret); + nfnetlink_subsys_unregister(&ip_set_netlink_subsys); + kfree(ip_set_list); + return ret; + } + + pr_notice("ip_set: protocol %u\n", IPSET_PROTOCOL); + return 0; +} + +static void __exit +ip_set_fini(void) +{ + /* There can't be any existing set */ + nf_unregister_sockopt(&so_set); + nfnetlink_subsys_unregister(&ip_set_netlink_subsys); + kfree(ip_set_list); + pr_debug("these are the famous last words\n"); +} + +module_init(ip_set_init); +module_exit(ip_set_fini); diff --git a/net/netfilter/ipset/ip_set_getport.c b/net/netfilter/ipset/ip_set_getport.c new file mode 100644 index 000000000000..76737bba28ee --- /dev/null +++ b/net/netfilter/ipset/ip_set_getport.c @@ -0,0 +1,136 @@ +/* Copyright (C) 2003-2011 Jozsef Kadlecsik + * + * 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. + */ + +/* Get Layer-4 data from the packets */ + +#include +#include +#include +#include +#include +#include + +#include + +/* We must handle non-linear skbs */ +static bool +get_port(const struct sk_buff *skb, int protocol, unsigned int protooff, + bool src, __be16 *port, u8 *proto) +{ + switch (protocol) { + case IPPROTO_TCP: { + struct tcphdr _tcph; + const struct tcphdr *th; + + th = skb_header_pointer(skb, protooff, sizeof(_tcph), &_tcph); + if (th == NULL) + /* No choice either */ + return false; + + *port = src ? th->source : th->dest; + break; + } + case IPPROTO_UDP: { + struct udphdr _udph; + const struct udphdr *uh; + + uh = skb_header_pointer(skb, protooff, sizeof(_udph), &_udph); + if (uh == NULL) + /* No choice either */ + return false; + + *port = src ? uh->source : uh->dest; + break; + } + case IPPROTO_ICMP: { + struct icmphdr _ich; + const struct icmphdr *ic; + + ic = skb_header_pointer(skb, protooff, sizeof(_ich), &_ich); + if (ic == NULL) + return false; + + *port = (__force __be16)htons((ic->type << 8) | ic->code); + break; + } + case IPPROTO_ICMPV6: { + struct icmp6hdr _ich; + const struct icmp6hdr *ic; + + ic = skb_header_pointer(skb, protooff, sizeof(_ich), &_ich); + if (ic == NULL) + return false; + + *port = (__force __be16) + htons((ic->icmp6_type << 8) | ic->icmp6_code); + break; + } + default: + break; + } + *proto = protocol; + + return true; +} + +bool +ip_set_get_ip4_port(const struct sk_buff *skb, bool src, + __be16 *port, u8 *proto) +{ + const struct iphdr *iph = ip_hdr(skb); + unsigned int protooff = ip_hdrlen(skb); + int protocol = iph->protocol; + + /* See comments at tcp_match in ip_tables.c */ + if (protocol <= 0 || (ntohs(iph->frag_off) & IP_OFFSET)) + return false; + + return get_port(skb, protocol, protooff, src, port, proto); +} +EXPORT_SYMBOL_GPL(ip_set_get_ip4_port); + +bool +ip_set_get_ip6_port(const struct sk_buff *skb, bool src, + __be16 *port, u8 *proto) +{ + unsigned int protooff = 0; + int protocol; + unsigned short fragoff; + + protocol = ipv6_find_hdr(skb, &protooff, -1, &fragoff); + if (protocol <= 0 || fragoff) + return false; + + return get_port(skb, protocol, protooff, src, port, proto); +} +EXPORT_SYMBOL_GPL(ip_set_get_ip6_port); + +bool +ip_set_get_ip_port(const struct sk_buff *skb, u8 pf, bool src, __be16 *port) +{ + bool ret; + u8 proto; + + switch (pf) { + case AF_INET: + ret = ip_set_get_ip4_port(skb, src, port, &proto); + case AF_INET6: + ret = ip_set_get_ip6_port(skb, src, port, &proto); + default: + return false; + } + if (!ret) + return ret; + switch (proto) { + case IPPROTO_TCP: + case IPPROTO_UDP: + return true; + default: + return false; + } +} +EXPORT_SYMBOL_GPL(ip_set_get_ip_port); diff --git a/net/netfilter/ipset/pfxlen.c b/net/netfilter/ipset/pfxlen.c new file mode 100644 index 000000000000..23f8c8162214 --- /dev/null +++ b/net/netfilter/ipset/pfxlen.c @@ -0,0 +1,291 @@ +#include + +/* + * Prefixlen maps for fast conversions, by Jan Engelhardt. + */ + +#define E(a, b, c, d) \ + {.ip6 = { \ + __constant_htonl(a), __constant_htonl(b), \ + __constant_htonl(c), __constant_htonl(d), \ + } } + +/* + * This table works for both IPv4 and IPv6; + * just use prefixlen_netmask_map[prefixlength].ip. + */ +const union nf_inet_addr ip_set_netmask_map[] = { + E(0x00000000, 0x00000000, 0x00000000, 0x00000000), + E(0x80000000, 0x00000000, 0x00000000, 0x00000000), + E(0xC0000000, 0x00000000, 0x00000000, 0x00000000), + E(0xE0000000, 0x00000000, 0x00000000, 0x00000000), + E(0xF0000000, 0x00000000, 0x00000000, 0x00000000), + E(0xF8000000, 0x00000000, 0x00000000, 0x00000000), + E(0xFC000000, 0x00000000, 0x00000000, 0x00000000), + E(0xFE000000, 0x00000000, 0x00000000, 0x00000000), + E(0xFF000000, 0x00000000, 0x00000000, 0x00000000), + E(0xFF800000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFC00000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFE00000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFF00000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFF80000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFC0000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFE0000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFF0000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFF8000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFC000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFE000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFF000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFF800, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFC00, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFE00, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFF00, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFF80, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFC0, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFE0, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFF0, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFF8, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFFC, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFFE, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0x80000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xC0000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xE0000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xF0000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xF8000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFC000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFE000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFF000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFF800000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFC00000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFE00000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFF00000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFF80000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFC0000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFE0000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFF0000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFF8000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFC000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFE000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFF000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFF800, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFC00, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFE00, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFF00, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFF80, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFC0, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFE0, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFF0, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFF8, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFC, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFE, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0x80000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xC0000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xE0000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xF0000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xF8000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFC000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFE000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFF000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFF800000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFC00000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFE00000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFF00000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFF80000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFC0000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFE0000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFF0000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFF8000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFC000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFE000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFF000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFF800, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFC00, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFE00, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFF00, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFF80, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFC0, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFE0, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFF0, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFF8, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFC, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFE, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0x80000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xC0000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xE0000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xF0000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xF8000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFC000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFE000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFF000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFF800000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFC00000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFE00000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFF00000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFF80000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFC0000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFE0000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFF0000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFF8000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFC000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFE000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFF000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFF800), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFC00), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFE00), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFF00), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFF80), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFC0), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFE0), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFF0), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFF8), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFC), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFE), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF), +}; +EXPORT_SYMBOL_GPL(ip_set_netmask_map); + +#undef E +#define E(a, b, c, d) \ + {.ip6 = { (__force __be32) a, (__force __be32) b, \ + (__force __be32) c, (__force __be32) d, \ + } } + +/* + * This table works for both IPv4 and IPv6; + * just use prefixlen_hostmask_map[prefixlength].ip. + */ +const union nf_inet_addr ip_set_hostmask_map[] = { + E(0x00000000, 0x00000000, 0x00000000, 0x00000000), + E(0x80000000, 0x00000000, 0x00000000, 0x00000000), + E(0xC0000000, 0x00000000, 0x00000000, 0x00000000), + E(0xE0000000, 0x00000000, 0x00000000, 0x00000000), + E(0xF0000000, 0x00000000, 0x00000000, 0x00000000), + E(0xF8000000, 0x00000000, 0x00000000, 0x00000000), + E(0xFC000000, 0x00000000, 0x00000000, 0x00000000), + E(0xFE000000, 0x00000000, 0x00000000, 0x00000000), + E(0xFF000000, 0x00000000, 0x00000000, 0x00000000), + E(0xFF800000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFC00000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFE00000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFF00000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFF80000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFC0000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFE0000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFF0000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFF8000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFC000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFE000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFF000, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFF800, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFC00, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFE00, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFF00, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFF80, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFC0, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFE0, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFF0, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFF8, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFFC, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFFE, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0x00000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0x80000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xC0000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xE0000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xF0000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xF8000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFC000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFE000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFF000000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFF800000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFC00000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFE00000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFF00000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFF80000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFC0000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFE0000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFF0000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFF8000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFC000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFE000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFF000, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFF800, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFC00, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFE00, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFF00, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFF80, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFC0, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFE0, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFF0, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFF8, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFC, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFE, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0x00000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0x80000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xC0000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xE0000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xF0000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xF8000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFC000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFE000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFF000000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFF800000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFC00000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFE00000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFF00000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFF80000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFC0000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFE0000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFF0000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFF8000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFC000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFE000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFF000, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFF800, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFC00, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFE00, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFF00, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFF80, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFC0, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFE0, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFF0, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFF8, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFC, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFE, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0x00000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0x80000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xC0000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xE0000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xF0000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xF8000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFC000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFE000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFF000000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFF800000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFC00000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFE00000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFF00000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFF80000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFC0000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFE0000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFF0000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFF8000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFC000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFE000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFF000), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFF800), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFC00), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFE00), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFF00), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFF80), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFC0), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFE0), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFF0), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFF8), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFC), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFE), + E(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF), +}; +EXPORT_SYMBOL_GPL(ip_set_hostmask_map); -- cgit v1.2.3 From 72205fc68bd13109576aa6c4c12c740962d28a6c Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 1 Feb 2011 15:33:17 +0100 Subject: netfilter: ipset: bitmap:ip set type support The module implements the bitmap:ip set type in two flavours, without and with timeout support. In this kind of set one can store IPv4 addresses (or network addresses) from a given range. In order not to waste memory, the timeout version does not rely on the kernel timer for every element to be timed out but on garbage collection. All set types use this mechanism. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/ipset/Kconfig | 9 + net/netfilter/ipset/Makefile | 3 + net/netfilter/ipset/ip_set_bitmap_ip.c | 588 +++++++++++++++++++++++++++++++++ 3 files changed, 600 insertions(+) create mode 100644 net/netfilter/ipset/ip_set_bitmap_ip.c (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/Kconfig b/net/netfilter/ipset/Kconfig index 5ade156dd470..b63a8eea183c 100644 --- a/net/netfilter/ipset/Kconfig +++ b/net/netfilter/ipset/Kconfig @@ -23,4 +23,13 @@ config IP_SET_MAX The value can be overriden by the 'max_sets' module parameter of the 'ip_set' module. +config IP_SET_BITMAP_IP + tristate "bitmap:ip set support" + depends on IP_SET + help + This option adds the bitmap:ip set type support, by which one + can store IPv4 addresses (or network addresse) from a range. + + To compile it as a module, choose M here. If unsure, say N. + endif # IP_SET diff --git a/net/netfilter/ipset/Makefile b/net/netfilter/ipset/Makefile index 910cd425c067..ea1c85e28af1 100644 --- a/net/netfilter/ipset/Makefile +++ b/net/netfilter/ipset/Makefile @@ -6,3 +6,6 @@ ip_set-y := ip_set_core.o ip_set_getport.o pfxlen.o # ipset core obj-$(CONFIG_IP_SET) += ip_set.o + +# bitmap types +obj-$(CONFIG_IP_SET_BITMAP_IP) += ip_set_bitmap_ip.o diff --git a/net/netfilter/ipset/ip_set_bitmap_ip.c b/net/netfilter/ipset/ip_set_bitmap_ip.c new file mode 100644 index 000000000000..047440085509 --- /dev/null +++ b/net/netfilter/ipset/ip_set_bitmap_ip.c @@ -0,0 +1,588 @@ +/* Copyright (C) 2000-2002 Joakim Axelsson + * Patrick Schaaf + * Copyright (C) 2003-2011 Jozsef Kadlecsik + * + * 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. + */ + +/* Kernel module implementing an IP set type: the bitmap:ip type */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#define IP_SET_BITMAP_TIMEOUT +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Jozsef Kadlecsik "); +MODULE_DESCRIPTION("bitmap:ip type of IP sets"); +MODULE_ALIAS("ip_set_bitmap:ip"); + +/* Type structure */ +struct bitmap_ip { + void *members; /* the set members */ + u32 first_ip; /* host byte order, included in range */ + u32 last_ip; /* host byte order, included in range */ + u32 elements; /* number of max elements in the set */ + u32 hosts; /* number of hosts in a subnet */ + size_t memsize; /* members size */ + u8 netmask; /* subnet netmask */ + u32 timeout; /* timeout parameter */ + struct timer_list gc; /* garbage collection */ +}; + +/* Base variant */ + +static inline u32 +ip_to_id(const struct bitmap_ip *m, u32 ip) +{ + return ((ip & ip_set_hostmask(m->netmask)) - m->first_ip)/m->hosts; +} + +static int +bitmap_ip_test(struct ip_set *set, void *value, u32 timeout) +{ + const struct bitmap_ip *map = set->data; + u16 id = *(u16 *)value; + + return !!test_bit(id, map->members); +} + +static int +bitmap_ip_add(struct ip_set *set, void *value, u32 timeout) +{ + struct bitmap_ip *map = set->data; + u16 id = *(u16 *)value; + + if (test_and_set_bit(id, map->members)) + return -IPSET_ERR_EXIST; + + return 0; +} + +static int +bitmap_ip_del(struct ip_set *set, void *value, u32 timeout) +{ + struct bitmap_ip *map = set->data; + u16 id = *(u16 *)value; + + if (!test_and_clear_bit(id, map->members)) + return -IPSET_ERR_EXIST; + + return 0; +} + +static int +bitmap_ip_list(const struct ip_set *set, + struct sk_buff *skb, struct netlink_callback *cb) +{ + const struct bitmap_ip *map = set->data; + struct nlattr *atd, *nested; + u32 id, first = cb->args[2]; + + atd = ipset_nest_start(skb, IPSET_ATTR_ADT); + if (!atd) + return -EMSGSIZE; + for (; cb->args[2] < map->elements; cb->args[2]++) { + id = cb->args[2]; + if (!test_bit(id, map->members)) + continue; + nested = ipset_nest_start(skb, IPSET_ATTR_DATA); + if (!nested) { + if (id == first) { + nla_nest_cancel(skb, atd); + return -EMSGSIZE; + } else + goto nla_put_failure; + } + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, + htonl(map->first_ip + id * map->hosts)); + ipset_nest_end(skb, nested); + } + ipset_nest_end(skb, atd); + /* Set listing finished */ + cb->args[2] = 0; + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nested); + ipset_nest_end(skb, atd); + if (unlikely(id == first)) { + cb->args[2] = 0; + return -EMSGSIZE; + } + return 0; +} + +/* Timeout variant */ + +static int +bitmap_ip_ttest(struct ip_set *set, void *value, u32 timeout) +{ + const struct bitmap_ip *map = set->data; + const unsigned long *members = map->members; + u16 id = *(u16 *)value; + + return ip_set_timeout_test(members[id]); +} + +static int +bitmap_ip_tadd(struct ip_set *set, void *value, u32 timeout) +{ + struct bitmap_ip *map = set->data; + unsigned long *members = map->members; + u16 id = *(u16 *)value; + + if (ip_set_timeout_test(members[id])) + return -IPSET_ERR_EXIST; + + members[id] = ip_set_timeout_set(timeout); + + return 0; +} + +static int +bitmap_ip_tdel(struct ip_set *set, void *value, u32 timeout) +{ + struct bitmap_ip *map = set->data; + unsigned long *members = map->members; + u16 id = *(u16 *)value; + int ret = -IPSET_ERR_EXIST; + + if (ip_set_timeout_test(members[id])) + ret = 0; + + members[id] = IPSET_ELEM_UNSET; + return ret; +} + +static int +bitmap_ip_tlist(const struct ip_set *set, + struct sk_buff *skb, struct netlink_callback *cb) +{ + const struct bitmap_ip *map = set->data; + struct nlattr *adt, *nested; + u32 id, first = cb->args[2]; + const unsigned long *members = map->members; + + adt = ipset_nest_start(skb, IPSET_ATTR_ADT); + if (!adt) + return -EMSGSIZE; + for (; cb->args[2] < map->elements; cb->args[2]++) { + id = cb->args[2]; + if (!ip_set_timeout_test(members[id])) + continue; + nested = ipset_nest_start(skb, IPSET_ATTR_DATA); + if (!nested) { + if (id == first) { + nla_nest_cancel(skb, adt); + return -EMSGSIZE; + } else + goto nla_put_failure; + } + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, + htonl(map->first_ip + id * map->hosts)); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(members[id]))); + ipset_nest_end(skb, nested); + } + ipset_nest_end(skb, adt); + + /* Set listing finished */ + cb->args[2] = 0; + + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nested); + ipset_nest_end(skb, adt); + if (unlikely(id == first)) { + cb->args[2] = 0; + return -EMSGSIZE; + } + return 0; +} + +static int +bitmap_ip_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + struct bitmap_ip *map = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + u32 ip; + + ip = ntohl(ip4addr(skb, flags & IPSET_DIM_ONE_SRC)); + if (ip < map->first_ip || ip > map->last_ip) + return -IPSET_ERR_BITMAP_RANGE; + + ip = ip_to_id(map, ip); + + return adtfn(set, &ip, map->timeout); +} + +static int +bitmap_ip_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + struct bitmap_ip *map = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + u32 timeout = map->timeout; + u32 ip, ip_to, id; + int ret = 0; + + if (unlikely(!tb[IPSET_ATTR_IP] || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP], &ip); + if (ret) + return ret; + + if (ip < map->first_ip || ip > map->last_ip) + return -IPSET_ERR_BITMAP_RANGE; + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(map->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + if (adt == IPSET_TEST) { + id = ip_to_id(map, ip); + return adtfn(set, &id, timeout); + } + + if (tb[IPSET_ATTR_IP_TO]) { + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP_TO], &ip_to); + if (ret) + return ret; + if (ip > ip_to) { + swap(ip, ip_to); + if (ip < map->first_ip) + return -IPSET_ERR_BITMAP_RANGE; + } + } else if (tb[IPSET_ATTR_CIDR]) { + u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); + + if (cidr > 32) + return -IPSET_ERR_INVALID_CIDR; + ip &= ip_set_hostmask(cidr); + ip_to = ip | ~ip_set_hostmask(cidr); + } else + ip_to = ip; + + if (ip_to > map->last_ip) + return -IPSET_ERR_BITMAP_RANGE; + + for (; !before(ip_to, ip); ip += map->hosts) { + id = ip_to_id(map, ip); + ret = adtfn(set, &id, timeout);; + + if (ret && !ip_set_eexist(ret, flags)) + return ret; + else + ret = 0; + } + return ret; +} + +static void +bitmap_ip_destroy(struct ip_set *set) +{ + struct bitmap_ip *map = set->data; + + if (with_timeout(map->timeout)) + del_timer_sync(&map->gc); + + ip_set_free(map->members); + kfree(map); + + set->data = NULL; +} + +static void +bitmap_ip_flush(struct ip_set *set) +{ + struct bitmap_ip *map = set->data; + + memset(map->members, 0, map->memsize); +} + +static int +bitmap_ip_head(struct ip_set *set, struct sk_buff *skb) +{ + const struct bitmap_ip *map = set->data; + struct nlattr *nested; + + nested = ipset_nest_start(skb, IPSET_ATTR_DATA); + if (!nested) + goto nla_put_failure; + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, htonl(map->first_ip)); + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP_TO, htonl(map->last_ip)); + if (map->netmask != 32) + NLA_PUT_U8(skb, IPSET_ATTR_NETMASK, map->netmask); + NLA_PUT_NET32(skb, IPSET_ATTR_REFERENCES, + htonl(atomic_read(&set->ref) - 1)); + NLA_PUT_NET32(skb, IPSET_ATTR_MEMSIZE, + htonl(sizeof(*map) + map->memsize)); + if (with_timeout(map->timeout)) + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, htonl(map->timeout)); + ipset_nest_end(skb, nested); + + return 0; +nla_put_failure: + return -EMSGSIZE; +} + +static bool +bitmap_ip_same_set(const struct ip_set *a, const struct ip_set *b) +{ + const struct bitmap_ip *x = a->data; + const struct bitmap_ip *y = b->data; + + return x->first_ip == y->first_ip && + x->last_ip == y->last_ip && + x->netmask == y->netmask && + x->timeout == y->timeout; +} + +static const struct ip_set_type_variant bitmap_ip = { + .kadt = bitmap_ip_kadt, + .uadt = bitmap_ip_uadt, + .adt = { + [IPSET_ADD] = bitmap_ip_add, + [IPSET_DEL] = bitmap_ip_del, + [IPSET_TEST] = bitmap_ip_test, + }, + .destroy = bitmap_ip_destroy, + .flush = bitmap_ip_flush, + .head = bitmap_ip_head, + .list = bitmap_ip_list, + .same_set = bitmap_ip_same_set, +}; + +static const struct ip_set_type_variant bitmap_tip = { + .kadt = bitmap_ip_kadt, + .uadt = bitmap_ip_uadt, + .adt = { + [IPSET_ADD] = bitmap_ip_tadd, + [IPSET_DEL] = bitmap_ip_tdel, + [IPSET_TEST] = bitmap_ip_ttest, + }, + .destroy = bitmap_ip_destroy, + .flush = bitmap_ip_flush, + .head = bitmap_ip_head, + .list = bitmap_ip_tlist, + .same_set = bitmap_ip_same_set, +}; + +static void +bitmap_ip_gc(unsigned long ul_set) +{ + struct ip_set *set = (struct ip_set *) ul_set; + struct bitmap_ip *map = set->data; + unsigned long *table = map->members; + u32 id; + + /* We run parallel with other readers (test element) + * but adding/deleting new entries is locked out */ + read_lock_bh(&set->lock); + for (id = 0; id < map->elements; id++) + if (ip_set_timeout_expired(table[id])) + table[id] = IPSET_ELEM_UNSET; + read_unlock_bh(&set->lock); + + map->gc.expires = jiffies + IPSET_GC_PERIOD(map->timeout) * HZ; + add_timer(&map->gc); +} + +static void +bitmap_ip_gc_init(struct ip_set *set) +{ + struct bitmap_ip *map = set->data; + + init_timer(&map->gc); + map->gc.data = (unsigned long) set; + map->gc.function = bitmap_ip_gc; + map->gc.expires = jiffies + IPSET_GC_PERIOD(map->timeout) * HZ; + add_timer(&map->gc); +} + +/* Create bitmap:ip type of sets */ + +static bool +init_map_ip(struct ip_set *set, struct bitmap_ip *map, + u32 first_ip, u32 last_ip, + u32 elements, u32 hosts, u8 netmask) +{ + map->members = ip_set_alloc(map->memsize); + if (!map->members) + return false; + map->first_ip = first_ip; + map->last_ip = last_ip; + map->elements = elements; + map->hosts = hosts; + map->netmask = netmask; + map->timeout = IPSET_NO_TIMEOUT; + + set->data = map; + set->family = AF_INET; + + return true; +} + +static int +bitmap_ip_create(struct ip_set *set, struct nlattr *tb[], u32 flags) +{ + struct bitmap_ip *map; + u32 first_ip, last_ip, hosts, elements; + u8 netmask = 32; + int ret; + + if (unlikely(!tb[IPSET_ATTR_IP] || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP], &first_ip); + if (ret) + return ret; + + if (tb[IPSET_ATTR_IP_TO]) { + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP_TO], &last_ip); + if (ret) + return ret; + if (first_ip > last_ip) { + u32 tmp = first_ip; + + first_ip = last_ip; + last_ip = tmp; + } + } else if (tb[IPSET_ATTR_CIDR]) { + u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); + + if (cidr >= 32) + return -IPSET_ERR_INVALID_CIDR; + last_ip = first_ip | ~ip_set_hostmask(cidr); + } else + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_NETMASK]) { + netmask = nla_get_u8(tb[IPSET_ATTR_NETMASK]); + + if (netmask > 32) + return -IPSET_ERR_INVALID_NETMASK; + + first_ip &= ip_set_hostmask(netmask); + last_ip |= ~ip_set_hostmask(netmask); + } + + if (netmask == 32) { + hosts = 1; + elements = last_ip - first_ip + 1; + } else { + u8 mask_bits; + u32 mask; + + mask = range_to_mask(first_ip, last_ip, &mask_bits); + + if ((!mask && (first_ip || last_ip != 0xFFFFFFFF)) || + netmask <= mask_bits) + return -IPSET_ERR_BITMAP_RANGE; + + pr_debug("mask_bits %u, netmask %u\n", mask_bits, netmask); + hosts = 2 << (32 - netmask - 1); + elements = 2 << (netmask - mask_bits - 1); + } + if (elements > IPSET_BITMAP_MAX_RANGE + 1) + return -IPSET_ERR_BITMAP_RANGE_SIZE; + + pr_debug("hosts %u, elements %u\n", hosts, elements); + + map = kzalloc(sizeof(*map), GFP_KERNEL); + if (!map) + return -ENOMEM; + + if (tb[IPSET_ATTR_TIMEOUT]) { + map->memsize = elements * sizeof(unsigned long); + + if (!init_map_ip(set, map, first_ip, last_ip, + elements, hosts, netmask)) { + kfree(map); + return -ENOMEM; + } + + map->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + set->variant = &bitmap_tip; + + bitmap_ip_gc_init(set); + } else { + map->memsize = bitmap_bytes(0, elements - 1); + + if (!init_map_ip(set, map, first_ip, last_ip, + elements, hosts, netmask)) { + kfree(map); + return -ENOMEM; + } + + set->variant = &bitmap_ip; + } + return 0; +} + +static struct ip_set_type bitmap_ip_type __read_mostly = { + .name = "bitmap:ip", + .protocol = IPSET_PROTOCOL, + .features = IPSET_TYPE_IP, + .dimension = IPSET_DIM_ONE, + .family = AF_INET, + .revision = 0, + .create = bitmap_ip_create, + .create_policy = { + [IPSET_ATTR_IP] = { .type = NLA_NESTED }, + [IPSET_ATTR_IP_TO] = { .type = NLA_NESTED }, + [IPSET_ATTR_CIDR] = { .type = NLA_U8 }, + [IPSET_ATTR_NETMASK] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + }, + .adt_policy = { + [IPSET_ATTR_IP] = { .type = NLA_NESTED }, + [IPSET_ATTR_IP_TO] = { .type = NLA_NESTED }, + [IPSET_ATTR_CIDR] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + [IPSET_ATTR_LINENO] = { .type = NLA_U32 }, + }, + .me = THIS_MODULE, +}; + +static int __init +bitmap_ip_init(void) +{ + return ip_set_type_register(&bitmap_ip_type); +} + +static void __exit +bitmap_ip_fini(void) +{ + ip_set_type_unregister(&bitmap_ip_type); +} + +module_init(bitmap_ip_init); +module_exit(bitmap_ip_fini); -- cgit v1.2.3 From de76021a1bb35e3560afccf741d1119a872aea49 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 1 Feb 2011 15:35:12 +0100 Subject: netfilter: ipset: bitmap:ip,mac type support The module implements the bitmap:ip,mac set type in two flavours, without and with timeout support. In this kind of set one can store IPv4 address and (source) MAC address pairs. The type supports elements added without the MAC part filled out: when the first matching from kernel happens, the MAC part is automatically filled out. The timing out of the elements stars when an element is complete in the IP,MAC pair. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/ipset/Kconfig | 9 + net/netfilter/ipset/Makefile | 1 + net/netfilter/ipset/ip_set_bitmap_ipmac.c | 655 ++++++++++++++++++++++++++++++ 3 files changed, 665 insertions(+) create mode 100644 net/netfilter/ipset/ip_set_bitmap_ipmac.c (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/Kconfig b/net/netfilter/ipset/Kconfig index b63a8eea183c..f18654c73af2 100644 --- a/net/netfilter/ipset/Kconfig +++ b/net/netfilter/ipset/Kconfig @@ -32,4 +32,13 @@ config IP_SET_BITMAP_IP To compile it as a module, choose M here. If unsure, say N. +config IP_SET_BITMAP_IPMAC + tristate "bitmap:ip,mac set support" + depends on IP_SET + help + This option adds the bitmap:ip,mac set type support, by which one + can store IPv4 address and (source) MAC address pairs from a range. + + To compile it as a module, choose M here. If unsure, say N. + endif # IP_SET diff --git a/net/netfilter/ipset/Makefile b/net/netfilter/ipset/Makefile index ea1c85e28af1..f7a099f40d0b 100644 --- a/net/netfilter/ipset/Makefile +++ b/net/netfilter/ipset/Makefile @@ -9,3 +9,4 @@ obj-$(CONFIG_IP_SET) += ip_set.o # bitmap types obj-$(CONFIG_IP_SET_BITMAP_IP) += ip_set_bitmap_ip.o +obj-$(CONFIG_IP_SET_BITMAP_IPMAC) += ip_set_bitmap_ipmac.o diff --git a/net/netfilter/ipset/ip_set_bitmap_ipmac.c b/net/netfilter/ipset/ip_set_bitmap_ipmac.c new file mode 100644 index 000000000000..d826332d2937 --- /dev/null +++ b/net/netfilter/ipset/ip_set_bitmap_ipmac.c @@ -0,0 +1,655 @@ +/* Copyright (C) 2000-2002 Joakim Axelsson + * Patrick Schaaf + * Martin Josefsson + * Copyright (C) 2003-2011 Jozsef Kadlecsik + * + * 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. + */ + +/* Kernel module implementing an IP set type: the bitmap:ip,mac type */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Jozsef Kadlecsik "); +MODULE_DESCRIPTION("bitmap:ip,mac type of IP sets"); +MODULE_ALIAS("ip_set_bitmap:ip,mac"); + +enum { + MAC_EMPTY, /* element is not set */ + MAC_FILLED, /* element is set with MAC */ + MAC_UNSET, /* element is set, without MAC */ +}; + +/* Type structure */ +struct bitmap_ipmac { + void *members; /* the set members */ + u32 first_ip; /* host byte order, included in range */ + u32 last_ip; /* host byte order, included in range */ + u32 timeout; /* timeout value */ + struct timer_list gc; /* garbage collector */ + size_t dsize; /* size of element */ +}; + +/* ADT structure for generic function args */ +struct ipmac { + u32 id; /* id in array */ + unsigned char *ether; /* ethernet address */ +}; + +/* Member element without and with timeout */ + +struct ipmac_elem { + unsigned char ether[ETH_ALEN]; + unsigned char match; +} __attribute__ ((aligned)); + +struct ipmac_telem { + unsigned char ether[ETH_ALEN]; + unsigned char match; + unsigned long timeout; +} __attribute__ ((aligned)); + +static inline void * +bitmap_ipmac_elem(const struct bitmap_ipmac *map, u32 id) +{ + return (void *)((char *)map->members + id * map->dsize); +} + +static inline bool +bitmap_timeout(const struct bitmap_ipmac *map, u32 id) +{ + const struct ipmac_telem *elem = bitmap_ipmac_elem(map, id); + + return ip_set_timeout_test(elem->timeout); +} + +static inline bool +bitmap_expired(const struct bitmap_ipmac *map, u32 id) +{ + const struct ipmac_telem *elem = bitmap_ipmac_elem(map, id); + + return ip_set_timeout_expired(elem->timeout); +} + +static inline int +bitmap_ipmac_exist(const struct ipmac_telem *elem) +{ + return elem->match == MAC_UNSET || + (elem->match == MAC_FILLED && + !ip_set_timeout_expired(elem->timeout)); +} + +/* Base variant */ + +static int +bitmap_ipmac_test(struct ip_set *set, void *value, u32 timeout) +{ + const struct bitmap_ipmac *map = set->data; + const struct ipmac *data = value; + const struct ipmac_elem *elem = bitmap_ipmac_elem(map, data->id); + + switch (elem->match) { + case MAC_UNSET: + /* Trigger kernel to fill out the ethernet address */ + return -EAGAIN; + case MAC_FILLED: + return data->ether == NULL || + compare_ether_addr(data->ether, elem->ether) == 0; + } + return 0; +} + +static int +bitmap_ipmac_add(struct ip_set *set, void *value, u32 timeout) +{ + struct bitmap_ipmac *map = set->data; + const struct ipmac *data = value; + struct ipmac_elem *elem = bitmap_ipmac_elem(map, data->id); + + switch (elem->match) { + case MAC_UNSET: + if (!data->ether) + /* Already added without ethernet address */ + return -IPSET_ERR_EXIST; + /* Fill the MAC address */ + memcpy(elem->ether, data->ether, ETH_ALEN); + elem->match = MAC_FILLED; + break; + case MAC_FILLED: + return -IPSET_ERR_EXIST; + case MAC_EMPTY: + if (data->ether) { + memcpy(elem->ether, data->ether, ETH_ALEN); + elem->match = MAC_FILLED; + } else + elem->match = MAC_UNSET; + } + + return 0; +} + +static int +bitmap_ipmac_del(struct ip_set *set, void *value, u32 timeout) +{ + struct bitmap_ipmac *map = set->data; + const struct ipmac *data = value; + struct ipmac_elem *elem = bitmap_ipmac_elem(map, data->id); + + if (elem->match == MAC_EMPTY) + return -IPSET_ERR_EXIST; + + elem->match = MAC_EMPTY; + + return 0; +} + +static int +bitmap_ipmac_list(const struct ip_set *set, + struct sk_buff *skb, struct netlink_callback *cb) +{ + const struct bitmap_ipmac *map = set->data; + const struct ipmac_elem *elem; + struct nlattr *atd, *nested; + u32 id, first = cb->args[2]; + u32 last = map->last_ip - map->first_ip; + + atd = ipset_nest_start(skb, IPSET_ATTR_ADT); + if (!atd) + return -EMSGSIZE; + for (; cb->args[2] <= last; cb->args[2]++) { + id = cb->args[2]; + elem = bitmap_ipmac_elem(map, id); + if (elem->match == MAC_EMPTY) + continue; + nested = ipset_nest_start(skb, IPSET_ATTR_DATA); + if (!nested) { + if (id == first) { + nla_nest_cancel(skb, atd); + return -EMSGSIZE; + } else + goto nla_put_failure; + } + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, + htonl(map->first_ip + id)); + if (elem->match == MAC_FILLED) + NLA_PUT(skb, IPSET_ATTR_ETHER, ETH_ALEN, + elem->ether); + ipset_nest_end(skb, nested); + } + ipset_nest_end(skb, atd); + /* Set listing finished */ + cb->args[2] = 0; + + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nested); + ipset_nest_end(skb, atd); + if (unlikely(id == first)) { + cb->args[2] = 0; + return -EMSGSIZE; + } + return 0; +} + +/* Timeout variant */ + +static int +bitmap_ipmac_ttest(struct ip_set *set, void *value, u32 timeout) +{ + const struct bitmap_ipmac *map = set->data; + const struct ipmac *data = value; + const struct ipmac_elem *elem = bitmap_ipmac_elem(map, data->id); + + switch (elem->match) { + case MAC_UNSET: + /* Trigger kernel to fill out the ethernet address */ + return -EAGAIN; + case MAC_FILLED: + return (data->ether == NULL || + compare_ether_addr(data->ether, elem->ether) == 0) && + !bitmap_expired(map, data->id); + } + return 0; +} + +static int +bitmap_ipmac_tadd(struct ip_set *set, void *value, u32 timeout) +{ + struct bitmap_ipmac *map = set->data; + const struct ipmac *data = value; + struct ipmac_telem *elem = bitmap_ipmac_elem(map, data->id); + + switch (elem->match) { + case MAC_UNSET: + if (!data->ether) + /* Already added without ethernet address */ + return -IPSET_ERR_EXIST; + /* Fill the MAC address and activate the timer */ + memcpy(elem->ether, data->ether, ETH_ALEN); + elem->match = MAC_FILLED; + if (timeout == map->timeout) + /* Timeout was not specified, get stored one */ + timeout = elem->timeout; + elem->timeout = ip_set_timeout_set(timeout); + break; + case MAC_FILLED: + if (!bitmap_expired(map, data->id)) + return -IPSET_ERR_EXIST; + /* Fall through */ + case MAC_EMPTY: + if (data->ether) { + memcpy(elem->ether, data->ether, ETH_ALEN); + elem->match = MAC_FILLED; + } else + elem->match = MAC_UNSET; + /* If MAC is unset yet, we store plain timeout value + * because the timer is not activated yet + * and we can reuse it later when MAC is filled out, + * possibly by the kernel */ + elem->timeout = data->ether ? ip_set_timeout_set(timeout) + : timeout; + break; + } + + return 0; +} + +static int +bitmap_ipmac_tdel(struct ip_set *set, void *value, u32 timeout) +{ + struct bitmap_ipmac *map = set->data; + const struct ipmac *data = value; + struct ipmac_telem *elem = bitmap_ipmac_elem(map, data->id); + + if (elem->match == MAC_EMPTY || bitmap_expired(map, data->id)) + return -IPSET_ERR_EXIST; + + elem->match = MAC_EMPTY; + + return 0; +} + +static int +bitmap_ipmac_tlist(const struct ip_set *set, + struct sk_buff *skb, struct netlink_callback *cb) +{ + const struct bitmap_ipmac *map = set->data; + const struct ipmac_telem *elem; + struct nlattr *atd, *nested; + u32 id, first = cb->args[2]; + u32 timeout, last = map->last_ip - map->first_ip; + + atd = ipset_nest_start(skb, IPSET_ATTR_ADT); + if (!atd) + return -EMSGSIZE; + for (; cb->args[2] <= last; cb->args[2]++) { + id = cb->args[2]; + elem = bitmap_ipmac_elem(map, id); + if (!bitmap_ipmac_exist(elem)) + continue; + nested = ipset_nest_start(skb, IPSET_ATTR_DATA); + if (!nested) { + if (id == first) { + nla_nest_cancel(skb, atd); + return -EMSGSIZE; + } else + goto nla_put_failure; + } + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, + htonl(map->first_ip + id)); + if (elem->match == MAC_FILLED) + NLA_PUT(skb, IPSET_ATTR_ETHER, ETH_ALEN, + elem->ether); + timeout = elem->match == MAC_UNSET ? elem->timeout + : ip_set_timeout_get(elem->timeout); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, htonl(timeout)); + ipset_nest_end(skb, nested); + } + ipset_nest_end(skb, atd); + /* Set listing finished */ + cb->args[2] = 0; + + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nested); + ipset_nest_end(skb, atd); + return -EMSGSIZE; +} + +static int +bitmap_ipmac_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + struct bitmap_ipmac *map = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct ipmac data; + + data.id = ntohl(ip4addr(skb, flags & IPSET_DIM_ONE_SRC)); + if (data.id < map->first_ip || data.id > map->last_ip) + return -IPSET_ERR_BITMAP_RANGE; + + /* Backward compatibility: we don't check the second flag */ + if (skb_mac_header(skb) < skb->head || + (skb_mac_header(skb) + ETH_HLEN) > skb->data) + return -EINVAL; + + data.id -= map->first_ip; + data.ether = eth_hdr(skb)->h_source; + + return adtfn(set, &data, map->timeout); +} + +static int +bitmap_ipmac_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + const struct bitmap_ipmac *map = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct ipmac data; + u32 timeout = map->timeout; + int ret = 0; + + if (unlikely(!tb[IPSET_ATTR_IP] || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP], &data.id); + if (ret) + return ret; + + if (data.id < map->first_ip || data.id > map->last_ip) + return -IPSET_ERR_BITMAP_RANGE; + + if (tb[IPSET_ATTR_ETHER]) + data.ether = nla_data(tb[IPSET_ATTR_ETHER]); + else + data.ether = NULL; + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(map->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + data.id -= map->first_ip; + + ret = adtfn(set, &data, timeout); + + return ip_set_eexist(ret, flags) ? 0 : ret; +} + +static void +bitmap_ipmac_destroy(struct ip_set *set) +{ + struct bitmap_ipmac *map = set->data; + + if (with_timeout(map->timeout)) + del_timer_sync(&map->gc); + + ip_set_free(map->members); + kfree(map); + + set->data = NULL; +} + +static void +bitmap_ipmac_flush(struct ip_set *set) +{ + struct bitmap_ipmac *map = set->data; + + memset(map->members, 0, + (map->last_ip - map->first_ip + 1) * map->dsize); +} + +static int +bitmap_ipmac_head(struct ip_set *set, struct sk_buff *skb) +{ + const struct bitmap_ipmac *map = set->data; + struct nlattr *nested; + + nested = ipset_nest_start(skb, IPSET_ATTR_DATA); + if (!nested) + goto nla_put_failure; + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, htonl(map->first_ip)); + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP_TO, htonl(map->last_ip)); + NLA_PUT_NET32(skb, IPSET_ATTR_REFERENCES, + htonl(atomic_read(&set->ref) - 1)); + NLA_PUT_NET32(skb, IPSET_ATTR_MEMSIZE, + htonl(sizeof(*map) + + (map->last_ip - map->first_ip + 1) * map->dsize)); + if (with_timeout(map->timeout)) + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, htonl(map->timeout)); + ipset_nest_end(skb, nested); + + return 0; +nla_put_failure: + return -EMSGSIZE; +} + +static bool +bitmap_ipmac_same_set(const struct ip_set *a, const struct ip_set *b) +{ + const struct bitmap_ipmac *x = a->data; + const struct bitmap_ipmac *y = b->data; + + return x->first_ip == y->first_ip && + x->last_ip == y->last_ip && + x->timeout == y->timeout; +} + +static const struct ip_set_type_variant bitmap_ipmac = { + .kadt = bitmap_ipmac_kadt, + .uadt = bitmap_ipmac_uadt, + .adt = { + [IPSET_ADD] = bitmap_ipmac_add, + [IPSET_DEL] = bitmap_ipmac_del, + [IPSET_TEST] = bitmap_ipmac_test, + }, + .destroy = bitmap_ipmac_destroy, + .flush = bitmap_ipmac_flush, + .head = bitmap_ipmac_head, + .list = bitmap_ipmac_list, + .same_set = bitmap_ipmac_same_set, +}; + +static const struct ip_set_type_variant bitmap_tipmac = { + .kadt = bitmap_ipmac_kadt, + .uadt = bitmap_ipmac_uadt, + .adt = { + [IPSET_ADD] = bitmap_ipmac_tadd, + [IPSET_DEL] = bitmap_ipmac_tdel, + [IPSET_TEST] = bitmap_ipmac_ttest, + }, + .destroy = bitmap_ipmac_destroy, + .flush = bitmap_ipmac_flush, + .head = bitmap_ipmac_head, + .list = bitmap_ipmac_tlist, + .same_set = bitmap_ipmac_same_set, +}; + +static void +bitmap_ipmac_gc(unsigned long ul_set) +{ + struct ip_set *set = (struct ip_set *) ul_set; + struct bitmap_ipmac *map = set->data; + struct ipmac_telem *elem; + u32 id, last = map->last_ip - map->first_ip; + + /* We run parallel with other readers (test element) + * but adding/deleting new entries is locked out */ + read_lock_bh(&set->lock); + for (id = 0; id <= last; id++) { + elem = bitmap_ipmac_elem(map, id); + if (elem->match == MAC_FILLED && + ip_set_timeout_expired(elem->timeout)) + elem->match = MAC_EMPTY; + } + read_unlock_bh(&set->lock); + + map->gc.expires = jiffies + IPSET_GC_PERIOD(map->timeout) * HZ; + add_timer(&map->gc); +} + +static void +bitmap_ipmac_gc_init(struct ip_set *set) +{ + struct bitmap_ipmac *map = set->data; + + init_timer(&map->gc); + map->gc.data = (unsigned long) set; + map->gc.function = bitmap_ipmac_gc; + map->gc.expires = jiffies + IPSET_GC_PERIOD(map->timeout) * HZ; + add_timer(&map->gc); +} + +/* Create bitmap:ip,mac type of sets */ + +static bool +init_map_ipmac(struct ip_set *set, struct bitmap_ipmac *map, + u32 first_ip, u32 last_ip) +{ + map->members = ip_set_alloc((last_ip - first_ip + 1) * map->dsize); + if (!map->members) + return false; + map->first_ip = first_ip; + map->last_ip = last_ip; + map->timeout = IPSET_NO_TIMEOUT; + + set->data = map; + set->family = AF_INET; + + return true; +} + +static int +bitmap_ipmac_create(struct ip_set *set, struct nlattr *tb[], + u32 flags) +{ + u32 first_ip, last_ip, elements; + struct bitmap_ipmac *map; + int ret; + + if (unlikely(!tb[IPSET_ATTR_IP] || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP], &first_ip); + if (ret) + return ret; + + if (tb[IPSET_ATTR_IP_TO]) { + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP_TO], &last_ip); + if (ret) + return ret; + if (first_ip > last_ip) { + u32 tmp = first_ip; + + first_ip = last_ip; + last_ip = tmp; + } + } else if (tb[IPSET_ATTR_CIDR]) { + u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); + + if (cidr >= 32) + return -IPSET_ERR_INVALID_CIDR; + last_ip = first_ip | ~ip_set_hostmask(cidr); + } else + return -IPSET_ERR_PROTOCOL; + + elements = last_ip - first_ip + 1; + + if (elements > IPSET_BITMAP_MAX_RANGE + 1) + return -IPSET_ERR_BITMAP_RANGE_SIZE; + + map = kzalloc(sizeof(*map), GFP_KERNEL); + if (!map) + return -ENOMEM; + + if (tb[IPSET_ATTR_TIMEOUT]) { + map->dsize = sizeof(struct ipmac_telem); + + if (!init_map_ipmac(set, map, first_ip, last_ip)) { + kfree(map); + return -ENOMEM; + } + + map->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + + set->variant = &bitmap_tipmac; + + bitmap_ipmac_gc_init(set); + } else { + map->dsize = sizeof(struct ipmac_elem); + + if (!init_map_ipmac(set, map, first_ip, last_ip)) { + kfree(map); + return -ENOMEM; + } + set->variant = &bitmap_ipmac; + + } + return 0; +} + +static struct ip_set_type bitmap_ipmac_type = { + .name = "bitmap:ip,mac", + .protocol = IPSET_PROTOCOL, + .features = IPSET_TYPE_IP | IPSET_TYPE_MAC, + .dimension = IPSET_DIM_TWO, + .family = AF_INET, + .revision = 0, + .create = bitmap_ipmac_create, + .create_policy = { + [IPSET_ATTR_IP] = { .type = NLA_NESTED }, + [IPSET_ATTR_IP_TO] = { .type = NLA_NESTED }, + [IPSET_ATTR_CIDR] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + }, + .adt_policy = { + [IPSET_ATTR_IP] = { .type = NLA_NESTED }, + [IPSET_ATTR_ETHER] = { .type = NLA_BINARY, .len = ETH_ALEN }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + [IPSET_ATTR_LINENO] = { .type = NLA_U32 }, + }, + .me = THIS_MODULE, +}; + +static int __init +bitmap_ipmac_init(void) +{ + return ip_set_type_register(&bitmap_ipmac_type); +} + +static void __exit +bitmap_ipmac_fini(void) +{ + ip_set_type_unregister(&bitmap_ipmac_type); +} + +module_init(bitmap_ipmac_init); +module_exit(bitmap_ipmac_fini); -- cgit v1.2.3 From 543261907dc3c4e90845acfcd602ebdbfdfcb4f0 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 1 Feb 2011 15:37:04 +0100 Subject: netfilter: ipset; bitmap:port set type support The module implements the bitmap:port type in two flavours, without and with timeout support to store TCP/UDP ports from a range. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/ipset/Kconfig | 9 + net/netfilter/ipset/Makefile | 1 + net/netfilter/ipset/ip_set_bitmap_port.c | 520 +++++++++++++++++++++++++++++++ 3 files changed, 530 insertions(+) create mode 100644 net/netfilter/ipset/ip_set_bitmap_port.c (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/Kconfig b/net/netfilter/ipset/Kconfig index f18654c73af2..f401e9112703 100644 --- a/net/netfilter/ipset/Kconfig +++ b/net/netfilter/ipset/Kconfig @@ -41,4 +41,13 @@ config IP_SET_BITMAP_IPMAC To compile it as a module, choose M here. If unsure, say N. +config IP_SET_BITMAP_PORT + tristate "bitmap:port set support" + depends on IP_SET + help + This option adds the bitmap:port set type support, by which one + can store TCP/UDP port numbers from a range. + + To compile it as a module, choose M here. If unsure, say N. + endif # IP_SET diff --git a/net/netfilter/ipset/Makefile b/net/netfilter/ipset/Makefile index f7a099f40d0b..40866e267752 100644 --- a/net/netfilter/ipset/Makefile +++ b/net/netfilter/ipset/Makefile @@ -10,3 +10,4 @@ obj-$(CONFIG_IP_SET) += ip_set.o # bitmap types obj-$(CONFIG_IP_SET_BITMAP_IP) += ip_set_bitmap_ip.o obj-$(CONFIG_IP_SET_BITMAP_IPMAC) += ip_set_bitmap_ipmac.o +obj-$(CONFIG_IP_SET_BITMAP_PORT) += ip_set_bitmap_port.o diff --git a/net/netfilter/ipset/ip_set_bitmap_port.c b/net/netfilter/ipset/ip_set_bitmap_port.c new file mode 100644 index 000000000000..92074bb64134 --- /dev/null +++ b/net/netfilter/ipset/ip_set_bitmap_port.c @@ -0,0 +1,520 @@ +/* Copyright (C) 2003-2011 Jozsef Kadlecsik + * + * 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. + */ + +/* Kernel module implementing an IP set type: the bitmap:port type */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#define IP_SET_BITMAP_TIMEOUT +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Jozsef Kadlecsik "); +MODULE_DESCRIPTION("bitmap:port type of IP sets"); +MODULE_ALIAS("ip_set_bitmap:port"); + +/* Type structure */ +struct bitmap_port { + void *members; /* the set members */ + u16 first_port; /* host byte order, included in range */ + u16 last_port; /* host byte order, included in range */ + size_t memsize; /* members size */ + u32 timeout; /* timeout parameter */ + struct timer_list gc; /* garbage collection */ +}; + +/* Base variant */ + +static int +bitmap_port_test(struct ip_set *set, void *value, u32 timeout) +{ + const struct bitmap_port *map = set->data; + u16 id = *(u16 *)value; + + return !!test_bit(id, map->members); +} + +static int +bitmap_port_add(struct ip_set *set, void *value, u32 timeout) +{ + struct bitmap_port *map = set->data; + u16 id = *(u16 *)value; + + if (test_and_set_bit(id, map->members)) + return -IPSET_ERR_EXIST; + + return 0; +} + +static int +bitmap_port_del(struct ip_set *set, void *value, u32 timeout) +{ + struct bitmap_port *map = set->data; + u16 id = *(u16 *)value; + + if (!test_and_clear_bit(id, map->members)) + return -IPSET_ERR_EXIST; + + return 0; +} + +static int +bitmap_port_list(const struct ip_set *set, + struct sk_buff *skb, struct netlink_callback *cb) +{ + const struct bitmap_port *map = set->data; + struct nlattr *atd, *nested; + u16 id, first = cb->args[2]; + u16 last = map->last_port - map->first_port; + + atd = ipset_nest_start(skb, IPSET_ATTR_ADT); + if (!atd) + return -EMSGSIZE; + for (; cb->args[2] <= last; cb->args[2]++) { + id = cb->args[2]; + if (!test_bit(id, map->members)) + continue; + nested = ipset_nest_start(skb, IPSET_ATTR_DATA); + if (!nested) { + if (id == first) { + nla_nest_cancel(skb, atd); + return -EMSGSIZE; + } else + goto nla_put_failure; + } + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, + htons(map->first_port + id)); + ipset_nest_end(skb, nested); + } + ipset_nest_end(skb, atd); + /* Set listing finished */ + cb->args[2] = 0; + + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nested); + ipset_nest_end(skb, atd); + if (unlikely(id == first)) { + cb->args[2] = 0; + return -EMSGSIZE; + } + return 0; +} + +/* Timeout variant */ + +static int +bitmap_port_ttest(struct ip_set *set, void *value, u32 timeout) +{ + const struct bitmap_port *map = set->data; + const unsigned long *members = map->members; + u16 id = *(u16 *)value; + + return ip_set_timeout_test(members[id]); +} + +static int +bitmap_port_tadd(struct ip_set *set, void *value, u32 timeout) +{ + struct bitmap_port *map = set->data; + unsigned long *members = map->members; + u16 id = *(u16 *)value; + + if (ip_set_timeout_test(members[id])) + return -IPSET_ERR_EXIST; + + members[id] = ip_set_timeout_set(timeout); + + return 0; +} + +static int +bitmap_port_tdel(struct ip_set *set, void *value, u32 timeout) +{ + struct bitmap_port *map = set->data; + unsigned long *members = map->members; + u16 id = *(u16 *)value; + int ret = -IPSET_ERR_EXIST; + + if (ip_set_timeout_test(members[id])) + ret = 0; + + members[id] = IPSET_ELEM_UNSET; + return ret; +} + +static int +bitmap_port_tlist(const struct ip_set *set, + struct sk_buff *skb, struct netlink_callback *cb) +{ + const struct bitmap_port *map = set->data; + struct nlattr *adt, *nested; + u16 id, first = cb->args[2]; + u16 last = map->last_port - map->first_port; + const unsigned long *members = map->members; + + adt = ipset_nest_start(skb, IPSET_ATTR_ADT); + if (!adt) + return -EMSGSIZE; + for (; cb->args[2] <= last; cb->args[2]++) { + id = cb->args[2]; + if (!ip_set_timeout_test(members[id])) + continue; + nested = ipset_nest_start(skb, IPSET_ATTR_DATA); + if (!nested) { + if (id == first) { + nla_nest_cancel(skb, adt); + return -EMSGSIZE; + } else + goto nla_put_failure; + } + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, + htons(map->first_port + id)); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(members[id]))); + ipset_nest_end(skb, nested); + } + ipset_nest_end(skb, adt); + + /* Set listing finished */ + cb->args[2] = 0; + + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nested); + ipset_nest_end(skb, adt); + if (unlikely(id == first)) { + cb->args[2] = 0; + return -EMSGSIZE; + } + return 0; +} + +static int +bitmap_port_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + struct bitmap_port *map = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + __be16 __port; + u16 port = 0; + + if (!ip_set_get_ip_port(skb, pf, flags & IPSET_DIM_ONE_SRC, &__port)) + return -EINVAL; + + port = ntohs(__port); + + if (port < map->first_port || port > map->last_port) + return -IPSET_ERR_BITMAP_RANGE; + + port -= map->first_port; + + return adtfn(set, &port, map->timeout); +} + +static int +bitmap_port_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + struct bitmap_port *map = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + u32 timeout = map->timeout; + u32 port; /* wraparound */ + u16 id, port_to; + int ret = 0; + + if (unlikely(!ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + port = ip_set_get_h16(tb[IPSET_ATTR_PORT]); + if (port < map->first_port || port > map->last_port) + return -IPSET_ERR_BITMAP_RANGE; + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(map->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + if (adt == IPSET_TEST) { + id = port - map->first_port; + return adtfn(set, &id, timeout); + } + + if (tb[IPSET_ATTR_PORT_TO]) { + port_to = ip_set_get_h16(tb[IPSET_ATTR_PORT_TO]); + if (port > port_to) { + swap(port, port_to); + if (port < map->first_port) + return -IPSET_ERR_BITMAP_RANGE; + } + } else + port_to = port; + + if (port_to > map->last_port) + return -IPSET_ERR_BITMAP_RANGE; + + for (; port <= port_to; port++) { + id = port - map->first_port; + ret = adtfn(set, &id, timeout); + + if (ret && !ip_set_eexist(ret, flags)) + return ret; + else + ret = 0; + } + return ret; +} + +static void +bitmap_port_destroy(struct ip_set *set) +{ + struct bitmap_port *map = set->data; + + if (with_timeout(map->timeout)) + del_timer_sync(&map->gc); + + ip_set_free(map->members); + kfree(map); + + set->data = NULL; +} + +static void +bitmap_port_flush(struct ip_set *set) +{ + struct bitmap_port *map = set->data; + + memset(map->members, 0, map->memsize); +} + +static int +bitmap_port_head(struct ip_set *set, struct sk_buff *skb) +{ + const struct bitmap_port *map = set->data; + struct nlattr *nested; + + nested = ipset_nest_start(skb, IPSET_ATTR_DATA); + if (!nested) + goto nla_put_failure; + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, htons(map->first_port)); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT_TO, htons(map->last_port)); + NLA_PUT_NET32(skb, IPSET_ATTR_REFERENCES, + htonl(atomic_read(&set->ref) - 1)); + NLA_PUT_NET32(skb, IPSET_ATTR_MEMSIZE, + htonl(sizeof(*map) + map->memsize)); + if (with_timeout(map->timeout)) + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, htonl(map->timeout)); + ipset_nest_end(skb, nested); + + return 0; +nla_put_failure: + return -EMSGSIZE; +} + +static bool +bitmap_port_same_set(const struct ip_set *a, const struct ip_set *b) +{ + const struct bitmap_port *x = a->data; + const struct bitmap_port *y = b->data; + + return x->first_port == y->first_port && + x->last_port == y->last_port && + x->timeout == y->timeout; +} + +static const struct ip_set_type_variant bitmap_port = { + .kadt = bitmap_port_kadt, + .uadt = bitmap_port_uadt, + .adt = { + [IPSET_ADD] = bitmap_port_add, + [IPSET_DEL] = bitmap_port_del, + [IPSET_TEST] = bitmap_port_test, + }, + .destroy = bitmap_port_destroy, + .flush = bitmap_port_flush, + .head = bitmap_port_head, + .list = bitmap_port_list, + .same_set = bitmap_port_same_set, +}; + +static const struct ip_set_type_variant bitmap_tport = { + .kadt = bitmap_port_kadt, + .uadt = bitmap_port_uadt, + .adt = { + [IPSET_ADD] = bitmap_port_tadd, + [IPSET_DEL] = bitmap_port_tdel, + [IPSET_TEST] = bitmap_port_ttest, + }, + .destroy = bitmap_port_destroy, + .flush = bitmap_port_flush, + .head = bitmap_port_head, + .list = bitmap_port_tlist, + .same_set = bitmap_port_same_set, +}; + +static void +bitmap_port_gc(unsigned long ul_set) +{ + struct ip_set *set = (struct ip_set *) ul_set; + struct bitmap_port *map = set->data; + unsigned long *table = map->members; + u32 id; /* wraparound */ + u16 last = map->last_port - map->first_port; + + /* We run parallel with other readers (test element) + * but adding/deleting new entries is locked out */ + read_lock_bh(&set->lock); + for (id = 0; id <= last; id++) + if (ip_set_timeout_expired(table[id])) + table[id] = IPSET_ELEM_UNSET; + read_unlock_bh(&set->lock); + + map->gc.expires = jiffies + IPSET_GC_PERIOD(map->timeout) * HZ; + add_timer(&map->gc); +} + +static void +bitmap_port_gc_init(struct ip_set *set) +{ + struct bitmap_port *map = set->data; + + init_timer(&map->gc); + map->gc.data = (unsigned long) set; + map->gc.function = bitmap_port_gc; + map->gc.expires = jiffies + IPSET_GC_PERIOD(map->timeout) * HZ; + add_timer(&map->gc); +} + +/* Create bitmap:ip type of sets */ + +static bool +init_map_port(struct ip_set *set, struct bitmap_port *map, + u16 first_port, u16 last_port) +{ + map->members = ip_set_alloc(map->memsize); + if (!map->members) + return false; + map->first_port = first_port; + map->last_port = last_port; + map->timeout = IPSET_NO_TIMEOUT; + + set->data = map; + set->family = AF_UNSPEC; + + return true; +} + +static int +bitmap_port_create(struct ip_set *set, struct nlattr *tb[], + u32 flags) +{ + struct bitmap_port *map; + u16 first_port, last_port; + + if (unlikely(!ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || + !ip_set_attr_netorder(tb, IPSET_ATTR_PORT_TO) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + first_port = ip_set_get_h16(tb[IPSET_ATTR_PORT]); + last_port = ip_set_get_h16(tb[IPSET_ATTR_PORT_TO]); + if (first_port > last_port) { + u16 tmp = first_port; + + first_port = last_port; + last_port = tmp; + } + + map = kzalloc(sizeof(*map), GFP_KERNEL); + if (!map) + return -ENOMEM; + + if (tb[IPSET_ATTR_TIMEOUT]) { + map->memsize = (last_port - first_port + 1) + * sizeof(unsigned long); + + if (!init_map_port(set, map, first_port, last_port)) { + kfree(map); + return -ENOMEM; + } + + map->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + set->variant = &bitmap_tport; + + bitmap_port_gc_init(set); + } else { + map->memsize = bitmap_bytes(0, last_port - first_port); + pr_debug("memsize: %zu\n", map->memsize); + if (!init_map_port(set, map, first_port, last_port)) { + kfree(map); + return -ENOMEM; + } + + set->variant = &bitmap_port; + } + return 0; +} + +static struct ip_set_type bitmap_port_type = { + .name = "bitmap:port", + .protocol = IPSET_PROTOCOL, + .features = IPSET_TYPE_PORT, + .dimension = IPSET_DIM_ONE, + .family = AF_UNSPEC, + .revision = 0, + .create = bitmap_port_create, + .create_policy = { + [IPSET_ATTR_PORT] = { .type = NLA_U16 }, + [IPSET_ATTR_PORT_TO] = { .type = NLA_U16 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + }, + .adt_policy = { + [IPSET_ATTR_PORT] = { .type = NLA_U16 }, + [IPSET_ATTR_PORT_TO] = { .type = NLA_U16 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + [IPSET_ATTR_LINENO] = { .type = NLA_U32 }, + }, + .me = THIS_MODULE, +}; + +static int __init +bitmap_port_init(void) +{ + return ip_set_type_register(&bitmap_port_type); +} + +static void __exit +bitmap_port_fini(void) +{ + ip_set_type_unregister(&bitmap_port_type); +} + +module_init(bitmap_port_init); +module_exit(bitmap_port_fini); -- cgit v1.2.3 From 6c027889696a7a694b0e2f6e3cabadefec7553b6 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 1 Feb 2011 15:38:36 +0100 Subject: netfilter: ipset: hash:ip set type support The module implements the hash:ip type support in four flavours: for IPv4 or IPv6, both without and with timeout support. All the hash types are based on the "array hash" or ahash structure and functions as a good compromise between minimal memory footprint and speed. The hashing uses arrays to resolve clashes. The hash table is resized (doubled) when searching becomes too long. Resizing can be triggered by userspace add commands only and those are serialized by the nfnl mutex. During resizing the set is read-locked, so the only possible concurrent operations are the kernel side readers. Those are protected by RCU locking. Because of the four flavours and the other hash types, the functions are implemented in general forms in the ip_set_ahash.h header file and the real functions are generated before compiling by macro expansion. Thus the dereferencing of low-level functions and void pointer arguments could be avoided: the low-level functions are inlined, the function arguments are pointers of type-specific structures. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/ipset/Kconfig | 10 + net/netfilter/ipset/Makefile | 3 + net/netfilter/ipset/ip_set_hash_ip.c | 467 +++++++++++++++++++++++++++++++++++ 3 files changed, 480 insertions(+) create mode 100644 net/netfilter/ipset/ip_set_hash_ip.c (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/Kconfig b/net/netfilter/ipset/Kconfig index f401e9112703..194d89caeb3d 100644 --- a/net/netfilter/ipset/Kconfig +++ b/net/netfilter/ipset/Kconfig @@ -50,4 +50,14 @@ config IP_SET_BITMAP_PORT To compile it as a module, choose M here. If unsure, say N. +config IP_SET_HASH_IP + tristate "hash:ip set support" + depends on IP_SET + help + This option adds the hash:ip set type support, by which one + can store arbitrary IPv4 or IPv6 addresses (or network addresses) + in a set. + + To compile it as a module, choose M here. If unsure, say N. + endif # IP_SET diff --git a/net/netfilter/ipset/Makefile b/net/netfilter/ipset/Makefile index 40866e267752..5cbf00c41a98 100644 --- a/net/netfilter/ipset/Makefile +++ b/net/netfilter/ipset/Makefile @@ -11,3 +11,6 @@ obj-$(CONFIG_IP_SET) += ip_set.o obj-$(CONFIG_IP_SET_BITMAP_IP) += ip_set_bitmap_ip.o obj-$(CONFIG_IP_SET_BITMAP_IPMAC) += ip_set_bitmap_ipmac.o obj-$(CONFIG_IP_SET_BITMAP_PORT) += ip_set_bitmap_port.o + +# hash types +obj-$(CONFIG_IP_SET_HASH_IP) += ip_set_hash_ip.o diff --git a/net/netfilter/ipset/ip_set_hash_ip.c b/net/netfilter/ipset/ip_set_hash_ip.c new file mode 100644 index 000000000000..53964bc831aa --- /dev/null +++ b/net/netfilter/ipset/ip_set_hash_ip.c @@ -0,0 +1,467 @@ +/* Copyright (C) 2003-2011 Jozsef Kadlecsik + * + * 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. + */ + +/* Kernel module implementing an IP set type: the hash:ip type */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Jozsef Kadlecsik "); +MODULE_DESCRIPTION("hash:ip type of IP sets"); +MODULE_ALIAS("ip_set_hash:ip"); + +/* Type specific function prefix */ +#define TYPE hash_ip + +static bool +hash_ip_same_set(const struct ip_set *a, const struct ip_set *b); + +#define hash_ip4_same_set hash_ip_same_set +#define hash_ip6_same_set hash_ip_same_set + +/* The type variant functions: IPv4 */ + +/* Member elements without timeout */ +struct hash_ip4_elem { + __be32 ip; +}; + +/* Member elements with timeout support */ +struct hash_ip4_telem { + __be32 ip; + unsigned long timeout; +}; + +static inline bool +hash_ip4_data_equal(const struct hash_ip4_elem *ip1, + const struct hash_ip4_elem *ip2) +{ + return ip1->ip == ip2->ip; +} + +static inline bool +hash_ip4_data_isnull(const struct hash_ip4_elem *elem) +{ + return elem->ip == 0; +} + +static inline void +hash_ip4_data_copy(struct hash_ip4_elem *dst, const struct hash_ip4_elem *src) +{ + dst->ip = src->ip; +} + +/* Zero valued IP addresses cannot be stored */ +static inline void +hash_ip4_data_zero_out(struct hash_ip4_elem *elem) +{ + elem->ip = 0; +} + +static inline bool +hash_ip4_data_list(struct sk_buff *skb, const struct hash_ip4_elem *data) +{ + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, data->ip); + return 0; + +nla_put_failure: + return 1; +} + +static bool +hash_ip4_data_tlist(struct sk_buff *skb, const struct hash_ip4_elem *data) +{ + const struct hash_ip4_telem *tdata = + (const struct hash_ip4_telem *)data; + + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, tdata->ip); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(tdata->timeout))); + + return 0; + +nla_put_failure: + return 1; +} + +#define IP_SET_HASH_WITH_NETMASK +#define PF 4 +#define HOST_MASK 32 +#include + +static int +hash_ip4_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + __be32 ip; + + ip4addrptr(skb, flags & IPSET_DIM_ONE_SRC, &ip); + ip &= ip_set_netmask(h->netmask); + if (ip == 0) + return -EINVAL; + + return adtfn(set, &ip, h->timeout); +} + +static int +hash_ip4_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + u32 ip, ip_to, hosts, timeout = h->timeout; + __be32 nip; + int ret = 0; + + if (unlikely(!tb[IPSET_ATTR_IP] || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP], &ip); + if (ret) + return ret; + + ip &= ip_set_hostmask(h->netmask); + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(h->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + if (adt == IPSET_TEST) { + nip = htonl(ip); + if (nip == 0) + return -IPSET_ERR_HASH_ELEM; + return adtfn(set, &nip, timeout); + } + + if (tb[IPSET_ATTR_IP_TO]) { + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP_TO], &ip_to); + if (ret) + return ret; + if (ip > ip_to) + swap(ip, ip_to); + } else if (tb[IPSET_ATTR_CIDR]) { + u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); + + if (cidr > 32) + return -IPSET_ERR_INVALID_CIDR; + ip &= ip_set_hostmask(cidr); + ip_to = ip | ~ip_set_hostmask(cidr); + } else + ip_to = ip; + + hosts = h->netmask == 32 ? 1 : 2 << (32 - h->netmask - 1); + + for (; !before(ip_to, ip); ip += hosts) { + nip = htonl(ip); + if (nip == 0) + return -IPSET_ERR_HASH_ELEM; + ret = adtfn(set, &nip, timeout); + + if (ret && !ip_set_eexist(ret, flags)) + return ret; + else + ret = 0; + } + return ret; +} + +static bool +hash_ip_same_set(const struct ip_set *a, const struct ip_set *b) +{ + const struct ip_set_hash *x = a->data; + const struct ip_set_hash *y = b->data; + + /* Resizing changes htable_bits, so we ignore it */ + return x->maxelem == y->maxelem && + x->timeout == y->timeout && + x->netmask == y->netmask; +} + +/* The type variant functions: IPv6 */ + +struct hash_ip6_elem { + union nf_inet_addr ip; +}; + +struct hash_ip6_telem { + union nf_inet_addr ip; + unsigned long timeout; +}; + +static inline bool +hash_ip6_data_equal(const struct hash_ip6_elem *ip1, + const struct hash_ip6_elem *ip2) +{ + return ipv6_addr_cmp(&ip1->ip.in6, &ip2->ip.in6) == 0; +} + +static inline bool +hash_ip6_data_isnull(const struct hash_ip6_elem *elem) +{ + return ipv6_addr_any(&elem->ip.in6); +} + +static inline void +hash_ip6_data_copy(struct hash_ip6_elem *dst, const struct hash_ip6_elem *src) +{ + ipv6_addr_copy(&dst->ip.in6, &src->ip.in6); +} + +static inline void +hash_ip6_data_zero_out(struct hash_ip6_elem *elem) +{ + ipv6_addr_set(&elem->ip.in6, 0, 0, 0, 0); +} + +static inline void +ip6_netmask(union nf_inet_addr *ip, u8 prefix) +{ + ip->ip6[0] &= ip_set_netmask6(prefix)[0]; + ip->ip6[1] &= ip_set_netmask6(prefix)[1]; + ip->ip6[2] &= ip_set_netmask6(prefix)[2]; + ip->ip6[3] &= ip_set_netmask6(prefix)[3]; +} + +static bool +hash_ip6_data_list(struct sk_buff *skb, const struct hash_ip6_elem *data) +{ + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP, &data->ip); + return 0; + +nla_put_failure: + return 1; +} + +static bool +hash_ip6_data_tlist(struct sk_buff *skb, const struct hash_ip6_elem *data) +{ + const struct hash_ip6_telem *e = + (const struct hash_ip6_telem *)data; + + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP, &e->ip); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(e->timeout))); + return 0; + +nla_put_failure: + return 1; +} + +#undef PF +#undef HOST_MASK + +#define PF 6 +#define HOST_MASK 128 +#include + +static int +hash_ip6_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + union nf_inet_addr ip; + + ip6addrptr(skb, flags & IPSET_DIM_ONE_SRC, &ip.in6); + ip6_netmask(&ip, h->netmask); + if (ipv6_addr_any(&ip.in6)) + return -EINVAL; + + return adtfn(set, &ip, h->timeout); +} + +static const struct nla_policy hash_ip6_adt_policy[IPSET_ATTR_ADT_MAX + 1] = { + [IPSET_ATTR_IP] = { .type = NLA_NESTED }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + [IPSET_ATTR_LINENO] = { .type = NLA_U32 }, +}; + +static int +hash_ip6_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + union nf_inet_addr ip; + u32 timeout = h->timeout; + int ret; + + if (unlikely(!tb[IPSET_ATTR_IP] || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || + tb[IPSET_ATTR_IP_TO] || + tb[IPSET_ATTR_CIDR])) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &ip); + if (ret) + return ret; + + ip6_netmask(&ip, h->netmask); + if (ipv6_addr_any(&ip.in6)) + return -IPSET_ERR_HASH_ELEM; + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(h->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + ret = adtfn(set, &ip, timeout); + + return ip_set_eexist(ret, flags) ? 0 : ret; +} + +/* Create hash:ip type of sets */ + +static int +hash_ip_create(struct ip_set *set, struct nlattr *tb[], u32 flags) +{ + u32 hashsize = IPSET_DEFAULT_HASHSIZE, maxelem = IPSET_DEFAULT_MAXELEM; + u8 netmask, hbits; + struct ip_set_hash *h; + + if (!(set->family == AF_INET || set->family == AF_INET6)) + return -IPSET_ERR_INVALID_FAMILY; + netmask = set->family == AF_INET ? 32 : 128; + pr_debug("Create set %s with family %s\n", + set->name, set->family == AF_INET ? "inet" : "inet6"); + + if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_HASHSIZE) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_MAXELEM) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_HASHSIZE]) { + hashsize = ip_set_get_h32(tb[IPSET_ATTR_HASHSIZE]); + if (hashsize < IPSET_MIMINAL_HASHSIZE) + hashsize = IPSET_MIMINAL_HASHSIZE; + } + + if (tb[IPSET_ATTR_MAXELEM]) + maxelem = ip_set_get_h32(tb[IPSET_ATTR_MAXELEM]); + + if (tb[IPSET_ATTR_NETMASK]) { + netmask = nla_get_u8(tb[IPSET_ATTR_NETMASK]); + + if ((set->family == AF_INET && netmask > 32) || + (set->family == AF_INET6 && netmask > 128) || + netmask == 0) + return -IPSET_ERR_INVALID_NETMASK; + } + + h = kzalloc(sizeof(*h), GFP_KERNEL); + if (!h) + return -ENOMEM; + + h->maxelem = maxelem; + h->netmask = netmask; + get_random_bytes(&h->initval, sizeof(h->initval)); + h->timeout = IPSET_NO_TIMEOUT; + + hbits = htable_bits(hashsize); + h->table = ip_set_alloc( + sizeof(struct htable) + + jhash_size(hbits) * sizeof(struct hbucket)); + if (!h->table) { + kfree(h); + return -ENOMEM; + } + h->table->htable_bits = hbits; + + set->data = h; + + if (tb[IPSET_ATTR_TIMEOUT]) { + h->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + + set->variant = set->family == AF_INET + ? &hash_ip4_tvariant : &hash_ip6_tvariant; + + if (set->family == AF_INET) + hash_ip4_gc_init(set); + else + hash_ip6_gc_init(set); + } else { + set->variant = set->family == AF_INET + ? &hash_ip4_variant : &hash_ip6_variant; + } + + pr_debug("create %s hashsize %u (%u) maxelem %u: %p(%p)\n", + set->name, jhash_size(h->table->htable_bits), + h->table->htable_bits, h->maxelem, set->data, h->table); + + return 0; +} + +static struct ip_set_type hash_ip_type __read_mostly = { + .name = "hash:ip", + .protocol = IPSET_PROTOCOL, + .features = IPSET_TYPE_IP, + .dimension = IPSET_DIM_ONE, + .family = AF_UNSPEC, + .revision = 0, + .create = hash_ip_create, + .create_policy = { + [IPSET_ATTR_HASHSIZE] = { .type = NLA_U32 }, + [IPSET_ATTR_MAXELEM] = { .type = NLA_U32 }, + [IPSET_ATTR_PROBES] = { .type = NLA_U8 }, + [IPSET_ATTR_RESIZE] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + [IPSET_ATTR_NETMASK] = { .type = NLA_U8 }, + }, + .adt_policy = { + [IPSET_ATTR_IP] = { .type = NLA_NESTED }, + [IPSET_ATTR_IP_TO] = { .type = NLA_NESTED }, + [IPSET_ATTR_CIDR] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + [IPSET_ATTR_LINENO] = { .type = NLA_U32 }, + }, + .me = THIS_MODULE, +}; + +static int __init +hash_ip_init(void) +{ + return ip_set_type_register(&hash_ip_type); +} + +static void __exit +hash_ip_fini(void) +{ + ip_set_type_unregister(&hash_ip_type); +} + +module_init(hash_ip_init); +module_exit(hash_ip_fini); -- cgit v1.2.3 From 07896ed37b94599a1b8ea97f4bd5766be71390f4 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 1 Feb 2011 15:39:52 +0100 Subject: netfilter: ipset: hash:ip,port set type support The module implements the hash:ip,port type support in four flavours: for IPv4 and IPv6, both without and with timeout support. The elements are two dimensional: IPv4/IPv6 address and protocol/port pairs. The port is interpeted for TCP, UPD, ICMP and ICMPv6 (at the latters as type/code of course). Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/ipset/Kconfig | 9 + net/netfilter/ipset/Makefile | 1 + net/netfilter/ipset/ip_set_hash_ipport.c | 547 +++++++++++++++++++++++++++++++ 3 files changed, 557 insertions(+) create mode 100644 net/netfilter/ipset/ip_set_hash_ipport.c (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/Kconfig b/net/netfilter/ipset/Kconfig index 194d89caeb3d..325b5312a47d 100644 --- a/net/netfilter/ipset/Kconfig +++ b/net/netfilter/ipset/Kconfig @@ -60,4 +60,13 @@ config IP_SET_HASH_IP To compile it as a module, choose M here. If unsure, say N. +config IP_SET_HASH_IPPORT + tristate "hash:ip,port set support" + depends on IP_SET + help + This option adds the hash:ip,port set type support, by which one + can store IPv4/IPv6 address and protocol/port pairs. + + To compile it as a module, choose M here. If unsure, say N. + endif # IP_SET diff --git a/net/netfilter/ipset/Makefile b/net/netfilter/ipset/Makefile index 5cbf00c41a98..6a3663e5cef1 100644 --- a/net/netfilter/ipset/Makefile +++ b/net/netfilter/ipset/Makefile @@ -14,3 +14,4 @@ obj-$(CONFIG_IP_SET_BITMAP_PORT) += ip_set_bitmap_port.o # hash types obj-$(CONFIG_IP_SET_HASH_IP) += ip_set_hash_ip.o +obj-$(CONFIG_IP_SET_HASH_IPPORT) += ip_set_hash_ipport.o diff --git a/net/netfilter/ipset/ip_set_hash_ipport.c b/net/netfilter/ipset/ip_set_hash_ipport.c new file mode 100644 index 000000000000..d9b192818e58 --- /dev/null +++ b/net/netfilter/ipset/ip_set_hash_ipport.c @@ -0,0 +1,547 @@ +/* Copyright (C) 2003-2011 Jozsef Kadlecsik + * + * 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. + */ + +/* Kernel module implementing an IP set type: the hash:ip,port type */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Jozsef Kadlecsik "); +MODULE_DESCRIPTION("hash:ip,port type of IP sets"); +MODULE_ALIAS("ip_set_hash:ip,port"); + +/* Type specific function prefix */ +#define TYPE hash_ipport + +static bool +hash_ipport_same_set(const struct ip_set *a, const struct ip_set *b); + +#define hash_ipport4_same_set hash_ipport_same_set +#define hash_ipport6_same_set hash_ipport_same_set + +/* The type variant functions: IPv4 */ + +/* Member elements without timeout */ +struct hash_ipport4_elem { + __be32 ip; + __be16 port; + u8 proto; + u8 padding; +}; + +/* Member elements with timeout support */ +struct hash_ipport4_telem { + __be32 ip; + __be16 port; + u8 proto; + u8 padding; + unsigned long timeout; +}; + +static inline bool +hash_ipport4_data_equal(const struct hash_ipport4_elem *ip1, + const struct hash_ipport4_elem *ip2) +{ + return ip1->ip == ip2->ip && + ip1->port == ip2->port && + ip1->proto == ip2->proto; +} + +static inline bool +hash_ipport4_data_isnull(const struct hash_ipport4_elem *elem) +{ + return elem->proto == 0; +} + +static inline void +hash_ipport4_data_copy(struct hash_ipport4_elem *dst, + const struct hash_ipport4_elem *src) +{ + dst->ip = src->ip; + dst->port = src->port; + dst->proto = src->proto; +} + +static inline void +hash_ipport4_data_zero_out(struct hash_ipport4_elem *elem) +{ + elem->proto = 0; +} + +static bool +hash_ipport4_data_list(struct sk_buff *skb, + const struct hash_ipport4_elem *data) +{ + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, data->ip); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, data->port); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + return 0; + +nla_put_failure: + return 1; +} + +static bool +hash_ipport4_data_tlist(struct sk_buff *skb, + const struct hash_ipport4_elem *data) +{ + const struct hash_ipport4_telem *tdata = + (const struct hash_ipport4_telem *)data; + + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, tdata->ip); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, tdata->port); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(tdata->timeout))); + + return 0; + +nla_put_failure: + return 1; +} + +#define PF 4 +#define HOST_MASK 32 +#include + +static int +hash_ipport4_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_ipport4_elem data = { }; + + if (!ip_set_get_ip4_port(skb, flags & IPSET_DIM_TWO_SRC, + &data.port, &data.proto)) + return -EINVAL; + + ip4addrptr(skb, flags & IPSET_DIM_ONE_SRC, &data.ip); + + return adtfn(set, &data, h->timeout); +} + +static int +hash_ipport4_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_ipport4_elem data = { }; + u32 ip, ip_to, p, port, port_to; + u32 timeout = h->timeout; + int ret; + + if (unlikely(!tb[IPSET_ATTR_IP] || + !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + ret = ip_set_get_ipaddr4(tb[IPSET_ATTR_IP], &data.ip); + if (ret) + return ret; + + if (tb[IPSET_ATTR_PORT]) + data.port = nla_get_be16(tb[IPSET_ATTR_PORT]); + else + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_PROTO]) { + data.proto = nla_get_u8(tb[IPSET_ATTR_PROTO]); + + if (data.proto == 0) + return -IPSET_ERR_INVALID_PROTO; + } else + return -IPSET_ERR_MISSING_PROTO; + + switch (data.proto) { + case IPPROTO_UDP: + case IPPROTO_TCP: + case IPPROTO_ICMP: + break; + default: + data.port = 0; + break; + } + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(h->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + if (adt == IPSET_TEST || + !(data.proto == IPPROTO_TCP || data.proto == IPPROTO_UDP) || + !(tb[IPSET_ATTR_IP_TO] || tb[IPSET_ATTR_CIDR] || + tb[IPSET_ATTR_PORT_TO])) { + ret = adtfn(set, &data, timeout); + return ip_set_eexist(ret, flags) ? 0 : ret; + } + + ip = ntohl(data.ip); + if (tb[IPSET_ATTR_IP_TO]) { + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP_TO], &ip_to); + if (ret) + return ret; + if (ip > ip_to) + swap(ip, ip_to); + } else if (tb[IPSET_ATTR_CIDR]) { + u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); + + if (cidr > 32) + return -IPSET_ERR_INVALID_CIDR; + ip &= ip_set_hostmask(cidr); + ip_to = ip | ~ip_set_hostmask(cidr); + } else + ip_to = ip; + + port = ntohs(data.port); + if (tb[IPSET_ATTR_PORT_TO]) { + port_to = ip_set_get_h16(tb[IPSET_ATTR_PORT_TO]); + if (port > port_to) + swap(port, port_to); + } else + port_to = port; + + for (; !before(ip_to, ip); ip++) + for (p = port; p <= port_to; p++) { + data.ip = htonl(ip); + data.port = htons(p); + ret = adtfn(set, &data, timeout); + + if (ret && !ip_set_eexist(ret, flags)) + return ret; + else + ret = 0; + } + return ret; +} + +static bool +hash_ipport_same_set(const struct ip_set *a, const struct ip_set *b) +{ + const struct ip_set_hash *x = a->data; + const struct ip_set_hash *y = b->data; + + /* Resizing changes htable_bits, so we ignore it */ + return x->maxelem == y->maxelem && + x->timeout == y->timeout; +} + +/* The type variant functions: IPv6 */ + +struct hash_ipport6_elem { + union nf_inet_addr ip; + __be16 port; + u8 proto; + u8 padding; +}; + +struct hash_ipport6_telem { + union nf_inet_addr ip; + __be16 port; + u8 proto; + u8 padding; + unsigned long timeout; +}; + +static inline bool +hash_ipport6_data_equal(const struct hash_ipport6_elem *ip1, + const struct hash_ipport6_elem *ip2) +{ + return ipv6_addr_cmp(&ip1->ip.in6, &ip2->ip.in6) == 0 && + ip1->port == ip2->port && + ip1->proto == ip2->proto; +} + +static inline bool +hash_ipport6_data_isnull(const struct hash_ipport6_elem *elem) +{ + return elem->proto == 0; +} + +static inline void +hash_ipport6_data_copy(struct hash_ipport6_elem *dst, + const struct hash_ipport6_elem *src) +{ + memcpy(dst, src, sizeof(*dst)); +} + +static inline void +hash_ipport6_data_zero_out(struct hash_ipport6_elem *elem) +{ + elem->proto = 0; +} + +static bool +hash_ipport6_data_list(struct sk_buff *skb, + const struct hash_ipport6_elem *data) +{ + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP, &data->ip); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, data->port); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + return 0; + +nla_put_failure: + return 1; +} + +static bool +hash_ipport6_data_tlist(struct sk_buff *skb, + const struct hash_ipport6_elem *data) +{ + const struct hash_ipport6_telem *e = + (const struct hash_ipport6_telem *)data; + + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP, &e->ip); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, data->port); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(e->timeout))); + return 0; + +nla_put_failure: + return 1; +} + +#undef PF +#undef HOST_MASK + +#define PF 6 +#define HOST_MASK 128 +#include + +static int +hash_ipport6_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_ipport6_elem data = { }; + + if (!ip_set_get_ip6_port(skb, flags & IPSET_DIM_TWO_SRC, + &data.port, &data.proto)) + return -EINVAL; + + ip6addrptr(skb, flags & IPSET_DIM_ONE_SRC, &data.ip.in6); + + return adtfn(set, &data, h->timeout); +} + +static int +hash_ipport6_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_ipport6_elem data = { }; + u32 port, port_to; + u32 timeout = h->timeout; + int ret; + + if (unlikely(!tb[IPSET_ATTR_IP] || + !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || + tb[IPSET_ATTR_IP_TO] || + tb[IPSET_ATTR_CIDR])) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &data.ip); + if (ret) + return ret; + + if (tb[IPSET_ATTR_PORT]) + data.port = nla_get_be16(tb[IPSET_ATTR_PORT]); + else + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_PROTO]) { + data.proto = nla_get_u8(tb[IPSET_ATTR_PROTO]); + + if (data.proto == 0) + return -IPSET_ERR_INVALID_PROTO; + } else + return -IPSET_ERR_MISSING_PROTO; + + switch (data.proto) { + case IPPROTO_UDP: + case IPPROTO_TCP: + case IPPROTO_ICMPV6: + break; + default: + data.port = 0; + break; + } + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(h->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + if (adt == IPSET_TEST || + !(data.proto == IPPROTO_TCP || data.proto == IPPROTO_UDP) || + !tb[IPSET_ATTR_PORT_TO]) { + ret = adtfn(set, &data, timeout); + return ip_set_eexist(ret, flags) ? 0 : ret; + } + + port = ntohs(data.port); + port_to = ip_set_get_h16(tb[IPSET_ATTR_PORT_TO]); + if (port > port_to) + swap(port, port_to); + + for (; port <= port_to; port++) { + data.port = htons(port); + ret = adtfn(set, &data, timeout); + + if (ret && !ip_set_eexist(ret, flags)) + return ret; + else + ret = 0; + } + return ret; +} + +/* Create hash:ip type of sets */ + +static int +hash_ipport_create(struct ip_set *set, struct nlattr *tb[], u32 flags) +{ + struct ip_set_hash *h; + u32 hashsize = IPSET_DEFAULT_HASHSIZE, maxelem = IPSET_DEFAULT_MAXELEM; + u8 hbits; + + if (!(set->family == AF_INET || set->family == AF_INET6)) + return -IPSET_ERR_INVALID_FAMILY; + + if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_HASHSIZE) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_MAXELEM) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_HASHSIZE]) { + hashsize = ip_set_get_h32(tb[IPSET_ATTR_HASHSIZE]); + if (hashsize < IPSET_MIMINAL_HASHSIZE) + hashsize = IPSET_MIMINAL_HASHSIZE; + } + + if (tb[IPSET_ATTR_MAXELEM]) + maxelem = ip_set_get_h32(tb[IPSET_ATTR_MAXELEM]); + + h = kzalloc(sizeof(*h), GFP_KERNEL); + if (!h) + return -ENOMEM; + + h->maxelem = maxelem; + get_random_bytes(&h->initval, sizeof(h->initval)); + h->timeout = IPSET_NO_TIMEOUT; + + hbits = htable_bits(hashsize); + h->table = ip_set_alloc( + sizeof(struct htable) + + jhash_size(hbits) * sizeof(struct hbucket)); + if (!h->table) { + kfree(h); + return -ENOMEM; + } + h->table->htable_bits = hbits; + + set->data = h; + + if (tb[IPSET_ATTR_TIMEOUT]) { + h->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + + set->variant = set->family == AF_INET + ? &hash_ipport4_tvariant : &hash_ipport6_tvariant; + + if (set->family == AF_INET) + hash_ipport4_gc_init(set); + else + hash_ipport6_gc_init(set); + } else { + set->variant = set->family == AF_INET + ? &hash_ipport4_variant : &hash_ipport6_variant; + } + + pr_debug("create %s hashsize %u (%u) maxelem %u: %p(%p)\n", + set->name, jhash_size(h->table->htable_bits), + h->table->htable_bits, h->maxelem, set->data, h->table); + + return 0; +} + +static struct ip_set_type hash_ipport_type __read_mostly = { + .name = "hash:ip,port", + .protocol = IPSET_PROTOCOL, + .features = IPSET_TYPE_IP | IPSET_TYPE_PORT, + .dimension = IPSET_DIM_TWO, + .family = AF_UNSPEC, + .revision = 0, + .create = hash_ipport_create, + .create_policy = { + [IPSET_ATTR_HASHSIZE] = { .type = NLA_U32 }, + [IPSET_ATTR_MAXELEM] = { .type = NLA_U32 }, + [IPSET_ATTR_PROBES] = { .type = NLA_U8 }, + [IPSET_ATTR_RESIZE] = { .type = NLA_U8 }, + [IPSET_ATTR_PROTO] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + }, + .adt_policy = { + [IPSET_ATTR_IP] = { .type = NLA_NESTED }, + [IPSET_ATTR_IP_TO] = { .type = NLA_NESTED }, + [IPSET_ATTR_PORT] = { .type = NLA_U16 }, + [IPSET_ATTR_PORT_TO] = { .type = NLA_U16 }, + [IPSET_ATTR_CIDR] = { .type = NLA_U8 }, + [IPSET_ATTR_PROTO] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + [IPSET_ATTR_LINENO] = { .type = NLA_U32 }, + }, + .me = THIS_MODULE, +}; + +static int __init +hash_ipport_init(void) +{ + return ip_set_type_register(&hash_ipport_type); +} + +static void __exit +hash_ipport_fini(void) +{ + ip_set_type_unregister(&hash_ipport_type); +} + +module_init(hash_ipport_init); +module_exit(hash_ipport_fini); -- cgit v1.2.3 From 5663bc30e6114b6ba88cc428619ede46a3246a7b Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 1 Feb 2011 15:41:26 +0100 Subject: netfilter: ipset: hash:ip,port,ip set type support The module implements the hash:ip,port,ip type support in four flavours: for IPv4 and IPv6, both without and with timeout support. The elements are three dimensional: IPv4/IPv6 address, protocol/port and IPv4/IPv6 address triples. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/ipset/Kconfig | 10 + net/netfilter/ipset/Makefile | 1 + net/netfilter/ipset/ip_set_hash_ipportip.c | 565 +++++++++++++++++++++++++++++ 3 files changed, 576 insertions(+) create mode 100644 net/netfilter/ipset/ip_set_hash_ipportip.c (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/Kconfig b/net/netfilter/ipset/Kconfig index 325b5312a47d..e69355334252 100644 --- a/net/netfilter/ipset/Kconfig +++ b/net/netfilter/ipset/Kconfig @@ -69,4 +69,14 @@ config IP_SET_HASH_IPPORT To compile it as a module, choose M here. If unsure, say N. +config IP_SET_HASH_IPPORTIP + tristate "hash:ip,port,ip set support" + depends on IP_SET + help + This option adds the hash:ip,port,ip set type support, by which + one can store IPv4/IPv6 address, protocol/port, and IPv4/IPv6 + address triples in a set. + + To compile it as a module, choose M here. If unsure, say N. + endif # IP_SET diff --git a/net/netfilter/ipset/Makefile b/net/netfilter/ipset/Makefile index 6a3663e5cef1..e9ddb25189d0 100644 --- a/net/netfilter/ipset/Makefile +++ b/net/netfilter/ipset/Makefile @@ -15,3 +15,4 @@ obj-$(CONFIG_IP_SET_BITMAP_PORT) += ip_set_bitmap_port.o # hash types obj-$(CONFIG_IP_SET_HASH_IP) += ip_set_hash_ip.o obj-$(CONFIG_IP_SET_HASH_IPPORT) += ip_set_hash_ipport.o +obj-$(CONFIG_IP_SET_HASH_IPPORTIP) += ip_set_hash_ipportip.o diff --git a/net/netfilter/ipset/ip_set_hash_ipportip.c b/net/netfilter/ipset/ip_set_hash_ipportip.c new file mode 100644 index 000000000000..80dae9de5d18 --- /dev/null +++ b/net/netfilter/ipset/ip_set_hash_ipportip.c @@ -0,0 +1,565 @@ +/* Copyright (C) 2003-2011 Jozsef Kadlecsik + * + * 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. + */ + +/* Kernel module implementing an IP set type: the hash:ip,port,ip type */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Jozsef Kadlecsik "); +MODULE_DESCRIPTION("hash:ip,port,ip type of IP sets"); +MODULE_ALIAS("ip_set_hash:ip,port,ip"); + +/* Type specific function prefix */ +#define TYPE hash_ipportip + +static bool +hash_ipportip_same_set(const struct ip_set *a, const struct ip_set *b); + +#define hash_ipportip4_same_set hash_ipportip_same_set +#define hash_ipportip6_same_set hash_ipportip_same_set + +/* The type variant functions: IPv4 */ + +/* Member elements without timeout */ +struct hash_ipportip4_elem { + __be32 ip; + __be32 ip2; + __be16 port; + u8 proto; + u8 padding; +}; + +/* Member elements with timeout support */ +struct hash_ipportip4_telem { + __be32 ip; + __be32 ip2; + __be16 port; + u8 proto; + u8 padding; + unsigned long timeout; +}; + +static inline bool +hash_ipportip4_data_equal(const struct hash_ipportip4_elem *ip1, + const struct hash_ipportip4_elem *ip2) +{ + return ip1->ip == ip2->ip && + ip1->ip2 == ip2->ip2 && + ip1->port == ip2->port && + ip1->proto == ip2->proto; +} + +static inline bool +hash_ipportip4_data_isnull(const struct hash_ipportip4_elem *elem) +{ + return elem->proto == 0; +} + +static inline void +hash_ipportip4_data_copy(struct hash_ipportip4_elem *dst, + const struct hash_ipportip4_elem *src) +{ + memcpy(dst, src, sizeof(*dst)); +} + +static inline void +hash_ipportip4_data_zero_out(struct hash_ipportip4_elem *elem) +{ + elem->proto = 0; +} + +static bool +hash_ipportip4_data_list(struct sk_buff *skb, + const struct hash_ipportip4_elem *data) +{ + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, data->ip); + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP2, data->ip2); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, data->port); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + return 0; + +nla_put_failure: + return 1; +} + +static bool +hash_ipportip4_data_tlist(struct sk_buff *skb, + const struct hash_ipportip4_elem *data) +{ + const struct hash_ipportip4_telem *tdata = + (const struct hash_ipportip4_telem *)data; + + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, tdata->ip); + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP2, tdata->ip2); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, tdata->port); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(tdata->timeout))); + + return 0; + +nla_put_failure: + return 1; +} + +#define PF 4 +#define HOST_MASK 32 +#include + +static int +hash_ipportip4_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_ipportip4_elem data = { }; + + if (!ip_set_get_ip4_port(skb, flags & IPSET_DIM_TWO_SRC, + &data.port, &data.proto)) + return -EINVAL; + + ip4addrptr(skb, flags & IPSET_DIM_ONE_SRC, &data.ip); + ip4addrptr(skb, flags & IPSET_DIM_THREE_SRC, &data.ip2); + + return adtfn(set, &data, h->timeout); +} + +static int +hash_ipportip4_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_ipportip4_elem data = { }; + u32 ip, ip_to, p, port, port_to; + u32 timeout = h->timeout; + int ret; + + if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IP2] || + !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + ret = ip_set_get_ipaddr4(tb[IPSET_ATTR_IP], &data.ip); + if (ret) + return ret; + + ret = ip_set_get_ipaddr4(tb[IPSET_ATTR_IP2], &data.ip2); + if (ret) + return ret; + + if (tb[IPSET_ATTR_PORT]) + data.port = nla_get_be16(tb[IPSET_ATTR_PORT]); + else + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_PROTO]) { + data.proto = nla_get_u8(tb[IPSET_ATTR_PROTO]); + + if (data.proto == 0) + return -IPSET_ERR_INVALID_PROTO; + } else + return -IPSET_ERR_MISSING_PROTO; + + switch (data.proto) { + case IPPROTO_UDP: + case IPPROTO_TCP: + case IPPROTO_ICMP: + break; + default: + data.port = 0; + break; + } + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(h->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + if (adt == IPSET_TEST || + !(data.proto == IPPROTO_TCP || data.proto == IPPROTO_UDP) || + !(tb[IPSET_ATTR_IP_TO] || tb[IPSET_ATTR_CIDR] || + tb[IPSET_ATTR_PORT_TO])) { + ret = adtfn(set, &data, timeout); + return ip_set_eexist(ret, flags) ? 0 : ret; + } + + ip = ntohl(data.ip); + if (tb[IPSET_ATTR_IP_TO]) { + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP_TO], &ip_to); + if (ret) + return ret; + if (ip > ip_to) + swap(ip, ip_to); + } else if (tb[IPSET_ATTR_CIDR]) { + u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); + + if (cidr > 32) + return -IPSET_ERR_INVALID_CIDR; + ip &= ip_set_hostmask(cidr); + ip_to = ip | ~ip_set_hostmask(cidr); + } else + ip_to = ip; + + port = ntohs(data.port); + if (tb[IPSET_ATTR_PORT_TO]) { + port_to = ip_set_get_h16(tb[IPSET_ATTR_PORT_TO]); + if (port > port_to) + swap(port, port_to); + } else + port_to = port; + + for (; !before(ip_to, ip); ip++) + for (p = port; p <= port_to; p++) { + data.ip = htonl(ip); + data.port = htons(p); + ret = adtfn(set, &data, timeout); + + if (ret && !ip_set_eexist(ret, flags)) + return ret; + else + ret = 0; + } + return ret; +} + +static bool +hash_ipportip_same_set(const struct ip_set *a, const struct ip_set *b) +{ + const struct ip_set_hash *x = a->data; + const struct ip_set_hash *y = b->data; + + /* Resizing changes htable_bits, so we ignore it */ + return x->maxelem == y->maxelem && + x->timeout == y->timeout; +} + +/* The type variant functions: IPv6 */ + +struct hash_ipportip6_elem { + union nf_inet_addr ip; + union nf_inet_addr ip2; + __be16 port; + u8 proto; + u8 padding; +}; + +struct hash_ipportip6_telem { + union nf_inet_addr ip; + union nf_inet_addr ip2; + __be16 port; + u8 proto; + u8 padding; + unsigned long timeout; +}; + +static inline bool +hash_ipportip6_data_equal(const struct hash_ipportip6_elem *ip1, + const struct hash_ipportip6_elem *ip2) +{ + return ipv6_addr_cmp(&ip1->ip.in6, &ip2->ip.in6) == 0 && + ipv6_addr_cmp(&ip1->ip2.in6, &ip2->ip2.in6) == 0 && + ip1->port == ip2->port && + ip1->proto == ip2->proto; +} + +static inline bool +hash_ipportip6_data_isnull(const struct hash_ipportip6_elem *elem) +{ + return elem->proto == 0; +} + +static inline void +hash_ipportip6_data_copy(struct hash_ipportip6_elem *dst, + const struct hash_ipportip6_elem *src) +{ + memcpy(dst, src, sizeof(*dst)); +} + +static inline void +hash_ipportip6_data_zero_out(struct hash_ipportip6_elem *elem) +{ + elem->proto = 0; +} + +static bool +hash_ipportip6_data_list(struct sk_buff *skb, + const struct hash_ipportip6_elem *data) +{ + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP, &data->ip); + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP2, &data->ip2); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, data->port); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + return 0; + +nla_put_failure: + return 1; +} + +static bool +hash_ipportip6_data_tlist(struct sk_buff *skb, + const struct hash_ipportip6_elem *data) +{ + const struct hash_ipportip6_telem *e = + (const struct hash_ipportip6_telem *)data; + + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP, &e->ip); + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP2, &data->ip2); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, data->port); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(e->timeout))); + return 0; + +nla_put_failure: + return 1; +} + +#undef PF +#undef HOST_MASK + +#define PF 6 +#define HOST_MASK 128 +#include + +static int +hash_ipportip6_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_ipportip6_elem data = { }; + + if (!ip_set_get_ip6_port(skb, flags & IPSET_DIM_TWO_SRC, + &data.port, &data.proto)) + return -EINVAL; + + ip6addrptr(skb, flags & IPSET_DIM_ONE_SRC, &data.ip.in6); + ip6addrptr(skb, flags & IPSET_DIM_THREE_SRC, &data.ip2.in6); + + return adtfn(set, &data, h->timeout); +} + +static int +hash_ipportip6_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_ipportip6_elem data = { }; + u32 port, port_to; + u32 timeout = h->timeout; + int ret; + + if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IP2] || + !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || + tb[IPSET_ATTR_IP_TO] || + tb[IPSET_ATTR_CIDR])) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &data.ip); + if (ret) + return ret; + + ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP2], &data.ip2); + if (ret) + return ret; + + if (tb[IPSET_ATTR_PORT]) + data.port = nla_get_be16(tb[IPSET_ATTR_PORT]); + else + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_PROTO]) { + data.proto = nla_get_u8(tb[IPSET_ATTR_PROTO]); + + if (data.proto == 0) + return -IPSET_ERR_INVALID_PROTO; + } else + return -IPSET_ERR_MISSING_PROTO; + + switch (data.proto) { + case IPPROTO_UDP: + case IPPROTO_TCP: + case IPPROTO_ICMPV6: + break; + default: + data.port = 0; + break; + } + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(h->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + if (adt == IPSET_TEST || + !(data.proto == IPPROTO_TCP || data.proto == IPPROTO_UDP) || + !tb[IPSET_ATTR_PORT_TO]) { + ret = adtfn(set, &data, timeout); + return ip_set_eexist(ret, flags) ? 0 : ret; + } + + port = ntohs(data.port); + port_to = ip_set_get_h16(tb[IPSET_ATTR_PORT_TO]); + if (port > port_to) + swap(port, port_to); + + for (; port <= port_to; port++) { + data.port = htons(port); + ret = adtfn(set, &data, timeout); + + if (ret && !ip_set_eexist(ret, flags)) + return ret; + else + ret = 0; + } + return ret; +} + +/* Create hash:ip type of sets */ + +static int +hash_ipportip_create(struct ip_set *set, struct nlattr *tb[], u32 flags) +{ + struct ip_set_hash *h; + u32 hashsize = IPSET_DEFAULT_HASHSIZE, maxelem = IPSET_DEFAULT_MAXELEM; + u8 hbits; + + if (!(set->family == AF_INET || set->family == AF_INET6)) + return -IPSET_ERR_INVALID_FAMILY; + + if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_HASHSIZE) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_MAXELEM) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_HASHSIZE]) { + hashsize = ip_set_get_h32(tb[IPSET_ATTR_HASHSIZE]); + if (hashsize < IPSET_MIMINAL_HASHSIZE) + hashsize = IPSET_MIMINAL_HASHSIZE; + } + + if (tb[IPSET_ATTR_MAXELEM]) + maxelem = ip_set_get_h32(tb[IPSET_ATTR_MAXELEM]); + + h = kzalloc(sizeof(*h), GFP_KERNEL); + if (!h) + return -ENOMEM; + + h->maxelem = maxelem; + get_random_bytes(&h->initval, sizeof(h->initval)); + h->timeout = IPSET_NO_TIMEOUT; + + hbits = htable_bits(hashsize); + h->table = ip_set_alloc( + sizeof(struct htable) + + jhash_size(hbits) * sizeof(struct hbucket)); + if (!h->table) { + kfree(h); + return -ENOMEM; + } + h->table->htable_bits = hbits; + + set->data = h; + + if (tb[IPSET_ATTR_TIMEOUT]) { + h->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + + set->variant = set->family == AF_INET + ? &hash_ipportip4_tvariant : &hash_ipportip6_tvariant; + + if (set->family == AF_INET) + hash_ipportip4_gc_init(set); + else + hash_ipportip6_gc_init(set); + } else { + set->variant = set->family == AF_INET + ? &hash_ipportip4_variant : &hash_ipportip6_variant; + } + + pr_debug("create %s hashsize %u (%u) maxelem %u: %p(%p)\n", + set->name, jhash_size(h->table->htable_bits), + h->table->htable_bits, h->maxelem, set->data, h->table); + + return 0; +} + +static struct ip_set_type hash_ipportip_type __read_mostly = { + .name = "hash:ip,port,ip", + .protocol = IPSET_PROTOCOL, + .features = IPSET_TYPE_IP | IPSET_TYPE_PORT | IPSET_TYPE_IP2, + .dimension = IPSET_DIM_THREE, + .family = AF_UNSPEC, + .revision = 0, + .create = hash_ipportip_create, + .create_policy = { + [IPSET_ATTR_HASHSIZE] = { .type = NLA_U32 }, + [IPSET_ATTR_MAXELEM] = { .type = NLA_U32 }, + [IPSET_ATTR_PROBES] = { .type = NLA_U8 }, + [IPSET_ATTR_RESIZE] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + }, + .adt_policy = { + [IPSET_ATTR_IP] = { .type = NLA_NESTED }, + [IPSET_ATTR_IP_TO] = { .type = NLA_NESTED }, + [IPSET_ATTR_IP2] = { .type = NLA_NESTED }, + [IPSET_ATTR_PORT] = { .type = NLA_U16 }, + [IPSET_ATTR_PORT_TO] = { .type = NLA_U16 }, + [IPSET_ATTR_CIDR] = { .type = NLA_U8 }, + [IPSET_ATTR_PROTO] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + [IPSET_ATTR_LINENO] = { .type = NLA_U32 }, + }, + .me = THIS_MODULE, +}; + +static int __init +hash_ipportip_init(void) +{ + return ip_set_type_register(&hash_ipportip_type); +} + +static void __exit +hash_ipportip_fini(void) +{ + ip_set_type_unregister(&hash_ipportip_type); +} + +module_init(hash_ipportip_init); +module_exit(hash_ipportip_fini); -- cgit v1.2.3 From 41d22f7b2e48c77175b62cec3797d7d7173a626e Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 1 Feb 2011 15:51:00 +0100 Subject: netfilter: ipset: hash:ip,port,net set type support The module implements the hash:ip,port,net type support in four flavours: for IPv4 and IPv6, both without and with timeout support. The elements are three dimensional: IPv4/IPv6 address, protocol/port and IPv4/IPv6 network address/prefix triples. The different prefixes are searched/matched from the longest prefix to the shortes one (most specific to least). In other words the processing time linearly grows with the number of different prefixes in the set. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/ipset/Kconfig | 10 + net/netfilter/ipset/Makefile | 1 + net/netfilter/ipset/ip_set_hash_ipportnet.c | 631 ++++++++++++++++++++++++++++ 3 files changed, 642 insertions(+) create mode 100644 net/netfilter/ipset/ip_set_hash_ipportnet.c (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/Kconfig b/net/netfilter/ipset/Kconfig index e69355334252..e2fbaa9d902a 100644 --- a/net/netfilter/ipset/Kconfig +++ b/net/netfilter/ipset/Kconfig @@ -79,4 +79,14 @@ config IP_SET_HASH_IPPORTIP To compile it as a module, choose M here. If unsure, say N. +config IP_SET_HASH_IPPORTNET + tristate "hash:ip,port,net set support" + depends on IP_SET + help + This option adds the hash:ip,port,net set type support, by which + one can store IPv4/IPv6 address, protocol/port, and IPv4/IPv6 + network address/prefix triples in a set. + + To compile it as a module, choose M here. If unsure, say N. + endif # IP_SET diff --git a/net/netfilter/ipset/Makefile b/net/netfilter/ipset/Makefile index e9ddb25189d0..9c5d857511c4 100644 --- a/net/netfilter/ipset/Makefile +++ b/net/netfilter/ipset/Makefile @@ -16,3 +16,4 @@ obj-$(CONFIG_IP_SET_BITMAP_PORT) += ip_set_bitmap_port.o obj-$(CONFIG_IP_SET_HASH_IP) += ip_set_hash_ip.o obj-$(CONFIG_IP_SET_HASH_IPPORT) += ip_set_hash_ipport.o obj-$(CONFIG_IP_SET_HASH_IPPORTIP) += ip_set_hash_ipportip.o +obj-$(CONFIG_IP_SET_HASH_IPPORTNET) += ip_set_hash_ipportnet.o diff --git a/net/netfilter/ipset/ip_set_hash_ipportnet.c b/net/netfilter/ipset/ip_set_hash_ipportnet.c new file mode 100644 index 000000000000..8eacd8a46c1c --- /dev/null +++ b/net/netfilter/ipset/ip_set_hash_ipportnet.c @@ -0,0 +1,631 @@ +/* Copyright (C) 2003-2011 Jozsef Kadlecsik + * + * 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. + */ + +/* Kernel module implementing an IP set type: the hash:ip,port,net type */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Jozsef Kadlecsik "); +MODULE_DESCRIPTION("hash:ip,port,net type of IP sets"); +MODULE_ALIAS("ip_set_hash:ip,port,net"); + +/* Type specific function prefix */ +#define TYPE hash_ipportnet + +static bool +hash_ipportnet_same_set(const struct ip_set *a, const struct ip_set *b); + +#define hash_ipportnet4_same_set hash_ipportnet_same_set +#define hash_ipportnet6_same_set hash_ipportnet_same_set + +/* The type variant functions: IPv4 */ + +/* Member elements without timeout */ +struct hash_ipportnet4_elem { + __be32 ip; + __be32 ip2; + __be16 port; + u8 cidr; + u8 proto; +}; + +/* Member elements with timeout support */ +struct hash_ipportnet4_telem { + __be32 ip; + __be32 ip2; + __be16 port; + u8 cidr; + u8 proto; + unsigned long timeout; +}; + +static inline bool +hash_ipportnet4_data_equal(const struct hash_ipportnet4_elem *ip1, + const struct hash_ipportnet4_elem *ip2) +{ + return ip1->ip == ip2->ip && + ip1->ip2 == ip2->ip2 && + ip1->cidr == ip2->cidr && + ip1->port == ip2->port && + ip1->proto == ip2->proto; +} + +static inline bool +hash_ipportnet4_data_isnull(const struct hash_ipportnet4_elem *elem) +{ + return elem->proto == 0; +} + +static inline void +hash_ipportnet4_data_copy(struct hash_ipportnet4_elem *dst, + const struct hash_ipportnet4_elem *src) +{ + memcpy(dst, src, sizeof(*dst)); +} + +static inline void +hash_ipportnet4_data_netmask(struct hash_ipportnet4_elem *elem, u8 cidr) +{ + elem->ip2 &= ip_set_netmask(cidr); + elem->cidr = cidr; +} + +static inline void +hash_ipportnet4_data_zero_out(struct hash_ipportnet4_elem *elem) +{ + elem->proto = 0; +} + +static bool +hash_ipportnet4_data_list(struct sk_buff *skb, + const struct hash_ipportnet4_elem *data) +{ + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, data->ip); + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP2, data->ip2); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, data->port); + NLA_PUT_U8(skb, IPSET_ATTR_CIDR2, data->cidr); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + return 0; + +nla_put_failure: + return 1; +} + +static bool +hash_ipportnet4_data_tlist(struct sk_buff *skb, + const struct hash_ipportnet4_elem *data) +{ + const struct hash_ipportnet4_telem *tdata = + (const struct hash_ipportnet4_telem *)data; + + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, tdata->ip); + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP2, tdata->ip2); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, tdata->port); + NLA_PUT_U8(skb, IPSET_ATTR_CIDR2, data->cidr); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(tdata->timeout))); + + return 0; + +nla_put_failure: + return 1; +} + +#define IP_SET_HASH_WITH_PROTO +#define IP_SET_HASH_WITH_NETS + +#define PF 4 +#define HOST_MASK 32 +#include + +static int +hash_ipportnet4_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_ipportnet4_elem data = + { .cidr = h->nets[0].cidr || HOST_MASK }; + + if (data.cidr == 0) + return -EINVAL; + if (adt == IPSET_TEST) + data.cidr = HOST_MASK; + + if (!ip_set_get_ip4_port(skb, flags & IPSET_DIM_TWO_SRC, + &data.port, &data.proto)) + return -EINVAL; + + ip4addrptr(skb, flags & IPSET_DIM_ONE_SRC, &data.ip); + ip4addrptr(skb, flags & IPSET_DIM_THREE_SRC, &data.ip2); + data.ip2 &= ip_set_netmask(data.cidr); + + return adtfn(set, &data, h->timeout); +} + +static int +hash_ipportnet4_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_ipportnet4_elem data = { .cidr = HOST_MASK }; + u32 ip, ip_to, p, port, port_to; + u32 timeout = h->timeout; + int ret; + + if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IP2] || + !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + ret = ip_set_get_ipaddr4(tb[IPSET_ATTR_IP], &data.ip); + if (ret) + return ret; + + ret = ip_set_get_ipaddr4(tb[IPSET_ATTR_IP2], &data.ip2); + if (ret) + return ret; + + if (tb[IPSET_ATTR_CIDR2]) + data.cidr = nla_get_u8(tb[IPSET_ATTR_CIDR2]); + + if (!data.cidr) + return -IPSET_ERR_INVALID_CIDR; + + data.ip2 &= ip_set_netmask(data.cidr); + + if (tb[IPSET_ATTR_PORT]) + data.port = nla_get_be16(tb[IPSET_ATTR_PORT]); + else + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_PROTO]) { + data.proto = nla_get_u8(tb[IPSET_ATTR_PROTO]); + + if (data.proto == 0) + return -IPSET_ERR_INVALID_PROTO; + } else + return -IPSET_ERR_MISSING_PROTO; + + switch (data.proto) { + case IPPROTO_UDP: + case IPPROTO_TCP: + case IPPROTO_ICMP: + break; + default: + data.port = 0; + break; + } + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(h->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + if (adt == IPSET_TEST || + !(data.proto == IPPROTO_TCP || data.proto == IPPROTO_UDP) || + !(tb[IPSET_ATTR_IP_TO] || tb[IPSET_ATTR_CIDR] || + tb[IPSET_ATTR_PORT_TO])) { + ret = adtfn(set, &data, timeout); + return ip_set_eexist(ret, flags) ? 0 : ret; + } + + ip = ntohl(data.ip); + if (tb[IPSET_ATTR_IP_TO]) { + ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP_TO], &ip_to); + if (ret) + return ret; + if (ip > ip_to) + swap(ip, ip_to); + } else if (tb[IPSET_ATTR_CIDR]) { + u8 cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); + + if (cidr > 32) + return -IPSET_ERR_INVALID_CIDR; + ip &= ip_set_hostmask(cidr); + ip_to = ip | ~ip_set_hostmask(cidr); + } else + ip_to = ip; + + port = ntohs(data.port); + if (tb[IPSET_ATTR_PORT_TO]) { + port_to = ip_set_get_h16(tb[IPSET_ATTR_PORT_TO]); + if (port > port_to) + swap(port, port_to); + } else + port_to = port; + + for (; !before(ip_to, ip); ip++) + for (p = port; p <= port_to; p++) { + data.ip = htonl(ip); + data.port = htons(p); + ret = adtfn(set, &data, timeout); + + if (ret && !ip_set_eexist(ret, flags)) + return ret; + else + ret = 0; + } + return ret; +} + +static bool +hash_ipportnet_same_set(const struct ip_set *a, const struct ip_set *b) +{ + const struct ip_set_hash *x = a->data; + const struct ip_set_hash *y = b->data; + + /* Resizing changes htable_bits, so we ignore it */ + return x->maxelem == y->maxelem && + x->timeout == y->timeout; +} + +/* The type variant functions: IPv6 */ + +struct hash_ipportnet6_elem { + union nf_inet_addr ip; + union nf_inet_addr ip2; + __be16 port; + u8 cidr; + u8 proto; +}; + +struct hash_ipportnet6_telem { + union nf_inet_addr ip; + union nf_inet_addr ip2; + __be16 port; + u8 cidr; + u8 proto; + unsigned long timeout; +}; + +static inline bool +hash_ipportnet6_data_equal(const struct hash_ipportnet6_elem *ip1, + const struct hash_ipportnet6_elem *ip2) +{ + return ipv6_addr_cmp(&ip1->ip.in6, &ip2->ip.in6) == 0 && + ipv6_addr_cmp(&ip1->ip2.in6, &ip2->ip2.in6) == 0 && + ip1->cidr == ip2->cidr && + ip1->port == ip2->port && + ip1->proto == ip2->proto; +} + +static inline bool +hash_ipportnet6_data_isnull(const struct hash_ipportnet6_elem *elem) +{ + return elem->proto == 0; +} + +static inline void +hash_ipportnet6_data_copy(struct hash_ipportnet6_elem *dst, + const struct hash_ipportnet6_elem *src) +{ + memcpy(dst, src, sizeof(*dst)); +} + +static inline void +hash_ipportnet6_data_zero_out(struct hash_ipportnet6_elem *elem) +{ + elem->proto = 0; +} + +static inline void +ip6_netmask(union nf_inet_addr *ip, u8 prefix) +{ + ip->ip6[0] &= ip_set_netmask6(prefix)[0]; + ip->ip6[1] &= ip_set_netmask6(prefix)[1]; + ip->ip6[2] &= ip_set_netmask6(prefix)[2]; + ip->ip6[3] &= ip_set_netmask6(prefix)[3]; +} + +static inline void +hash_ipportnet6_data_netmask(struct hash_ipportnet6_elem *elem, u8 cidr) +{ + ip6_netmask(&elem->ip2, cidr); + elem->cidr = cidr; +} + +static bool +hash_ipportnet6_data_list(struct sk_buff *skb, + const struct hash_ipportnet6_elem *data) +{ + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP, &data->ip); + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP2, &data->ip2); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, data->port); + NLA_PUT_U8(skb, IPSET_ATTR_CIDR2, data->cidr); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + return 0; + +nla_put_failure: + return 1; +} + +static bool +hash_ipportnet6_data_tlist(struct sk_buff *skb, + const struct hash_ipportnet6_elem *data) +{ + const struct hash_ipportnet6_telem *e = + (const struct hash_ipportnet6_telem *)data; + + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP, &e->ip); + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP2, &data->ip2); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, data->port); + NLA_PUT_U8(skb, IPSET_ATTR_CIDR2, data->cidr); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(e->timeout))); + return 0; + +nla_put_failure: + return 1; +} + +#undef PF +#undef HOST_MASK + +#define PF 6 +#define HOST_MASK 128 +#include + +static int +hash_ipportnet6_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_ipportnet6_elem data = + { .cidr = h->nets[0].cidr || HOST_MASK }; + + if (data.cidr == 0) + return -EINVAL; + if (adt == IPSET_TEST) + data.cidr = HOST_MASK; + + if (!ip_set_get_ip6_port(skb, flags & IPSET_DIM_TWO_SRC, + &data.port, &data.proto)) + return -EINVAL; + + ip6addrptr(skb, flags & IPSET_DIM_ONE_SRC, &data.ip.in6); + ip6addrptr(skb, flags & IPSET_DIM_THREE_SRC, &data.ip2.in6); + ip6_netmask(&data.ip2, data.cidr); + + return adtfn(set, &data, h->timeout); +} + +static int +hash_ipportnet6_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_ipportnet6_elem data = { .cidr = HOST_MASK }; + u32 port, port_to; + u32 timeout = h->timeout; + int ret; + + if (unlikely(!tb[IPSET_ATTR_IP] || !tb[IPSET_ATTR_IP2] || + !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || + tb[IPSET_ATTR_IP_TO] || + tb[IPSET_ATTR_CIDR])) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &data.ip); + if (ret) + return ret; + + ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP2], &data.ip2); + if (ret) + return ret; + + if (tb[IPSET_ATTR_CIDR2]) + data.cidr = nla_get_u8(tb[IPSET_ATTR_CIDR2]); + + if (!data.cidr) + return -IPSET_ERR_INVALID_CIDR; + + ip6_netmask(&data.ip2, data.cidr); + + if (tb[IPSET_ATTR_PORT]) + data.port = nla_get_be16(tb[IPSET_ATTR_PORT]); + else + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_PROTO]) { + data.proto = nla_get_u8(tb[IPSET_ATTR_PROTO]); + + if (data.proto == 0) + return -IPSET_ERR_INVALID_PROTO; + } else + return -IPSET_ERR_MISSING_PROTO; + + switch (data.proto) { + case IPPROTO_UDP: + case IPPROTO_TCP: + case IPPROTO_ICMPV6: + break; + default: + data.port = 0; + break; + } + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(h->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + if (adt == IPSET_TEST || + !(data.proto == IPPROTO_TCP || data.proto == IPPROTO_UDP) || + !tb[IPSET_ATTR_PORT_TO]) { + ret = adtfn(set, &data, timeout); + return ip_set_eexist(ret, flags) ? 0 : ret; + } + + port = ntohs(data.port); + port_to = ip_set_get_h16(tb[IPSET_ATTR_PORT_TO]); + if (port > port_to) + swap(port, port_to); + + for (; port <= port_to; port++) { + data.port = htons(port); + ret = adtfn(set, &data, timeout); + + if (ret && !ip_set_eexist(ret, flags)) + return ret; + else + ret = 0; + } + return ret; +} + +/* Create hash:ip type of sets */ + +static int +hash_ipportnet_create(struct ip_set *set, struct nlattr *tb[], u32 flags) +{ + struct ip_set_hash *h; + u32 hashsize = IPSET_DEFAULT_HASHSIZE, maxelem = IPSET_DEFAULT_MAXELEM; + u8 hbits; + + if (!(set->family == AF_INET || set->family == AF_INET6)) + return -IPSET_ERR_INVALID_FAMILY; + + if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_HASHSIZE) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_MAXELEM) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_HASHSIZE]) { + hashsize = ip_set_get_h32(tb[IPSET_ATTR_HASHSIZE]); + if (hashsize < IPSET_MIMINAL_HASHSIZE) + hashsize = IPSET_MIMINAL_HASHSIZE; + } + + if (tb[IPSET_ATTR_MAXELEM]) + maxelem = ip_set_get_h32(tb[IPSET_ATTR_MAXELEM]); + + h = kzalloc(sizeof(*h) + + sizeof(struct ip_set_hash_nets) + * (set->family == AF_INET ? 32 : 128), GFP_KERNEL); + if (!h) + return -ENOMEM; + + h->maxelem = maxelem; + get_random_bytes(&h->initval, sizeof(h->initval)); + h->timeout = IPSET_NO_TIMEOUT; + + hbits = htable_bits(hashsize); + h->table = ip_set_alloc( + sizeof(struct htable) + + jhash_size(hbits) * sizeof(struct hbucket)); + if (!h->table) { + kfree(h); + return -ENOMEM; + } + h->table->htable_bits = hbits; + + set->data = h; + + if (tb[IPSET_ATTR_TIMEOUT]) { + h->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + + set->variant = set->family == AF_INET + ? &hash_ipportnet4_tvariant + : &hash_ipportnet6_tvariant; + + if (set->family == AF_INET) + hash_ipportnet4_gc_init(set); + else + hash_ipportnet6_gc_init(set); + } else { + set->variant = set->family == AF_INET + ? &hash_ipportnet4_variant : &hash_ipportnet6_variant; + } + + pr_debug("create %s hashsize %u (%u) maxelem %u: %p(%p)\n", + set->name, jhash_size(h->table->htable_bits), + h->table->htable_bits, h->maxelem, set->data, h->table); + + return 0; +} + +static struct ip_set_type hash_ipportnet_type __read_mostly = { + .name = "hash:ip,port,net", + .protocol = IPSET_PROTOCOL, + .features = IPSET_TYPE_IP | IPSET_TYPE_PORT | IPSET_TYPE_IP2, + .dimension = IPSET_DIM_THREE, + .family = AF_UNSPEC, + .revision = 0, + .create = hash_ipportnet_create, + .create_policy = { + [IPSET_ATTR_HASHSIZE] = { .type = NLA_U32 }, + [IPSET_ATTR_MAXELEM] = { .type = NLA_U32 }, + [IPSET_ATTR_PROBES] = { .type = NLA_U8 }, + [IPSET_ATTR_RESIZE] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + }, + .adt_policy = { + [IPSET_ATTR_IP] = { .type = NLA_NESTED }, + [IPSET_ATTR_IP_TO] = { .type = NLA_NESTED }, + [IPSET_ATTR_IP2] = { .type = NLA_NESTED }, + [IPSET_ATTR_PORT] = { .type = NLA_U16 }, + [IPSET_ATTR_PORT_TO] = { .type = NLA_U16 }, + [IPSET_ATTR_CIDR] = { .type = NLA_U8 }, + [IPSET_ATTR_CIDR2] = { .type = NLA_U8 }, + [IPSET_ATTR_PROTO] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + [IPSET_ATTR_LINENO] = { .type = NLA_U32 }, + }, + .me = THIS_MODULE, +}; + +static int __init +hash_ipportnet_init(void) +{ + return ip_set_type_register(&hash_ipportnet_type); +} + +static void __exit +hash_ipportnet_fini(void) +{ + ip_set_type_unregister(&hash_ipportnet_type); +} + +module_init(hash_ipportnet_init); +module_exit(hash_ipportnet_fini); -- cgit v1.2.3 From b38370299eeaba4cf8a9e0c5c6acc2a1e049be23 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 1 Feb 2011 15:52:54 +0100 Subject: netfilter: ipset: hash:net set type support The module implements the hash:net type support in four flavours: for IPv4 and IPv6, both without and with timeout support. The elements are one dimensional: IPv4/IPv6 network address/prefixes. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/ipset/Kconfig | 9 + net/netfilter/ipset/Makefile | 1 + net/netfilter/ipset/ip_set_hash_net.c | 461 ++++++++++++++++++++++++++++++++++ 3 files changed, 471 insertions(+) create mode 100644 net/netfilter/ipset/ip_set_hash_net.c (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/Kconfig b/net/netfilter/ipset/Kconfig index e2fbaa9d902a..8d85de00e9e0 100644 --- a/net/netfilter/ipset/Kconfig +++ b/net/netfilter/ipset/Kconfig @@ -89,4 +89,13 @@ config IP_SET_HASH_IPPORTNET To compile it as a module, choose M here. If unsure, say N. +config IP_SET_HASH_NET + tristate "hash:net set support" + depends on IP_SET + help + This option adds the hash:net set type support, by which + one can store IPv4/IPv6 network address/prefix elements in a set. + + To compile it as a module, choose M here. If unsure, say N. + endif # IP_SET diff --git a/net/netfilter/ipset/Makefile b/net/netfilter/ipset/Makefile index 9c5d857511c4..fd5eeb6a7e5b 100644 --- a/net/netfilter/ipset/Makefile +++ b/net/netfilter/ipset/Makefile @@ -17,3 +17,4 @@ obj-$(CONFIG_IP_SET_HASH_IP) += ip_set_hash_ip.o obj-$(CONFIG_IP_SET_HASH_IPPORT) += ip_set_hash_ipport.o obj-$(CONFIG_IP_SET_HASH_IPPORTIP) += ip_set_hash_ipportip.o obj-$(CONFIG_IP_SET_HASH_IPPORTNET) += ip_set_hash_ipportnet.o +obj-$(CONFIG_IP_SET_HASH_NET) += ip_set_hash_net.o diff --git a/net/netfilter/ipset/ip_set_hash_net.c b/net/netfilter/ipset/ip_set_hash_net.c new file mode 100644 index 000000000000..fb0e6a68d13e --- /dev/null +++ b/net/netfilter/ipset/ip_set_hash_net.c @@ -0,0 +1,461 @@ +/* Copyright (C) 2003-2011 Jozsef Kadlecsik + * + * 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. + */ + +/* Kernel module implementing an IP set type: the hash:net type */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Jozsef Kadlecsik "); +MODULE_DESCRIPTION("hash:net type of IP sets"); +MODULE_ALIAS("ip_set_hash:net"); + +/* Type specific function prefix */ +#define TYPE hash_net + +static bool +hash_net_same_set(const struct ip_set *a, const struct ip_set *b); + +#define hash_net4_same_set hash_net_same_set +#define hash_net6_same_set hash_net_same_set + +/* The type variant functions: IPv4 */ + +/* Member elements without timeout */ +struct hash_net4_elem { + __be32 ip; + u16 padding0; + u8 padding1; + u8 cidr; +}; + +/* Member elements with timeout support */ +struct hash_net4_telem { + __be32 ip; + u16 padding0; + u8 padding1; + u8 cidr; + unsigned long timeout; +}; + +static inline bool +hash_net4_data_equal(const struct hash_net4_elem *ip1, + const struct hash_net4_elem *ip2) +{ + return ip1->ip == ip2->ip && ip1->cidr == ip2->cidr; +} + +static inline bool +hash_net4_data_isnull(const struct hash_net4_elem *elem) +{ + return elem->cidr == 0; +} + +static inline void +hash_net4_data_copy(struct hash_net4_elem *dst, + const struct hash_net4_elem *src) +{ + dst->ip = src->ip; + dst->cidr = src->cidr; +} + +static inline void +hash_net4_data_netmask(struct hash_net4_elem *elem, u8 cidr) +{ + elem->ip &= ip_set_netmask(cidr); + elem->cidr = cidr; +} + +/* Zero CIDR values cannot be stored */ +static inline void +hash_net4_data_zero_out(struct hash_net4_elem *elem) +{ + elem->cidr = 0; +} + +static bool +hash_net4_data_list(struct sk_buff *skb, const struct hash_net4_elem *data) +{ + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, data->ip); + NLA_PUT_U8(skb, IPSET_ATTR_CIDR, data->cidr); + return 0; + +nla_put_failure: + return 1; +} + +static bool +hash_net4_data_tlist(struct sk_buff *skb, const struct hash_net4_elem *data) +{ + const struct hash_net4_telem *tdata = + (const struct hash_net4_telem *)data; + + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, tdata->ip); + NLA_PUT_U8(skb, IPSET_ATTR_CIDR, tdata->cidr); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(tdata->timeout))); + + return 0; + +nla_put_failure: + return 1; +} + +#define IP_SET_HASH_WITH_NETS + +#define PF 4 +#define HOST_MASK 32 +#include + +static int +hash_net4_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_net4_elem data = { .cidr = h->nets[0].cidr || HOST_MASK }; + + if (data.cidr == 0) + return -EINVAL; + if (adt == IPSET_TEST) + data.cidr = HOST_MASK; + + ip4addrptr(skb, flags & IPSET_DIM_ONE_SRC, &data.ip); + data.ip &= ip_set_netmask(data.cidr); + + return adtfn(set, &data, h->timeout); +} + +static int +hash_net4_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_net4_elem data = { .cidr = HOST_MASK }; + u32 timeout = h->timeout; + int ret; + + if (unlikely(!tb[IPSET_ATTR_IP] || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + ret = ip_set_get_ipaddr4(tb[IPSET_ATTR_IP], &data.ip); + if (ret) + return ret; + + if (tb[IPSET_ATTR_CIDR]) + data.cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); + + if (!data.cidr) + return -IPSET_ERR_INVALID_CIDR; + + data.ip &= ip_set_netmask(data.cidr); + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(h->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + ret = adtfn(set, &data, timeout); + + return ip_set_eexist(ret, flags) ? 0 : ret; +} + +static bool +hash_net_same_set(const struct ip_set *a, const struct ip_set *b) +{ + const struct ip_set_hash *x = a->data; + const struct ip_set_hash *y = b->data; + + /* Resizing changes htable_bits, so we ignore it */ + return x->maxelem == y->maxelem && + x->timeout == y->timeout; +} + +/* The type variant functions: IPv6 */ + +struct hash_net6_elem { + union nf_inet_addr ip; + u16 padding0; + u8 padding1; + u8 cidr; +}; + +struct hash_net6_telem { + union nf_inet_addr ip; + u16 padding0; + u8 padding1; + u8 cidr; + unsigned long timeout; +}; + +static inline bool +hash_net6_data_equal(const struct hash_net6_elem *ip1, + const struct hash_net6_elem *ip2) +{ + return ipv6_addr_cmp(&ip1->ip.in6, &ip2->ip.in6) == 0 && + ip1->cidr == ip2->cidr; +} + +static inline bool +hash_net6_data_isnull(const struct hash_net6_elem *elem) +{ + return elem->cidr == 0; +} + +static inline void +hash_net6_data_copy(struct hash_net6_elem *dst, + const struct hash_net6_elem *src) +{ + ipv6_addr_copy(&dst->ip.in6, &src->ip.in6); + dst->cidr = src->cidr; +} + +static inline void +hash_net6_data_zero_out(struct hash_net6_elem *elem) +{ + elem->cidr = 0; +} + +static inline void +ip6_netmask(union nf_inet_addr *ip, u8 prefix) +{ + ip->ip6[0] &= ip_set_netmask6(prefix)[0]; + ip->ip6[1] &= ip_set_netmask6(prefix)[1]; + ip->ip6[2] &= ip_set_netmask6(prefix)[2]; + ip->ip6[3] &= ip_set_netmask6(prefix)[3]; +} + +static inline void +hash_net6_data_netmask(struct hash_net6_elem *elem, u8 cidr) +{ + ip6_netmask(&elem->ip, cidr); + elem->cidr = cidr; +} + +static bool +hash_net6_data_list(struct sk_buff *skb, const struct hash_net6_elem *data) +{ + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP, &data->ip); + NLA_PUT_U8(skb, IPSET_ATTR_CIDR, data->cidr); + return 0; + +nla_put_failure: + return 1; +} + +static bool +hash_net6_data_tlist(struct sk_buff *skb, const struct hash_net6_elem *data) +{ + const struct hash_net6_telem *e = + (const struct hash_net6_telem *)data; + + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP, &e->ip); + NLA_PUT_U8(skb, IPSET_ATTR_CIDR, e->cidr); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(e->timeout))); + return 0; + +nla_put_failure: + return 1; +} + +#undef PF +#undef HOST_MASK + +#define PF 6 +#define HOST_MASK 128 +#include + +static int +hash_net6_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_net6_elem data = { .cidr = h->nets[0].cidr || HOST_MASK }; + + if (data.cidr == 0) + return -EINVAL; + if (adt == IPSET_TEST) + data.cidr = HOST_MASK; + + ip6addrptr(skb, flags & IPSET_DIM_ONE_SRC, &data.ip.in6); + ip6_netmask(&data.ip, data.cidr); + + return adtfn(set, &data, h->timeout); +} + +static int +hash_net6_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_net6_elem data = { .cidr = HOST_MASK }; + u32 timeout = h->timeout; + int ret; + + if (unlikely(!tb[IPSET_ATTR_IP] || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &data.ip); + if (ret) + return ret; + + if (tb[IPSET_ATTR_CIDR]) + data.cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); + + if (!data.cidr) + return -IPSET_ERR_INVALID_CIDR; + + ip6_netmask(&data.ip, data.cidr); + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(h->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + ret = adtfn(set, &data, timeout); + + return ip_set_eexist(ret, flags) ? 0 : ret; +} + +/* Create hash:ip type of sets */ + +static int +hash_net_create(struct ip_set *set, struct nlattr *tb[], u32 flags) +{ + u32 hashsize = IPSET_DEFAULT_HASHSIZE, maxelem = IPSET_DEFAULT_MAXELEM; + struct ip_set_hash *h; + u8 hbits; + + if (!(set->family == AF_INET || set->family == AF_INET6)) + return -IPSET_ERR_INVALID_FAMILY; + + if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_HASHSIZE) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_MAXELEM) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_HASHSIZE]) { + hashsize = ip_set_get_h32(tb[IPSET_ATTR_HASHSIZE]); + if (hashsize < IPSET_MIMINAL_HASHSIZE) + hashsize = IPSET_MIMINAL_HASHSIZE; + } + + if (tb[IPSET_ATTR_MAXELEM]) + maxelem = ip_set_get_h32(tb[IPSET_ATTR_MAXELEM]); + + h = kzalloc(sizeof(*h) + + sizeof(struct ip_set_hash_nets) + * (set->family == AF_INET ? 32 : 128), GFP_KERNEL); + if (!h) + return -ENOMEM; + + h->maxelem = maxelem; + get_random_bytes(&h->initval, sizeof(h->initval)); + h->timeout = IPSET_NO_TIMEOUT; + + hbits = htable_bits(hashsize); + h->table = ip_set_alloc( + sizeof(struct htable) + + jhash_size(hbits) * sizeof(struct hbucket)); + if (!h->table) { + kfree(h); + return -ENOMEM; + } + h->table->htable_bits = hbits; + + set->data = h; + + if (tb[IPSET_ATTR_TIMEOUT]) { + h->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + + set->variant = set->family == AF_INET + ? &hash_net4_tvariant : &hash_net6_tvariant; + + if (set->family == AF_INET) + hash_net4_gc_init(set); + else + hash_net6_gc_init(set); + } else { + set->variant = set->family == AF_INET + ? &hash_net4_variant : &hash_net6_variant; + } + + pr_debug("create %s hashsize %u (%u) maxelem %u: %p(%p)\n", + set->name, jhash_size(h->table->htable_bits), + h->table->htable_bits, h->maxelem, set->data, h->table); + + return 0; +} + +static struct ip_set_type hash_net_type __read_mostly = { + .name = "hash:net", + .protocol = IPSET_PROTOCOL, + .features = IPSET_TYPE_IP, + .dimension = IPSET_DIM_ONE, + .family = AF_UNSPEC, + .revision = 0, + .create = hash_net_create, + .create_policy = { + [IPSET_ATTR_HASHSIZE] = { .type = NLA_U32 }, + [IPSET_ATTR_MAXELEM] = { .type = NLA_U32 }, + [IPSET_ATTR_PROBES] = { .type = NLA_U8 }, + [IPSET_ATTR_RESIZE] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + }, + .adt_policy = { + [IPSET_ATTR_IP] = { .type = NLA_NESTED }, + [IPSET_ATTR_CIDR] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + }, + .me = THIS_MODULE, +}; + +static int __init +hash_net_init(void) +{ + return ip_set_type_register(&hash_net_type); +} + +static void __exit +hash_net_fini(void) +{ + ip_set_type_unregister(&hash_net_type); +} + +module_init(hash_net_init); +module_exit(hash_net_fini); -- cgit v1.2.3 From 21f45020a3084f80fcdd5f056a0c6389f5406399 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 1 Feb 2011 15:53:55 +0100 Subject: netfilter: ipset: hash:net,port set type support The module implements the hash:net,port type support in four flavours: for IPv4 and IPv6, both without and with timeout support. The elements are two dimensional: IPv4/IPv6 network address/prefix and protocol/port pairs. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/ipset/Kconfig | 10 + net/netfilter/ipset/Makefile | 1 + net/netfilter/ipset/ip_set_hash_netport.c | 581 ++++++++++++++++++++++++++++++ 3 files changed, 592 insertions(+) create mode 100644 net/netfilter/ipset/ip_set_hash_netport.c (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/Kconfig b/net/netfilter/ipset/Kconfig index 8d85de00e9e0..2512e7b82d12 100644 --- a/net/netfilter/ipset/Kconfig +++ b/net/netfilter/ipset/Kconfig @@ -98,4 +98,14 @@ config IP_SET_HASH_NET To compile it as a module, choose M here. If unsure, say N. +config IP_SET_HASH_NETPORT + tristate "hash:net,port set support" + depends on IP_SET + help + This option adds the hash:net,port set type support, by which + one can store IPv4/IPv6 network address/prefix and + protocol/port pairs as elements in a set. + + To compile it as a module, choose M here. If unsure, say N. + endif # IP_SET diff --git a/net/netfilter/ipset/Makefile b/net/netfilter/ipset/Makefile index fd5eeb6a7e5b..fbbebd62bb2a 100644 --- a/net/netfilter/ipset/Makefile +++ b/net/netfilter/ipset/Makefile @@ -18,3 +18,4 @@ obj-$(CONFIG_IP_SET_HASH_IPPORT) += ip_set_hash_ipport.o obj-$(CONFIG_IP_SET_HASH_IPPORTIP) += ip_set_hash_ipportip.o obj-$(CONFIG_IP_SET_HASH_IPPORTNET) += ip_set_hash_ipportnet.o obj-$(CONFIG_IP_SET_HASH_NET) += ip_set_hash_net.o +obj-$(CONFIG_IP_SET_HASH_NETPORT) += ip_set_hash_netport.o diff --git a/net/netfilter/ipset/ip_set_hash_netport.c b/net/netfilter/ipset/ip_set_hash_netport.c new file mode 100644 index 000000000000..342250f9b114 --- /dev/null +++ b/net/netfilter/ipset/ip_set_hash_netport.c @@ -0,0 +1,581 @@ +/* Copyright (C) 2003-2011 Jozsef Kadlecsik + * + * 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. + */ + +/* Kernel module implementing an IP set type: the hash:net,port type */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Jozsef Kadlecsik "); +MODULE_DESCRIPTION("hash:net,port type of IP sets"); +MODULE_ALIAS("ip_set_hash:net,port"); + +/* Type specific function prefix */ +#define TYPE hash_netport + +static bool +hash_netport_same_set(const struct ip_set *a, const struct ip_set *b); + +#define hash_netport4_same_set hash_netport_same_set +#define hash_netport6_same_set hash_netport_same_set + +/* The type variant functions: IPv4 */ + +/* Member elements without timeout */ +struct hash_netport4_elem { + __be32 ip; + __be16 port; + u8 proto; + u8 cidr; +}; + +/* Member elements with timeout support */ +struct hash_netport4_telem { + __be32 ip; + __be16 port; + u8 proto; + u8 cidr; + unsigned long timeout; +}; + +static inline bool +hash_netport4_data_equal(const struct hash_netport4_elem *ip1, + const struct hash_netport4_elem *ip2) +{ + return ip1->ip == ip2->ip && + ip1->port == ip2->port && + ip1->proto == ip2->proto && + ip1->cidr == ip2->cidr; +} + +static inline bool +hash_netport4_data_isnull(const struct hash_netport4_elem *elem) +{ + return elem->proto == 0; +} + +static inline void +hash_netport4_data_copy(struct hash_netport4_elem *dst, + const struct hash_netport4_elem *src) +{ + dst->ip = src->ip; + dst->port = src->port; + dst->proto = src->proto; + dst->cidr = src->cidr; +} + +static inline void +hash_netport4_data_netmask(struct hash_netport4_elem *elem, u8 cidr) +{ + elem->ip &= ip_set_netmask(cidr); + elem->cidr = cidr; +} + +static inline void +hash_netport4_data_zero_out(struct hash_netport4_elem *elem) +{ + elem->proto = 0; +} + +static bool +hash_netport4_data_list(struct sk_buff *skb, + const struct hash_netport4_elem *data) +{ + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, data->ip); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, data->port); + NLA_PUT_U8(skb, IPSET_ATTR_CIDR, data->cidr); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + return 0; + +nla_put_failure: + return 1; +} + +static bool +hash_netport4_data_tlist(struct sk_buff *skb, + const struct hash_netport4_elem *data) +{ + const struct hash_netport4_telem *tdata = + (const struct hash_netport4_telem *)data; + + NLA_PUT_IPADDR4(skb, IPSET_ATTR_IP, tdata->ip); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, tdata->port); + NLA_PUT_U8(skb, IPSET_ATTR_CIDR, data->cidr); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(tdata->timeout))); + + return 0; + +nla_put_failure: + return 1; +} + +#define IP_SET_HASH_WITH_PROTO +#define IP_SET_HASH_WITH_NETS + +#define PF 4 +#define HOST_MASK 32 +#include + +static int +hash_netport4_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_netport4_elem data = { + .cidr = h->nets[0].cidr || HOST_MASK }; + + if (data.cidr == 0) + return -EINVAL; + if (adt == IPSET_TEST) + data.cidr = HOST_MASK; + + if (!ip_set_get_ip4_port(skb, flags & IPSET_DIM_TWO_SRC, + &data.port, &data.proto)) + return -EINVAL; + + ip4addrptr(skb, flags & IPSET_DIM_ONE_SRC, &data.ip); + data.ip &= ip_set_netmask(data.cidr); + + return adtfn(set, &data, h->timeout); +} + +static int +hash_netport4_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_netport4_elem data = { .cidr = HOST_MASK }; + u32 port, port_to; + u32 timeout = h->timeout; + int ret; + + if (unlikely(!tb[IPSET_ATTR_IP] || + !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + ret = ip_set_get_ipaddr4(tb[IPSET_ATTR_IP], &data.ip); + if (ret) + return ret; + + if (tb[IPSET_ATTR_CIDR]) + data.cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); + if (!data.cidr) + return -IPSET_ERR_INVALID_CIDR; + data.ip &= ip_set_netmask(data.cidr); + + if (tb[IPSET_ATTR_PORT]) + data.port = nla_get_be16(tb[IPSET_ATTR_PORT]); + else + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_PROTO]) { + data.proto = nla_get_u8(tb[IPSET_ATTR_PROTO]); + + if (data.proto == 0) + return -IPSET_ERR_INVALID_PROTO; + } else + return -IPSET_ERR_MISSING_PROTO; + + switch (data.proto) { + case IPPROTO_UDP: + case IPPROTO_TCP: + case IPPROTO_ICMP: + break; + default: + data.port = 0; + break; + } + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(h->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + if (adt == IPSET_TEST || + !(data.proto == IPPROTO_TCP || data.proto == IPPROTO_UDP) || + !tb[IPSET_ATTR_PORT_TO]) { + ret = adtfn(set, &data, timeout); + return ip_set_eexist(ret, flags) ? 0 : ret; + } + + port = ntohs(data.port); + port_to = ip_set_get_h16(tb[IPSET_ATTR_PORT_TO]); + if (port > port_to) + swap(port, port_to); + + for (; port <= port_to; port++) { + data.port = htons(port); + ret = adtfn(set, &data, timeout); + + if (ret && !ip_set_eexist(ret, flags)) + return ret; + else + ret = 0; + } + return ret; +} + +static bool +hash_netport_same_set(const struct ip_set *a, const struct ip_set *b) +{ + const struct ip_set_hash *x = a->data; + const struct ip_set_hash *y = b->data; + + /* Resizing changes htable_bits, so we ignore it */ + return x->maxelem == y->maxelem && + x->timeout == y->timeout; +} + +/* The type variant functions: IPv6 */ + +struct hash_netport6_elem { + union nf_inet_addr ip; + __be16 port; + u8 proto; + u8 cidr; +}; + +struct hash_netport6_telem { + union nf_inet_addr ip; + __be16 port; + u8 proto; + u8 cidr; + unsigned long timeout; +}; + +static inline bool +hash_netport6_data_equal(const struct hash_netport6_elem *ip1, + const struct hash_netport6_elem *ip2) +{ + return ipv6_addr_cmp(&ip1->ip.in6, &ip2->ip.in6) == 0 && + ip1->port == ip2->port && + ip1->proto == ip2->proto && + ip1->cidr == ip2->cidr; +} + +static inline bool +hash_netport6_data_isnull(const struct hash_netport6_elem *elem) +{ + return elem->proto == 0; +} + +static inline void +hash_netport6_data_copy(struct hash_netport6_elem *dst, + const struct hash_netport6_elem *src) +{ + memcpy(dst, src, sizeof(*dst)); +} + +static inline void +hash_netport6_data_zero_out(struct hash_netport6_elem *elem) +{ + elem->proto = 0; +} + +static inline void +ip6_netmask(union nf_inet_addr *ip, u8 prefix) +{ + ip->ip6[0] &= ip_set_netmask6(prefix)[0]; + ip->ip6[1] &= ip_set_netmask6(prefix)[1]; + ip->ip6[2] &= ip_set_netmask6(prefix)[2]; + ip->ip6[3] &= ip_set_netmask6(prefix)[3]; +} + +static inline void +hash_netport6_data_netmask(struct hash_netport6_elem *elem, u8 cidr) +{ + ip6_netmask(&elem->ip, cidr); + elem->cidr = cidr; +} + +static bool +hash_netport6_data_list(struct sk_buff *skb, + const struct hash_netport6_elem *data) +{ + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP, &data->ip); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, data->port); + NLA_PUT_U8(skb, IPSET_ATTR_CIDR, data->cidr); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + return 0; + +nla_put_failure: + return 1; +} + +static bool +hash_netport6_data_tlist(struct sk_buff *skb, + const struct hash_netport6_elem *data) +{ + const struct hash_netport6_telem *e = + (const struct hash_netport6_telem *)data; + + NLA_PUT_IPADDR6(skb, IPSET_ATTR_IP, &e->ip); + NLA_PUT_NET16(skb, IPSET_ATTR_PORT, data->port); + NLA_PUT_U8(skb, IPSET_ATTR_CIDR, data->cidr); + NLA_PUT_U8(skb, IPSET_ATTR_PROTO, data->proto); + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(e->timeout))); + return 0; + +nla_put_failure: + return 1; +} + +#undef PF +#undef HOST_MASK + +#define PF 6 +#define HOST_MASK 128 +#include + +static int +hash_netport6_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_netport6_elem data = { + .cidr = h->nets[0].cidr || HOST_MASK }; + + if (data.cidr == 0) + return -EINVAL; + if (adt == IPSET_TEST) + data.cidr = HOST_MASK; + + if (!ip_set_get_ip6_port(skb, flags & IPSET_DIM_TWO_SRC, + &data.port, &data.proto)) + return -EINVAL; + + ip6addrptr(skb, flags & IPSET_DIM_ONE_SRC, &data.ip.in6); + ip6_netmask(&data.ip, data.cidr); + + return adtfn(set, &data, h->timeout); +} + +static int +hash_netport6_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + const struct ip_set_hash *h = set->data; + ipset_adtfn adtfn = set->variant->adt[adt]; + struct hash_netport6_elem data = { .cidr = HOST_MASK }; + u32 port, port_to; + u32 timeout = h->timeout; + int ret; + + if (unlikely(!tb[IPSET_ATTR_IP] || + !ip_set_attr_netorder(tb, IPSET_ATTR_PORT) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_PORT_TO) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + ret = ip_set_get_ipaddr6(tb[IPSET_ATTR_IP], &data.ip); + if (ret) + return ret; + + if (tb[IPSET_ATTR_CIDR]) + data.cidr = nla_get_u8(tb[IPSET_ATTR_CIDR]); + if (!data.cidr) + return -IPSET_ERR_INVALID_CIDR; + ip6_netmask(&data.ip, data.cidr); + + if (tb[IPSET_ATTR_PORT]) + data.port = nla_get_be16(tb[IPSET_ATTR_PORT]); + else + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_PROTO]) { + data.proto = nla_get_u8(tb[IPSET_ATTR_PROTO]); + + if (data.proto == 0) + return -IPSET_ERR_INVALID_PROTO; + } else + return -IPSET_ERR_MISSING_PROTO; + + switch (data.proto) { + case IPPROTO_UDP: + case IPPROTO_TCP: + case IPPROTO_ICMPV6: + break; + default: + data.port = 0; + break; + } + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout(h->timeout)) + return -IPSET_ERR_TIMEOUT; + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + if (adt == IPSET_TEST || + !(data.proto == IPPROTO_TCP || data.proto == IPPROTO_UDP) || + !tb[IPSET_ATTR_PORT_TO]) { + ret = adtfn(set, &data, timeout); + return ip_set_eexist(ret, flags) ? 0 : ret; + } + + port = ntohs(data.port); + port_to = ip_set_get_h16(tb[IPSET_ATTR_PORT_TO]); + if (port > port_to) + swap(port, port_to); + + for (; port <= port_to; port++) { + data.port = htons(port); + ret = adtfn(set, &data, timeout); + + if (ret && !ip_set_eexist(ret, flags)) + return ret; + else + ret = 0; + } + return ret; +} + +/* Create hash:ip type of sets */ + +static int +hash_netport_create(struct ip_set *set, struct nlattr *tb[], u32 flags) +{ + struct ip_set_hash *h; + u32 hashsize = IPSET_DEFAULT_HASHSIZE, maxelem = IPSET_DEFAULT_MAXELEM; + u8 hbits; + + if (!(set->family == AF_INET || set->family == AF_INET6)) + return -IPSET_ERR_INVALID_FAMILY; + + if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_HASHSIZE) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_MAXELEM) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_HASHSIZE]) { + hashsize = ip_set_get_h32(tb[IPSET_ATTR_HASHSIZE]); + if (hashsize < IPSET_MIMINAL_HASHSIZE) + hashsize = IPSET_MIMINAL_HASHSIZE; + } + + if (tb[IPSET_ATTR_MAXELEM]) + maxelem = ip_set_get_h32(tb[IPSET_ATTR_MAXELEM]); + + h = kzalloc(sizeof(*h) + + sizeof(struct ip_set_hash_nets) + * (set->family == AF_INET ? 32 : 128), GFP_KERNEL); + if (!h) + return -ENOMEM; + + h->maxelem = maxelem; + get_random_bytes(&h->initval, sizeof(h->initval)); + h->timeout = IPSET_NO_TIMEOUT; + + hbits = htable_bits(hashsize); + h->table = ip_set_alloc( + sizeof(struct htable) + + jhash_size(hbits) * sizeof(struct hbucket)); + if (!h->table) { + kfree(h); + return -ENOMEM; + } + h->table->htable_bits = hbits; + + set->data = h; + + if (tb[IPSET_ATTR_TIMEOUT]) { + h->timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + + set->variant = set->family == AF_INET + ? &hash_netport4_tvariant : &hash_netport6_tvariant; + + if (set->family == AF_INET) + hash_netport4_gc_init(set); + else + hash_netport6_gc_init(set); + } else { + set->variant = set->family == AF_INET + ? &hash_netport4_variant : &hash_netport6_variant; + } + + pr_debug("create %s hashsize %u (%u) maxelem %u: %p(%p)\n", + set->name, jhash_size(h->table->htable_bits), + h->table->htable_bits, h->maxelem, set->data, h->table); + + return 0; +} + +static struct ip_set_type hash_netport_type __read_mostly = { + .name = "hash:net,port", + .protocol = IPSET_PROTOCOL, + .features = IPSET_TYPE_IP | IPSET_TYPE_PORT, + .dimension = IPSET_DIM_TWO, + .family = AF_UNSPEC, + .revision = 0, + .create = hash_netport_create, + .create_policy = { + [IPSET_ATTR_HASHSIZE] = { .type = NLA_U32 }, + [IPSET_ATTR_MAXELEM] = { .type = NLA_U32 }, + [IPSET_ATTR_PROBES] = { .type = NLA_U8 }, + [IPSET_ATTR_RESIZE] = { .type = NLA_U8 }, + [IPSET_ATTR_PROTO] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + }, + .adt_policy = { + [IPSET_ATTR_IP] = { .type = NLA_NESTED }, + [IPSET_ATTR_PORT] = { .type = NLA_U16 }, + [IPSET_ATTR_PORT_TO] = { .type = NLA_U16 }, + [IPSET_ATTR_PROTO] = { .type = NLA_U8 }, + [IPSET_ATTR_CIDR] = { .type = NLA_U8 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + [IPSET_ATTR_LINENO] = { .type = NLA_U32 }, + }, + .me = THIS_MODULE, +}; + +static int __init +hash_netport_init(void) +{ + return ip_set_type_register(&hash_netport_type); +} + +static void __exit +hash_netport_fini(void) +{ + ip_set_type_unregister(&hash_netport_type); +} + +module_init(hash_netport_init); +module_exit(hash_netport_fini); -- cgit v1.2.3 From f830837f0eed0f9e371b8fd65169365780814bb1 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 1 Feb 2011 15:54:59 +0100 Subject: netfilter: ipset: list:set set type support The module implements the list:set type support in two flavours: without and with timeout. The sets has two sides: for the userspace, they store the names of other (non list:set type of) sets: one can add, delete and test set names. For the kernel, it forms an ordered union of the member sets: the members sets are tried in order when elements are added, deleted and tested and the process stops at the first success. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/ipset/Kconfig | 10 + net/netfilter/ipset/Makefile | 3 + net/netfilter/ipset/ip_set_list_set.c | 584 ++++++++++++++++++++++++++++++++++ 3 files changed, 597 insertions(+) create mode 100644 net/netfilter/ipset/ip_set_list_set.c (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/Kconfig b/net/netfilter/ipset/Kconfig index 2512e7b82d12..3b970d343023 100644 --- a/net/netfilter/ipset/Kconfig +++ b/net/netfilter/ipset/Kconfig @@ -108,4 +108,14 @@ config IP_SET_HASH_NETPORT To compile it as a module, choose M here. If unsure, say N. +config IP_SET_LIST_SET + tristate "list:set set support" + depends on IP_SET + help + This option adds the list:set set type support. In this + kind of set one can store the name of other sets and it forms + an ordered union of the member sets. + + To compile it as a module, choose M here. If unsure, say N. + endif # IP_SET diff --git a/net/netfilter/ipset/Makefile b/net/netfilter/ipset/Makefile index fbbebd62bb2a..5adbdab67bd2 100644 --- a/net/netfilter/ipset/Makefile +++ b/net/netfilter/ipset/Makefile @@ -19,3 +19,6 @@ obj-$(CONFIG_IP_SET_HASH_IPPORTIP) += ip_set_hash_ipportip.o obj-$(CONFIG_IP_SET_HASH_IPPORTNET) += ip_set_hash_ipportnet.o obj-$(CONFIG_IP_SET_HASH_NET) += ip_set_hash_net.o obj-$(CONFIG_IP_SET_HASH_NETPORT) += ip_set_hash_netport.o + +# list types +obj-$(CONFIG_IP_SET_LIST_SET) += ip_set_list_set.o diff --git a/net/netfilter/ipset/ip_set_list_set.c b/net/netfilter/ipset/ip_set_list_set.c new file mode 100644 index 000000000000..a47c32982f06 --- /dev/null +++ b/net/netfilter/ipset/ip_set_list_set.c @@ -0,0 +1,584 @@ +/* Copyright (C) 2008-2011 Jozsef Kadlecsik + * + * 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. + */ + +/* Kernel module implementing an IP set type: the list:set type */ + +#include +#include +#include +#include + +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Jozsef Kadlecsik "); +MODULE_DESCRIPTION("list:set type of IP sets"); +MODULE_ALIAS("ip_set_list:set"); + +/* Member elements without and with timeout */ +struct set_elem { + ip_set_id_t id; +}; + +struct set_telem { + ip_set_id_t id; + unsigned long timeout; +}; + +/* Type structure */ +struct list_set { + size_t dsize; /* element size */ + u32 size; /* size of set list array */ + u32 timeout; /* timeout value */ + struct timer_list gc; /* garbage collection */ + struct set_elem members[0]; /* the set members */ +}; + +static inline struct set_elem * +list_set_elem(const struct list_set *map, u32 id) +{ + return (struct set_elem *)((char *)map->members + id * map->dsize); +} + +static inline bool +list_set_timeout(const struct list_set *map, u32 id) +{ + const struct set_telem *elem = + (const struct set_telem *) list_set_elem(map, id); + + return ip_set_timeout_test(elem->timeout); +} + +static inline bool +list_set_expired(const struct list_set *map, u32 id) +{ + const struct set_telem *elem = + (const struct set_telem *) list_set_elem(map, id); + + return ip_set_timeout_expired(elem->timeout); +} + +static inline int +list_set_exist(const struct set_telem *elem) +{ + return elem->id != IPSET_INVALID_ID && + !ip_set_timeout_expired(elem->timeout); +} + +/* Set list without and with timeout */ + +static int +list_set_kadt(struct ip_set *set, const struct sk_buff *skb, + enum ipset_adt adt, u8 pf, u8 dim, u8 flags) +{ + struct list_set *map = set->data; + struct set_elem *elem; + u32 i; + int ret; + + for (i = 0; i < map->size; i++) { + elem = list_set_elem(map, i); + if (elem->id == IPSET_INVALID_ID) + return 0; + if (with_timeout(map->timeout) && list_set_expired(map, i)) + continue; + switch (adt) { + case IPSET_TEST: + ret = ip_set_test(elem->id, skb, pf, dim, flags); + if (ret > 0) + return ret; + break; + case IPSET_ADD: + ret = ip_set_add(elem->id, skb, pf, dim, flags); + if (ret == 0) + return ret; + break; + case IPSET_DEL: + ret = ip_set_del(elem->id, skb, pf, dim, flags); + if (ret == 0) + return ret; + break; + default: + break; + } + } + return -EINVAL; +} + +static bool +next_id_eq(const struct list_set *map, u32 i, ip_set_id_t id) +{ + const struct set_elem *elem; + + if (i + 1 < map->size) { + elem = list_set_elem(map, i + 1); + return !!(elem->id == id && + !(with_timeout(map->timeout) && + list_set_expired(map, i + 1))); + } + + return 0; +} + +static void +list_elem_add(struct list_set *map, u32 i, ip_set_id_t id) +{ + struct set_elem *e; + + for (; i < map->size; i++) { + e = list_set_elem(map, i); + swap(e->id, id); + if (e->id == IPSET_INVALID_ID) + break; + } +} + +static void +list_elem_tadd(struct list_set *map, u32 i, ip_set_id_t id, + unsigned long timeout) +{ + struct set_telem *e; + + for (; i < map->size; i++) { + e = (struct set_telem *)list_set_elem(map, i); + swap(e->id, id); + if (e->id == IPSET_INVALID_ID) + break; + swap(e->timeout, timeout); + } +} + +static int +list_set_add(struct list_set *map, u32 i, ip_set_id_t id, + unsigned long timeout) +{ + const struct set_elem *e = list_set_elem(map, i); + + if (i == map->size - 1 && e->id != IPSET_INVALID_ID) + /* Last element replaced: e.g. add new,before,last */ + ip_set_put_byindex(e->id); + if (with_timeout(map->timeout)) + list_elem_tadd(map, i, id, timeout); + else + list_elem_add(map, i, id); + + return 0; +} + +static int +list_set_del(struct list_set *map, ip_set_id_t id, u32 i) +{ + struct set_elem *a = list_set_elem(map, i), *b; + + ip_set_put_byindex(id); + + for (; i < map->size - 1; i++) { + b = list_set_elem(map, i + 1); + a->id = b->id; + if (with_timeout(map->timeout)) + ((struct set_telem *)a)->timeout = + ((struct set_telem *)b)->timeout; + a = b; + if (a->id == IPSET_INVALID_ID) + break; + } + /* Last element */ + a->id = IPSET_INVALID_ID; + return 0; +} + +static int +list_set_uadt(struct ip_set *set, struct nlattr *tb[], + enum ipset_adt adt, u32 *lineno, u32 flags) +{ + struct list_set *map = set->data; + bool with_timeout = with_timeout(map->timeout); + int before = 0; + u32 timeout = map->timeout; + ip_set_id_t id, refid = IPSET_INVALID_ID; + const struct set_elem *elem; + struct ip_set *s; + u32 i; + int ret = 0; + + if (unlikely(!tb[IPSET_ATTR_NAME] || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_CADT_FLAGS))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_LINENO]) + *lineno = nla_get_u32(tb[IPSET_ATTR_LINENO]); + + id = ip_set_get_byname(nla_data(tb[IPSET_ATTR_NAME]), &s); + if (id == IPSET_INVALID_ID) + return -IPSET_ERR_NAME; + /* "Loop detection" */ + if (s->type->features & IPSET_TYPE_NAME) { + ret = -IPSET_ERR_LOOP; + goto finish; + } + + if (tb[IPSET_ATTR_CADT_FLAGS]) { + u32 f = ip_set_get_h32(tb[IPSET_ATTR_CADT_FLAGS]); + before = f & IPSET_FLAG_BEFORE; + } + + if (before && !tb[IPSET_ATTR_NAMEREF]) { + ret = -IPSET_ERR_BEFORE; + goto finish; + } + + if (tb[IPSET_ATTR_NAMEREF]) { + refid = ip_set_get_byname(nla_data(tb[IPSET_ATTR_NAMEREF]), + &s); + if (refid == IPSET_INVALID_ID) { + ret = -IPSET_ERR_NAMEREF; + goto finish; + } + if (!before) + before = -1; + } + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!with_timeout) { + ret = -IPSET_ERR_TIMEOUT; + goto finish; + } + timeout = ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]); + } + + switch (adt) { + case IPSET_TEST: + for (i = 0; i < map->size && !ret; i++) { + elem = list_set_elem(map, i); + if (elem->id == IPSET_INVALID_ID || + (before != 0 && i + 1 >= map->size)) + break; + else if (with_timeout && list_set_expired(map, i)) + continue; + else if (before > 0 && elem->id == id) + ret = next_id_eq(map, i, refid); + else if (before < 0 && elem->id == refid) + ret = next_id_eq(map, i, id); + else if (before == 0 && elem->id == id) + ret = 1; + } + break; + case IPSET_ADD: + for (i = 0; i < map->size && !ret; i++) { + elem = list_set_elem(map, i); + if (elem->id == id && + !(with_timeout && list_set_expired(map, i))) + ret = -IPSET_ERR_EXIST; + } + if (ret == -IPSET_ERR_EXIST) + break; + ret = -IPSET_ERR_LIST_FULL; + for (i = 0; i < map->size && ret == -IPSET_ERR_LIST_FULL; i++) { + elem = list_set_elem(map, i); + if (elem->id == IPSET_INVALID_ID) + ret = before != 0 ? -IPSET_ERR_REF_EXIST + : list_set_add(map, i, id, timeout); + else if (elem->id != refid) + continue; + else if (with_timeout && list_set_expired(map, i)) + ret = -IPSET_ERR_REF_EXIST; + else if (before) + ret = list_set_add(map, i, id, timeout); + else if (i + 1 < map->size) + ret = list_set_add(map, i + 1, id, timeout); + } + break; + case IPSET_DEL: + ret = -IPSET_ERR_EXIST; + for (i = 0; i < map->size && ret == -IPSET_ERR_EXIST; i++) { + elem = list_set_elem(map, i); + if (elem->id == IPSET_INVALID_ID) { + ret = before != 0 ? -IPSET_ERR_REF_EXIST + : -IPSET_ERR_EXIST; + break; + } else if (with_timeout && list_set_expired(map, i)) + continue; + else if (elem->id == id && + (before == 0 || + (before > 0 && + next_id_eq(map, i, refid)))) + ret = list_set_del(map, id, i); + else if (before < 0 && + elem->id == refid && + next_id_eq(map, i, id)) + ret = list_set_del(map, id, i + 1); + } + break; + default: + break; + } + +finish: + if (refid != IPSET_INVALID_ID) + ip_set_put_byindex(refid); + if (adt != IPSET_ADD || ret) + ip_set_put_byindex(id); + + return ip_set_eexist(ret, flags) ? 0 : ret; +} + +static void +list_set_flush(struct ip_set *set) +{ + struct list_set *map = set->data; + struct set_elem *elem; + u32 i; + + for (i = 0; i < map->size; i++) { + elem = list_set_elem(map, i); + if (elem->id != IPSET_INVALID_ID) { + ip_set_put_byindex(elem->id); + elem->id = IPSET_INVALID_ID; + } + } +} + +static void +list_set_destroy(struct ip_set *set) +{ + struct list_set *map = set->data; + + if (with_timeout(map->timeout)) + del_timer_sync(&map->gc); + list_set_flush(set); + kfree(map); + + set->data = NULL; +} + +static int +list_set_head(struct ip_set *set, struct sk_buff *skb) +{ + const struct list_set *map = set->data; + struct nlattr *nested; + + nested = ipset_nest_start(skb, IPSET_ATTR_DATA); + if (!nested) + goto nla_put_failure; + NLA_PUT_NET32(skb, IPSET_ATTR_SIZE, htonl(map->size)); + if (with_timeout(map->timeout)) + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, htonl(map->timeout)); + NLA_PUT_NET32(skb, IPSET_ATTR_REFERENCES, + htonl(atomic_read(&set->ref) - 1)); + NLA_PUT_NET32(skb, IPSET_ATTR_MEMSIZE, + htonl(sizeof(*map) + map->size * map->dsize)); + ipset_nest_end(skb, nested); + + return 0; +nla_put_failure: + return -EMSGSIZE; +} + +static int +list_set_list(const struct ip_set *set, + struct sk_buff *skb, struct netlink_callback *cb) +{ + const struct list_set *map = set->data; + struct nlattr *atd, *nested; + u32 i, first = cb->args[2]; + const struct set_elem *e; + + atd = ipset_nest_start(skb, IPSET_ATTR_ADT); + if (!atd) + return -EMSGSIZE; + for (; cb->args[2] < map->size; cb->args[2]++) { + i = cb->args[2]; + e = list_set_elem(map, i); + if (e->id == IPSET_INVALID_ID) + goto finish; + if (with_timeout(map->timeout) && list_set_expired(map, i)) + continue; + nested = ipset_nest_start(skb, IPSET_ATTR_DATA); + if (!nested) { + if (i == first) { + nla_nest_cancel(skb, atd); + return -EMSGSIZE; + } else + goto nla_put_failure; + } + NLA_PUT_STRING(skb, IPSET_ATTR_NAME, + ip_set_name_byindex(e->id)); + if (with_timeout(map->timeout)) { + const struct set_telem *te = + (const struct set_telem *) e; + NLA_PUT_NET32(skb, IPSET_ATTR_TIMEOUT, + htonl(ip_set_timeout_get(te->timeout))); + } + ipset_nest_end(skb, nested); + } +finish: + ipset_nest_end(skb, atd); + /* Set listing finished */ + cb->args[2] = 0; + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nested); + ipset_nest_end(skb, atd); + if (unlikely(i == first)) { + cb->args[2] = 0; + return -EMSGSIZE; + } + return 0; +} + +static bool +list_set_same_set(const struct ip_set *a, const struct ip_set *b) +{ + const struct list_set *x = a->data; + const struct list_set *y = b->data; + + return x->size == y->size && + x->timeout == y->timeout; +} + +static const struct ip_set_type_variant list_set = { + .kadt = list_set_kadt, + .uadt = list_set_uadt, + .destroy = list_set_destroy, + .flush = list_set_flush, + .head = list_set_head, + .list = list_set_list, + .same_set = list_set_same_set, +}; + +static void +list_set_gc(unsigned long ul_set) +{ + struct ip_set *set = (struct ip_set *) ul_set; + struct list_set *map = set->data; + struct set_telem *e; + u32 i; + + /* We run parallel with other readers (test element) + * but adding/deleting new entries is locked out */ + read_lock_bh(&set->lock); + for (i = map->size - 1; i >= 0; i--) { + e = (struct set_telem *) list_set_elem(map, i); + if (e->id != IPSET_INVALID_ID && + list_set_expired(map, i)) + list_set_del(map, e->id, i); + } + read_unlock_bh(&set->lock); + + map->gc.expires = jiffies + IPSET_GC_PERIOD(map->timeout) * HZ; + add_timer(&map->gc); +} + +static void +list_set_gc_init(struct ip_set *set) +{ + struct list_set *map = set->data; + + init_timer(&map->gc); + map->gc.data = (unsigned long) set; + map->gc.function = list_set_gc; + map->gc.expires = jiffies + IPSET_GC_PERIOD(map->timeout) * HZ; + add_timer(&map->gc); +} + +/* Create list:set type of sets */ + +static bool +init_list_set(struct ip_set *set, u32 size, size_t dsize, + unsigned long timeout) +{ + struct list_set *map; + struct set_elem *e; + u32 i; + + map = kzalloc(sizeof(*map) + size * dsize, GFP_KERNEL); + if (!map) + return false; + + map->size = size; + map->dsize = dsize; + map->timeout = timeout; + set->data = map; + + for (i = 0; i < size; i++) { + e = list_set_elem(map, i); + e->id = IPSET_INVALID_ID; + } + + return true; +} + +static int +list_set_create(struct ip_set *set, struct nlattr *tb[], u32 flags) +{ + u32 size = IP_SET_LIST_DEFAULT_SIZE; + + if (unlikely(!ip_set_optattr_netorder(tb, IPSET_ATTR_SIZE) || + !ip_set_optattr_netorder(tb, IPSET_ATTR_TIMEOUT))) + return -IPSET_ERR_PROTOCOL; + + if (tb[IPSET_ATTR_SIZE]) + size = ip_set_get_h32(tb[IPSET_ATTR_SIZE]); + if (size < IP_SET_LIST_MIN_SIZE) + size = IP_SET_LIST_MIN_SIZE; + + if (tb[IPSET_ATTR_TIMEOUT]) { + if (!init_list_set(set, size, sizeof(struct set_telem), + ip_set_timeout_uget(tb[IPSET_ATTR_TIMEOUT]))) + return -ENOMEM; + + list_set_gc_init(set); + } else { + if (!init_list_set(set, size, sizeof(struct set_elem), + IPSET_NO_TIMEOUT)) + return -ENOMEM; + } + set->variant = &list_set; + return 0; +} + +static struct ip_set_type list_set_type __read_mostly = { + .name = "list:set", + .protocol = IPSET_PROTOCOL, + .features = IPSET_TYPE_NAME | IPSET_DUMP_LAST, + .dimension = IPSET_DIM_ONE, + .family = AF_UNSPEC, + .revision = 0, + .create = list_set_create, + .create_policy = { + [IPSET_ATTR_SIZE] = { .type = NLA_U32 }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + }, + .adt_policy = { + [IPSET_ATTR_NAME] = { .type = NLA_STRING, + .len = IPSET_MAXNAMELEN }, + [IPSET_ATTR_NAMEREF] = { .type = NLA_STRING, + .len = IPSET_MAXNAMELEN }, + [IPSET_ATTR_TIMEOUT] = { .type = NLA_U32 }, + [IPSET_ATTR_LINENO] = { .type = NLA_U32 }, + [IPSET_ATTR_CADT_FLAGS] = { .type = NLA_U32 }, + }, + .me = THIS_MODULE, +}; + +static int __init +list_set_init(void) +{ + return ip_set_type_register(&list_set_type); +} + +static void __exit +list_set_fini(void) +{ + ip_set_type_unregister(&list_set_type); +} + +module_init(list_set_init); +module_exit(list_set_fini); -- cgit v1.2.3 From d956798d82d2d331c031301965d69e17a1a48a2b Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Tue, 1 Feb 2011 15:56:00 +0100 Subject: netfilter: xtables: "set" match and "SET" target support The patch adds the combined module of the "SET" target and "set" match to netfilter. Both the previous and the current revisions are supported. Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/Kconfig | 12 ++ net/netfilter/Makefile | 1 + net/netfilter/xt_set.c | 359 +++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 372 insertions(+) create mode 100644 net/netfilter/xt_set.c (limited to 'net/netfilter') diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 351abf8ace13..06fa9e4e45c7 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -352,6 +352,18 @@ config NETFILTER_XT_CONNMARK ctmark), similarly to the packet mark (nfmark). Using this target and match, you can set and match on this mark. +config NETFILTER_XT_SET + tristate 'set target and match support' + depends on IP_SET + depends on NETFILTER_ADVANCED + help + This option adds the "SET" target and "set" match. + + Using this target and match, you can add/delete and match + elements in the sets created by ipset(8). + + To compile it as a module, choose M here. If unsure, say N. + # alphabetically ordered list of targets comment "Xtables targets" diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 510b586ccb7f..1148643559cb 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -46,6 +46,7 @@ obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o # combos obj-$(CONFIG_NETFILTER_XT_MARK) += xt_mark.o obj-$(CONFIG_NETFILTER_XT_CONNMARK) += xt_connmark.o +obj-$(CONFIG_NETFILTER_XT_SET) += xt_set.o # targets obj-$(CONFIG_NETFILTER_XT_TARGET_AUDIT) += xt_AUDIT.o diff --git a/net/netfilter/xt_set.c b/net/netfilter/xt_set.c new file mode 100644 index 000000000000..061d48cec137 --- /dev/null +++ b/net/netfilter/xt_set.c @@ -0,0 +1,359 @@ +/* Copyright (C) 2000-2002 Joakim Axelsson + * Patrick Schaaf + * Martin Josefsson + * Copyright (C) 2003-2011 Jozsef Kadlecsik + * + * 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. + */ + +/* Kernel module which implements the set match and SET target + * for netfilter/iptables. */ + +#include +#include +#include + +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Jozsef Kadlecsik "); +MODULE_DESCRIPTION("Xtables: IP set match and target module"); +MODULE_ALIAS("xt_SET"); +MODULE_ALIAS("ipt_set"); +MODULE_ALIAS("ip6t_set"); +MODULE_ALIAS("ipt_SET"); +MODULE_ALIAS("ip6t_SET"); + +static inline int +match_set(ip_set_id_t index, const struct sk_buff *skb, + u8 pf, u8 dim, u8 flags, int inv) +{ + if (ip_set_test(index, skb, pf, dim, flags)) + inv = !inv; + return inv; +} + +/* Revision 0 interface: backward compatible with netfilter/iptables */ + +static bool +set_match_v0(const struct sk_buff *skb, struct xt_action_param *par) +{ + const struct xt_set_info_match_v0 *info = par->matchinfo; + + return match_set(info->match_set.index, skb, par->family, + info->match_set.u.compat.dim, + info->match_set.u.compat.flags, + info->match_set.u.compat.flags & IPSET_INV_MATCH); +} + +static void +compat_flags(struct xt_set_info_v0 *info) +{ + u_int8_t i; + + /* Fill out compatibility data according to enum ip_set_kopt */ + info->u.compat.dim = IPSET_DIM_ZERO; + if (info->u.flags[0] & IPSET_MATCH_INV) + info->u.compat.flags |= IPSET_INV_MATCH; + for (i = 0; i < IPSET_DIM_MAX-1 && info->u.flags[i]; i++) { + info->u.compat.dim++; + if (info->u.flags[i] & IPSET_SRC) + info->u.compat.flags |= (1<u.compat.dim); + } +} + +static int +set_match_v0_checkentry(const struct xt_mtchk_param *par) +{ + struct xt_set_info_match_v0 *info = par->matchinfo; + ip_set_id_t index; + + index = ip_set_nfnl_get_byindex(info->match_set.index); + + if (index == IPSET_INVALID_ID) { + pr_warning("Cannot find set indentified by id %u to match\n", + info->match_set.index); + return -ENOENT; + } + if (info->match_set.u.flags[IPSET_DIM_MAX-1] != 0) { + pr_warning("Protocol error: set match dimension " + "is over the limit!\n"); + return -ERANGE; + } + + /* Fill out compatibility data */ + compat_flags(&info->match_set); + + return 0; +} + +static void +set_match_v0_destroy(const struct xt_mtdtor_param *par) +{ + struct xt_set_info_match_v0 *info = par->matchinfo; + + ip_set_nfnl_put(info->match_set.index); +} + +static unsigned int +set_target_v0(struct sk_buff *skb, const struct xt_action_param *par) +{ + const struct xt_set_info_target_v0 *info = par->targinfo; + + if (info->add_set.index != IPSET_INVALID_ID) + ip_set_add(info->add_set.index, skb, par->family, + info->add_set.u.compat.dim, + info->add_set.u.compat.flags); + if (info->del_set.index != IPSET_INVALID_ID) + ip_set_del(info->del_set.index, skb, par->family, + info->del_set.u.compat.dim, + info->del_set.u.compat.flags); + + return XT_CONTINUE; +} + +static int +set_target_v0_checkentry(const struct xt_tgchk_param *par) +{ + struct xt_set_info_target_v0 *info = par->targinfo; + ip_set_id_t index; + + if (info->add_set.index != IPSET_INVALID_ID) { + index = ip_set_nfnl_get_byindex(info->add_set.index); + if (index == IPSET_INVALID_ID) { + pr_warning("Cannot find add_set index %u as target\n", + info->add_set.index); + return -ENOENT; + } + } + + if (info->del_set.index != IPSET_INVALID_ID) { + index = ip_set_nfnl_get_byindex(info->del_set.index); + if (index == IPSET_INVALID_ID) { + pr_warning("Cannot find del_set index %u as target\n", + info->del_set.index); + return -ENOENT; + } + } + if (info->add_set.u.flags[IPSET_DIM_MAX-1] != 0 || + info->del_set.u.flags[IPSET_DIM_MAX-1] != 0) { + pr_warning("Protocol error: SET target dimension " + "is over the limit!\n"); + return -ERANGE; + } + + /* Fill out compatibility data */ + compat_flags(&info->add_set); + compat_flags(&info->del_set); + + return 0; +} + +static void +set_target_v0_destroy(const struct xt_tgdtor_param *par) +{ + const struct xt_set_info_target_v0 *info = par->targinfo; + + if (info->add_set.index != IPSET_INVALID_ID) + ip_set_nfnl_put(info->add_set.index); + if (info->del_set.index != IPSET_INVALID_ID) + ip_set_nfnl_put(info->del_set.index); +} + +/* Revision 1: current interface to netfilter/iptables */ + +static bool +set_match(const struct sk_buff *skb, struct xt_action_param *par) +{ + const struct xt_set_info_match *info = par->matchinfo; + + return match_set(info->match_set.index, skb, par->family, + info->match_set.dim, + info->match_set.flags, + info->match_set.flags & IPSET_INV_MATCH); +} + +static int +set_match_checkentry(const struct xt_mtchk_param *par) +{ + struct xt_set_info_match *info = par->matchinfo; + ip_set_id_t index; + + index = ip_set_nfnl_get_byindex(info->match_set.index); + + if (index == IPSET_INVALID_ID) { + pr_warning("Cannot find set indentified by id %u to match\n", + info->match_set.index); + return -ENOENT; + } + if (info->match_set.dim > IPSET_DIM_MAX) { + pr_warning("Protocol error: set match dimension " + "is over the limit!\n"); + return -ERANGE; + } + + return 0; +} + +static void +set_match_destroy(const struct xt_mtdtor_param *par) +{ + struct xt_set_info_match *info = par->matchinfo; + + ip_set_nfnl_put(info->match_set.index); +} + +static unsigned int +set_target(struct sk_buff *skb, const struct xt_action_param *par) +{ + const struct xt_set_info_target *info = par->targinfo; + + if (info->add_set.index != IPSET_INVALID_ID) + ip_set_add(info->add_set.index, + skb, par->family, + info->add_set.dim, + info->add_set.flags); + if (info->del_set.index != IPSET_INVALID_ID) + ip_set_del(info->del_set.index, + skb, par->family, + info->add_set.dim, + info->del_set.flags); + + return XT_CONTINUE; +} + +static int +set_target_checkentry(const struct xt_tgchk_param *par) +{ + const struct xt_set_info_target *info = par->targinfo; + ip_set_id_t index; + + if (info->add_set.index != IPSET_INVALID_ID) { + index = ip_set_nfnl_get_byindex(info->add_set.index); + if (index == IPSET_INVALID_ID) { + pr_warning("Cannot find add_set index %u as target\n", + info->add_set.index); + return -ENOENT; + } + } + + if (info->del_set.index != IPSET_INVALID_ID) { + index = ip_set_nfnl_get_byindex(info->del_set.index); + if (index == IPSET_INVALID_ID) { + pr_warning("Cannot find del_set index %u as target\n", + info->del_set.index); + return -ENOENT; + } + } + if (info->add_set.dim > IPSET_DIM_MAX || + info->del_set.flags > IPSET_DIM_MAX) { + pr_warning("Protocol error: SET target dimension " + "is over the limit!\n"); + return -ERANGE; + } + + return 0; +} + +static void +set_target_destroy(const struct xt_tgdtor_param *par) +{ + const struct xt_set_info_target *info = par->targinfo; + + if (info->add_set.index != IPSET_INVALID_ID) + ip_set_nfnl_put(info->add_set.index); + if (info->del_set.index != IPSET_INVALID_ID) + ip_set_nfnl_put(info->del_set.index); +} + +static struct xt_match set_matches[] __read_mostly = { + { + .name = "set", + .family = NFPROTO_IPV4, + .revision = 0, + .match = set_match_v0, + .matchsize = sizeof(struct xt_set_info_match_v0), + .checkentry = set_match_v0_checkentry, + .destroy = set_match_v0_destroy, + .me = THIS_MODULE + }, + { + .name = "set", + .family = NFPROTO_IPV4, + .revision = 1, + .match = set_match, + .matchsize = sizeof(struct xt_set_info_match), + .checkentry = set_match_checkentry, + .destroy = set_match_destroy, + .me = THIS_MODULE + }, + { + .name = "set", + .family = NFPROTO_IPV6, + .revision = 1, + .match = set_match, + .matchsize = sizeof(struct xt_set_info_match), + .checkentry = set_match_checkentry, + .destroy = set_match_destroy, + .me = THIS_MODULE + }, +}; + +static struct xt_target set_targets[] __read_mostly = { + { + .name = "SET", + .revision = 0, + .family = NFPROTO_IPV4, + .target = set_target_v0, + .targetsize = sizeof(struct xt_set_info_target_v0), + .checkentry = set_target_v0_checkentry, + .destroy = set_target_v0_destroy, + .me = THIS_MODULE + }, + { + .name = "SET", + .revision = 1, + .family = NFPROTO_IPV4, + .target = set_target, + .targetsize = sizeof(struct xt_set_info_target), + .checkentry = set_target_checkentry, + .destroy = set_target_destroy, + .me = THIS_MODULE + }, + { + .name = "SET", + .revision = 1, + .family = NFPROTO_IPV6, + .target = set_target, + .targetsize = sizeof(struct xt_set_info_target), + .checkentry = set_target_checkentry, + .destroy = set_target_destroy, + .me = THIS_MODULE + }, +}; + +static int __init xt_set_init(void) +{ + int ret = xt_register_matches(set_matches, ARRAY_SIZE(set_matches)); + + if (!ret) { + ret = xt_register_targets(set_targets, + ARRAY_SIZE(set_targets)); + if (ret) + xt_unregister_matches(set_matches, + ARRAY_SIZE(set_matches)); + } + return ret; +} + +static void __exit xt_set_fini(void) +{ + xt_unregister_matches(set_matches, ARRAY_SIZE(set_matches)); + xt_unregister_targets(set_targets, ARRAY_SIZE(set_targets)); +} + +module_init(xt_set_init); +module_exit(xt_set_fini); -- cgit v1.2.3 From 8da560ced56c423cd6d35803cd0244c944c676bd Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 1 Feb 2011 16:27:25 +0100 Subject: netfilter: ipset: use nla_parse_nested() Replace calls of the form: nla_parse(tb, ATTR_MAX, nla_data(attr), nla_len(attr), policy) by: nla_parse_nested(tb, ATTR_MAX, attr, policy) Signed-off-by: Patrick McHardy --- net/netfilter/ipset/ip_set_core.c | 42 +++++++++++++++------------------------ 1 file changed, 16 insertions(+), 26 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c index 8a736247e85f..ae0f8b595106 100644 --- a/net/netfilter/ipset/ip_set_core.c +++ b/net/netfilter/ipset/ip_set_core.c @@ -246,8 +246,7 @@ ip_set_get_ipaddr4(struct nlattr *nla, __be32 *ipaddr) if (unlikely(!flag_nested(nla))) return -IPSET_ERR_PROTOCOL; - if (nla_parse(tb, IPSET_ATTR_IPADDR_MAX, nla_data(nla), nla_len(nla), - ipaddr_policy)) + if (nla_parse_nested(tb, IPSET_ATTR_IPADDR_MAX, nla, ipaddr_policy)) return -IPSET_ERR_PROTOCOL; if (unlikely(!ip_set_attr_netorder(tb, IPSET_ATTR_IPADDR_IPV4))) return -IPSET_ERR_PROTOCOL; @@ -265,8 +264,7 @@ ip_set_get_ipaddr6(struct nlattr *nla, union nf_inet_addr *ipaddr) if (unlikely(!flag_nested(nla))) return -IPSET_ERR_PROTOCOL; - if (nla_parse(tb, IPSET_ATTR_IPADDR_MAX, nla_data(nla), nla_len(nla), - ipaddr_policy)) + if (nla_parse_nested(tb, IPSET_ATTR_IPADDR_MAX, nla, ipaddr_policy)) return -IPSET_ERR_PROTOCOL; if (unlikely(!ip_set_attr_netorder(tb, IPSET_ATTR_IPADDR_IPV6))) return -IPSET_ERR_PROTOCOL; @@ -666,10 +664,8 @@ ip_set_create(struct sock *ctnl, struct sk_buff *skb, * Without holding any locks, create private part. */ if (attr[IPSET_ATTR_DATA] && - nla_parse(tb, IPSET_ATTR_CREATE_MAX, - nla_data(attr[IPSET_ATTR_DATA]), - nla_len(attr[IPSET_ATTR_DATA]), - set->type->create_policy)) { + nla_parse_nested(tb, IPSET_ATTR_CREATE_MAX, attr[IPSET_ATTR_DATA], + set->type->create_policy)) { ret = -IPSET_ERR_PROTOCOL; goto put_out; } @@ -1169,10 +1165,9 @@ ip_set_uadd(struct sock *ctnl, struct sk_buff *skb, use_lineno = !!attr[IPSET_ATTR_LINENO]; if (attr[IPSET_ATTR_DATA]) { - if (nla_parse(tb, IPSET_ATTR_ADT_MAX, - nla_data(attr[IPSET_ATTR_DATA]), - nla_len(attr[IPSET_ATTR_DATA]), - set->type->adt_policy)) + if (nla_parse_nested(tb, IPSET_ATTR_ADT_MAX, + attr[IPSET_ATTR_DATA], + set->type->adt_policy)) return -IPSET_ERR_PROTOCOL; ret = call_ad(skb, set, tb, IPSET_ADD, flags, use_lineno); } else { @@ -1182,9 +1177,8 @@ ip_set_uadd(struct sock *ctnl, struct sk_buff *skb, memset(tb, 0, sizeof(tb)); if (nla_type(nla) != IPSET_ATTR_DATA || !flag_nested(nla) || - nla_parse(tb, IPSET_ATTR_ADT_MAX, - nla_data(nla), nla_len(nla), - set->type->adt_policy)) + nla_parse_nested(tb, IPSET_ATTR_ADT_MAX, nla, + set->type->adt_policy)) return -IPSET_ERR_PROTOCOL; ret = call_ad(skb, set, tb, IPSET_ADD, flags, use_lineno); @@ -1224,10 +1218,9 @@ ip_set_udel(struct sock *ctnl, struct sk_buff *skb, use_lineno = !!attr[IPSET_ATTR_LINENO]; if (attr[IPSET_ATTR_DATA]) { - if (nla_parse(tb, IPSET_ATTR_ADT_MAX, - nla_data(attr[IPSET_ATTR_DATA]), - nla_len(attr[IPSET_ATTR_DATA]), - set->type->adt_policy)) + if (nla_parse_nested(tb, IPSET_ATTR_ADT_MAX, + attr[IPSET_ATTR_DATA], + set->type->adt_policy)) return -IPSET_ERR_PROTOCOL; ret = call_ad(skb, set, tb, IPSET_DEL, flags, use_lineno); } else { @@ -1237,9 +1230,8 @@ ip_set_udel(struct sock *ctnl, struct sk_buff *skb, memset(tb, 0, sizeof(*tb)); if (nla_type(nla) != IPSET_ATTR_DATA || !flag_nested(nla) || - nla_parse(tb, IPSET_ATTR_ADT_MAX, - nla_data(nla), nla_len(nla), - set->type->adt_policy)) + nla_parse_nested(tb, IPSET_ATTR_ADT_MAX, nla, + set->type->adt_policy)) return -IPSET_ERR_PROTOCOL; ret = call_ad(skb, set, tb, IPSET_DEL, flags, use_lineno); @@ -1269,10 +1261,8 @@ ip_set_utest(struct sock *ctnl, struct sk_buff *skb, if (set == NULL) return -ENOENT; - if (nla_parse(tb, IPSET_ATTR_ADT_MAX, - nla_data(attr[IPSET_ATTR_DATA]), - nla_len(attr[IPSET_ATTR_DATA]), - set->type->adt_policy)) + if (nla_parse_nested(tb, IPSET_ATTR_ADT_MAX, attr[IPSET_ATTR_DATA], + set->type->adt_policy)) return -IPSET_ERR_PROTOCOL; read_lock_bh(&set->lock); -- cgit v1.2.3 From 582e1fc85ca3727abd4e99109a267c514ea5c260 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 1 Feb 2011 16:57:37 +0100 Subject: netfilter: ipset: remove unnecessary includes None of the set types need uaccess.h since this is handled centrally in ip_set_core. Most set types additionally don't need bitops.h and spinlock.h since they use neither. tcp.h is only needed by those using before(), udp.h is not needed at all. Signed-off-by: Patrick McHardy --- net/netfilter/ipset/ip_set_bitmap_ip.c | 1 - net/netfilter/ipset/ip_set_bitmap_ipmac.c | 3 --- net/netfilter/ipset/ip_set_bitmap_port.c | 5 ----- net/netfilter/ipset/ip_set_hash_ip.c | 3 --- net/netfilter/ipset/ip_set_hash_ipport.c | 3 --- net/netfilter/ipset/ip_set_hash_ipportip.c | 3 --- net/netfilter/ipset/ip_set_hash_ipportnet.c | 3 --- net/netfilter/ipset/ip_set_hash_net.c | 3 --- net/netfilter/ipset/ip_set_hash_netport.c | 3 --- 9 files changed, 27 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/ip_set_bitmap_ip.c b/net/netfilter/ipset/ip_set_bitmap_ip.c index 047440085509..bca96990218d 100644 --- a/net/netfilter/ipset/ip_set_bitmap_ip.c +++ b/net/netfilter/ipset/ip_set_bitmap_ip.c @@ -13,7 +13,6 @@ #include #include #include -#include #include #include #include diff --git a/net/netfilter/ipset/ip_set_bitmap_ipmac.c b/net/netfilter/ipset/ip_set_bitmap_ipmac.c index d826332d2937..5e790172deff 100644 --- a/net/netfilter/ipset/ip_set_bitmap_ipmac.c +++ b/net/netfilter/ipset/ip_set_bitmap_ipmac.c @@ -15,9 +15,6 @@ #include #include #include -#include -#include -#include #include #include #include diff --git a/net/netfilter/ipset/ip_set_bitmap_port.c b/net/netfilter/ipset/ip_set_bitmap_port.c index 92074bb64134..165f09b1a9cb 100644 --- a/net/netfilter/ipset/ip_set_bitmap_port.c +++ b/net/netfilter/ipset/ip_set_bitmap_port.c @@ -9,13 +9,8 @@ #include #include -#include -#include #include #include -#include -#include -#include #include #include #include diff --git a/net/netfilter/ipset/ip_set_hash_ip.c b/net/netfilter/ipset/ip_set_hash_ip.c index 53964bc831aa..43bcce200129 100644 --- a/net/netfilter/ipset/ip_set_hash_ip.c +++ b/net/netfilter/ipset/ip_set_hash_ip.c @@ -12,9 +12,6 @@ #include #include #include -#include -#include -#include #include #include #include diff --git a/net/netfilter/ipset/ip_set_hash_ipport.c b/net/netfilter/ipset/ip_set_hash_ipport.c index d9b192818e58..adbe787ea5dc 100644 --- a/net/netfilter/ipset/ip_set_hash_ipport.c +++ b/net/netfilter/ipset/ip_set_hash_ipport.c @@ -12,9 +12,6 @@ #include #include #include -#include -#include -#include #include #include #include diff --git a/net/netfilter/ipset/ip_set_hash_ipportip.c b/net/netfilter/ipset/ip_set_hash_ipportip.c index 80dae9de5d18..22e23abb86c6 100644 --- a/net/netfilter/ipset/ip_set_hash_ipportip.c +++ b/net/netfilter/ipset/ip_set_hash_ipportip.c @@ -12,9 +12,6 @@ #include #include #include -#include -#include -#include #include #include #include diff --git a/net/netfilter/ipset/ip_set_hash_ipportnet.c b/net/netfilter/ipset/ip_set_hash_ipportnet.c index 8eacd8a46c1c..6033e8b54bbd 100644 --- a/net/netfilter/ipset/ip_set_hash_ipportnet.c +++ b/net/netfilter/ipset/ip_set_hash_ipportnet.c @@ -12,9 +12,6 @@ #include #include #include -#include -#include -#include #include #include #include diff --git a/net/netfilter/ipset/ip_set_hash_net.c b/net/netfilter/ipset/ip_set_hash_net.c index fb0e6a68d13e..c4db202b7da4 100644 --- a/net/netfilter/ipset/ip_set_hash_net.c +++ b/net/netfilter/ipset/ip_set_hash_net.c @@ -12,9 +12,6 @@ #include #include #include -#include -#include -#include #include #include #include diff --git a/net/netfilter/ipset/ip_set_hash_netport.c b/net/netfilter/ipset/ip_set_hash_netport.c index 342250f9b114..34a165626ee9 100644 --- a/net/netfilter/ipset/ip_set_hash_netport.c +++ b/net/netfilter/ipset/ip_set_hash_netport.c @@ -12,9 +12,6 @@ #include #include #include -#include -#include -#include #include #include #include -- cgit v1.2.3 From a00f1f3686d6a062b5295c092a9dff059adbdbf5 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 1 Feb 2011 17:26:37 +0100 Subject: netfilter: ctnetlink: fix ctnetlink_parse_tuple() warning net/netfilter/nf_conntrack_netlink.c: In function 'ctnetlink_parse_tuple': net/netfilter/nf_conntrack_netlink.c:832:11: warning: comparison between 'enum ctattr_tuple' and 'enum ctattr_type' Use ctattr_type for the 'type' parameter since that's the type of all attributes passed to this function. Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_netlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 61c73945bb94..cb1a8190ba29 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -803,7 +803,7 @@ static const struct nla_policy tuple_nla_policy[CTA_TUPLE_MAX+1] = { static int ctnetlink_parse_tuple(const struct nlattr * const cda[], struct nf_conntrack_tuple *tuple, - enum ctattr_tuple type, u_int8_t l3num) + enum ctattr_type type, u_int8_t l3num) { struct nlattr *tb[CTA_TUPLE_MAX+1]; int err; -- cgit v1.2.3 From a870c8c5cbe41bcf42cf4fa9f43d969b5134090b Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Tue, 1 Feb 2011 18:21:53 +0100 Subject: IPVS: use z modifier for sizeof() argument Reported-by: Randy Dunlap Signed-off-by: Simon Horman Acked-by: Randy Dunlap Signed-off-by: Hans Schillstrom Tested-by: Hans Schillstrom Signed-off-by: Patrick McHardy --- net/netfilter/ipvs/ip_vs_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index d889f4f6be99..4d06617fab6c 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -1887,7 +1887,7 @@ static int __net_init __ip_vs_init(struct net *net) ipvs->gen = atomic_read(&ipvs_netns_cnt); atomic_inc(&ipvs_netns_cnt); net->ipvs = ipvs; - printk(KERN_INFO "IPVS: Creating netns size=%lu id=%d\n", + printk(KERN_INFO "IPVS: Creating netns size=%zu id=%d\n", sizeof(struct netns_ipvs), ipvs->gen); return 0; } -- cgit v1.2.3 From 258e958b85cef23b1598515504426e8d0576d223 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Tue, 1 Feb 2011 18:24:09 +0100 Subject: IPVS: remove duplicate initialisation or rs_table Signed-off-by: Simon Horman Acked-by: Randy Dunlap Signed-off-by: Hans Schillstrom Tested-by: Hans Schillstrom Signed-off-by: Patrick McHardy --- net/netfilter/ipvs/ip_vs_ctl.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 98df59a12453..d7c2fa80cee3 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -3515,9 +3515,6 @@ int __net_init __ip_vs_control_init(struct net *net) } spin_lock_init(&ipvs->tot_stats->lock); - for (idx = 0; idx < IP_VS_RTAB_SIZE; idx++) - INIT_LIST_HEAD(&ipvs->rs_table[idx]); - proc_net_fops_create(net, "ip_vs", 0, &ip_vs_info_fops); proc_net_fops_create(net, "ip_vs_stats", 0, &ip_vs_stats_fops); proc_net_fops_create(net, "ip_vs_stats_percpu", 0, -- cgit v1.2.3 From 0443929ff0ecc4d1e690edaffa338cabe0863d3b Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Tue, 1 Feb 2011 18:29:04 +0100 Subject: IPVS: Allow compilation with CONFIG_SYSCTL disabled This is a rather naieve approach to allowing PVS to compile with CONFIG_SYSCTL disabled. I am working on a more comprehensive patch which will remove compilation of all sysctl-related IPVS code when CONFIG_SYSCTL is disabled. Reported-by: Randy Dunlap Signed-off-by: Simon Horman Acked-by: Randy Dunlap Signed-off-by: Hans Schillstrom Tested-by: Hans Schillstrom Signed-off-by: Patrick McHardy --- net/netfilter/ipvs/ip_vs_ctl.c | 14 +++++++++----- net/netfilter/ipvs/ip_vs_lblc.c | 20 ++++++++++---------- net/netfilter/ipvs/ip_vs_lblcr.c | 20 ++++++++++---------- 3 files changed, 29 insertions(+), 25 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index d7c2fa80cee3..c73b0c831a2d 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -3552,10 +3552,15 @@ int __net_init __ip_vs_control_init(struct net *net) tbl[idx++].data = &ipvs->sysctl_nat_icmp_send; +#ifdef CONFIG_SYSCTL ipvs->sysctl_hdr = register_net_sysctl_table(net, net_vs_ctl_path, tbl); - if (ipvs->sysctl_hdr == NULL) - goto err_reg; + if (ipvs->sysctl_hdr == NULL) { + if (!net_eq(net, &init_net)) + kfree(tbl); + goto err_dup; + } +#endif ip_vs_new_estimator(net, ipvs->tot_stats); ipvs->sysctl_tbl = tbl; /* Schedule defense work */ @@ -3563,9 +3568,6 @@ int __net_init __ip_vs_control_init(struct net *net) schedule_delayed_work(&ipvs->defense_work, DEFENSE_TIMER_PERIOD); return 0; -err_reg: - if (!net_eq(net, &init_net)) - kfree(tbl); err_dup: free_percpu(ipvs->cpustats); err_alloc: @@ -3581,7 +3583,9 @@ static void __net_exit __ip_vs_control_cleanup(struct net *net) ip_vs_kill_estimator(net, ipvs->tot_stats); cancel_delayed_work_sync(&ipvs->defense_work); cancel_work_sync(&ipvs->defense_work.work); +#ifdef CONFIG_SYSCTL unregister_net_sysctl_table(ipvs->sysctl_hdr); +#endif proc_net_remove(net, "ip_vs_stats_percpu"); proc_net_remove(net, "ip_vs_stats"); proc_net_remove(net, "ip_vs"); diff --git a/net/netfilter/ipvs/ip_vs_lblc.c b/net/netfilter/ipvs/ip_vs_lblc.c index d5bec3371871..00b5ffab3768 100644 --- a/net/netfilter/ipvs/ip_vs_lblc.c +++ b/net/netfilter/ipvs/ip_vs_lblc.c @@ -554,33 +554,33 @@ static int __net_init __ip_vs_lblc_init(struct net *net) sizeof(vs_vars_table), GFP_KERNEL); if (ipvs->lblc_ctl_table == NULL) - goto err_dup; + return -ENOMEM; } else ipvs->lblc_ctl_table = vs_vars_table; ipvs->sysctl_lblc_expiration = 24*60*60*HZ; ipvs->lblc_ctl_table[0].data = &ipvs->sysctl_lblc_expiration; +#ifdef CONFIG_SYSCTL ipvs->lblc_ctl_header = register_net_sysctl_table(net, net_vs_ctl_path, ipvs->lblc_ctl_table); - if (!ipvs->lblc_ctl_header) - goto err_reg; + if (!ipvs->lblc_ctl_header) { + if (!net_eq(net, &init_net)) + kfree(ipvs->lblc_ctl_table); + return -ENOMEM; + } +#endif return 0; - -err_reg: - if (!net_eq(net, &init_net)) - kfree(ipvs->lblc_ctl_table); - -err_dup: - return -ENOMEM; } static void __net_exit __ip_vs_lblc_exit(struct net *net) { struct netns_ipvs *ipvs = net_ipvs(net); +#ifdef CONFIG_SYSCTL unregister_net_sysctl_table(ipvs->lblc_ctl_header); +#endif if (!net_eq(net, &init_net)) kfree(ipvs->lblc_ctl_table); diff --git a/net/netfilter/ipvs/ip_vs_lblcr.c b/net/netfilter/ipvs/ip_vs_lblcr.c index 61ae8cfcf0b4..bfa25f1ea9e4 100644 --- a/net/netfilter/ipvs/ip_vs_lblcr.c +++ b/net/netfilter/ipvs/ip_vs_lblcr.c @@ -754,33 +754,33 @@ static int __net_init __ip_vs_lblcr_init(struct net *net) sizeof(vs_vars_table), GFP_KERNEL); if (ipvs->lblcr_ctl_table == NULL) - goto err_dup; + return -ENOMEM; } else ipvs->lblcr_ctl_table = vs_vars_table; ipvs->sysctl_lblcr_expiration = 24*60*60*HZ; ipvs->lblcr_ctl_table[0].data = &ipvs->sysctl_lblcr_expiration; +#ifdef CONFIG_SYSCTL ipvs->lblcr_ctl_header = register_net_sysctl_table(net, net_vs_ctl_path, ipvs->lblcr_ctl_table); - if (!ipvs->lblcr_ctl_header) - goto err_reg; + if (!ipvs->lblcr_ctl_header) { + if (!net_eq(net, &init_net)) + kfree(ipvs->lblcr_ctl_table); + return -ENOMEM; + } +#endif return 0; - -err_reg: - if (!net_eq(net, &init_net)) - kfree(ipvs->lblcr_ctl_table); - -err_dup: - return -ENOMEM; } static void __net_exit __ip_vs_lblcr_exit(struct net *net) { struct netns_ipvs *ipvs = net_ipvs(net); +#ifdef CONFIG_SYSCTL unregister_net_sysctl_table(ipvs->lblcr_ctl_header); +#endif if (!net_eq(net, &init_net)) kfree(ipvs->lblcr_ctl_table); -- cgit v1.2.3 From ed3d1e7b72069a3463b7e227b18cae4a09b0ddad Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Tue, 1 Feb 2011 18:30:26 +0100 Subject: IPVS: Remove ip_vs_sync_cleanup from section __exit ip_vs_sync_cleanup() may be called from ip_vs_init() on error and thus needs to be accesible from section __init Reporte-by: Randy Dunlap Signed-off-by: Simon Horman Acked-by: Randy Dunlap Signed-off-by: Hans Schillstrom Tested-by: Hans Schillstrom Signed-off-by: Patrick McHardy --- net/netfilter/ipvs/ip_vs_sync.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index d5a6e640ea45..2a2a8363ca16 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -1686,7 +1686,7 @@ int __init ip_vs_sync_init(void) return register_pernet_subsys(&ipvs_sync_ops); } -void __exit ip_vs_sync_cleanup(void) +void ip_vs_sync_cleanup(void) { unregister_pernet_subsys(&ipvs_sync_ops); } -- cgit v1.2.3 From 316ed388802533bcfd3dffb38d2ba29ac5428456 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 2 Feb 2011 09:31:37 +0100 Subject: netfilter: ipset: add missing break statemtns in ip_set_get_ip_port() Don't fall through in the switch statement, otherwise IPv4 headers are incorrectly parsed again as IPv6 and the return value will always be 'false'. Signed-off-by: Patrick McHardy --- net/netfilter/ipset/ip_set_getport.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/ip_set_getport.c b/net/netfilter/ipset/ip_set_getport.c index 76737bba28ee..4dd2785a5c72 100644 --- a/net/netfilter/ipset/ip_set_getport.c +++ b/net/netfilter/ipset/ip_set_getport.c @@ -118,8 +118,10 @@ ip_set_get_ip_port(const struct sk_buff *skb, u8 pf, bool src, __be16 *port) switch (pf) { case AF_INET: ret = ip_set_get_ip4_port(skb, src, port, &proto); + break; case AF_INET6: ret = ip_set_get_ip6_port(skb, src, port, &proto); + break; default: return false; } -- cgit v1.2.3 From 724bab476bcac9f7d0b5204cb06e346216d42166 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 2 Feb 2011 23:50:01 +0100 Subject: netfilter: ipset: fix linking with CONFIG_IPV6=n Add a dummy ip_set_get_ip6_port function that unconditionally returns false for CONFIG_IPV6=n and convert the real function to ipv6_skip_exthdr() to avoid pulling in the ip6_tables module when loading ipset. Signed-off-by: Patrick McHardy --- net/netfilter/ipset/ip_set_getport.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/ip_set_getport.c b/net/netfilter/ipset/ip_set_getport.c index 4dd2785a5c72..8d5227212686 100644 --- a/net/netfilter/ipset/ip_set_getport.c +++ b/net/netfilter/ipset/ip_set_getport.c @@ -13,6 +13,7 @@ #include #include #include +#include #include @@ -93,21 +94,23 @@ ip_set_get_ip4_port(const struct sk_buff *skb, bool src, } EXPORT_SYMBOL_GPL(ip_set_get_ip4_port); +#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE) bool ip_set_get_ip6_port(const struct sk_buff *skb, bool src, __be16 *port, u8 *proto) { - unsigned int protooff = 0; - int protocol; - unsigned short fragoff; + int protoff; + u8 nexthdr; - protocol = ipv6_find_hdr(skb, &protooff, -1, &fragoff); - if (protocol <= 0 || fragoff) + nexthdr = ipv6_hdr(skb)->nexthdr; + protoff = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr), &nexthdr); + if (protoff < 0) return false; - return get_port(skb, protocol, protooff, src, port, proto); + return get_port(skb, nexthdr, protoff, src, port, proto); } EXPORT_SYMBOL_GPL(ip_set_get_ip6_port); +#endif bool ip_set_get_ip_port(const struct sk_buff *skb, u8 pf, bool src, __be16 *port) -- cgit v1.2.3 From 5f52bc3cdd1bb2e12e61639df19d9dcd530c4568 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Wed, 2 Feb 2011 23:56:00 +0100 Subject: netfilter: ipset: send error message manually When a message carries multiple commands and one of them triggers an error, we have to report to the userspace which one was that. The line number of the command plays this role and there's an attribute reserved in the header part of the message to be filled out with the error line number. In order not to modify the original message received from the userspace, we construct a new, complete netlink error message and modifies the attribute there, then send it. Netlink is notified not to send its ACK/error message. Signed-off-by: Jozsef Kadlecsik --- net/netfilter/ipset/ip_set_core.c | 33 ++++++++++++++++++++++++++------- 1 file changed, 26 insertions(+), 7 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c index ae0f8b595106..8b1a54c1e400 100644 --- a/net/netfilter/ipset/ip_set_core.c +++ b/net/netfilter/ipset/ip_set_core.c @@ -1098,7 +1098,7 @@ static const struct nla_policy ip_set_adt_policy[IPSET_ATTR_CMD_MAX + 1] = { }; static int -call_ad(struct sk_buff *skb, struct ip_set *set, +call_ad(struct sock *ctnl, struct sk_buff *skb, struct ip_set *set, struct nlattr *tb[], enum ipset_adt adt, u32 flags, bool use_lineno) { @@ -1118,12 +1118,25 @@ call_ad(struct sk_buff *skb, struct ip_set *set, return 0; if (lineno && use_lineno) { /* Error in restore/batch mode: send back lineno */ - struct nlmsghdr *nlh = nlmsg_hdr(skb); + struct nlmsghdr *rep, *nlh = nlmsg_hdr(skb); + struct sk_buff *skb2; + struct nlmsgerr *errmsg; + size_t payload = sizeof(*errmsg) + nlmsg_len(nlh); int min_len = NLMSG_SPACE(sizeof(struct nfgenmsg)); struct nlattr *cda[IPSET_ATTR_CMD_MAX+1]; - struct nlattr *cmdattr = (void *)nlh + min_len; + struct nlattr *cmdattr; u32 *errline; + skb2 = nlmsg_new(payload, GFP_KERNEL); + if (skb2 == NULL) + return -ENOMEM; + rep = __nlmsg_put(skb2, NETLINK_CB(skb).pid, + nlh->nlmsg_seq, NLMSG_ERROR, payload, 0); + errmsg = nlmsg_data(rep); + errmsg->error = ret; + memcpy(&errmsg->msg, nlh, nlh->nlmsg_len); + cmdattr = (void *)&errmsg->msg + min_len; + nla_parse(cda, IPSET_ATTR_CMD_MAX, cmdattr, nlh->nlmsg_len - min_len, ip_set_adt_policy); @@ -1131,6 +1144,10 @@ call_ad(struct sk_buff *skb, struct ip_set *set, errline = nla_data(cda[IPSET_ATTR_LINENO]); *errline = lineno; + + netlink_unicast(ctnl, skb2, NETLINK_CB(skb).pid, MSG_DONTWAIT); + /* Signal netlink not to send its ACK/errmsg. */ + return -EINTR; } return ret; @@ -1169,7 +1186,8 @@ ip_set_uadd(struct sock *ctnl, struct sk_buff *skb, attr[IPSET_ATTR_DATA], set->type->adt_policy)) return -IPSET_ERR_PROTOCOL; - ret = call_ad(skb, set, tb, IPSET_ADD, flags, use_lineno); + ret = call_ad(ctnl, skb, set, tb, IPSET_ADD, flags, + use_lineno); } else { int nla_rem; @@ -1180,7 +1198,7 @@ ip_set_uadd(struct sock *ctnl, struct sk_buff *skb, nla_parse_nested(tb, IPSET_ATTR_ADT_MAX, nla, set->type->adt_policy)) return -IPSET_ERR_PROTOCOL; - ret = call_ad(skb, set, tb, IPSET_ADD, + ret = call_ad(ctnl, skb, set, tb, IPSET_ADD, flags, use_lineno); if (ret < 0) return ret; @@ -1222,7 +1240,8 @@ ip_set_udel(struct sock *ctnl, struct sk_buff *skb, attr[IPSET_ATTR_DATA], set->type->adt_policy)) return -IPSET_ERR_PROTOCOL; - ret = call_ad(skb, set, tb, IPSET_DEL, flags, use_lineno); + ret = call_ad(ctnl, skb, set, tb, IPSET_DEL, flags, + use_lineno); } else { int nla_rem; @@ -1233,7 +1252,7 @@ ip_set_udel(struct sock *ctnl, struct sk_buff *skb, nla_parse_nested(tb, IPSET_ATTR_ADT_MAX, nla, set->type->adt_policy)) return -IPSET_ERR_PROTOCOL; - ret = call_ad(skb, set, tb, IPSET_DEL, + ret = call_ad(ctnl, skb, set, tb, IPSET_DEL, flags, use_lineno); if (ret < 0) return ret; -- cgit v1.2.3 From 9291747f118d6404e509747b85ff5f6dfec368d2 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 3 Feb 2011 00:05:43 +0100 Subject: netfilter: xtables: add device group match Add a new 'devgroup' match to match on the device group of the incoming and outgoing network device of a packet. Signed-off-by: Patrick McHardy --- net/netfilter/Kconfig | 9 +++++ net/netfilter/Makefile | 1 + net/netfilter/xt_devgroup.c | 82 +++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 92 insertions(+) create mode 100644 net/netfilter/xt_devgroup.c (limited to 'net/netfilter') diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 06fa9e4e45c7..82a6e0d80f05 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -738,6 +738,15 @@ config NETFILTER_XT_MATCH_DCCP If you want to compile it as a module, say M here and read . If unsure, say `N'. +config NETFILTER_XT_MATCH_DEVGROUP + tristate '"devgroup" match support' + depends on NETFILTER_ADVANCED + help + This options adds a `devgroup' match, which allows to match on the + device group a network device is assigned to. + + To compile it as a module, choose M here. If unsure, say N. + config NETFILTER_XT_MATCH_DSCP tristate '"dscp" and "tos" match support' depends on NETFILTER_ADVANCED diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 1148643559cb..d57a890eaee5 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -77,6 +77,7 @@ obj-$(CONFIG_NETFILTER_XT_MATCH_CONNLIMIT) += xt_connlimit.o obj-$(CONFIG_NETFILTER_XT_MATCH_CONNTRACK) += xt_conntrack.o obj-$(CONFIG_NETFILTER_XT_MATCH_CPU) += xt_cpu.o obj-$(CONFIG_NETFILTER_XT_MATCH_DCCP) += xt_dccp.o +obj-$(CONFIG_NETFILTER_XT_MATCH_DEVGROUP) += xt_devgroup.o obj-$(CONFIG_NETFILTER_XT_MATCH_DSCP) += xt_dscp.o obj-$(CONFIG_NETFILTER_XT_MATCH_ESP) += xt_esp.o obj-$(CONFIG_NETFILTER_XT_MATCH_HASHLIMIT) += xt_hashlimit.o diff --git a/net/netfilter/xt_devgroup.c b/net/netfilter/xt_devgroup.c new file mode 100644 index 000000000000..d9202cdd25c9 --- /dev/null +++ b/net/netfilter/xt_devgroup.c @@ -0,0 +1,82 @@ +/* + * Copyright (c) 2011 Patrick McHardy + * + * 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. + */ + +#include +#include +#include + +#include +#include + +MODULE_AUTHOR("Patrick McHardy "); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("Xtables: Device group match"); +MODULE_ALIAS("ipt_devgroup"); +MODULE_ALIAS("ip6t_devgroup"); + +static bool devgroup_mt(const struct sk_buff *skb, struct xt_action_param *par) +{ + const struct xt_devgroup_info *info = par->matchinfo; + + if (info->flags & XT_DEVGROUP_MATCH_SRC && + (((info->src_group ^ par->in->group) & info->src_mask ? 1 : 0) ^ + ((info->flags & XT_DEVGROUP_INVERT_SRC) ? 1 : 0))) + return false; + + if (info->flags & XT_DEVGROUP_MATCH_DST && + (((info->dst_group ^ par->out->group) & info->dst_mask ? 1 : 0) ^ + ((info->flags & XT_DEVGROUP_INVERT_DST) ? 1 : 0))) + return false; + + return true; +} + +static int devgroup_mt_checkentry(const struct xt_mtchk_param *par) +{ + const struct xt_devgroup_info *info = par->matchinfo; + + if (info->flags & ~(XT_DEVGROUP_MATCH_SRC | XT_DEVGROUP_INVERT_SRC | + XT_DEVGROUP_MATCH_DST | XT_DEVGROUP_INVERT_DST)) + return -EINVAL; + + if (info->flags & XT_DEVGROUP_MATCH_SRC && + par->hook_mask & ~((1 << NF_INET_PRE_ROUTING) | + (1 << NF_INET_LOCAL_IN) | + (1 << NF_INET_FORWARD))) + return -EINVAL; + + if (info->flags & XT_DEVGROUP_MATCH_DST && + par->hook_mask & ~((1 << NF_INET_FORWARD) | + (1 << NF_INET_LOCAL_OUT) | + (1 << NF_INET_POST_ROUTING))) + return -EINVAL; + + return 0; +} + +static struct xt_match devgroup_mt_reg __read_mostly = { + .name = "devgroup", + .match = devgroup_mt, + .checkentry = devgroup_mt_checkentry, + .matchsize = sizeof(struct xt_devgroup_info), + .family = NFPROTO_UNSPEC, + .me = THIS_MODULE +}; + +static int __init devgroup_mt_init(void) +{ + return xt_register_match(&devgroup_mt_reg); +} + +static void __exit devgroup_mt_exit(void) +{ + xt_unregister_match(&devgroup_mt_reg); +} + +module_init(devgroup_mt_init); +module_exit(devgroup_mt_exit); -- cgit v1.2.3 From 8525d6f84f576402278a552ed17d2ba3b61f8e3c Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Thu, 3 Feb 2011 07:22:43 +0900 Subject: IPVS: Use correct lock in SCTP module Use sctp_app_lock instead of tcp_app_lock in the SCTP protocol module. This appears to be a typo introduced by the netns changes. Signed-off-by: Simon Horman Signed-off-by: Hans Schillstrom --- net/netfilter/ipvs/ip_vs_proto_sctp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_proto_sctp.c b/net/netfilter/ipvs/ip_vs_proto_sctp.c index fb2d04ac5d4e..b027ccc49f43 100644 --- a/net/netfilter/ipvs/ip_vs_proto_sctp.c +++ b/net/netfilter/ipvs/ip_vs_proto_sctp.c @@ -1101,7 +1101,7 @@ static void __ip_vs_sctp_init(struct net *net, struct ip_vs_proto_data *pd) struct netns_ipvs *ipvs = net_ipvs(net); ip_vs_init_hash_table(ipvs->sctp_apps, SCTP_APP_TAB_SIZE); - spin_lock_init(&ipvs->tcp_app_lock); + spin_lock_init(&ipvs->sctp_app_lock); pd->timeout_table = ip_vs_create_timeout_table((int *)sctp_timeouts, sizeof(sctp_timeouts)); } -- cgit v1.2.3 From 7c9989a76e62ceca90e5f31f8920fd6b7b8b6525 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 7 Feb 2011 11:38:55 +0300 Subject: IPVS: precedence bug in ip_vs_sync_switch_mode() '!' has higher precedence than '&'. IP_VS_STATE_MASTER is 0x1 so the original code is equivelent to if (!ipvs->sync_state) ... Signed-off-by: Dan Carpenter Signed-off-by: Hans Schillstrom Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_sync.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index 2a2a8363ca16..d1b7298e5894 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -392,7 +392,7 @@ void ip_vs_sync_switch_mode(struct net *net, int mode) { struct netns_ipvs *ipvs = net_ipvs(net); - if (!ipvs->sync_state & IP_VS_STATE_MASTER) + if (!(ipvs->sync_state & IP_VS_STATE_MASTER)) return; if (mode == ipvs->sysctl_sync_ver || !ipvs->sync_buff) return; -- cgit v1.2.3 From c16e19c11730199c1df686b160c9c972ad28baf8 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 10 Feb 2011 10:13:07 +0100 Subject: netfilter: ipset: add dependency on CONFIG_NETFILTER_NETLINK When SYSCTL and PROC_FS and NETFILTER_NETLINK are not enabled: net/built-in.o: In function `try_to_load_type': ip_set_core.c:(.text+0x3ab49): undefined reference to `nfnl_unlock' ip_set_core.c:(.text+0x3ab4e): undefined reference to `nfnl_lock' ... Reported-by: Randy Dunlap Acked-by: Randy Dunlap Signed-off-by: Patrick McHardy --- net/netfilter/ipset/Kconfig | 1 + 1 file changed, 1 insertion(+) (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/Kconfig b/net/netfilter/ipset/Kconfig index 3b970d343023..2c5b348eb3a8 100644 --- a/net/netfilter/ipset/Kconfig +++ b/net/netfilter/ipset/Kconfig @@ -1,6 +1,7 @@ menuconfig IP_SET tristate "IP set support" depends on INET && NETFILTER + depends on NETFILTER_NETLINK help This option adds IP set support to the kernel. In order to define and use the sets, you need the userspace utility -- cgit v1.2.3 From 44bd4de9c2270b22c3c898310102bc6be9ed2978 Mon Sep 17 00:00:00 2001 From: Stefan Berger Date: Fri, 11 Feb 2011 18:00:07 +0100 Subject: netfilter: xt_connlimit: connlimit-above early loop termination The patch below introduces an early termination of the loop that is counting matches. It terminates once the counter has exceeded the threshold provided by the user. There's no point in continuing the loop afterwards and looking at other entries. It plays together with the following code further below: return (connections > info->limit) ^ info->inverse; where connections is the result of the counted connection, which in turn is the matches variable in the loop. So once -> matches = info->limit + 1 alias -> matches > info->limit alias -> matches > threshold we can terminate the loop. Signed-off-by: Stefan Berger Signed-off-by: Patrick McHardy --- net/netfilter/xt_connlimit.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c index e029c4807404..82ce7c5fbbc2 100644 --- a/net/netfilter/xt_connlimit.c +++ b/net/netfilter/xt_connlimit.c @@ -97,7 +97,8 @@ static int count_them(struct net *net, const struct nf_conntrack_tuple *tuple, const union nf_inet_addr *addr, const union nf_inet_addr *mask, - u_int8_t family) + u_int8_t family, + unsigned int threshold) { const struct nf_conntrack_tuple_hash *found; struct xt_connlimit_conn *conn; @@ -151,9 +152,14 @@ static int count_them(struct net *net, continue; } - if (same_source_net(addr, mask, &conn->tuple.src.u3, family)) + if (same_source_net(addr, mask, &conn->tuple.src.u3, family)) { /* same source network -> be counted! */ ++matches; + if (matches > threshold) { + nf_ct_put(found_ct); + break; + } + } nf_ct_put(found_ct); } @@ -207,7 +213,8 @@ connlimit_mt(const struct sk_buff *skb, struct xt_action_param *par) spin_lock_bh(&info->data->lock); connections = count_them(net, info->data, tuple_ptr, &addr, - &info->mask, par->family); + &info->mask, par->family, + info->limit); spin_unlock_bh(&info->data->lock); if (connections < 0) -- cgit v1.2.3 From 20b7975e5aefc7fd08b7f582f3901b1669725cd0 Mon Sep 17 00:00:00 2001 From: Stefan Berger Date: Mon, 14 Feb 2011 16:54:33 +0100 Subject: Revert "netfilter: xt_connlimit: connlimit-above early loop termination" This reverts commit 44bd4de9c2270b22c3c898310102bc6be9ed2978. I have to revert the early loop termination in connlimit since it generates problems when an iptables statement does not use -m state --state NEW before the connlimit match extension. Signed-off-by: Stefan Berger Signed-off-by: Patrick McHardy --- net/netfilter/xt_connlimit.c | 13 +++---------- 1 file changed, 3 insertions(+), 10 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c index 82ce7c5fbbc2..e029c4807404 100644 --- a/net/netfilter/xt_connlimit.c +++ b/net/netfilter/xt_connlimit.c @@ -97,8 +97,7 @@ static int count_them(struct net *net, const struct nf_conntrack_tuple *tuple, const union nf_inet_addr *addr, const union nf_inet_addr *mask, - u_int8_t family, - unsigned int threshold) + u_int8_t family) { const struct nf_conntrack_tuple_hash *found; struct xt_connlimit_conn *conn; @@ -152,14 +151,9 @@ static int count_them(struct net *net, continue; } - if (same_source_net(addr, mask, &conn->tuple.src.u3, family)) { + if (same_source_net(addr, mask, &conn->tuple.src.u3, family)) /* same source network -> be counted! */ ++matches; - if (matches > threshold) { - nf_ct_put(found_ct); - break; - } - } nf_ct_put(found_ct); } @@ -213,8 +207,7 @@ connlimit_mt(const struct sk_buff *skb, struct xt_action_param *par) spin_lock_bh(&info->data->lock); connections = count_them(net, info->data, tuple_ptr, &addr, - &info->mask, par->family, - info->limit); + &info->mask, par->family); spin_unlock_bh(&info->data->lock); if (connections < 0) -- cgit v1.2.3 From a2361c8735e07322023aedc36e4938b35af31eb0 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 14 Feb 2011 17:28:55 +0100 Subject: netfilter: xt_conntrack: warn about use in raw table nfct happens to run after the raw table only. Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- net/netfilter/xt_conntrack.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_conntrack.c b/net/netfilter/xt_conntrack.c index 4ef1b63ad73f..2c0086a4751e 100644 --- a/net/netfilter/xt_conntrack.c +++ b/net/netfilter/xt_conntrack.c @@ -272,6 +272,11 @@ static int conntrack_mt_check(const struct xt_mtchk_param *par) { int ret; + if (strcmp(par->table, "raw") == 0) { + pr_info("state is undetermined at the time of raw table\n"); + return -EINVAL; + } + ret = nf_ct_l3proto_try_module_get(par->family); if (ret < 0) pr_info("cannot load conntrack support for proto=%u\n", -- cgit v1.2.3 From 8248779b1878f17cce2bb809831f4f2a252bdb77 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 15 Feb 2011 21:59:37 +0100 Subject: netfilter: nfnetlink_log: remove unused parameter Signed-off-by: Florian Westphal Signed-off-by: Patrick McHardy --- net/netfilter/nfnetlink_log.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index 91592da504b9..985e9b76c916 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -376,7 +376,6 @@ __build_packet_message(struct nfulnl_instance *inst, unsigned int hooknum, const struct net_device *indev, const struct net_device *outdev, - const struct nf_loginfo *li, const char *prefix, unsigned int plen) { struct nfulnl_msg_packet_hdr pmsg; @@ -652,7 +651,7 @@ nfulnl_log_packet(u_int8_t pf, inst->qlen++; __build_packet_message(inst, skb, data_len, pf, - hooknum, in, out, li, prefix, plen); + hooknum, in, out, prefix, plen); if (inst->qlen >= qthreshold) __nfulnl_flush(inst); -- cgit v1.2.3 From 16a7fd323f93eab88df79fc647575ae9789037c2 Mon Sep 17 00:00:00 2001 From: Tinggong Wang Date: Wed, 9 Feb 2011 02:21:59 +0200 Subject: ipvs: fix timer in get_curr_sync_buff Fix get_curr_sync_buff to keep buffer for 2 seconds as intended, not just for the current jiffie. By this way we will sync more connection structures with single packet. Signed-off-by: Tinggong Wang Signed-off-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_sync.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index d1b7298e5894..fecf24de4af3 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -374,8 +374,8 @@ get_curr_sync_buff(struct netns_ipvs *ipvs, unsigned long time) struct ip_vs_sync_buff *sb; spin_lock_bh(&ipvs->sync_buff_lock); - if (ipvs->sync_buff && (time == 0 || - time_before(jiffies - ipvs->sync_buff->firstuse, time))) { + if (ipvs->sync_buff && + time_after_eq(jiffies - ipvs->sync_buff->firstuse, time)) { sb = ipvs->sync_buff; ipvs->sync_buff = NULL; } else -- cgit v1.2.3 From 6cb90db502c5f276c8d6256762cc3acde4d3bd9d Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Wed, 9 Feb 2011 02:26:38 +0200 Subject: ipvs: remove extra lookups for ICMP packets Remove code that should not be called anymore. Now when ip_vs_out handles replies for local clients at LOCAL_IN hook we do not need to call conn_out_get and handle_response_icmp from ip_vs_in_icmp* because such lookups were already performed for the ICMP packet and no connection was found. Signed-off-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 28 +++------------------------- 1 file changed, 3 insertions(+), 25 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 4d06617fab6c..2d1f932add46 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -729,7 +729,7 @@ void ip_vs_nat_icmp_v6(struct sk_buff *skb, struct ip_vs_protocol *pp, #endif /* Handle relevant response ICMP messages - forward to the right - * destination host. Used for NAT and local client. + * destination host. */ static int handle_response_icmp(int af, struct sk_buff *skb, union nf_inet_addr *snet, @@ -979,7 +979,6 @@ static inline int is_tcp_reset(const struct sk_buff *skb, int nh_len) } /* Handle response packets: rewrite addresses and send away... - * Used for NAT and local client. */ static unsigned int handle_response(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, @@ -1280,7 +1279,6 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum) struct ip_vs_protocol *pp; struct ip_vs_proto_data *pd; unsigned int offset, ihl, verdict; - union nf_inet_addr snet; *related = 1; @@ -1339,17 +1337,8 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum) ip_vs_fill_iphdr(AF_INET, cih, &ciph); /* The embedded headers contain source and dest in reverse order */ cp = pp->conn_in_get(AF_INET, skb, &ciph, offset, 1); - if (!cp) { - /* The packet could also belong to a local client */ - cp = pp->conn_out_get(AF_INET, skb, &ciph, offset, 1); - if (cp) { - snet.ip = iph->saddr; - return handle_response_icmp(AF_INET, skb, &snet, - cih->protocol, cp, pp, - offset, ihl); - } + if (!cp) return NF_ACCEPT; - } verdict = NF_DROP; @@ -1395,7 +1384,6 @@ ip_vs_in_icmp_v6(struct sk_buff *skb, int *related, unsigned int hooknum) struct ip_vs_protocol *pp; struct ip_vs_proto_data *pd; unsigned int offset, verdict; - union nf_inet_addr snet; struct rt6_info *rt; *related = 1; @@ -1455,18 +1443,8 @@ ip_vs_in_icmp_v6(struct sk_buff *skb, int *related, unsigned int hooknum) ip_vs_fill_iphdr(AF_INET6, cih, &ciph); /* The embedded headers contain source and dest in reverse order */ cp = pp->conn_in_get(AF_INET6, skb, &ciph, offset, 1); - if (!cp) { - /* The packet could also belong to a local client */ - cp = pp->conn_out_get(AF_INET6, skb, &ciph, offset, 1); - if (cp) { - ipv6_addr_copy(&snet.in6, &iph->saddr); - return handle_response_icmp(AF_INET6, skb, &snet, - cih->nexthdr, - cp, pp, offset, - sizeof(struct ipv6hdr)); - } + if (!cp) return NF_ACCEPT; - } verdict = NF_DROP; -- cgit v1.2.3 From 41ac51eeda58a85b8a06d748cce7035cc77deebd Mon Sep 17 00:00:00 2001 From: Patrick Schaaf Date: Fri, 11 Feb 2011 14:01:12 +0100 Subject: ipvs: make "no destination available" message more informative When IP_VS schedulers do not find a destination, they output a terse "WLC: no destination available" message through kernel syslog, which I can not only make sense of because syslog puts them in a logfile together with keepalived checker results. This patch makes the output a bit more informative, by telling you which virtual service failed to find a destination. Example output: kernel: [1539214.552233] IPVS: wlc: TCP 192.168.8.30:22 - no destination available kernel: [1539299.674418] IPVS: wlc: FWM 22 0x00000016 - no destination available I have tested the code for IPv4 and FWM services, as you can see from the example; I do not have an IPv6 setup to test the third code path with. To avoid code duplication, I put a new function ip_vs_scheduler_err() into ip_vs_sched.c, and use that from the schedulers instead of calling IP_VS_ERR_RL directly. Signed-off-by: Patrick Schaaf Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_lblc.c | 2 +- net/netfilter/ipvs/ip_vs_lblcr.c | 2 +- net/netfilter/ipvs/ip_vs_lc.c | 2 +- net/netfilter/ipvs/ip_vs_nq.c | 2 +- net/netfilter/ipvs/ip_vs_rr.c | 2 +- net/netfilter/ipvs/ip_vs_sched.c | 25 +++++++++++++++++++++++++ net/netfilter/ipvs/ip_vs_sed.c | 2 +- net/netfilter/ipvs/ip_vs_sh.c | 2 +- net/netfilter/ipvs/ip_vs_wlc.c | 2 +- net/netfilter/ipvs/ip_vs_wrr.c | 14 ++++++++------ 10 files changed, 41 insertions(+), 14 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_lblc.c b/net/netfilter/ipvs/ip_vs_lblc.c index 00b5ffab3768..4a9c8cd19690 100644 --- a/net/netfilter/ipvs/ip_vs_lblc.c +++ b/net/netfilter/ipvs/ip_vs_lblc.c @@ -510,7 +510,7 @@ ip_vs_lblc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) /* No cache entry or it is invalid, time to schedule */ dest = __ip_vs_lblc_schedule(svc); if (!dest) { - IP_VS_ERR_RL("LBLC: no destination available\n"); + ip_vs_scheduler_err(svc, "no destination available"); return NULL; } diff --git a/net/netfilter/ipvs/ip_vs_lblcr.c b/net/netfilter/ipvs/ip_vs_lblcr.c index bfa25f1ea9e4..bd329b1e9589 100644 --- a/net/netfilter/ipvs/ip_vs_lblcr.c +++ b/net/netfilter/ipvs/ip_vs_lblcr.c @@ -692,7 +692,7 @@ ip_vs_lblcr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) /* The cache entry is invalid, time to schedule */ dest = __ip_vs_lblcr_schedule(svc); if (!dest) { - IP_VS_ERR_RL("LBLCR: no destination available\n"); + ip_vs_scheduler_err(svc, "no destination available"); read_unlock(&svc->sched_lock); return NULL; } diff --git a/net/netfilter/ipvs/ip_vs_lc.c b/net/netfilter/ipvs/ip_vs_lc.c index 4f69db1fac56..60638007c6c7 100644 --- a/net/netfilter/ipvs/ip_vs_lc.c +++ b/net/netfilter/ipvs/ip_vs_lc.c @@ -70,7 +70,7 @@ ip_vs_lc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) } if (!least) - IP_VS_ERR_RL("LC: no destination available\n"); + ip_vs_scheduler_err(svc, "no destination available"); else IP_VS_DBG_BUF(6, "LC: server %s:%u activeconns %d " "inactconns %d\n", diff --git a/net/netfilter/ipvs/ip_vs_nq.c b/net/netfilter/ipvs/ip_vs_nq.c index c413e1830823..984d9c137d84 100644 --- a/net/netfilter/ipvs/ip_vs_nq.c +++ b/net/netfilter/ipvs/ip_vs_nq.c @@ -99,7 +99,7 @@ ip_vs_nq_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) } if (!least) { - IP_VS_ERR_RL("NQ: no destination available\n"); + ip_vs_scheduler_err(svc, "no destination available"); return NULL; } diff --git a/net/netfilter/ipvs/ip_vs_rr.c b/net/netfilter/ipvs/ip_vs_rr.c index e210f37d8ea2..c49b388d1085 100644 --- a/net/netfilter/ipvs/ip_vs_rr.c +++ b/net/netfilter/ipvs/ip_vs_rr.c @@ -72,7 +72,7 @@ ip_vs_rr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) q = q->next; } while (q != p); write_unlock(&svc->sched_lock); - IP_VS_ERR_RL("RR: no destination available\n"); + ip_vs_scheduler_err(svc, "no destination available"); return NULL; out: diff --git a/net/netfilter/ipvs/ip_vs_sched.c b/net/netfilter/ipvs/ip_vs_sched.c index 076ebe00435d..08dbdd5bc18f 100644 --- a/net/netfilter/ipvs/ip_vs_sched.c +++ b/net/netfilter/ipvs/ip_vs_sched.c @@ -29,6 +29,7 @@ #include +EXPORT_SYMBOL(ip_vs_scheduler_err); /* * IPVS scheduler list */ @@ -146,6 +147,30 @@ void ip_vs_scheduler_put(struct ip_vs_scheduler *scheduler) module_put(scheduler->module); } +/* + * Common error output helper for schedulers + */ + +void ip_vs_scheduler_err(struct ip_vs_service *svc, const char *msg) +{ + if (svc->fwmark) { + IP_VS_ERR_RL("%s: FWM %u 0x%08X - %s\n", + svc->scheduler->name, svc->fwmark, + svc->fwmark, msg); +#ifdef CONFIG_IP_VS_IPV6 + } else if (svc->af == AF_INET6) { + IP_VS_ERR_RL("%s: %s [%pI6]:%d - %s\n", + svc->scheduler->name, + ip_vs_proto_name(svc->protocol), + &svc->addr.in6, ntohs(svc->port), msg); +#endif + } else { + IP_VS_ERR_RL("%s: %s %pI4:%d - %s\n", + svc->scheduler->name, + ip_vs_proto_name(svc->protocol), + &svc->addr.ip, ntohs(svc->port), msg); + } +} /* * Register a scheduler in the scheduler list diff --git a/net/netfilter/ipvs/ip_vs_sed.c b/net/netfilter/ipvs/ip_vs_sed.c index 1ab75a9dc400..89ead246ed3d 100644 --- a/net/netfilter/ipvs/ip_vs_sed.c +++ b/net/netfilter/ipvs/ip_vs_sed.c @@ -87,7 +87,7 @@ ip_vs_sed_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) goto nextstage; } } - IP_VS_ERR_RL("SED: no destination available\n"); + ip_vs_scheduler_err(svc, "no destination available"); return NULL; /* diff --git a/net/netfilter/ipvs/ip_vs_sh.c b/net/netfilter/ipvs/ip_vs_sh.c index e6cc174fbc06..b5e2556c581a 100644 --- a/net/netfilter/ipvs/ip_vs_sh.c +++ b/net/netfilter/ipvs/ip_vs_sh.c @@ -223,7 +223,7 @@ ip_vs_sh_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) || !(dest->flags & IP_VS_DEST_F_AVAILABLE) || atomic_read(&dest->weight) <= 0 || is_overloaded(dest)) { - IP_VS_ERR_RL("SH: no destination available\n"); + ip_vs_scheduler_err(svc, "no destination available"); return NULL; } diff --git a/net/netfilter/ipvs/ip_vs_wlc.c b/net/netfilter/ipvs/ip_vs_wlc.c index bbddfdb10db2..fdf0f58962a4 100644 --- a/net/netfilter/ipvs/ip_vs_wlc.c +++ b/net/netfilter/ipvs/ip_vs_wlc.c @@ -75,7 +75,7 @@ ip_vs_wlc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) goto nextstage; } } - IP_VS_ERR_RL("WLC: no destination available\n"); + ip_vs_scheduler_err(svc, "no destination available"); return NULL; /* diff --git a/net/netfilter/ipvs/ip_vs_wrr.c b/net/netfilter/ipvs/ip_vs_wrr.c index 30db633f88f1..1ef41f50723c 100644 --- a/net/netfilter/ipvs/ip_vs_wrr.c +++ b/net/netfilter/ipvs/ip_vs_wrr.c @@ -147,8 +147,9 @@ ip_vs_wrr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) if (mark->cl == mark->cl->next) { /* no dest entry */ - IP_VS_ERR_RL("WRR: no destination available: " - "no destinations present\n"); + ip_vs_scheduler_err(svc, + "no destination available: " + "no destinations present"); dest = NULL; goto out; } @@ -162,8 +163,8 @@ ip_vs_wrr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) */ if (mark->cw == 0) { mark->cl = &svc->destinations; - IP_VS_ERR_RL("WRR: no destination " - "available\n"); + ip_vs_scheduler_err(svc, + "no destination available"); dest = NULL; goto out; } @@ -185,8 +186,9 @@ ip_vs_wrr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) /* back to the start, and no dest is found. It is only possible when all dests are OVERLOADED */ dest = NULL; - IP_VS_ERR_RL("WRR: no destination available: " - "all destinations are overloaded\n"); + ip_vs_scheduler_err(svc, + "no destination available: " + "all destinations are overloaded"); goto out; } } -- cgit v1.2.3 From 731109e78415b4cc6c2f8de6c11b37f0e40741f8 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Sat, 19 Feb 2011 18:05:08 +0800 Subject: ipvs: use hlist instead of list Signed-off-by: Changli Gao Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_conn.c | 52 +++++++++++++++++++++++------------------ 1 file changed, 29 insertions(+), 23 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 83233fe24a08..9c2a517b69c8 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -59,7 +59,7 @@ static int ip_vs_conn_tab_mask __read_mostly; /* * Connection hash table: for input and output packets lookups of IPVS */ -static struct list_head *ip_vs_conn_tab __read_mostly; +static struct hlist_head *ip_vs_conn_tab __read_mostly; /* SLAB cache for IPVS connections */ static struct kmem_cache *ip_vs_conn_cachep __read_mostly; @@ -201,7 +201,7 @@ static inline int ip_vs_conn_hash(struct ip_vs_conn *cp) spin_lock(&cp->lock); if (!(cp->flags & IP_VS_CONN_F_HASHED)) { - list_add(&cp->c_list, &ip_vs_conn_tab[hash]); + hlist_add_head(&cp->c_list, &ip_vs_conn_tab[hash]); cp->flags |= IP_VS_CONN_F_HASHED; atomic_inc(&cp->refcnt); ret = 1; @@ -234,7 +234,7 @@ static inline int ip_vs_conn_unhash(struct ip_vs_conn *cp) spin_lock(&cp->lock); if (cp->flags & IP_VS_CONN_F_HASHED) { - list_del(&cp->c_list); + hlist_del(&cp->c_list); cp->flags &= ~IP_VS_CONN_F_HASHED; atomic_dec(&cp->refcnt); ret = 1; @@ -259,12 +259,13 @@ __ip_vs_conn_in_get(const struct ip_vs_conn_param *p) { unsigned hash; struct ip_vs_conn *cp; + struct hlist_node *n; hash = ip_vs_conn_hashkey_param(p, false); ct_read_lock(hash); - list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { + hlist_for_each_entry(cp, n, &ip_vs_conn_tab[hash], c_list) { if (cp->af == p->af && p->cport == cp->cport && p->vport == cp->vport && ip_vs_addr_equal(p->af, p->caddr, &cp->caddr) && @@ -345,12 +346,13 @@ struct ip_vs_conn *ip_vs_ct_in_get(const struct ip_vs_conn_param *p) { unsigned hash; struct ip_vs_conn *cp; + struct hlist_node *n; hash = ip_vs_conn_hashkey_param(p, false); ct_read_lock(hash); - list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { + hlist_for_each_entry(cp, n, &ip_vs_conn_tab[hash], c_list) { if (!ip_vs_conn_net_eq(cp, p->net)) continue; if (p->pe_data && p->pe->ct_match) { @@ -394,6 +396,7 @@ struct ip_vs_conn *ip_vs_conn_out_get(const struct ip_vs_conn_param *p) { unsigned hash; struct ip_vs_conn *cp, *ret=NULL; + struct hlist_node *n; /* * Check for "full" addressed entries @@ -402,7 +405,7 @@ struct ip_vs_conn *ip_vs_conn_out_get(const struct ip_vs_conn_param *p) ct_read_lock(hash); - list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { + hlist_for_each_entry(cp, n, &ip_vs_conn_tab[hash], c_list) { if (cp->af == p->af && p->vport == cp->cport && p->cport == cp->dport && ip_vs_addr_equal(p->af, p->vaddr, &cp->caddr) && @@ -818,7 +821,7 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p, return NULL; } - INIT_LIST_HEAD(&cp->c_list); + INIT_HLIST_NODE(&cp->c_list); setup_timer(&cp->timer, ip_vs_conn_expire, (unsigned long)cp); ip_vs_conn_net_set(cp, p->net); cp->af = p->af; @@ -894,8 +897,8 @@ ip_vs_conn_new(const struct ip_vs_conn_param *p, */ #ifdef CONFIG_PROC_FS struct ip_vs_iter_state { - struct seq_net_private p; - struct list_head *l; + struct seq_net_private p; + struct hlist_head *l; }; static void *ip_vs_conn_array(struct seq_file *seq, loff_t pos) @@ -903,13 +906,14 @@ static void *ip_vs_conn_array(struct seq_file *seq, loff_t pos) int idx; struct ip_vs_conn *cp; struct ip_vs_iter_state *iter = seq->private; + struct hlist_node *n; for (idx = 0; idx < ip_vs_conn_tab_size; idx++) { ct_read_lock_bh(idx); - list_for_each_entry(cp, &ip_vs_conn_tab[idx], c_list) { + hlist_for_each_entry(cp, n, &ip_vs_conn_tab[idx], c_list) { if (pos-- == 0) { iter->l = &ip_vs_conn_tab[idx]; - return cp; + return cp; } } ct_read_unlock_bh(idx); @@ -930,7 +934,8 @@ static void *ip_vs_conn_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct ip_vs_conn *cp = v; struct ip_vs_iter_state *iter = seq->private; - struct list_head *e, *l = iter->l; + struct hlist_node *e; + struct hlist_head *l = iter->l; int idx; ++*pos; @@ -938,15 +943,15 @@ static void *ip_vs_conn_seq_next(struct seq_file *seq, void *v, loff_t *pos) return ip_vs_conn_array(seq, 0); /* more on same hash chain? */ - if ((e = cp->c_list.next) != l) - return list_entry(e, struct ip_vs_conn, c_list); + if ((e = cp->c_list.next)) + return hlist_entry(e, struct ip_vs_conn, c_list); idx = l - ip_vs_conn_tab; ct_read_unlock_bh(idx); while (++idx < ip_vs_conn_tab_size) { ct_read_lock_bh(idx); - list_for_each_entry(cp, &ip_vs_conn_tab[idx], c_list) { + hlist_for_each_entry(cp, e, &ip_vs_conn_tab[idx], c_list) { iter->l = &ip_vs_conn_tab[idx]; return cp; } @@ -959,7 +964,7 @@ static void *ip_vs_conn_seq_next(struct seq_file *seq, void *v, loff_t *pos) static void ip_vs_conn_seq_stop(struct seq_file *seq, void *v) { struct ip_vs_iter_state *iter = seq->private; - struct list_head *l = iter->l; + struct hlist_head *l = iter->l; if (l) ct_read_unlock_bh(l - ip_vs_conn_tab); @@ -1148,13 +1153,14 @@ void ip_vs_random_dropentry(struct net *net) */ for (idx = 0; idx < (ip_vs_conn_tab_size>>5); idx++) { unsigned hash = net_random() & ip_vs_conn_tab_mask; + struct hlist_node *n; /* * Lock is actually needed in this loop. */ ct_write_lock_bh(hash); - list_for_each_entry(cp, &ip_vs_conn_tab[hash], c_list) { + hlist_for_each_entry(cp, n, &ip_vs_conn_tab[hash], c_list) { if (cp->flags & IP_VS_CONN_F_TEMPLATE) /* connection template */ continue; @@ -1202,12 +1208,14 @@ static void ip_vs_conn_flush(struct net *net) flush_again: for (idx = 0; idx < ip_vs_conn_tab_size; idx++) { + struct hlist_node *n; + /* * Lock is actually needed in this loop. */ ct_write_lock_bh(idx); - list_for_each_entry(cp, &ip_vs_conn_tab[idx], c_list) { + hlist_for_each_entry(cp, n, &ip_vs_conn_tab[idx], c_list) { if (!ip_vs_conn_net_eq(cp, net)) continue; IP_VS_DBG(4, "del connection\n"); @@ -1265,8 +1273,7 @@ int __init ip_vs_conn_init(void) /* * Allocate the connection hash table and initialize its list heads */ - ip_vs_conn_tab = vmalloc(ip_vs_conn_tab_size * - sizeof(struct list_head)); + ip_vs_conn_tab = vmalloc(ip_vs_conn_tab_size * sizeof(*ip_vs_conn_tab)); if (!ip_vs_conn_tab) return -ENOMEM; @@ -1286,9 +1293,8 @@ int __init ip_vs_conn_init(void) IP_VS_DBG(0, "Each connection entry needs %Zd bytes at least\n", sizeof(struct ip_vs_conn)); - for (idx = 0; idx < ip_vs_conn_tab_size; idx++) { - INIT_LIST_HEAD(&ip_vs_conn_tab[idx]); - } + for (idx = 0; idx < ip_vs_conn_tab_size; idx++) + INIT_HLIST_HEAD(&ip_vs_conn_tab[idx]); for (idx = 0; idx < CT_LOCKARRAY_SIZE; idx++) { rwlock_init(&__ip_vs_conntbl_lock_array[idx].l); -- cgit v1.2.3 From 17a8f8e3734920cf2f030f2fa521a0b940ef6f90 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Thu, 24 Feb 2011 08:19:57 +0800 Subject: ipvs: use enum to instead of magic numbers Signed-off-by: Changli Gao Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_xmit.c | 41 +++++++++++++++++++++++++++-------------- 1 file changed, 27 insertions(+), 14 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 1f2a4e35fb11..a48239aba33b 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -43,6 +43,13 @@ #include +enum { + IP_VS_RT_MODE_LOCAL = 1, /* Allow local dest */ + IP_VS_RT_MODE_NON_LOCAL = 2, /* Allow non-local dest */ + IP_VS_RT_MODE_RDR = 4, /* Allow redirect from remote daddr to + * local + */ +}; /* * Destination cache to speed up outgoing route lookup @@ -77,11 +84,7 @@ __ip_vs_dst_check(struct ip_vs_dest *dest, u32 rtos) return dst; } -/* - * Get route to destination or remote server - * rt_mode: flags, &1=Allow local dest, &2=Allow non-local dest, - * &4=Allow redirect from remote daddr to local - */ +/* Get route to destination or remote server */ static struct rtable * __ip_vs_get_out_rt(struct sk_buff *skb, struct ip_vs_dest *dest, __be32 daddr, u32 rtos, int rt_mode) @@ -126,15 +129,16 @@ __ip_vs_get_out_rt(struct sk_buff *skb, struct ip_vs_dest *dest, } local = rt->rt_flags & RTCF_LOCAL; - if (!((local ? 1 : 2) & rt_mode)) { + if (!((local ? IP_VS_RT_MODE_LOCAL : IP_VS_RT_MODE_NON_LOCAL) & + rt_mode)) { IP_VS_DBG_RL("Stopping traffic to %s address, dest: %pI4\n", (rt->rt_flags & RTCF_LOCAL) ? "local":"non-local", &rt->rt_dst); ip_rt_put(rt); return NULL; } - if (local && !(rt_mode & 4) && !((ort = skb_rtable(skb)) && - ort->rt_flags & RTCF_LOCAL)) { + if (local && !(rt_mode & IP_VS_RT_MODE_RDR) && + !((ort = skb_rtable(skb)) && ort->rt_flags & RTCF_LOCAL)) { IP_VS_DBG_RL("Redirect from non-local address %pI4 to local " "requires NAT method, dest: %pI4\n", &ip_hdr(skb)->daddr, &rt->rt_dst); @@ -383,8 +387,8 @@ ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, EnterFunction(10); - if (!(rt = __ip_vs_get_out_rt(skb, NULL, iph->daddr, - RT_TOS(iph->tos), 2))) + if (!(rt = __ip_vs_get_out_rt(skb, NULL, iph->daddr, RT_TOS(iph->tos), + IP_VS_RT_MODE_NON_LOCAL))) goto tx_error_icmp; /* MTU checking */ @@ -512,7 +516,10 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, } if (!(rt = __ip_vs_get_out_rt(skb, cp->dest, cp->daddr.ip, - RT_TOS(iph->tos), 1|2|4))) + RT_TOS(iph->tos), + IP_VS_RT_MODE_LOCAL | + IP_VS_RT_MODE_NON_LOCAL | + IP_VS_RT_MODE_RDR))) goto tx_error_icmp; local = rt->rt_flags & RTCF_LOCAL; /* @@ -755,7 +762,8 @@ ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, EnterFunction(10); if (!(rt = __ip_vs_get_out_rt(skb, cp->dest, cp->daddr.ip, - RT_TOS(tos), 1|2))) + RT_TOS(tos), IP_VS_RT_MODE_LOCAL | + IP_VS_RT_MODE_NON_LOCAL))) goto tx_error_icmp; if (rt->rt_flags & RTCF_LOCAL) { ip_rt_put(rt); @@ -984,7 +992,9 @@ ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, EnterFunction(10); if (!(rt = __ip_vs_get_out_rt(skb, cp->dest, cp->daddr.ip, - RT_TOS(iph->tos), 1|2))) + RT_TOS(iph->tos), + IP_VS_RT_MODE_LOCAL | + IP_VS_RT_MODE_NON_LOCAL))) goto tx_error_icmp; if (rt->rt_flags & RTCF_LOCAL) { ip_rt_put(rt); @@ -1128,7 +1138,10 @@ ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, */ if (!(rt = __ip_vs_get_out_rt(skb, cp->dest, cp->daddr.ip, - RT_TOS(ip_hdr(skb)->tos), 1|2|4))) + RT_TOS(ip_hdr(skb)->tos), + IP_VS_RT_MODE_LOCAL | + IP_VS_RT_MODE_NON_LOCAL | + IP_VS_RT_MODE_RDR))) goto tx_error_icmp; local = rt->rt_flags & RTCF_LOCAL; -- cgit v1.2.3 From b552f7e3a9524abcbcdf86f0a99b2be58e55a9c6 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Sat, 19 Feb 2011 17:32:28 +0800 Subject: ipvs: unify the formula to estimate the overhead of processing connections lc and wlc use the same formula, but lblc and lblcr use another one. There is no reason for using two different formulas for the lc variants. The formula used by lc is used by all the lc variants in this patch. Signed-off-by: Changli Gao Acked-by: Wensong Zhang Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_lblc.c | 13 +++---------- net/netfilter/ipvs/ip_vs_lblcr.c | 25 +++++++------------------ net/netfilter/ipvs/ip_vs_lc.c | 18 +----------------- net/netfilter/ipvs/ip_vs_wlc.c | 20 ++------------------ 4 files changed, 13 insertions(+), 63 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_lblc.c b/net/netfilter/ipvs/ip_vs_lblc.c index 4a9c8cd19690..6bf7a807649c 100644 --- a/net/netfilter/ipvs/ip_vs_lblc.c +++ b/net/netfilter/ipvs/ip_vs_lblc.c @@ -389,12 +389,7 @@ __ip_vs_lblc_schedule(struct ip_vs_service *svc) int loh, doh; /* - * We think the overhead of processing active connections is fifty - * times higher than that of inactive connections in average. (This - * fifty times might not be accurate, we will change it later.) We - * use the following formula to estimate the overhead: - * dest->activeconns*50 + dest->inactconns - * and the load: + * We use the following formula to estimate the load: * (dest overhead) / dest->weight * * Remember -- no floats in kernel mode!!! @@ -410,8 +405,7 @@ __ip_vs_lblc_schedule(struct ip_vs_service *svc) continue; if (atomic_read(&dest->weight) > 0) { least = dest; - loh = atomic_read(&least->activeconns) * 50 - + atomic_read(&least->inactconns); + loh = ip_vs_dest_conn_overhead(least); goto nextstage; } } @@ -425,8 +419,7 @@ __ip_vs_lblc_schedule(struct ip_vs_service *svc) if (dest->flags & IP_VS_DEST_F_OVERLOAD) continue; - doh = atomic_read(&dest->activeconns) * 50 - + atomic_read(&dest->inactconns); + doh = ip_vs_dest_conn_overhead(dest); if (loh * atomic_read(&dest->weight) > doh * atomic_read(&least->weight)) { least = dest; diff --git a/net/netfilter/ipvs/ip_vs_lblcr.c b/net/netfilter/ipvs/ip_vs_lblcr.c index bd329b1e9589..00631765b92a 100644 --- a/net/netfilter/ipvs/ip_vs_lblcr.c +++ b/net/netfilter/ipvs/ip_vs_lblcr.c @@ -178,8 +178,7 @@ static inline struct ip_vs_dest *ip_vs_dest_set_min(struct ip_vs_dest_set *set) if ((atomic_read(&least->weight) > 0) && (least->flags & IP_VS_DEST_F_AVAILABLE)) { - loh = atomic_read(&least->activeconns) * 50 - + atomic_read(&least->inactconns); + loh = ip_vs_dest_conn_overhead(least); goto nextstage; } } @@ -192,8 +191,7 @@ static inline struct ip_vs_dest *ip_vs_dest_set_min(struct ip_vs_dest_set *set) if (dest->flags & IP_VS_DEST_F_OVERLOAD) continue; - doh = atomic_read(&dest->activeconns) * 50 - + atomic_read(&dest->inactconns); + doh = ip_vs_dest_conn_overhead(dest); if ((loh * atomic_read(&dest->weight) > doh * atomic_read(&least->weight)) && (dest->flags & IP_VS_DEST_F_AVAILABLE)) { @@ -228,8 +226,7 @@ static inline struct ip_vs_dest *ip_vs_dest_set_max(struct ip_vs_dest_set *set) list_for_each_entry(e, &set->list, list) { most = e->dest; if (atomic_read(&most->weight) > 0) { - moh = atomic_read(&most->activeconns) * 50 - + atomic_read(&most->inactconns); + moh = ip_vs_dest_conn_overhead(most); goto nextstage; } } @@ -239,8 +236,7 @@ static inline struct ip_vs_dest *ip_vs_dest_set_max(struct ip_vs_dest_set *set) nextstage: list_for_each_entry(e, &set->list, list) { dest = e->dest; - doh = atomic_read(&dest->activeconns) * 50 - + atomic_read(&dest->inactconns); + doh = ip_vs_dest_conn_overhead(dest); /* moh/mw < doh/dw ==> moh*dw < doh*mw, where mw,dw>0 */ if ((moh * atomic_read(&dest->weight) < doh * atomic_read(&most->weight)) @@ -563,12 +559,7 @@ __ip_vs_lblcr_schedule(struct ip_vs_service *svc) int loh, doh; /* - * We think the overhead of processing active connections is fifty - * times higher than that of inactive connections in average. (This - * fifty times might not be accurate, we will change it later.) We - * use the following formula to estimate the overhead: - * dest->activeconns*50 + dest->inactconns - * and the load: + * We use the following formula to estimate the load: * (dest overhead) / dest->weight * * Remember -- no floats in kernel mode!!! @@ -585,8 +576,7 @@ __ip_vs_lblcr_schedule(struct ip_vs_service *svc) if (atomic_read(&dest->weight) > 0) { least = dest; - loh = atomic_read(&least->activeconns) * 50 - + atomic_read(&least->inactconns); + loh = ip_vs_dest_conn_overhead(least); goto nextstage; } } @@ -600,8 +590,7 @@ __ip_vs_lblcr_schedule(struct ip_vs_service *svc) if (dest->flags & IP_VS_DEST_F_OVERLOAD) continue; - doh = atomic_read(&dest->activeconns) * 50 - + atomic_read(&dest->inactconns); + doh = ip_vs_dest_conn_overhead(dest); if (loh * atomic_read(&dest->weight) > doh * atomic_read(&least->weight)) { least = dest; diff --git a/net/netfilter/ipvs/ip_vs_lc.c b/net/netfilter/ipvs/ip_vs_lc.c index 60638007c6c7..f391819c0cca 100644 --- a/net/netfilter/ipvs/ip_vs_lc.c +++ b/net/netfilter/ipvs/ip_vs_lc.c @@ -22,22 +22,6 @@ #include - -static inline unsigned int -ip_vs_lc_dest_overhead(struct ip_vs_dest *dest) -{ - /* - * We think the overhead of processing active connections is 256 - * times higher than that of inactive connections in average. (This - * 256 times might not be accurate, we will change it later) We - * use the following formula to estimate the overhead now: - * dest->activeconns*256 + dest->inactconns - */ - return (atomic_read(&dest->activeconns) << 8) + - atomic_read(&dest->inactconns); -} - - /* * Least Connection scheduling */ @@ -62,7 +46,7 @@ ip_vs_lc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) if ((dest->flags & IP_VS_DEST_F_OVERLOAD) || atomic_read(&dest->weight) == 0) continue; - doh = ip_vs_lc_dest_overhead(dest); + doh = ip_vs_dest_conn_overhead(dest); if (!least || doh < loh) { least = dest; loh = doh; diff --git a/net/netfilter/ipvs/ip_vs_wlc.c b/net/netfilter/ipvs/ip_vs_wlc.c index fdf0f58962a4..bc1bfc48a17f 100644 --- a/net/netfilter/ipvs/ip_vs_wlc.c +++ b/net/netfilter/ipvs/ip_vs_wlc.c @@ -27,22 +27,6 @@ #include - -static inline unsigned int -ip_vs_wlc_dest_overhead(struct ip_vs_dest *dest) -{ - /* - * We think the overhead of processing active connections is 256 - * times higher than that of inactive connections in average. (This - * 256 times might not be accurate, we will change it later) We - * use the following formula to estimate the overhead now: - * dest->activeconns*256 + dest->inactconns - */ - return (atomic_read(&dest->activeconns) << 8) + - atomic_read(&dest->inactconns); -} - - /* * Weighted Least Connection scheduling */ @@ -71,7 +55,7 @@ ip_vs_wlc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) if (!(dest->flags & IP_VS_DEST_F_OVERLOAD) && atomic_read(&dest->weight) > 0) { least = dest; - loh = ip_vs_wlc_dest_overhead(least); + loh = ip_vs_dest_conn_overhead(least); goto nextstage; } } @@ -85,7 +69,7 @@ ip_vs_wlc_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) list_for_each_entry_continue(dest, &svc->destinations, n_list) { if (dest->flags & IP_VS_DEST_F_OVERLOAD) continue; - doh = ip_vs_wlc_dest_overhead(dest); + doh = ip_vs_dest_conn_overhead(dest); if (loh * atomic_read(&dest->weight) > doh * atomic_read(&least->weight)) { least = dest; -- cgit v1.2.3 From 8a80c79a776d1b1b54895314ffaf53d0c7604c80 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 28 Feb 2011 17:59:15 +0100 Subject: netfilter: nf_ct_tcp: fix out of sync scenario while in SYN_RECV This patch fixes the out of sync scenarios while in SYN_RECV state. Quoting Jozsef, what it happens if we are out of sync if the following: > > b. conntrack entry is outdated, new SYN received > > - (b1) we ignore it but save the initialization data from it > > - (b2) when the reply SYN/ACK receives and it matches the saved data, > > we pick up the new connection This is what it should happen if we are in SYN_RECV state. Initially, the SYN packet hits b1, thus we save data from it. But the SYN/ACK packet is considered a retransmission given that we're in SYN_RECV state. Therefore, we never hit b2 and we don't get in sync. To fix this, we ignore SYN/ACK if we are in SYN_RECV. If the previous packet was a SYN, then we enter the ignore case that get us in sync. This patch helps a lot to conntrackd in stress scenarios (assumming a client that generates lots of small TCP connections). During the failover, consider that the new primary has injected one outdated flow in SYN_RECV state (this is likely to happen if the conntrack event rate is high because the backup will be a bit delayed from the primary). With the current code, if the client starts a new fresh connection that matches the tuple, the SYN packet will be ignored without updating the state tracking, and the SYN+ACK in reply will blocked as it will not pass checkings III or IV (since all state tracking in the original direction is not initialized because of the SYN packet was ignored and the ignore case that get us in sync is not applied). I posted a couple of patches before this one. Changli Gao spotted a simpler way to fix this problem. This patch implements his idea. Cc: Changli Gao Cc: Jozsef Kadlecsik Signed-off-by: Pablo Neira Ayuso Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_proto_tcp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 6f38d0e2ea4a..37bf94394be0 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -227,11 +227,11 @@ static const u8 tcp_conntracks[2][6][TCP_CONNTRACK_MAX] = { * sCL -> sIV */ /* sNO, sSS, sSR, sES, sFW, sCW, sLA, sTW, sCL, sS2 */ -/*synack*/ { sIV, sSR, sSR, sIG, sIG, sIG, sIG, sIG, sIG, sSR }, +/*synack*/ { sIV, sSR, sIG, sIG, sIG, sIG, sIG, sIG, sIG, sSR }, /* * sSS -> sSR Standard open. * sS2 -> sSR Simultaneous open - * sSR -> sSR Retransmitted SYN/ACK. + * sSR -> sIG Retransmitted SYN/ACK, ignore it. * sES -> sIG Late retransmitted SYN/ACK? * sFW -> sIG Might be SYN/ACK answering ignored SYN * sCW -> sIG -- cgit v1.2.3 From 452edd598f60522c11f7f88fdbab27eb36509d1a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 2 Mar 2011 13:27:41 -0800 Subject: xfrm: Return dst directly from xfrm_lookup() Instead of on the stack. Signed-off-by: David S. Miller --- net/netfilter/ipvs/ip_vs_xmit.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index a48239aba33b..6264219f0a42 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -218,8 +218,13 @@ __ip_vs_route_output_v6(struct net *net, struct in6_addr *daddr, ipv6_dev_get_saddr(net, ip6_dst_idev(dst)->dev, &fl.fl6_dst, 0, &fl.fl6_src) < 0) goto out_err; - if (do_xfrm && xfrm_lookup(net, &dst, &fl, NULL, 0) < 0) - goto out_err; + if (do_xfrm) { + dst = xfrm_lookup(net, dst, &fl, NULL, 0); + if (IS_ERR(dst)) { + dst = NULL; + goto out_err; + } + } ipv6_addr_copy(ret_saddr, &fl.fl6_src); return dst; -- cgit v1.2.3 From b23dd4fe42b455af5c6e20966b7d6959fa8352ea Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 2 Mar 2011 14:31:35 -0800 Subject: ipv4: Make output route lookup return rtable directly. Instead of on the stack. Signed-off-by: David S. Miller --- net/netfilter/ipvs/ip_vs_xmit.c | 9 ++++++--- net/netfilter/xt_TEE.c | 3 ++- 2 files changed, 8 insertions(+), 4 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 6264219f0a42..878f6dd9dbad 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -103,7 +103,8 @@ __ip_vs_get_out_rt(struct sk_buff *skb, struct ip_vs_dest *dest, .fl4_tos = rtos, }; - if (ip_route_output_key(net, &rt, &fl)) { + rt = ip_route_output_key(net, &fl); + if (IS_ERR(rt)) { spin_unlock(&dest->dst_lock); IP_VS_DBG_RL("ip_route_output error, dest: %pI4\n", &dest->addr.ip); @@ -121,7 +122,8 @@ __ip_vs_get_out_rt(struct sk_buff *skb, struct ip_vs_dest *dest, .fl4_tos = rtos, }; - if (ip_route_output_key(net, &rt, &fl)) { + rt = ip_route_output_key(net, &fl); + if (IS_ERR(rt)) { IP_VS_DBG_RL("ip_route_output error, dest: %pI4\n", &daddr); return NULL; @@ -180,7 +182,8 @@ __ip_vs_reroute_locally(struct sk_buff *skb) .mark = skb->mark, }; - if (ip_route_output_key(net, &rt, &fl)) + rt = ip_route_output_key(net, &fl); + if (IS_ERR(rt)) return 0; if (!(rt->rt_flags & RTCF_LOCAL)) { ip_rt_put(rt); diff --git a/net/netfilter/xt_TEE.c b/net/netfilter/xt_TEE.c index 5128a6c4cb2c..624725b5286f 100644 --- a/net/netfilter/xt_TEE.c +++ b/net/netfilter/xt_TEE.c @@ -73,7 +73,8 @@ tee_tg_route4(struct sk_buff *skb, const struct xt_tee_tginfo *info) fl.fl4_dst = info->gw.ip; fl.fl4_tos = RT_TOS(iph->tos); fl.fl4_scope = RT_SCOPE_UNIVERSE; - if (ip_route_output_key(net, &rt, &fl) != 0) + rt = ip_route_output_key(net, &fl); + if (IS_ERR(rt)) return false; skb_dst_drop(skb); -- cgit v1.2.3 From 9846ada138accc63994b57ebdfa76e3e137729e2 Mon Sep 17 00:00:00 2001 From: Shan Wei Date: Tue, 8 Mar 2011 15:37:27 +0100 Subject: netfilter: ipset: fix the compile warning in ip_set_create MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit net/netfilter/ipset/ip_set_core.c:615: warning: ‘clash’ may be used uninitialized in this function Signed-off-by: Shan Wei Signed-off-by: Jozsef Kadlecsik Signed-off-by: Patrick McHardy --- net/netfilter/ipset/ip_set_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c index 8b1a54c1e400..618a615acc9d 100644 --- a/net/netfilter/ipset/ip_set_core.c +++ b/net/netfilter/ipset/ip_set_core.c @@ -612,7 +612,7 @@ ip_set_create(struct sock *ctnl, struct sk_buff *skb, const struct nlmsghdr *nlh, const struct nlattr * const attr[]) { - struct ip_set *set, *clash; + struct ip_set *set, *clash = NULL; ip_set_id_t index = IPSET_INVALID_ID; struct nlattr *tb[IPSET_ATTR_CREATE_MAX+1] = {}; const char *name, *typename; -- cgit v1.2.3 From adb00ae2ea0ec65f9d3d06079950c0f0ade3b614 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 9 Mar 2011 14:14:26 +0100 Subject: netfilter: x_tables: misuse of try_then_request_module Since xt_find_match() returns ERR_PTR(xx) on error not NULL, the macro try_then_request_module won't work correctly here. The macro expects its first argument will be zero if condition fails. But ERR_PTR(-ENOENT) is not zero. The correct solution is to propagate the error value back. Found by inspection, and compile tested only. Signed-off-by: Stephen Hemminger Signed-off-by: Patrick McHardy --- net/netfilter/x_tables.c | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 0a77d2ff2154..271eed32a6a1 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -183,7 +183,7 @@ EXPORT_SYMBOL(xt_unregister_matches); /* * These are weird, but module loading must not be done with mutex * held (since they will register), and we have to have a single - * function to use try_then_request_module(). + * function to use. */ /* Find match, grabs ref. Returns ERR_PTR() on error. */ @@ -221,9 +221,13 @@ xt_request_find_match(uint8_t nfproto, const char *name, uint8_t revision) { struct xt_match *match; - match = try_then_request_module(xt_find_match(nfproto, name, revision), - "%st_%s", xt_prefix[nfproto], name); - return (match != NULL) ? match : ERR_PTR(-ENOENT); + match = xt_find_match(nfproto, name, revision); + if (IS_ERR(match)) { + request_module("%st_%s", xt_prefix[nfproto], name); + match = xt_find_match(nfproto, name, revision); + } + + return match; } EXPORT_SYMBOL_GPL(xt_request_find_match); @@ -261,9 +265,13 @@ struct xt_target *xt_request_find_target(u8 af, const char *name, u8 revision) { struct xt_target *target; - target = try_then_request_module(xt_find_target(af, name, revision), - "%st_%s", xt_prefix[af], name); - return (target != NULL) ? target : ERR_PTR(-ENOENT); + target = xt_find_target(af, name, revision); + if (IS_ERR(target)) { + request_module("%st_%s", xt_prefix[af], name); + target = xt_find_target(af, name, revision); + } + + return target; } EXPORT_SYMBOL_GPL(xt_request_find_target); -- cgit v1.2.3 From 78fbfd8a653ca972afe479517a40661bfff6d8c3 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sat, 12 Mar 2011 00:00:52 -0500 Subject: ipv4: Create and use route lookup helpers. The idea here is this minimizes the number of places one has to edit in order to make changes to how flows are defined and used. Signed-off-by: David S. Miller --- net/netfilter/ipvs/ip_vs_xmit.c | 14 ++------------ 1 file changed, 2 insertions(+), 12 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 878f6dd9dbad..faf381d9da7c 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -98,12 +98,7 @@ __ip_vs_get_out_rt(struct sk_buff *skb, struct ip_vs_dest *dest, spin_lock(&dest->dst_lock); if (!(rt = (struct rtable *) __ip_vs_dst_check(dest, rtos))) { - struct flowi fl = { - .fl4_dst = dest->addr.ip, - .fl4_tos = rtos, - }; - - rt = ip_route_output_key(net, &fl); + rt = ip_route_output(net, dest->addr.ip, 0, rtos, 0); if (IS_ERR(rt)) { spin_unlock(&dest->dst_lock); IP_VS_DBG_RL("ip_route_output error, dest: %pI4\n", @@ -117,12 +112,7 @@ __ip_vs_get_out_rt(struct sk_buff *skb, struct ip_vs_dest *dest, } spin_unlock(&dest->dst_lock); } else { - struct flowi fl = { - .fl4_dst = daddr, - .fl4_tos = rtos, - }; - - rt = ip_route_output_key(net, &fl); + rt = ip_route_output(net, daddr, 0, rtos, 0); if (IS_ERR(rt)) { IP_VS_DBG_RL("ip_route_output error, dest: %pI4\n", &daddr); -- cgit v1.2.3 From 1d28f42c1bd4bb2363d88df74d0128b4da135b4a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sat, 12 Mar 2011 00:29:39 -0500 Subject: net: Put flowi_* prefix on AF independent members of struct flowi I intend to turn struct flowi into a union of AF specific flowi structs. There will be a common structure that each variant includes first, much like struct sock_common. This is the first step to move in that direction. Signed-off-by: David S. Miller --- net/netfilter/ipvs/ip_vs_ctl.c | 2 +- net/netfilter/ipvs/ip_vs_xmit.c | 2 +- net/netfilter/xt_TEE.c | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index d69ec26b6bd4..d07a32aa07b6 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -76,7 +76,7 @@ static int __ip_vs_addr_is_local_v6(struct net *net, { struct rt6_info *rt; struct flowi fl = { - .oif = 0, + .flowi_oif = 0, .fl6_dst = *addr, .fl6_src = { .s6_addr32 = {0, 0, 0, 0} }, }; diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index faf381d9da7c..cc8071f68903 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -169,7 +169,7 @@ __ip_vs_reroute_locally(struct sk_buff *skb) .fl4_dst = iph->daddr, .fl4_src = iph->saddr, .fl4_tos = RT_TOS(iph->tos), - .mark = skb->mark, + .flowi_mark = skb->mark, }; rt = ip_route_output_key(net, &fl); diff --git a/net/netfilter/xt_TEE.c b/net/netfilter/xt_TEE.c index 624725b5286f..cb14ae2de15d 100644 --- a/net/netfilter/xt_TEE.c +++ b/net/netfilter/xt_TEE.c @@ -68,7 +68,7 @@ tee_tg_route4(struct sk_buff *skb, const struct xt_tee_tginfo *info) if (info->priv) { if (info->priv->oif == -1) return false; - fl.oif = info->priv->oif; + fl.flowi_oif = info->priv->oif; } fl.fl4_dst = info->gw.ip; fl.fl4_tos = RT_TOS(iph->tos); @@ -149,7 +149,7 @@ tee_tg_route6(struct sk_buff *skb, const struct xt_tee_tginfo *info) if (info->priv) { if (info->priv->oif == -1) return false; - fl.oif = info->priv->oif; + fl.flowi_oif = info->priv->oif; } fl.fl6_dst = info->gw.in6; fl.fl6_flowlabel = ((iph->flow_lbl[0] & 0xF) << 16) | -- cgit v1.2.3 From 9d6ec938019c6b16cb9ec96598ebe8f20de435fe Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sat, 12 Mar 2011 01:12:47 -0500 Subject: ipv4: Use flowi4 in public route lookup interfaces. Signed-off-by: David S. Miller --- net/netfilter/ipvs/ip_vs_xmit.c | 12 ++++++------ net/netfilter/xt_TEE.c | 14 +++++++------- 2 files changed, 13 insertions(+), 13 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index cc8071f68903..7dc00e313611 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -165,14 +165,14 @@ __ip_vs_reroute_locally(struct sk_buff *skb) return 0; refdst_drop(orefdst); } else { - struct flowi fl = { - .fl4_dst = iph->daddr, - .fl4_src = iph->saddr, - .fl4_tos = RT_TOS(iph->tos), - .flowi_mark = skb->mark, + struct flowi4 fl4 = { + .daddr = iph->daddr, + .saddr = iph->saddr, + .flowi4_tos = RT_TOS(iph->tos), + .flowi4_mark = skb->mark, }; - rt = ip_route_output_key(net, &fl); + rt = ip_route_output_key(net, &fl4); if (IS_ERR(rt)) return 0; if (!(rt->rt_flags & RTCF_LOCAL)) { diff --git a/net/netfilter/xt_TEE.c b/net/netfilter/xt_TEE.c index cb14ae2de15d..d8c00f9342ae 100644 --- a/net/netfilter/xt_TEE.c +++ b/net/netfilter/xt_TEE.c @@ -62,18 +62,18 @@ tee_tg_route4(struct sk_buff *skb, const struct xt_tee_tginfo *info) const struct iphdr *iph = ip_hdr(skb); struct net *net = pick_net(skb); struct rtable *rt; - struct flowi fl; + struct flowi4 fl4; - memset(&fl, 0, sizeof(fl)); + memset(&fl4, 0, sizeof(fl4)); if (info->priv) { if (info->priv->oif == -1) return false; - fl.flowi_oif = info->priv->oif; + fl4.flowi4_oif = info->priv->oif; } - fl.fl4_dst = info->gw.ip; - fl.fl4_tos = RT_TOS(iph->tos); - fl.fl4_scope = RT_SCOPE_UNIVERSE; - rt = ip_route_output_key(net, &fl); + fl4.daddr = info->gw.ip; + fl4.flowi4_tos = RT_TOS(iph->tos); + fl4.flowi4_scope = RT_SCOPE_UNIVERSE; + rt = ip_route_output_key(net, &fl4); if (IS_ERR(rt)) return false; -- cgit v1.2.3 From 5a49d0e04d62ab3f85aea4d15e0ca8be9b0ee89b Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sat, 12 Mar 2011 02:14:05 -0500 Subject: netfilter: Use flowi4 and flowi6 in nf_conntrack_h323_main Signed-off-by: David S. Miller --- net/netfilter/nf_conntrack_h323_main.c | 32 ++++++++++++++++++++------------ 1 file changed, 20 insertions(+), 12 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c index b969025cf82f..533a183e6661 100644 --- a/net/netfilter/nf_conntrack_h323_main.c +++ b/net/netfilter/nf_conntrack_h323_main.c @@ -714,7 +714,6 @@ static int callforward_do_filter(const union nf_inet_addr *src, u_int8_t family) { const struct nf_afinfo *afinfo; - struct flowi fl1, fl2; int ret = 0; /* rcu_read_lock()ed by nf_hook_slow() */ @@ -722,17 +721,20 @@ static int callforward_do_filter(const union nf_inet_addr *src, if (!afinfo) return 0; - memset(&fl1, 0, sizeof(fl1)); - memset(&fl2, 0, sizeof(fl2)); - switch (family) { case AF_INET: { + struct flowi4 fl1, fl2; struct rtable *rt1, *rt2; - fl1.fl4_dst = src->ip; - fl2.fl4_dst = dst->ip; - if (!afinfo->route((struct dst_entry **)&rt1, &fl1)) { - if (!afinfo->route((struct dst_entry **)&rt2, &fl2)) { + memset(&fl1, 0, sizeof(fl1)); + fl1.daddr = src->ip; + + memset(&fl2, 0, sizeof(fl2)); + fl2.daddr = dst->ip; + if (!afinfo->route((struct dst_entry **)&rt1, + flowi4_to_flowi(&fl1))) { + if (!afinfo->route((struct dst_entry **)&rt2, + flowi4_to_flowi(&fl2))) { if (rt1->rt_gateway == rt2->rt_gateway && rt1->dst.dev == rt2->dst.dev) ret = 1; @@ -745,12 +747,18 @@ static int callforward_do_filter(const union nf_inet_addr *src, #if defined(CONFIG_NF_CONNTRACK_IPV6) || \ defined(CONFIG_NF_CONNTRACK_IPV6_MODULE) case AF_INET6: { + struct flowi6 fl1, fl2; struct rt6_info *rt1, *rt2; - memcpy(&fl1.fl6_dst, src, sizeof(fl1.fl6_dst)); - memcpy(&fl2.fl6_dst, dst, sizeof(fl2.fl6_dst)); - if (!afinfo->route((struct dst_entry **)&rt1, &fl1)) { - if (!afinfo->route((struct dst_entry **)&rt2, &fl2)) { + memset(&fl1, 0, sizeof(fl1)); + ipv6_addr_copy(&fl1.daddr, &src->in6); + + memset(&fl2, 0, sizeof(fl2)); + ipv6_addr_copy(&fl2.daddr, &dst->in6); + if (!afinfo->route((struct dst_entry **)&rt1, + flowi6_to_flowi(&fl1))) { + if (!afinfo->route((struct dst_entry **)&rt2, + flowi6_to_flowi(&fl2))) { if (!memcmp(&rt1->rt6i_gateway, &rt2->rt6i_gateway, sizeof(rt1->rt6i_gateway)) && rt1->dst.dev == rt2->dst.dev) -- cgit v1.2.3 From a1bbb0e698b4ba18c6356564923bb395bed0e576 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sat, 12 Mar 2011 02:16:48 -0500 Subject: netfilter: Use flowi4 and flowi6 in xt_TCPMSS Signed-off-by: David S. Miller --- net/netfilter/xt_TCPMSS.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_TCPMSS.c b/net/netfilter/xt_TCPMSS.c index eb81c380da1b..6e6b46cb1db9 100644 --- a/net/netfilter/xt_TCPMSS.c +++ b/net/netfilter/xt_TCPMSS.c @@ -148,16 +148,21 @@ tcpmss_mangle_packet(struct sk_buff *skb, static u_int32_t tcpmss_reverse_mtu(const struct sk_buff *skb, unsigned int family) { - struct flowi fl = {}; + struct flowi fl; const struct nf_afinfo *ai; struct rtable *rt = NULL; u_int32_t mtu = ~0U; - if (family == PF_INET) - fl.fl4_dst = ip_hdr(skb)->saddr; - else - fl.fl6_dst = ipv6_hdr(skb)->saddr; + if (family == PF_INET) { + struct flowi4 *fl4 = &fl.u.ip4; + memset(fl4, 0, sizeof(*fl4)); + fl4->daddr = ip_hdr(skb)->saddr; + } else { + struct flowi6 *fl6 = &fl.u.ip6; + memset(fl6, 0, sizeof(*fl6)); + ipv6_addr_copy(&fl6->daddr, &ipv6_hdr(skb)->saddr); + } rcu_read_lock(); ai = nf_get_afinfo(family); if (ai != NULL) -- cgit v1.2.3 From 4c9483b2fb5d2548c3cc1fe03cdd4484ceeb5d1c Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sat, 12 Mar 2011 16:22:43 -0500 Subject: ipv6: Convert to use flowi6 where applicable. Signed-off-by: David S. Miller --- net/netfilter/ipvs/ip_vs_ctl.c | 10 ++++------ net/netfilter/ipvs/ip_vs_xmit.c | 14 +++++++------- net/netfilter/xt_TEE.c | 12 ++++++------ 3 files changed, 17 insertions(+), 19 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index d07a32aa07b6..a60b20fa142e 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -75,15 +75,13 @@ static int __ip_vs_addr_is_local_v6(struct net *net, const struct in6_addr *addr) { struct rt6_info *rt; - struct flowi fl = { - .flowi_oif = 0, - .fl6_dst = *addr, - .fl6_src = { .s6_addr32 = {0, 0, 0, 0} }, + struct flowi6 fl6 = { + .daddr = *addr, }; - rt = (struct rt6_info *)ip6_route_output(net, NULL, &fl); + rt = (struct rt6_info *)ip6_route_output(net, NULL, &fl6); if (rt && rt->rt6i_dev && (rt->rt6i_dev->flags & IFF_LOOPBACK)) - return 1; + return 1; return 0; } diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 7dc00e313611..6132b213eddc 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -198,27 +198,27 @@ __ip_vs_route_output_v6(struct net *net, struct in6_addr *daddr, struct in6_addr *ret_saddr, int do_xfrm) { struct dst_entry *dst; - struct flowi fl = { - .fl6_dst = *daddr, + struct flowi6 fl6 = { + .daddr = *daddr, }; - dst = ip6_route_output(net, NULL, &fl); + dst = ip6_route_output(net, NULL, &fl6); if (dst->error) goto out_err; if (!ret_saddr) return dst; - if (ipv6_addr_any(&fl.fl6_src) && + if (ipv6_addr_any(&fl6.saddr) && ipv6_dev_get_saddr(net, ip6_dst_idev(dst)->dev, - &fl.fl6_dst, 0, &fl.fl6_src) < 0) + &fl6.daddr, 0, &fl6.saddr) < 0) goto out_err; if (do_xfrm) { - dst = xfrm_lookup(net, dst, &fl, NULL, 0); + dst = xfrm_lookup(net, dst, flowi6_to_flowi(&fl6), NULL, 0); if (IS_ERR(dst)) { dst = NULL; goto out_err; } } - ipv6_addr_copy(ret_saddr, &fl.fl6_src); + ipv6_addr_copy(ret_saddr, &fl6.saddr); return dst; out_err: diff --git a/net/netfilter/xt_TEE.c b/net/netfilter/xt_TEE.c index d8c00f9342ae..5f054a0dbbb1 100644 --- a/net/netfilter/xt_TEE.c +++ b/net/netfilter/xt_TEE.c @@ -143,18 +143,18 @@ tee_tg_route6(struct sk_buff *skb, const struct xt_tee_tginfo *info) const struct ipv6hdr *iph = ipv6_hdr(skb); struct net *net = pick_net(skb); struct dst_entry *dst; - struct flowi fl; + struct flowi6 fl6; - memset(&fl, 0, sizeof(fl)); + memset(&fl6, 0, sizeof(fl6)); if (info->priv) { if (info->priv->oif == -1) return false; - fl.flowi_oif = info->priv->oif; + fl6.flowi6_oif = info->priv->oif; } - fl.fl6_dst = info->gw.in6; - fl.fl6_flowlabel = ((iph->flow_lbl[0] & 0xF) << 16) | + fl6.daddr = info->gw.in6; + fl6.flowlabel = ((iph->flow_lbl[0] & 0xF) << 16) | (iph->flow_lbl[1] << 8) | iph->flow_lbl[2]; - dst = ip6_route_output(net, NULL, &fl); + dst = ip6_route_output(net, NULL, &fl6); if (dst == NULL) return false; -- cgit v1.2.3 From 42046e2e45c109ba703993c510401a11f716c8df Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 14 Mar 2011 19:11:44 +0100 Subject: netfilter: x_tables: return -ENOENT for non-existant matches/targets As Stephen correctly points out, we need to return -ENOENT in xt_find_match()/xt_find_target() after the patch "netfilter: x_tables: misuse of try_then_request_module" in order to properly indicate a non-existant module to the caller. Signed-off-by: Patrick McHardy --- net/netfilter/x_tables.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 271eed32a6a1..a9adf4c6b299 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -190,7 +190,7 @@ EXPORT_SYMBOL(xt_unregister_matches); struct xt_match *xt_find_match(u8 af, const char *name, u8 revision) { struct xt_match *m; - int err = 0; + int err = -ENOENT; if (mutex_lock_interruptible(&xt[af].mutex) != 0) return ERR_PTR(-EINTR); @@ -235,7 +235,7 @@ EXPORT_SYMBOL_GPL(xt_request_find_match); struct xt_target *xt_find_target(u8 af, const char *name, u8 revision) { struct xt_target *t; - int err = 0; + int err = -ENOENT; if (mutex_lock_interruptible(&xt[af].mutex) != 0) return ERR_PTR(-EINTR); -- cgit v1.2.3 From fe8f661f2c2bb058822f13f6f232e121bde1338f Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Mon, 14 Mar 2011 19:20:44 +0100 Subject: netfilter: nf_conntrack: fix sysctl memory leak Message in log because sysctl table was not empty at netns exit WARNING: at net/sysctl_net.c:84 sysctl_net_exit+0x2a/0x2c() Instrumenting showed that the nf_conntrack_timestamp was the entry that was being created but not cleared. Signed-off-by: Stephen Hemminger Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_core.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net/netfilter') diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 1909311c392a..118123606a21 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -1294,6 +1294,7 @@ static void nf_conntrack_cleanup_net(struct net *net) nf_ct_free_hashtable(net->ct.hash, net->ct.htable_size); nf_conntrack_ecache_fini(net); + nf_conntrack_tstamp_fini(net); nf_conntrack_acct_fini(net); nf_conntrack_expect_fini(net); kmem_cache_destroy(net->ct.nf_conntrack_cachep); -- cgit v1.2.3 From 097fc76a0805bdca17baf12cad9d3bcb215716a9 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Fri, 4 Mar 2011 12:26:17 +0200 Subject: ipvs: avoid lookup for fwmark 0 Restore the previous behaviour to lookup for fwmark service only when fwmark is non-null. This saves only CPU. Signed-off-by: Julian Anastasov Signed-off-by: Hans Schillstrom Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_ctl.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index c73b0c831a2d..f0369d665088 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -411,9 +411,11 @@ ip_vs_service_get(struct net *net, int af, __u32 fwmark, __u16 protocol, /* * Check the table hashed by fwmark first */ - svc = __ip_vs_svc_fwm_find(net, af, fwmark); - if (fwmark && svc) - goto out; + if (fwmark) { + svc = __ip_vs_svc_fwm_find(net, af, fwmark); + if (svc) + goto out; + } /* * Check the table hashed by -- cgit v1.2.3 From 4a569c0c0f833adace1e3aadaa38780ec2fcdf9e Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Fri, 4 Mar 2011 12:28:20 +0200 Subject: ipvs: remove _bh from percpu stats reading ip_vs_read_cpu_stats is called only from timer, so no need for _bh locks. Signed-off-by: Julian Anastasov Signed-off-by: Hans Schillstrom Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_est.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_est.c b/net/netfilter/ipvs/ip_vs_est.c index f560a05c965a..88bd71647bf5 100644 --- a/net/netfilter/ipvs/ip_vs_est.c +++ b/net/netfilter/ipvs/ip_vs_est.c @@ -69,10 +69,10 @@ static void ip_vs_read_cpu_stats(struct ip_vs_stats_user *sum, sum->inpkts += s->ustats.inpkts; sum->outpkts += s->ustats.outpkts; do { - start = u64_stats_fetch_begin_bh(&s->syncp); + start = u64_stats_fetch_begin(&s->syncp); inbytes = s->ustats.inbytes; outbytes = s->ustats.outbytes; - } while (u64_stats_fetch_retry_bh(&s->syncp, start)); + } while (u64_stats_fetch_retry(&s->syncp, start)); sum->inbytes += inbytes; sum->outbytes += outbytes; } else { @@ -80,10 +80,10 @@ static void ip_vs_read_cpu_stats(struct ip_vs_stats_user *sum, sum->inpkts = s->ustats.inpkts; sum->outpkts = s->ustats.outpkts; do { - start = u64_stats_fetch_begin_bh(&s->syncp); + start = u64_stats_fetch_begin(&s->syncp); sum->inbytes = s->ustats.inbytes; sum->outbytes = s->ustats.outbytes; - } while (u64_stats_fetch_retry_bh(&s->syncp, start)); + } while (u64_stats_fetch_retry(&s->syncp, start)); } } } -- cgit v1.2.3 From 6060c74a3de8ed142c78133e2829e74711f77387 Mon Sep 17 00:00:00 2001 From: Shan Wei Date: Mon, 7 Mar 2011 10:11:34 +0800 Subject: netfilter:ipvs: use kmemdup The semantic patch that makes this output is available in scripts/coccinelle/api/memdup.cocci. More information about semantic patching is available at http://coccinelle.lip6.fr/ Signed-off-by: Shan Wei Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_pe_sip.c | 9 ++++----- net/netfilter/ipvs/ip_vs_sync.c | 3 +-- 2 files changed, 5 insertions(+), 7 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_pe_sip.c b/net/netfilter/ipvs/ip_vs_pe_sip.c index 0d83bc01fed4..13d607ae9c52 100644 --- a/net/netfilter/ipvs/ip_vs_pe_sip.c +++ b/net/netfilter/ipvs/ip_vs_pe_sip.c @@ -92,14 +92,13 @@ ip_vs_sip_fill_param(struct ip_vs_conn_param *p, struct sk_buff *skb) if (get_callid(dptr, dataoff, datalen, &matchoff, &matchlen)) return -EINVAL; - p->pe_data = kmalloc(matchlen, GFP_ATOMIC); - if (!p->pe_data) - return -ENOMEM; - /* N.B: pe_data is only set on success, * this allows fallback to the default persistence logic on failure */ - memcpy(p->pe_data, dptr + matchoff, matchlen); + p->pe_data = kmemdup(dptr + matchoff, matchlen, GFP_ATOMIC); + if (!p->pe_data) + return -ENOMEM; + p->pe_data_len = matchlen; return 0; diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index fecf24de4af3..c5d13b05275a 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -697,13 +697,12 @@ ip_vs_conn_fill_param_sync(struct net *net, int af, union ip_vs_sync_conn *sc, return 1; } - p->pe_data = kmalloc(pe_data_len, GFP_ATOMIC); + p->pe_data = kmemdup(pe_data, pe_data_len, GFP_ATOMIC); if (!p->pe_data) { if (p->pe->module) module_put(p->pe->module); return -ENOMEM; } - memcpy(p->pe_data, pe_data, pe_data_len); p->pe_data_len = pe_data_len; } return 0; -- cgit v1.2.3 From 2a0751af09c3099cf2837c623ca5d0436317d02d Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Fri, 4 Mar 2011 12:20:35 +0200 Subject: ipvs: reorganize tot_stats The global tot_stats contains cpustats field just like the stats for dest and svc, so better use it to simplify the usage in estimation_timer. As tot_stats is registered as estimator we can remove the special ip_vs_read_cpu_stats call for tot_stats. Fix ip_vs_read_cpu_stats to be called under stats lock because it is still used as synchronization between estimation timer and user context (the stats readers). Also, make sure ip_vs_stats_percpu_show reads properly the u64 stats from user context. Signed-off-by: Julian Anastasov Eric Dumazet Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 6 +++--- net/netfilter/ipvs/ip_vs_ctl.c | 45 ++++++++++++++++++++++------------------- net/netfilter/ipvs/ip_vs_est.c | 3 +-- 3 files changed, 28 insertions(+), 26 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 2d1f932add46..6f4940e86fc9 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -132,7 +132,7 @@ ip_vs_in_stats(struct ip_vs_conn *cp, struct sk_buff *skb) s->ustats.inbytes += skb->len; u64_stats_update_end(&s->syncp); - s = this_cpu_ptr(ipvs->cpustats); + s = this_cpu_ptr(ipvs->tot_stats.cpustats); s->ustats.inpkts++; u64_stats_update_begin(&s->syncp); s->ustats.inbytes += skb->len; @@ -162,7 +162,7 @@ ip_vs_out_stats(struct ip_vs_conn *cp, struct sk_buff *skb) s->ustats.outbytes += skb->len; u64_stats_update_end(&s->syncp); - s = this_cpu_ptr(ipvs->cpustats); + s = this_cpu_ptr(ipvs->tot_stats.cpustats); s->ustats.outpkts++; u64_stats_update_begin(&s->syncp); s->ustats.outbytes += skb->len; @@ -183,7 +183,7 @@ ip_vs_conn_stats(struct ip_vs_conn *cp, struct ip_vs_service *svc) s = this_cpu_ptr(svc->stats.cpustats); s->ustats.conns++; - s = this_cpu_ptr(ipvs->cpustats); + s = this_cpu_ptr(ipvs->tot_stats.cpustats); s->ustats.conns++; } diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index f0369d665088..a2a67ad7e094 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -1481,7 +1481,7 @@ static int ip_vs_zero_all(struct net *net) } } - ip_vs_zero_stats(net_ipvs(net)->tot_stats); + ip_vs_zero_stats(&net_ipvs(net)->tot_stats); return 0; } @@ -1963,7 +1963,7 @@ static const struct file_operations ip_vs_info_fops = { static int ip_vs_stats_show(struct seq_file *seq, void *v) { struct net *net = seq_file_single_net(seq); - struct ip_vs_stats *tot_stats = net_ipvs(net)->tot_stats; + struct ip_vs_stats *tot_stats = &net_ipvs(net)->tot_stats; /* 01234567 01234567 01234567 0123456701234567 0123456701234567 */ seq_puts(seq, @@ -2007,7 +2007,8 @@ static const struct file_operations ip_vs_stats_fops = { static int ip_vs_stats_percpu_show(struct seq_file *seq, void *v) { struct net *net = seq_file_single_net(seq); - struct ip_vs_stats *tot_stats = net_ipvs(net)->tot_stats; + struct ip_vs_stats *tot_stats = &net_ipvs(net)->tot_stats; + struct ip_vs_cpu_stats *cpustats = tot_stats->cpustats; int i; /* 01234567 01234567 01234567 0123456701234567 0123456701234567 */ @@ -2017,11 +2018,20 @@ static int ip_vs_stats_percpu_show(struct seq_file *seq, void *v) "CPU Conns Packets Packets Bytes Bytes\n"); for_each_possible_cpu(i) { - struct ip_vs_cpu_stats *u = per_cpu_ptr(net->ipvs->cpustats, i); + struct ip_vs_cpu_stats *u = per_cpu_ptr(cpustats, i); + unsigned int start; + __u64 inbytes, outbytes; + + do { + start = u64_stats_fetch_begin_bh(&u->syncp); + inbytes = u->ustats.inbytes; + outbytes = u->ustats.outbytes; + } while (u64_stats_fetch_retry_bh(&u->syncp, start)); + seq_printf(seq, "%3X %8X %8X %8X %16LX %16LX\n", - i, u->ustats.conns, u->ustats.inpkts, - u->ustats.outpkts, (__u64)u->ustats.inbytes, - (__u64)u->ustats.outbytes); + i, u->ustats.conns, u->ustats.inpkts, + u->ustats.outpkts, (__u64)inbytes, + (__u64)outbytes); } spin_lock_bh(&tot_stats->lock); @@ -3505,17 +3515,12 @@ int __net_init __ip_vs_control_init(struct net *net) atomic_set(&ipvs->nullsvc_counter, 0); /* procfs stats */ - ipvs->tot_stats = kzalloc(sizeof(struct ip_vs_stats), GFP_KERNEL); - if (ipvs->tot_stats == NULL) { - pr_err("%s(): no memory.\n", __func__); - return -ENOMEM; - } - ipvs->cpustats = alloc_percpu(struct ip_vs_cpu_stats); - if (!ipvs->cpustats) { + ipvs->tot_stats.cpustats = alloc_percpu(struct ip_vs_cpu_stats); + if (!ipvs->tot_stats.cpustats) { pr_err("%s() alloc_percpu failed\n", __func__); goto err_alloc; } - spin_lock_init(&ipvs->tot_stats->lock); + spin_lock_init(&ipvs->tot_stats.lock); proc_net_fops_create(net, "ip_vs", 0, &ip_vs_info_fops); proc_net_fops_create(net, "ip_vs_stats", 0, &ip_vs_stats_fops); @@ -3563,7 +3568,7 @@ int __net_init __ip_vs_control_init(struct net *net) goto err_dup; } #endif - ip_vs_new_estimator(net, ipvs->tot_stats); + ip_vs_new_estimator(net, &ipvs->tot_stats); ipvs->sysctl_tbl = tbl; /* Schedule defense work */ INIT_DELAYED_WORK(&ipvs->defense_work, defense_work_handler); @@ -3571,9 +3576,8 @@ int __net_init __ip_vs_control_init(struct net *net) return 0; err_dup: - free_percpu(ipvs->cpustats); + free_percpu(ipvs->tot_stats.cpustats); err_alloc: - kfree(ipvs->tot_stats); return -ENOMEM; } @@ -3582,7 +3586,7 @@ static void __net_exit __ip_vs_control_cleanup(struct net *net) struct netns_ipvs *ipvs = net_ipvs(net); ip_vs_trash_cleanup(net); - ip_vs_kill_estimator(net, ipvs->tot_stats); + ip_vs_kill_estimator(net, &ipvs->tot_stats); cancel_delayed_work_sync(&ipvs->defense_work); cancel_work_sync(&ipvs->defense_work.work); #ifdef CONFIG_SYSCTL @@ -3591,8 +3595,7 @@ static void __net_exit __ip_vs_control_cleanup(struct net *net) proc_net_remove(net, "ip_vs_stats_percpu"); proc_net_remove(net, "ip_vs_stats"); proc_net_remove(net, "ip_vs"); - free_percpu(ipvs->cpustats); - kfree(ipvs->tot_stats); + free_percpu(ipvs->tot_stats.cpustats); } static struct pernet_operations ipvs_control_ops = { diff --git a/net/netfilter/ipvs/ip_vs_est.c b/net/netfilter/ipvs/ip_vs_est.c index 88bd71647bf5..b3751cfede2c 100644 --- a/net/netfilter/ipvs/ip_vs_est.c +++ b/net/netfilter/ipvs/ip_vs_est.c @@ -101,13 +101,12 @@ static void estimation_timer(unsigned long arg) struct netns_ipvs *ipvs; ipvs = net_ipvs(net); - ip_vs_read_cpu_stats(&ipvs->tot_stats->ustats, ipvs->cpustats); spin_lock(&ipvs->est_lock); list_for_each_entry(e, &ipvs->est_list, list) { s = container_of(e, struct ip_vs_stats, est); - ip_vs_read_cpu_stats(&s->ustats, s->cpustats); spin_lock(&s->lock); + ip_vs_read_cpu_stats(&s->ustats, s->cpustats); n_conns = s->ustats.conns; n_inpkts = s->ustats.inpkts; n_outpkts = s->ustats.outpkts; -- cgit v1.2.3 From 55a3d4e15c7c953ecc55b96b83d2679abf8a7899 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Mon, 14 Mar 2011 01:37:49 +0200 Subject: ipvs: properly zero stats and rates Currently, the new percpu counters are not zeroed and the zero commands do not work as expected, we still show the old sum of percpu values. OTOH, we can not reset the percpu counters from user context without causing the incrementing to use old and bogus values. So, as Eric Dumazet suggested fix that by moving all overhead to stats reading in user context. Do not introduce overhead in timer context (estimator) and incrementing (packet handling in softirqs). The new ustats0 field holds the zero point for all counter values, the rates always use 0 as base value as before. When showing the values to user space just give the difference between counters and the base values. The only drawback is that percpu stats are not zeroed, they are accessible only from /proc and are new interface, so it should not be a compatibility problem as long as the sum stats are correct after zeroing. Signed-off-by: Julian Anastasov Acked-by: Eric Dumazet Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_ctl.c | 96 ++++++++++++++++++++++++++---------------- net/netfilter/ipvs/ip_vs_est.c | 15 ++++--- 2 files changed, 68 insertions(+), 43 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index a2a67ad7e094..804fee7be694 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -711,13 +711,51 @@ static void ip_vs_trash_cleanup(struct net *net) } } +static void +ip_vs_copy_stats(struct ip_vs_stats_user *dst, struct ip_vs_stats *src) +{ +#define IP_VS_SHOW_STATS_COUNTER(c) dst->c = src->ustats.c - src->ustats0.c +#define IP_VS_SHOW_STATS_RATE(r) dst->r = src->ustats.r + + spin_lock_bh(&src->lock); + + IP_VS_SHOW_STATS_COUNTER(conns); + IP_VS_SHOW_STATS_COUNTER(inpkts); + IP_VS_SHOW_STATS_COUNTER(outpkts); + IP_VS_SHOW_STATS_COUNTER(inbytes); + IP_VS_SHOW_STATS_COUNTER(outbytes); + + IP_VS_SHOW_STATS_RATE(cps); + IP_VS_SHOW_STATS_RATE(inpps); + IP_VS_SHOW_STATS_RATE(outpps); + IP_VS_SHOW_STATS_RATE(inbps); + IP_VS_SHOW_STATS_RATE(outbps); + + spin_unlock_bh(&src->lock); +} static void ip_vs_zero_stats(struct ip_vs_stats *stats) { spin_lock_bh(&stats->lock); - memset(&stats->ustats, 0, sizeof(stats->ustats)); + /* get current counters as zero point, rates are zeroed */ + +#define IP_VS_ZERO_STATS_COUNTER(c) stats->ustats0.c = stats->ustats.c +#define IP_VS_ZERO_STATS_RATE(r) stats->ustats.r = 0 + + IP_VS_ZERO_STATS_COUNTER(conns); + IP_VS_ZERO_STATS_COUNTER(inpkts); + IP_VS_ZERO_STATS_COUNTER(outpkts); + IP_VS_ZERO_STATS_COUNTER(inbytes); + IP_VS_ZERO_STATS_COUNTER(outbytes); + + IP_VS_ZERO_STATS_RATE(cps); + IP_VS_ZERO_STATS_RATE(inpps); + IP_VS_ZERO_STATS_RATE(outpps); + IP_VS_ZERO_STATS_RATE(inbps); + IP_VS_ZERO_STATS_RATE(outbps); + ip_vs_zero_estimator(stats); spin_unlock_bh(&stats->lock); @@ -1963,7 +2001,7 @@ static const struct file_operations ip_vs_info_fops = { static int ip_vs_stats_show(struct seq_file *seq, void *v) { struct net *net = seq_file_single_net(seq); - struct ip_vs_stats *tot_stats = &net_ipvs(net)->tot_stats; + struct ip_vs_stats_user show; /* 01234567 01234567 01234567 0123456701234567 0123456701234567 */ seq_puts(seq, @@ -1971,22 +2009,18 @@ static int ip_vs_stats_show(struct seq_file *seq, void *v) seq_printf(seq, " Conns Packets Packets Bytes Bytes\n"); - spin_lock_bh(&tot_stats->lock); - seq_printf(seq, "%8X %8X %8X %16LX %16LX\n\n", tot_stats->ustats.conns, - tot_stats->ustats.inpkts, tot_stats->ustats.outpkts, - (unsigned long long) tot_stats->ustats.inbytes, - (unsigned long long) tot_stats->ustats.outbytes); + ip_vs_copy_stats(&show, &net_ipvs(net)->tot_stats); + seq_printf(seq, "%8X %8X %8X %16LX %16LX\n\n", show.conns, + show.inpkts, show.outpkts, + (unsigned long long) show.inbytes, + (unsigned long long) show.outbytes); /* 01234567 01234567 01234567 0123456701234567 0123456701234567 */ seq_puts(seq, " Conns/s Pkts/s Pkts/s Bytes/s Bytes/s\n"); - seq_printf(seq,"%8X %8X %8X %16X %16X\n", - tot_stats->ustats.cps, - tot_stats->ustats.inpps, - tot_stats->ustats.outpps, - tot_stats->ustats.inbps, - tot_stats->ustats.outbps); - spin_unlock_bh(&tot_stats->lock); + seq_printf(seq, "%8X %8X %8X %16X %16X\n", + show.cps, show.inpps, show.outpps, + show.inbps, show.outbps); return 0; } @@ -2297,14 +2331,6 @@ do_ip_vs_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) } -static void -ip_vs_copy_stats(struct ip_vs_stats_user *dst, struct ip_vs_stats *src) -{ - spin_lock_bh(&src->lock); - memcpy(dst, &src->ustats, sizeof(*dst)); - spin_unlock_bh(&src->lock); -} - static void ip_vs_copy_service(struct ip_vs_service_entry *dst, struct ip_vs_service *src) { @@ -2691,31 +2717,29 @@ static const struct nla_policy ip_vs_dest_policy[IPVS_DEST_ATTR_MAX + 1] = { static int ip_vs_genl_fill_stats(struct sk_buff *skb, int container_type, struct ip_vs_stats *stats) { + struct ip_vs_stats_user ustats; struct nlattr *nl_stats = nla_nest_start(skb, container_type); if (!nl_stats) return -EMSGSIZE; - spin_lock_bh(&stats->lock); - - NLA_PUT_U32(skb, IPVS_STATS_ATTR_CONNS, stats->ustats.conns); - NLA_PUT_U32(skb, IPVS_STATS_ATTR_INPKTS, stats->ustats.inpkts); - NLA_PUT_U32(skb, IPVS_STATS_ATTR_OUTPKTS, stats->ustats.outpkts); - NLA_PUT_U64(skb, IPVS_STATS_ATTR_INBYTES, stats->ustats.inbytes); - NLA_PUT_U64(skb, IPVS_STATS_ATTR_OUTBYTES, stats->ustats.outbytes); - NLA_PUT_U32(skb, IPVS_STATS_ATTR_CPS, stats->ustats.cps); - NLA_PUT_U32(skb, IPVS_STATS_ATTR_INPPS, stats->ustats.inpps); - NLA_PUT_U32(skb, IPVS_STATS_ATTR_OUTPPS, stats->ustats.outpps); - NLA_PUT_U32(skb, IPVS_STATS_ATTR_INBPS, stats->ustats.inbps); - NLA_PUT_U32(skb, IPVS_STATS_ATTR_OUTBPS, stats->ustats.outbps); + ip_vs_copy_stats(&ustats, stats); - spin_unlock_bh(&stats->lock); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_CONNS, ustats.conns); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_INPKTS, ustats.inpkts); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_OUTPKTS, ustats.outpkts); + NLA_PUT_U64(skb, IPVS_STATS_ATTR_INBYTES, ustats.inbytes); + NLA_PUT_U64(skb, IPVS_STATS_ATTR_OUTBYTES, ustats.outbytes); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_CPS, ustats.cps); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_INPPS, ustats.inpps); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_OUTPPS, ustats.outpps); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_INBPS, ustats.inbps); + NLA_PUT_U32(skb, IPVS_STATS_ATTR_OUTBPS, ustats.outbps); nla_nest_end(skb, nl_stats); return 0; nla_put_failure: - spin_unlock_bh(&stats->lock); nla_nest_cancel(skb, nl_stats); return -EMSGSIZE; } diff --git a/net/netfilter/ipvs/ip_vs_est.c b/net/netfilter/ipvs/ip_vs_est.c index b3751cfede2c..a85008796370 100644 --- a/net/netfilter/ipvs/ip_vs_est.c +++ b/net/netfilter/ipvs/ip_vs_est.c @@ -184,13 +184,14 @@ void ip_vs_kill_estimator(struct net *net, struct ip_vs_stats *stats) void ip_vs_zero_estimator(struct ip_vs_stats *stats) { struct ip_vs_estimator *est = &stats->est; - - /* set counters zero, caller must hold the stats->lock lock */ - est->last_inbytes = 0; - est->last_outbytes = 0; - est->last_conns = 0; - est->last_inpkts = 0; - est->last_outpkts = 0; + struct ip_vs_stats_user *u = &stats->ustats; + + /* reset counters, caller must hold the stats->lock lock */ + est->last_inbytes = u->inbytes; + est->last_outbytes = u->outbytes; + est->last_conns = u->conns; + est->last_inpkts = u->inpkts; + est->last_outpkts = u->outpkts; est->cps = 0; est->inpps = 0; est->outpps = 0; -- cgit v1.2.3 From ea9f22cce9c2530d659f9122819940b69506b2d9 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Mon, 14 Mar 2011 01:41:54 +0200 Subject: ipvs: optimize rates reading Move the estimator reading from estimation_timer to user context. ip_vs_read_estimator() will be used to decode the rate values. As the decoded rates are not set by estimation timer there is no need to reset them in ip_vs_zero_stats. There is no need ip_vs_new_estimator() to encode stats to rates, if the destination is in trash both the stats and the rates are inactive. Signed-off-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_ctl.c | 31 ++++++++++++------------------- net/netfilter/ipvs/ip_vs_est.c | 33 +++++++++++++-------------------- 2 files changed, 25 insertions(+), 39 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 804fee7be694..c93d806e73bd 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -715,7 +715,6 @@ static void ip_vs_copy_stats(struct ip_vs_stats_user *dst, struct ip_vs_stats *src) { #define IP_VS_SHOW_STATS_COUNTER(c) dst->c = src->ustats.c - src->ustats0.c -#define IP_VS_SHOW_STATS_RATE(r) dst->r = src->ustats.r spin_lock_bh(&src->lock); @@ -725,11 +724,7 @@ ip_vs_copy_stats(struct ip_vs_stats_user *dst, struct ip_vs_stats *src) IP_VS_SHOW_STATS_COUNTER(inbytes); IP_VS_SHOW_STATS_COUNTER(outbytes); - IP_VS_SHOW_STATS_RATE(cps); - IP_VS_SHOW_STATS_RATE(inpps); - IP_VS_SHOW_STATS_RATE(outpps); - IP_VS_SHOW_STATS_RATE(inbps); - IP_VS_SHOW_STATS_RATE(outbps); + ip_vs_read_estimator(dst, src); spin_unlock_bh(&src->lock); } @@ -742,7 +737,6 @@ ip_vs_zero_stats(struct ip_vs_stats *stats) /* get current counters as zero point, rates are zeroed */ #define IP_VS_ZERO_STATS_COUNTER(c) stats->ustats0.c = stats->ustats.c -#define IP_VS_ZERO_STATS_RATE(r) stats->ustats.r = 0 IP_VS_ZERO_STATS_COUNTER(conns); IP_VS_ZERO_STATS_COUNTER(inpkts); @@ -750,12 +744,6 @@ ip_vs_zero_stats(struct ip_vs_stats *stats) IP_VS_ZERO_STATS_COUNTER(inbytes); IP_VS_ZERO_STATS_COUNTER(outbytes); - IP_VS_ZERO_STATS_RATE(cps); - IP_VS_ZERO_STATS_RATE(inpps); - IP_VS_ZERO_STATS_RATE(outpps); - IP_VS_ZERO_STATS_RATE(inbps); - IP_VS_ZERO_STATS_RATE(outbps); - ip_vs_zero_estimator(stats); spin_unlock_bh(&stats->lock); @@ -2043,6 +2031,7 @@ static int ip_vs_stats_percpu_show(struct seq_file *seq, void *v) struct net *net = seq_file_single_net(seq); struct ip_vs_stats *tot_stats = &net_ipvs(net)->tot_stats; struct ip_vs_cpu_stats *cpustats = tot_stats->cpustats; + struct ip_vs_stats_user rates; int i; /* 01234567 01234567 01234567 0123456701234567 0123456701234567 */ @@ -2069,22 +2058,26 @@ static int ip_vs_stats_percpu_show(struct seq_file *seq, void *v) } spin_lock_bh(&tot_stats->lock); + seq_printf(seq, " ~ %8X %8X %8X %16LX %16LX\n\n", tot_stats->ustats.conns, tot_stats->ustats.inpkts, tot_stats->ustats.outpkts, (unsigned long long) tot_stats->ustats.inbytes, (unsigned long long) tot_stats->ustats.outbytes); + ip_vs_read_estimator(&rates, tot_stats); + + spin_unlock_bh(&tot_stats->lock); + /* 01234567 01234567 01234567 0123456701234567 0123456701234567 */ seq_puts(seq, " Conns/s Pkts/s Pkts/s Bytes/s Bytes/s\n"); seq_printf(seq, " %8X %8X %8X %16X %16X\n", - tot_stats->ustats.cps, - tot_stats->ustats.inpps, - tot_stats->ustats.outpps, - tot_stats->ustats.inbps, - tot_stats->ustats.outbps); - spin_unlock_bh(&tot_stats->lock); + rates.cps, + rates.inpps, + rates.outpps, + rates.inbps, + rates.outbps); return 0; } diff --git a/net/netfilter/ipvs/ip_vs_est.c b/net/netfilter/ipvs/ip_vs_est.c index a85008796370..fda75be231e8 100644 --- a/net/netfilter/ipvs/ip_vs_est.c +++ b/net/netfilter/ipvs/ip_vs_est.c @@ -117,27 +117,22 @@ static void estimation_timer(unsigned long arg) rate = (n_conns - e->last_conns) << 9; e->last_conns = n_conns; e->cps += ((long)rate - (long)e->cps) >> 2; - s->ustats.cps = (e->cps + 0x1FF) >> 10; rate = (n_inpkts - e->last_inpkts) << 9; e->last_inpkts = n_inpkts; e->inpps += ((long)rate - (long)e->inpps) >> 2; - s->ustats.inpps = (e->inpps + 0x1FF) >> 10; rate = (n_outpkts - e->last_outpkts) << 9; e->last_outpkts = n_outpkts; e->outpps += ((long)rate - (long)e->outpps) >> 2; - s->ustats.outpps = (e->outpps + 0x1FF) >> 10; rate = (n_inbytes - e->last_inbytes) << 4; e->last_inbytes = n_inbytes; e->inbps += ((long)rate - (long)e->inbps) >> 2; - s->ustats.inbps = (e->inbps + 0xF) >> 5; rate = (n_outbytes - e->last_outbytes) << 4; e->last_outbytes = n_outbytes; e->outbps += ((long)rate - (long)e->outbps) >> 2; - s->ustats.outbps = (e->outbps + 0xF) >> 5; spin_unlock(&s->lock); } spin_unlock(&ipvs->est_lock); @@ -151,21 +146,6 @@ void ip_vs_new_estimator(struct net *net, struct ip_vs_stats *stats) INIT_LIST_HEAD(&est->list); - est->last_conns = stats->ustats.conns; - est->cps = stats->ustats.cps<<10; - - est->last_inpkts = stats->ustats.inpkts; - est->inpps = stats->ustats.inpps<<10; - - est->last_outpkts = stats->ustats.outpkts; - est->outpps = stats->ustats.outpps<<10; - - est->last_inbytes = stats->ustats.inbytes; - est->inbps = stats->ustats.inbps<<5; - - est->last_outbytes = stats->ustats.outbytes; - est->outbps = stats->ustats.outbps<<5; - spin_lock_bh(&ipvs->est_lock); list_add(&est->list, &ipvs->est_list); spin_unlock_bh(&ipvs->est_lock); @@ -199,6 +179,19 @@ void ip_vs_zero_estimator(struct ip_vs_stats *stats) est->outbps = 0; } +/* Get decoded rates */ +void ip_vs_read_estimator(struct ip_vs_stats_user *dst, + struct ip_vs_stats *stats) +{ + struct ip_vs_estimator *e = &stats->est; + + dst->cps = (e->cps + 0x1FF) >> 10; + dst->inpps = (e->inpps + 0x1FF) >> 10; + dst->outpps = (e->outpps + 0x1FF) >> 10; + dst->inbps = (e->inbps + 0xF) >> 5; + dst->outbps = (e->outbps + 0xF) >> 5; +} + static int __net_init __ip_vs_estimator_init(struct net *net) { struct netns_ipvs *ipvs = net_ipvs(net); -- cgit v1.2.3 From 6ef757f965c9133e82116475eab7f30df391c7fa Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Mon, 14 Mar 2011 01:44:28 +0200 Subject: ipvs: rename estimator functions Rename ip_vs_new_estimator to ip_vs_start_estimator and ip_vs_kill_estimator to ip_vs_stop_estimator to better match their logic. Signed-off-by: Julian Anastasov Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_ctl.c | 12 ++++++------ net/netfilter/ipvs/ip_vs_est.c | 4 ++-- 2 files changed, 8 insertions(+), 8 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index c93d806e73bd..c5b1234b630e 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -802,7 +802,7 @@ __ip_vs_update_dest(struct ip_vs_service *svc, struct ip_vs_dest *dest, spin_unlock(&dest->dst_lock); if (add) - ip_vs_new_estimator(svc->net, &dest->stats); + ip_vs_start_estimator(svc->net, &dest->stats); write_lock_bh(&__ip_vs_svc_lock); @@ -1008,7 +1008,7 @@ static void __ip_vs_del_dest(struct net *net, struct ip_vs_dest *dest) { struct netns_ipvs *ipvs = net_ipvs(net); - ip_vs_kill_estimator(net, &dest->stats); + ip_vs_stop_estimator(net, &dest->stats); /* * Remove it from the d-linked list with the real services. @@ -1201,7 +1201,7 @@ ip_vs_add_service(struct net *net, struct ip_vs_service_user_kern *u, else if (svc->port == 0) atomic_inc(&ipvs->nullsvc_counter); - ip_vs_new_estimator(net, &svc->stats); + ip_vs_start_estimator(net, &svc->stats); /* Count only IPv4 services for old get/setsockopt interface */ if (svc->af == AF_INET) @@ -1353,7 +1353,7 @@ static void __ip_vs_del_service(struct ip_vs_service *svc) if (svc->af == AF_INET) ipvs->num_services--; - ip_vs_kill_estimator(svc->net, &svc->stats); + ip_vs_stop_estimator(svc->net, &svc->stats); /* Unbind scheduler */ old_sched = svc->scheduler; @@ -3585,7 +3585,7 @@ int __net_init __ip_vs_control_init(struct net *net) goto err_dup; } #endif - ip_vs_new_estimator(net, &ipvs->tot_stats); + ip_vs_start_estimator(net, &ipvs->tot_stats); ipvs->sysctl_tbl = tbl; /* Schedule defense work */ INIT_DELAYED_WORK(&ipvs->defense_work, defense_work_handler); @@ -3603,7 +3603,7 @@ static void __net_exit __ip_vs_control_cleanup(struct net *net) struct netns_ipvs *ipvs = net_ipvs(net); ip_vs_trash_cleanup(net); - ip_vs_kill_estimator(net, &ipvs->tot_stats); + ip_vs_stop_estimator(net, &ipvs->tot_stats); cancel_delayed_work_sync(&ipvs->defense_work); cancel_work_sync(&ipvs->defense_work.work); #ifdef CONFIG_SYSCTL diff --git a/net/netfilter/ipvs/ip_vs_est.c b/net/netfilter/ipvs/ip_vs_est.c index fda75be231e8..8c8766ca56ad 100644 --- a/net/netfilter/ipvs/ip_vs_est.c +++ b/net/netfilter/ipvs/ip_vs_est.c @@ -139,7 +139,7 @@ static void estimation_timer(unsigned long arg) mod_timer(&ipvs->est_timer, jiffies + 2*HZ); } -void ip_vs_new_estimator(struct net *net, struct ip_vs_stats *stats) +void ip_vs_start_estimator(struct net *net, struct ip_vs_stats *stats) { struct netns_ipvs *ipvs = net_ipvs(net); struct ip_vs_estimator *est = &stats->est; @@ -151,7 +151,7 @@ void ip_vs_new_estimator(struct net *net, struct ip_vs_stats *stats) spin_unlock_bh(&ipvs->est_lock); } -void ip_vs_kill_estimator(struct net *net, struct ip_vs_stats *stats) +void ip_vs_stop_estimator(struct net *net, struct ip_vs_stats *stats) { struct netns_ipvs *ipvs = net_ipvs(net); struct ip_vs_estimator *est = &stats->est; -- cgit v1.2.3 From ba4fd7e966fa837557a3ec846c5fd15328b212af Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Fri, 4 Feb 2011 18:33:01 +0900 Subject: IPVS: Add ip_vs_route_me_harder() Add ip_vs_route_me_harder() to avoid repeating the same code twice. Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 48 +++++++++++++++++++---------------------- 1 file changed, 22 insertions(+), 26 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 6f4940e86fc9..299c7f33e18b 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -631,6 +631,24 @@ static inline int ip_vs_gather_frags_v6(struct sk_buff *skb, u_int32_t user) } #endif +static int ip_vs_route_me_harder(int af, struct sk_buff *skb) +{ + struct netns_ipvs *ipvs = net_ipvs(skb_net(skb)); + +#ifdef CONFIG_IP_VS_IPV6 + if (af == AF_INET6) { + if (ipvs->sysctl_snat_reroute && ip6_route_me_harder(skb) != 0) + return 1; + } else +#endif + if ((ipvs->sysctl_snat_reroute || + skb_rtable(skb)->rt_flags & RTCF_LOCAL) && + ip_route_me_harder(skb, RTN_LOCAL) != 0) + return 1; + + return 0; +} + /* * Packet has been made sufficiently writable in caller * - inout: 1=in->out, 0=out->in @@ -737,7 +755,6 @@ static int handle_response_icmp(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, unsigned int offset, unsigned int ihl) { - struct netns_ipvs *ipvs; unsigned int verdict = NF_DROP; if (IP_VS_FWD_METHOD(cp) != 0) { @@ -759,8 +776,6 @@ static int handle_response_icmp(int af, struct sk_buff *skb, if (!skb_make_writable(skb, offset)) goto out; - ipvs = net_ipvs(skb_net(skb)); - #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) ip_vs_nat_icmp_v6(skb, pp, cp, 1); @@ -768,16 +783,8 @@ static int handle_response_icmp(int af, struct sk_buff *skb, #endif ip_vs_nat_icmp(skb, pp, cp, 1); -#ifdef CONFIG_IP_VS_IPV6 - if (af == AF_INET6) { - if (ipvs->sysctl_snat_reroute && ip6_route_me_harder(skb) != 0) - goto out; - } else -#endif - if ((ipvs->sysctl_snat_reroute || - skb_rtable(skb)->rt_flags & RTCF_LOCAL) && - ip_route_me_harder(skb, RTN_LOCAL) != 0) - goto out; + if (ip_vs_route_me_harder(af, skb)) + goto out; /* do the statistics and put it back */ ip_vs_out_stats(cp, skb); @@ -985,7 +992,6 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, struct ip_vs_conn *cp, int ihl) { struct ip_vs_protocol *pp = pd->pp; - struct netns_ipvs *ipvs; IP_VS_DBG_PKT(11, af, pp, skb, 0, "Outgoing packet"); @@ -1021,18 +1027,8 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_proto_data *pd, * if it came from this machine itself. So re-compute * the routing information. */ - ipvs = net_ipvs(skb_net(skb)); - -#ifdef CONFIG_IP_VS_IPV6 - if (af == AF_INET6) { - if (ipvs->sysctl_snat_reroute && ip6_route_me_harder(skb) != 0) - goto drop; - } else -#endif - if ((ipvs->sysctl_snat_reroute || - skb_rtable(skb)->rt_flags & RTCF_LOCAL) && - ip_route_me_harder(skb, RTN_LOCAL) != 0) - goto drop; + if (ip_vs_route_me_harder(af, skb)) + goto drop; IP_VS_DBG_PKT(10, af, pp, skb, 0, "After SNAT"); -- cgit v1.2.3 From 84b3cee39ff1ffc97f4f6fba8ad26786c1f6d8f5 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Fri, 4 Feb 2011 18:33:01 +0900 Subject: IPVS: Add sysctl_snat_reroute() In preparation for not including sysctl_snat_reroute in struct netns_ipvs when CONFIG_SYCTL is not defined. Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 299c7f33e18b..1d8a2a226151 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -599,6 +599,20 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, return NF_DROP; } +#ifdef CONFIG_SYSCTL + +static int sysctl_snat_reroute(struct sk_buff *skb) +{ + struct netns_ipvs *ipvs = net_ipvs(skb_net(skb)); + return ipvs->sysctl_snat_reroute; +} + +#else + +static int sysctl_snat_reroute(struct sk_buff *skb) { return 0; } + +#endif + __sum16 ip_vs_checksum_complete(struct sk_buff *skb, int offset) { return csum_fold(skb_checksum(skb, offset, skb->len - offset, 0)); @@ -633,15 +647,13 @@ static inline int ip_vs_gather_frags_v6(struct sk_buff *skb, u_int32_t user) static int ip_vs_route_me_harder(int af, struct sk_buff *skb) { - struct netns_ipvs *ipvs = net_ipvs(skb_net(skb)); - #ifdef CONFIG_IP_VS_IPV6 if (af == AF_INET6) { - if (ipvs->sysctl_snat_reroute && ip6_route_me_harder(skb) != 0) + if (sysctl_snat_reroute(skb) && ip6_route_me_harder(skb) != 0) return 1; } else #endif - if ((ipvs->sysctl_snat_reroute || + if ((sysctl_snat_reroute(skb) || skb_rtable(skb)->rt_flags & RTCF_LOCAL) && ip_route_me_harder(skb, RTN_LOCAL) != 0) return 1; -- cgit v1.2.3 From 0cfa558e2c21644a0dd6c21cfadd8bbeaf9fe1a0 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Fri, 4 Feb 2011 18:33:01 +0900 Subject: IPVS: Add sysctl_nat_icmp_send() In preparation for not including sysctl_nat_icmp_send in struct netns_ipvs when CONFIG_SYCTL is not defined. Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 1d8a2a226151..c9b83728f3ce 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -607,9 +607,16 @@ static int sysctl_snat_reroute(struct sk_buff *skb) return ipvs->sysctl_snat_reroute; } +static int sysctl_nat_icmp_send(struct net *net) +{ + struct netns_ipvs *ipvs = net_ipvs(net); + return ipvs->sysctl_nat_icmp_send; +} + #else static int sysctl_snat_reroute(struct sk_buff *skb) { return 0; } +static int sysctl_nat_icmp_send(struct net *net) { return 0; } #endif @@ -1074,7 +1081,6 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af) struct ip_vs_protocol *pp; struct ip_vs_proto_data *pd; struct ip_vs_conn *cp; - struct netns_ipvs *ipvs; EnterFunction(11); @@ -1149,11 +1155,10 @@ ip_vs_out(unsigned int hooknum, struct sk_buff *skb, int af) * Check if the packet belongs to an existing entry */ cp = pp->conn_out_get(af, skb, &iph, iph.len, 0); - ipvs = net_ipvs(net); if (likely(cp)) return handle_response(af, skb, pd, cp, iph.len); - if (ipvs->sysctl_nat_icmp_send && + if (sysctl_nat_icmp_send(net) && (pp->protocol == IPPROTO_TCP || pp->protocol == IPPROTO_UDP || pp->protocol == IPPROTO_SCTP)) { -- cgit v1.2.3 From 59e0350eada0516a810cb780db37746165f1d516 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Fri, 4 Feb 2011 18:33:01 +0900 Subject: IPVS: Add {sysctl_sync_threshold,period}() In preparation for not including sysctl_sync_threshold in struct netns_ipvs when CONFIG_SYCTL is not defined. Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 10 +++++----- net/netfilter/ipvs/ip_vs_ctl.c | 4 ++-- net/netfilter/ipvs/ip_vs_sync.c | 4 ++-- 3 files changed, 9 insertions(+), 9 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index c9b83728f3ce..6a0053d91741 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -1613,15 +1613,15 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) */ if (cp->flags & IP_VS_CONN_F_ONE_PACKET) - pkts = ipvs->sysctl_sync_threshold[0]; + pkts = sysctl_sync_threshold(ipvs); else pkts = atomic_add_return(1, &cp->in_pkts); if ((ipvs->sync_state & IP_VS_STATE_MASTER) && cp->protocol == IPPROTO_SCTP) { if ((cp->state == IP_VS_SCTP_S_ESTABLISHED && - (pkts % ipvs->sysctl_sync_threshold[1] - == ipvs->sysctl_sync_threshold[0])) || + (pkts % sysctl_sync_period(ipvs) + == sysctl_sync_threshold(ipvs))) || (cp->old_state != cp->state && ((cp->state == IP_VS_SCTP_S_CLOSED) || (cp->state == IP_VS_SCTP_S_SHUT_ACK_CLI) || @@ -1635,8 +1635,8 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) else if ((ipvs->sync_state & IP_VS_STATE_MASTER) && (((cp->protocol != IPPROTO_TCP || cp->state == IP_VS_TCP_S_ESTABLISHED) && - (pkts % ipvs->sysctl_sync_threshold[1] - == ipvs->sysctl_sync_threshold[0])) || + (pkts % sysctl_sync_period(ipvs) + == sysctl_sync_threshold(ipvs))) || ((cp->protocol == IPPROTO_TCP) && (cp->old_state != cp->state) && ((cp->state == IP_VS_TCP_S_FIN_WAIT) || (cp->state == IP_VS_TCP_S_CLOSE) || diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index c5b1234b630e..364520f66b7a 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -3569,8 +3569,8 @@ int __net_init __ip_vs_control_init(struct net *net) tbl[idx++].data = &ipvs->sysctl_cache_bypass; tbl[idx++].data = &ipvs->sysctl_expire_nodest_conn; tbl[idx++].data = &ipvs->sysctl_expire_quiescent_template; - ipvs->sysctl_sync_threshold[0] = 3; - ipvs->sysctl_sync_threshold[1] = 50; + ipvs->sysctl_sync_threshold[0] = DEFAULT_SYNC_THRESHOLD; + ipvs->sysctl_sync_threshold[1] = DEFAULT_SYNC_PERIOD; tbl[idx].data = &ipvs->sysctl_sync_threshold; tbl[idx++].maxlen = sizeof(ipvs->sysctl_sync_threshold); tbl[idx++].data = &ipvs->sysctl_nat_icmp_send; diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index c5d13b05275a..e84987fa1bf8 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -650,7 +650,7 @@ control: if (cp->flags & IP_VS_CONN_F_TEMPLATE) { int pkts = atomic_add_return(1, &cp->in_pkts); - if (pkts % ipvs->sysctl_sync_threshold[1] != 1) + if (pkts % sysctl_sync_period(ipvs) != 1) return; } goto sloop; @@ -794,7 +794,7 @@ static void ip_vs_proc_conn(struct net *net, struct ip_vs_conn_param *param, if (opt) memcpy(&cp->in_seq, opt, sizeof(*opt)); - atomic_set(&cp->in_pkts, ipvs->sysctl_sync_threshold[0]); + atomic_set(&cp->in_pkts, sysctl_sync_threshold(ipvs)); cp->state = state; cp->old_state = cp->state; /* -- cgit v1.2.3 From 7532e8d40ccfdde6667169eeac4fd7778d6eb462 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Fri, 4 Feb 2011 18:33:01 +0900 Subject: IPVS: Add sysctl_sync_ver() In preparation for not including sysctl_sync_ver in struct netns_ipvs when CONFIG_SYCTL is not defined. Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_sync.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index e84987fa1bf8..3e7961e85e9c 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -394,7 +394,7 @@ void ip_vs_sync_switch_mode(struct net *net, int mode) if (!(ipvs->sync_state & IP_VS_STATE_MASTER)) return; - if (mode == ipvs->sysctl_sync_ver || !ipvs->sync_buff) + if (mode == sysctl_sync_ver(ipvs) || !ipvs->sync_buff) return; spin_lock_bh(&ipvs->sync_buff_lock); @@ -521,7 +521,7 @@ void ip_vs_sync_conn(struct net *net, struct ip_vs_conn *cp) unsigned int len, pe_name_len, pad; /* Handle old version of the protocol */ - if (ipvs->sysctl_sync_ver == 0) { + if (sysctl_sync_ver(ipvs) == 0) { ip_vs_sync_conn_v0(net, cp); return; } -- cgit v1.2.3 From 71a8ab6cad63b4816711f2ea518755677a870f6f Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Fri, 4 Feb 2011 18:33:01 +0900 Subject: IPVS: Add sysctl_expire_nodest_conn() In preparation for not including sysctl_expire_nodest_conn in struct netns_ipvs when CONFIG_SYCTL is not defined. Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 6a0053d91741..d418bc60d00d 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -613,10 +613,16 @@ static int sysctl_nat_icmp_send(struct net *net) return ipvs->sysctl_nat_icmp_send; } +static int sysctl_expire_nodest_conn(struct netns_ipvs *ipvs) +{ + return ipvs->sysctl_expire_nodest_conn; +} + #else static int sysctl_snat_reroute(struct sk_buff *skb) { return 0; } static int sysctl_nat_icmp_send(struct net *net) { return 0; } +static int sysctl_expire_nodest_conn(struct netns_ipvs *ipvs) { return 0; } #endif @@ -1583,7 +1589,7 @@ ip_vs_in(unsigned int hooknum, struct sk_buff *skb, int af) if (cp->dest && !(cp->dest->flags & IP_VS_DEST_F_AVAILABLE)) { /* the destination server is not available */ - if (ipvs->sysctl_expire_nodest_conn) { + if (sysctl_expire_nodest_conn(ipvs)) { /* try to expire the connection immediately */ ip_vs_conn_expire_now(cp); } -- cgit v1.2.3 From 8e1b0b1b560019cafebe45a7d9e6ec1122fedc7b Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Fri, 4 Feb 2011 18:33:01 +0900 Subject: IPVS: Add expire_quiescent_template() In preparation for not including sysctl_expire_quiescent_template in struct netns_ipvs when CONFIG_SYCTL is not defined. Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_conn.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 9c2a517b69c8..f289306cbf12 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -680,6 +680,16 @@ static inline void ip_vs_unbind_dest(struct ip_vs_conn *cp) atomic_dec(&dest->refcnt); } +static int expire_quiescent_template(struct netns_ipvs *ipvs, + struct ip_vs_dest *dest) +{ +#ifdef CONFIG_SYSCTL + return ipvs->sysctl_expire_quiescent_template && + (atomic_read(&dest->weight) == 0); +#else + return 0; +#endif +} /* * Checking if the destination of a connection template is available. @@ -696,8 +706,7 @@ int ip_vs_check_template(struct ip_vs_conn *ct) */ if ((dest == NULL) || !(dest->flags & IP_VS_DEST_F_AVAILABLE) || - (ipvs->sysctl_expire_quiescent_template && - (atomic_read(&dest->weight) == 0))) { + expire_quiescent_template(ipvs, dest)) { IP_VS_DBG_BUF(9, "check_template: dest not available for " "protocol %s s:%s:%d v:%s:%d " "-> d:%s:%d\n", -- cgit v1.2.3 From b27d777ec54205eb56cf4e873d043a426881c629 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Fri, 4 Feb 2011 18:33:01 +0900 Subject: IPVS: Conditinally use sysctl_lblc{r}_expiration In preparation for not including sysctl_lblc{r}_expiration in struct netns_ipvs when CONFIG_SYCTL is not defined. Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_lblc.c | 16 +++++++++++++--- net/netfilter/ipvs/ip_vs_lblcr.c | 21 +++++++++++++++------ 2 files changed, 28 insertions(+), 9 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_lblc.c b/net/netfilter/ipvs/ip_vs_lblc.c index 6bf7a807649c..51a27f58030d 100644 --- a/net/netfilter/ipvs/ip_vs_lblc.c +++ b/net/netfilter/ipvs/ip_vs_lblc.c @@ -63,6 +63,8 @@ #define CHECK_EXPIRE_INTERVAL (60*HZ) #define ENTRY_TIMEOUT (6*60*HZ) +#define DEFAULT_EXPIRATION (24*60*60*HZ) + /* * It is for full expiration check. * When there is no partial expiration check (garbage collection) @@ -238,6 +240,15 @@ static void ip_vs_lblc_flush(struct ip_vs_lblc_table *tbl) } } +static int sysctl_lblc_expiration(struct ip_vs_service *svc) +{ +#ifdef CONFIG_SYSCTL + struct netns_ipvs *ipvs = net_ipvs(svc->net); + return ipvs->sysctl_lblc_expiration; +#else + return DEFAULT_EXPIRATION; +#endif +} static inline void ip_vs_lblc_full_check(struct ip_vs_service *svc) { @@ -245,7 +256,6 @@ static inline void ip_vs_lblc_full_check(struct ip_vs_service *svc) struct ip_vs_lblc_entry *en, *nxt; unsigned long now = jiffies; int i, j; - struct netns_ipvs *ipvs = net_ipvs(svc->net); for (i=0, j=tbl->rover; ibucket[j], list) { if (time_before(now, en->lastuse + - ipvs->sysctl_lblc_expiration)) + sysctl_lblc_expiration(svc))) continue; ip_vs_lblc_free(en); @@ -550,7 +560,7 @@ static int __net_init __ip_vs_lblc_init(struct net *net) return -ENOMEM; } else ipvs->lblc_ctl_table = vs_vars_table; - ipvs->sysctl_lblc_expiration = 24*60*60*HZ; + ipvs->sysctl_lblc_expiration = DEFAULT_EXPIRATION; ipvs->lblc_ctl_table[0].data = &ipvs->sysctl_lblc_expiration; #ifdef CONFIG_SYSCTL diff --git a/net/netfilter/ipvs/ip_vs_lblcr.c b/net/netfilter/ipvs/ip_vs_lblcr.c index 00631765b92a..7fb919061296 100644 --- a/net/netfilter/ipvs/ip_vs_lblcr.c +++ b/net/netfilter/ipvs/ip_vs_lblcr.c @@ -63,6 +63,8 @@ #define CHECK_EXPIRE_INTERVAL (60*HZ) #define ENTRY_TIMEOUT (6*60*HZ) +#define DEFAULT_EXPIRATION (24*60*60*HZ) + /* * It is for full expiration check. * When there is no partial expiration check (garbage collection) @@ -410,6 +412,15 @@ static void ip_vs_lblcr_flush(struct ip_vs_lblcr_table *tbl) } } +static int sysctl_lblcr_expiration(struct ip_vs_service *svc) +{ +#ifdef CONFIG_SYSCTL + struct netns_ipvs *ipvs = net_ipvs(svc->net); + return ipvs->sysctl_lblcr_expiration; +#else + return DEFAULT_EXPIRATION; +#endif +} static inline void ip_vs_lblcr_full_check(struct ip_vs_service *svc) { @@ -417,15 +428,14 @@ static inline void ip_vs_lblcr_full_check(struct ip_vs_service *svc) unsigned long now = jiffies; int i, j; struct ip_vs_lblcr_entry *en, *nxt; - struct netns_ipvs *ipvs = net_ipvs(svc->net); for (i=0, j=tbl->rover; isched_lock); list_for_each_entry_safe(en, nxt, &tbl->bucket[j], list) { - if (time_after(en->lastuse - + ipvs->sysctl_lblcr_expiration, now)) + if (time_after(en->lastuse + + sysctl_lblcr_expiration(svc), now)) continue; ip_vs_lblcr_free(en); @@ -650,7 +660,6 @@ ip_vs_lblcr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) read_lock(&svc->sched_lock); en = ip_vs_lblcr_get(svc->af, tbl, &iph.daddr); if (en) { - struct netns_ipvs *ipvs = net_ipvs(svc->net); /* We only hold a read lock, but this is atomic */ en->lastuse = jiffies; @@ -662,7 +671,7 @@ ip_vs_lblcr_schedule(struct ip_vs_service *svc, const struct sk_buff *skb) /* More than one destination + enough time passed by, cleanup */ if (atomic_read(&en->set.size) > 1 && time_after(jiffies, en->set.lastmod + - ipvs->sysctl_lblcr_expiration)) { + sysctl_lblcr_expiration(svc))) { struct ip_vs_dest *m; write_lock(&en->set.lock); @@ -746,7 +755,7 @@ static int __net_init __ip_vs_lblcr_init(struct net *net) return -ENOMEM; } else ipvs->lblcr_ctl_table = vs_vars_table; - ipvs->sysctl_lblcr_expiration = 24*60*60*HZ; + ipvs->sysctl_lblcr_expiration = DEFAULT_EXPIRATION; ipvs->lblcr_ctl_table[0].data = &ipvs->sysctl_lblcr_expiration; #ifdef CONFIG_SYSCTL -- cgit v1.2.3 From a7a86b8616bc1595c4f5f109b7c39d4eb5d55e32 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Fri, 4 Feb 2011 18:33:02 +0900 Subject: IPVS: Minimise ip_vs_leave when CONFIG_SYSCTL is undefined Much of ip_vs_leave() is unnecessary if CONFIG_SYSCTL is undefined. I tried an approach of breaking the now #ifdef'ed portions out into a separate function. However this appeared to grow the compiled code on x86_64 by about 200 bytes in the case where CONFIG_SYSCTL is defined. So I have gone with the simpler though less elegant #ifdef'ed solution for now. Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_core.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index d418bc60d00d..07accf6b2401 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -499,11 +499,13 @@ ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb, int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, struct ip_vs_proto_data *pd) { - struct net *net; - struct netns_ipvs *ipvs; __be16 _ports[2], *pptr; struct ip_vs_iphdr iph; +#ifdef CONFIG_SYSCTL + struct net *net; + struct netns_ipvs *ipvs; int unicast; +#endif ip_vs_fill_iphdr(svc->af, skb_network_header(skb), &iph); @@ -512,6 +514,8 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, ip_vs_service_put(svc); return NF_DROP; } + +#ifdef CONFIG_SYSCTL net = skb_net(skb); #ifdef CONFIG_IP_VS_IPV6 @@ -563,6 +567,7 @@ int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb, ip_vs_conn_put(cp); return ret; } +#endif /* * When the virtual ftp service is presented, packets destined -- cgit v1.2.3 From fb1de432c1c7c26afb2e86d166fc37888b6a4423 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Fri, 4 Feb 2011 18:33:02 +0900 Subject: IPVS: Conditionally define and use ip_vs_lblc{r}_table ip_vs_lblc_table and ip_vs_lblcr_table, and code that uses them are unnecessary when CONFIG_SYSCTL is undefined. Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_lblc.c | 15 ++++++++++----- net/netfilter/ipvs/ip_vs_lblcr.c | 14 ++++++++++---- 2 files changed, 20 insertions(+), 9 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_lblc.c b/net/netfilter/ipvs/ip_vs_lblc.c index 51a27f58030d..f276df9896b3 100644 --- a/net/netfilter/ipvs/ip_vs_lblc.c +++ b/net/netfilter/ipvs/ip_vs_lblc.c @@ -114,7 +114,7 @@ struct ip_vs_lblc_table { /* * IPVS LBLC sysctl table */ - +#ifdef CONFIG_SYSCTL static ctl_table vs_vars_table[] = { { .procname = "lblc_expiration", @@ -125,6 +125,7 @@ static ctl_table vs_vars_table[] = { }, { } }; +#endif static inline void ip_vs_lblc_free(struct ip_vs_lblc_entry *en) { @@ -548,6 +549,7 @@ static struct ip_vs_scheduler ip_vs_lblc_scheduler = /* * per netns init. */ +#ifdef CONFIG_SYSCTL static int __net_init __ip_vs_lblc_init(struct net *net) { struct netns_ipvs *ipvs = net_ipvs(net); @@ -563,7 +565,6 @@ static int __net_init __ip_vs_lblc_init(struct net *net) ipvs->sysctl_lblc_expiration = DEFAULT_EXPIRATION; ipvs->lblc_ctl_table[0].data = &ipvs->sysctl_lblc_expiration; -#ifdef CONFIG_SYSCTL ipvs->lblc_ctl_header = register_net_sysctl_table(net, net_vs_ctl_path, ipvs->lblc_ctl_table); @@ -572,7 +573,6 @@ static int __net_init __ip_vs_lblc_init(struct net *net) kfree(ipvs->lblc_ctl_table); return -ENOMEM; } -#endif return 0; } @@ -581,14 +581,19 @@ static void __net_exit __ip_vs_lblc_exit(struct net *net) { struct netns_ipvs *ipvs = net_ipvs(net); -#ifdef CONFIG_SYSCTL unregister_net_sysctl_table(ipvs->lblc_ctl_header); -#endif if (!net_eq(net, &init_net)) kfree(ipvs->lblc_ctl_table); } +#else + +static int __net_init __ip_vs_lblc_init(struct net *net) { return 0; } +static void __net_exit __ip_vs_lblc_exit(struct net *net) { } + +#endif + static struct pernet_operations ip_vs_lblc_ops = { .init = __ip_vs_lblc_init, .exit = __ip_vs_lblc_exit, diff --git a/net/netfilter/ipvs/ip_vs_lblcr.c b/net/netfilter/ipvs/ip_vs_lblcr.c index 7fb919061296..cb1c9913d38b 100644 --- a/net/netfilter/ipvs/ip_vs_lblcr.c +++ b/net/netfilter/ipvs/ip_vs_lblcr.c @@ -285,6 +285,7 @@ struct ip_vs_lblcr_table { }; +#ifdef CONFIG_SYSCTL /* * IPVS LBLCR sysctl table */ @@ -299,6 +300,7 @@ static ctl_table vs_vars_table[] = { }, { } }; +#endif static inline void ip_vs_lblcr_free(struct ip_vs_lblcr_entry *en) { @@ -743,6 +745,7 @@ static struct ip_vs_scheduler ip_vs_lblcr_scheduler = /* * per netns init. */ +#ifdef CONFIG_SYSCTL static int __net_init __ip_vs_lblcr_init(struct net *net) { struct netns_ipvs *ipvs = net_ipvs(net); @@ -758,7 +761,6 @@ static int __net_init __ip_vs_lblcr_init(struct net *net) ipvs->sysctl_lblcr_expiration = DEFAULT_EXPIRATION; ipvs->lblcr_ctl_table[0].data = &ipvs->sysctl_lblcr_expiration; -#ifdef CONFIG_SYSCTL ipvs->lblcr_ctl_header = register_net_sysctl_table(net, net_vs_ctl_path, ipvs->lblcr_ctl_table); @@ -767,7 +769,6 @@ static int __net_init __ip_vs_lblcr_init(struct net *net) kfree(ipvs->lblcr_ctl_table); return -ENOMEM; } -#endif return 0; } @@ -776,14 +777,19 @@ static void __net_exit __ip_vs_lblcr_exit(struct net *net) { struct netns_ipvs *ipvs = net_ipvs(net); -#ifdef CONFIG_SYSCTL unregister_net_sysctl_table(ipvs->lblcr_ctl_header); -#endif if (!net_eq(net, &init_net)) kfree(ipvs->lblcr_ctl_table); } +#else + +static int __net_init __ip_vs_lblcr_init(struct net *net) { return 0; } +static void __net_exit __ip_vs_lblcr_exit(struct net *net) { } + +#endif + static struct pernet_operations ip_vs_lblcr_ops = { .init = __ip_vs_lblcr_init, .exit = __ip_vs_lblcr_exit, -- cgit v1.2.3 From 14e405461e664b777e2a5636e10b2ebf36a686ec Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Fri, 4 Feb 2011 18:33:02 +0900 Subject: IPVS: Add __ip_vs_control_{init,cleanup}_sysctl() Break out the portions of __ip_vs_control_init() and __ip_vs_control_cleanup() where aren't necessary when CONFIG_SYSCTL is undefined. Signed-off-by: Simon Horman --- net/netfilter/ipvs/ip_vs_ctl.c | 98 ++++++++++++++++++++++++++---------------- 1 file changed, 62 insertions(+), 36 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 364520f66b7a..fa6d44c62de3 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -88,6 +88,8 @@ static int __ip_vs_addr_is_local_v6(struct net *net, return 0; } #endif + +#ifdef CONFIG_SYSCTL /* * update_defense_level is called from keventd and from sysctl, * so it needs to protect itself from softirqs @@ -229,6 +231,7 @@ static void defense_work_handler(struct work_struct *work) ip_vs_random_dropentry(ipvs->net); schedule_delayed_work(&ipvs->defense_work, DEFENSE_TIMER_PERIOD); } +#endif int ip_vs_use_count_inc(void) @@ -1511,7 +1514,7 @@ static int ip_vs_zero_all(struct net *net) return 0; } - +#ifdef CONFIG_SYSCTL static int proc_do_defense_mode(ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) @@ -1533,7 +1536,6 @@ proc_do_defense_mode(ctl_table *table, int write, return rc; } - static int proc_do_sync_threshold(ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) @@ -1767,6 +1769,7 @@ const struct ctl_path net_vs_ctl_path[] = { { } }; EXPORT_SYMBOL_GPL(net_vs_ctl_path); +#endif #ifdef CONFIG_PROC_FS @@ -3511,7 +3514,8 @@ static void ip_vs_genl_unregister(void) /* * per netns intit/exit func. */ -int __net_init __ip_vs_control_init(struct net *net) +#ifdef CONFIG_SYSCTL +int __net_init __ip_vs_control_init_sysctl(struct net *net) { int idx; struct netns_ipvs *ipvs = net_ipvs(net); @@ -3521,33 +3525,11 @@ int __net_init __ip_vs_control_init(struct net *net) spin_lock_init(&ipvs->dropentry_lock); spin_lock_init(&ipvs->droppacket_lock); spin_lock_init(&ipvs->securetcp_lock); - ipvs->rs_lock = __RW_LOCK_UNLOCKED(ipvs->rs_lock); - - /* Initialize rs_table */ - for (idx = 0; idx < IP_VS_RTAB_SIZE; idx++) - INIT_LIST_HEAD(&ipvs->rs_table[idx]); - - INIT_LIST_HEAD(&ipvs->dest_trash); - atomic_set(&ipvs->ftpsvc_counter, 0); - atomic_set(&ipvs->nullsvc_counter, 0); - - /* procfs stats */ - ipvs->tot_stats.cpustats = alloc_percpu(struct ip_vs_cpu_stats); - if (!ipvs->tot_stats.cpustats) { - pr_err("%s() alloc_percpu failed\n", __func__); - goto err_alloc; - } - spin_lock_init(&ipvs->tot_stats.lock); - - proc_net_fops_create(net, "ip_vs", 0, &ip_vs_info_fops); - proc_net_fops_create(net, "ip_vs_stats", 0, &ip_vs_stats_fops); - proc_net_fops_create(net, "ip_vs_stats_percpu", 0, - &ip_vs_stats_percpu_fops); if (!net_eq(net, &init_net)) { tbl = kmemdup(vs_vars, sizeof(vs_vars), GFP_KERNEL); if (tbl == NULL) - goto err_dup; + return -ENOMEM; } else tbl = vs_vars; /* Initialize sysctl defaults */ @@ -3576,25 +3558,73 @@ int __net_init __ip_vs_control_init(struct net *net) tbl[idx++].data = &ipvs->sysctl_nat_icmp_send; -#ifdef CONFIG_SYSCTL ipvs->sysctl_hdr = register_net_sysctl_table(net, net_vs_ctl_path, tbl); if (ipvs->sysctl_hdr == NULL) { if (!net_eq(net, &init_net)) kfree(tbl); - goto err_dup; + return -ENOMEM; } -#endif ip_vs_start_estimator(net, &ipvs->tot_stats); ipvs->sysctl_tbl = tbl; /* Schedule defense work */ INIT_DELAYED_WORK(&ipvs->defense_work, defense_work_handler); schedule_delayed_work(&ipvs->defense_work, DEFENSE_TIMER_PERIOD); + return 0; +} + +void __net_init __ip_vs_control_cleanup_sysctl(struct net *net) +{ + struct netns_ipvs *ipvs = net_ipvs(net); + + cancel_delayed_work_sync(&ipvs->defense_work); + cancel_work_sync(&ipvs->defense_work.work); + unregister_net_sysctl_table(ipvs->sysctl_hdr); +} -err_dup: +#else + +int __net_init __ip_vs_control_init_sysctl(struct net *net) { return 0; } +void __net_init __ip_vs_control_cleanup_sysctl(struct net *net) { } + +#endif + +int __net_init __ip_vs_control_init(struct net *net) +{ + int idx; + struct netns_ipvs *ipvs = net_ipvs(net); + + ipvs->rs_lock = __RW_LOCK_UNLOCKED(ipvs->rs_lock); + + /* Initialize rs_table */ + for (idx = 0; idx < IP_VS_RTAB_SIZE; idx++) + INIT_LIST_HEAD(&ipvs->rs_table[idx]); + + INIT_LIST_HEAD(&ipvs->dest_trash); + atomic_set(&ipvs->ftpsvc_counter, 0); + atomic_set(&ipvs->nullsvc_counter, 0); + + /* procfs stats */ + ipvs->tot_stats.cpustats = alloc_percpu(struct ip_vs_cpu_stats); + if (ipvs->tot_stats.cpustats) { + pr_err("%s(): alloc_percpu.\n", __func__); + return -ENOMEM; + } + spin_lock_init(&ipvs->tot_stats.lock); + + proc_net_fops_create(net, "ip_vs", 0, &ip_vs_info_fops); + proc_net_fops_create(net, "ip_vs_stats", 0, &ip_vs_stats_fops); + proc_net_fops_create(net, "ip_vs_stats_percpu", 0, + &ip_vs_stats_percpu_fops); + + if (__ip_vs_control_init_sysctl(net)) + goto err; + + return 0; + +err: free_percpu(ipvs->tot_stats.cpustats); -err_alloc: return -ENOMEM; } @@ -3604,11 +3634,7 @@ static void __net_exit __ip_vs_control_cleanup(struct net *net) ip_vs_trash_cleanup(net); ip_vs_stop_estimator(net, &ipvs->tot_stats); - cancel_delayed_work_sync(&ipvs->defense_work); - cancel_work_sync(&ipvs->defense_work.work); -#ifdef CONFIG_SYSCTL - unregister_net_sysctl_table(ipvs->sysctl_hdr); -#endif + __ip_vs_control_cleanup_sysctl(net); proc_net_remove(net, "ip_vs_stats_percpu"); proc_net_remove(net, "ip_vs_stats"); proc_net_remove(net, "ip_vs"); -- cgit v1.2.3 From 8183e3a88aced228ab9770762692be6cc3786e80 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Tue, 15 Mar 2011 13:23:28 +0100 Subject: netfilter: xt_connlimit: fix daddr connlimit in SNAT scenario We use the reply tuples when limiting the connections by the destination addresses, however, in SNAT scenario, the final reply tuples won't be ready until SNAT is done in POSTROUING or INPUT chain, and the following nf_conntrack_find_get() in count_tem() will get nothing, so connlimit can't work as expected. In this patch, the original tuples are always used, and an additional member addr is appended to save the address in either end. Signed-off-by: Changli Gao Signed-off-by: Patrick McHardy --- net/netfilter/xt_connlimit.c | 20 +++++++++----------- 1 file changed, 9 insertions(+), 11 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c index e029c4807404..1f4b9f9da496 100644 --- a/net/netfilter/xt_connlimit.c +++ b/net/netfilter/xt_connlimit.c @@ -33,8 +33,9 @@ /* we will save the tuples of all connections we care about */ struct xt_connlimit_conn { - struct list_head list; - struct nf_conntrack_tuple tuple; + struct list_head list; + struct nf_conntrack_tuple tuple; + union nf_inet_addr addr; }; struct xt_connlimit_data { @@ -151,7 +152,7 @@ static int count_them(struct net *net, continue; } - if (same_source_net(addr, mask, &conn->tuple.src.u3, family)) + if (same_source_net(addr, mask, &conn->addr, family)) /* same source network -> be counted! */ ++matches; nf_ct_put(found_ct); @@ -165,6 +166,7 @@ static int count_them(struct net *net, if (conn == NULL) return -ENOMEM; conn->tuple = *tuple; + conn->addr = *addr; list_add(&conn->list, hash); ++matches; } @@ -185,15 +187,11 @@ connlimit_mt(const struct sk_buff *skb, struct xt_action_param *par) int connections; ct = nf_ct_get(skb, &ctinfo); - if (ct != NULL) { - if (info->flags & XT_CONNLIMIT_DADDR) - tuple_ptr = &ct->tuplehash[IP_CT_DIR_REPLY].tuple; - else - tuple_ptr = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; - } else if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), - par->family, &tuple)) { + if (ct != NULL) + tuple_ptr = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; + else if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb), + par->family, &tuple)) goto hotdrop; - } if (par->family == NFPROTO_IPV6) { const struct ipv6hdr *iph = ipv6_hdr(skb); -- cgit v1.2.3 From 0e23ca14f8e76091b402c01e2b169aba3d187b98 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Tue, 15 Mar 2011 13:24:56 +0100 Subject: netfilter: xt_connlimit: use kmalloc() instead of kzalloc() All the members are initialized after kzalloc(). Signed-off-by: Changli Gao Signed-off-by: Patrick McHardy --- net/netfilter/xt_connlimit.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c index 1f4b9f9da496..ade2a806a48e 100644 --- a/net/netfilter/xt_connlimit.c +++ b/net/netfilter/xt_connlimit.c @@ -162,7 +162,7 @@ static int count_them(struct net *net, if (addit) { /* save the new connection in our list */ - conn = kzalloc(sizeof(*conn), GFP_ATOMIC); + conn = kmalloc(sizeof(*conn), GFP_ATOMIC); if (conn == NULL) return -ENOMEM; conn->tuple = *tuple; -- cgit v1.2.3 From 3e0d5149e6dcbe7111a63773a07c5b33f7ca7236 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Tue, 15 Mar 2011 13:25:42 +0100 Subject: netfilter: xt_connlimit: use hlist instead The header of hlist is smaller than list. Signed-off-by: Changli Gao Signed-off-by: Patrick McHardy --- net/netfilter/xt_connlimit.c | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c index ade2a806a48e..da56d6ec13c9 100644 --- a/net/netfilter/xt_connlimit.c +++ b/net/netfilter/xt_connlimit.c @@ -33,14 +33,14 @@ /* we will save the tuples of all connections we care about */ struct xt_connlimit_conn { - struct list_head list; + struct hlist_node node; struct nf_conntrack_tuple tuple; union nf_inet_addr addr; }; struct xt_connlimit_data { - struct list_head iphash[256]; - spinlock_t lock; + struct hlist_head iphash[256]; + spinlock_t lock; }; static u_int32_t connlimit_rnd __read_mostly; @@ -102,9 +102,9 @@ static int count_them(struct net *net, { const struct nf_conntrack_tuple_hash *found; struct xt_connlimit_conn *conn; - struct xt_connlimit_conn *tmp; + struct hlist_node *pos, *n; struct nf_conn *found_ct; - struct list_head *hash; + struct hlist_head *hash; bool addit = true; int matches = 0; @@ -116,7 +116,7 @@ static int count_them(struct net *net, rcu_read_lock(); /* check the saved connections */ - list_for_each_entry_safe(conn, tmp, hash, list) { + hlist_for_each_entry_safe(conn, pos, n, hash, node) { found = nf_conntrack_find_get(net, NF_CT_DEFAULT_ZONE, &conn->tuple); found_ct = NULL; @@ -136,7 +136,7 @@ static int count_them(struct net *net, if (found == NULL) { /* this one is gone */ - list_del(&conn->list); + hlist_del(&conn->node); kfree(conn); continue; } @@ -147,7 +147,7 @@ static int count_them(struct net *net, * closed already -> ditch it */ nf_ct_put(found_ct); - list_del(&conn->list); + hlist_del(&conn->node); kfree(conn); continue; } @@ -167,7 +167,7 @@ static int count_them(struct net *net, return -ENOMEM; conn->tuple = *tuple; conn->addr = *addr; - list_add(&conn->list, hash); + hlist_add_head(&conn->node, hash); ++matches; } @@ -246,7 +246,7 @@ static int connlimit_mt_check(const struct xt_mtchk_param *par) spin_lock_init(&info->data->lock); for (i = 0; i < ARRAY_SIZE(info->data->iphash); ++i) - INIT_LIST_HEAD(&info->data->iphash[i]); + INIT_HLIST_HEAD(&info->data->iphash[i]); return 0; } @@ -255,15 +255,15 @@ static void connlimit_mt_destroy(const struct xt_mtdtor_param *par) { const struct xt_connlimit_info *info = par->matchinfo; struct xt_connlimit_conn *conn; - struct xt_connlimit_conn *tmp; - struct list_head *hash = info->data->iphash; + struct hlist_node *pos, *n; + struct hlist_head *hash = info->data->iphash; unsigned int i; nf_ct_l3proto_module_put(par->family); for (i = 0; i < ARRAY_SIZE(info->data->iphash); ++i) { - list_for_each_entry_safe(conn, tmp, &hash[i], list) { - list_del(&conn->list); + hlist_for_each_entry_safe(conn, pos, n, &hash[i], node) { + hlist_del(&conn->node); kfree(conn); } } -- cgit v1.2.3 From 4656c4d61adb8dc3ee04c08f57a5cc7598814420 Mon Sep 17 00:00:00 2001 From: Changli Gao Date: Tue, 15 Mar 2011 13:26:32 +0100 Subject: netfilter: xt_connlimit: remove connlimit_rnd_inited A potential race condition when generating connlimit_rnd is also fixed. Signed-off-by: Changli Gao Signed-off-by: Patrick McHardy --- net/netfilter/xt_connlimit.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c index da56d6ec13c9..c6d5a83450c9 100644 --- a/net/netfilter/xt_connlimit.c +++ b/net/netfilter/xt_connlimit.c @@ -44,7 +44,6 @@ struct xt_connlimit_data { }; static u_int32_t connlimit_rnd __read_mostly; -static bool connlimit_rnd_inited __read_mostly; static inline unsigned int connlimit_iphash(__be32 addr) { @@ -226,9 +225,13 @@ static int connlimit_mt_check(const struct xt_mtchk_param *par) unsigned int i; int ret; - if (unlikely(!connlimit_rnd_inited)) { - get_random_bytes(&connlimit_rnd, sizeof(connlimit_rnd)); - connlimit_rnd_inited = true; + if (unlikely(!connlimit_rnd)) { + u_int32_t rand; + + do { + get_random_bytes(&rand, sizeof(rand)); + } while (!rand); + cmpxchg(&connlimit_rnd, 0, rand); } ret = nf_ct_l3proto_try_module_get(par->family); if (ret < 0) { -- cgit v1.2.3 From de81bbea17650769882bc625d6b5df11ee7c4b24 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 15 Mar 2011 20:16:20 +0100 Subject: netfilter: ipt_addrtype: rename to xt_addrtype Followup patch will add ipv6 support. ipt_addrtype.h is retained for compatibility reasons, but no longer used by the kernel. Signed-off-by: Florian Westphal Signed-off-by: Patrick McHardy --- net/netfilter/Kconfig | 10 ++++ net/netfilter/Makefile | 1 + net/netfilter/xt_addrtype.c | 135 ++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 146 insertions(+) create mode 100644 net/netfilter/xt_addrtype.c (limited to 'net/netfilter') diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 82a6e0d80f05..32bff6d86cb2 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -649,6 +649,16 @@ config NETFILTER_XT_TARGET_TCPOPTSTRIP comment "Xtables matches" +config NETFILTER_XT_MATCH_ADDRTYPE + tristate '"addrtype" address type match support' + depends on NETFILTER_ADVANCED + ---help--- + This option allows you to match what routing thinks of an address, + eg. UNICAST, LOCAL, BROADCAST, ... + + If you want to compile it as a module, say M here and read + . If unsure, say `N'. + config NETFILTER_XT_MATCH_CLUSTER tristate '"cluster" match support' depends on NF_CONNTRACK diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index d57a890eaee5..1a02853df863 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -70,6 +70,7 @@ obj-$(CONFIG_NETFILTER_XT_TARGET_TRACE) += xt_TRACE.o obj-$(CONFIG_NETFILTER_XT_TARGET_IDLETIMER) += xt_IDLETIMER.o # matches +obj-$(CONFIG_NETFILTER_XT_MATCH_ADDRTYPE) += xt_addrtype.o obj-$(CONFIG_NETFILTER_XT_MATCH_CLUSTER) += xt_cluster.o obj-$(CONFIG_NETFILTER_XT_MATCH_COMMENT) += xt_comment.o obj-$(CONFIG_NETFILTER_XT_MATCH_CONNBYTES) += xt_connbytes.o diff --git a/net/netfilter/xt_addrtype.c b/net/netfilter/xt_addrtype.c new file mode 100644 index 000000000000..e89c0b84583c --- /dev/null +++ b/net/netfilter/xt_addrtype.c @@ -0,0 +1,135 @@ +/* + * iptables module to match inet_addr_type() of an ip. + * + * Copyright (c) 2004 Patrick McHardy + * (C) 2007 Laszlo Attila Toth + * + * 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. + */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt +#include +#include +#include +#include +#include +#include + +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Patrick McHardy "); +MODULE_DESCRIPTION("Xtables: address type match"); +MODULE_ALIAS("ipt_addrtype"); + +static inline bool match_type(struct net *net, const struct net_device *dev, + __be32 addr, u_int16_t mask) +{ + return !!(mask & (1 << inet_dev_addr_type(net, dev, addr))); +} + +static bool +addrtype_mt_v0(const struct sk_buff *skb, struct xt_action_param *par) +{ + struct net *net = dev_net(par->in ? par->in : par->out); + const struct xt_addrtype_info *info = par->matchinfo; + const struct iphdr *iph = ip_hdr(skb); + bool ret = true; + + if (info->source) + ret &= match_type(net, NULL, iph->saddr, info->source) ^ + info->invert_source; + if (info->dest) + ret &= match_type(net, NULL, iph->daddr, info->dest) ^ + info->invert_dest; + + return ret; +} + +static bool +addrtype_mt_v1(const struct sk_buff *skb, struct xt_action_param *par) +{ + struct net *net = dev_net(par->in ? par->in : par->out); + const struct xt_addrtype_info_v1 *info = par->matchinfo; + const struct iphdr *iph = ip_hdr(skb); + const struct net_device *dev = NULL; + bool ret = true; + + if (info->flags & XT_ADDRTYPE_LIMIT_IFACE_IN) + dev = par->in; + else if (info->flags & XT_ADDRTYPE_LIMIT_IFACE_OUT) + dev = par->out; + + if (info->source) + ret &= match_type(net, dev, iph->saddr, info->source) ^ + (info->flags & XT_ADDRTYPE_INVERT_SOURCE); + if (ret && info->dest) + ret &= match_type(net, dev, iph->daddr, info->dest) ^ + !!(info->flags & XT_ADDRTYPE_INVERT_DEST); + return ret; +} + +static int addrtype_mt_checkentry_v1(const struct xt_mtchk_param *par) +{ + struct xt_addrtype_info_v1 *info = par->matchinfo; + + if (info->flags & XT_ADDRTYPE_LIMIT_IFACE_IN && + info->flags & XT_ADDRTYPE_LIMIT_IFACE_OUT) { + pr_info("both incoming and outgoing " + "interface limitation cannot be selected\n"); + return -EINVAL; + } + + if (par->hook_mask & ((1 << NF_INET_PRE_ROUTING) | + (1 << NF_INET_LOCAL_IN)) && + info->flags & XT_ADDRTYPE_LIMIT_IFACE_OUT) { + pr_info("output interface limitation " + "not valid in PREROUTING and INPUT\n"); + return -EINVAL; + } + + if (par->hook_mask & ((1 << NF_INET_POST_ROUTING) | + (1 << NF_INET_LOCAL_OUT)) && + info->flags & XT_ADDRTYPE_LIMIT_IFACE_IN) { + pr_info("input interface limitation " + "not valid in POSTROUTING and OUTPUT\n"); + return -EINVAL; + } + + return 0; +} + +static struct xt_match addrtype_mt_reg[] __read_mostly = { + { + .name = "addrtype", + .family = NFPROTO_IPV4, + .match = addrtype_mt_v0, + .matchsize = sizeof(struct xt_addrtype_info), + .me = THIS_MODULE + }, + { + .name = "addrtype", + .family = NFPROTO_IPV4, + .revision = 1, + .match = addrtype_mt_v1, + .checkentry = addrtype_mt_checkentry_v1, + .matchsize = sizeof(struct xt_addrtype_info_v1), + .me = THIS_MODULE + } +}; + +static int __init addrtype_mt_init(void) +{ + return xt_register_matches(addrtype_mt_reg, + ARRAY_SIZE(addrtype_mt_reg)); +} + +static void __exit addrtype_mt_exit(void) +{ + xt_unregister_matches(addrtype_mt_reg, ARRAY_SIZE(addrtype_mt_reg)); +} + +module_init(addrtype_mt_init); +module_exit(addrtype_mt_exit); -- cgit v1.2.3 From 2f5dc63123905a89d4260ab8ee08d19ec104db04 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 15 Mar 2011 20:17:44 +0100 Subject: netfilter: xt_addrtype: ipv6 support The kernel will refuse certain types that do not work in ipv6 mode. We can then add these features incrementally without risk of userspace breakage. Signed-off-by: Florian Westphal Signed-off-by: Patrick McHardy --- net/netfilter/Kconfig | 1 + net/netfilter/xt_addrtype.c | 98 ++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 97 insertions(+), 2 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 32bff6d86cb2..c3f988aa1152 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -652,6 +652,7 @@ comment "Xtables matches" config NETFILTER_XT_MATCH_ADDRTYPE tristate '"addrtype" address type match support' depends on NETFILTER_ADVANCED + depends on (IPV6 || IPV6=n) ---help--- This option allows you to match what routing thinks of an address, eg. UNICAST, LOCAL, BROADCAST, ... diff --git a/net/netfilter/xt_addrtype.c b/net/netfilter/xt_addrtype.c index e89c0b84583c..2220b85e9519 100644 --- a/net/netfilter/xt_addrtype.c +++ b/net/netfilter/xt_addrtype.c @@ -16,6 +16,12 @@ #include #include +#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE) +#include +#include +#include +#endif + #include #include @@ -23,6 +29,73 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Patrick McHardy "); MODULE_DESCRIPTION("Xtables: address type match"); MODULE_ALIAS("ipt_addrtype"); +MODULE_ALIAS("ip6t_addrtype"); + +#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE) +static u32 xt_addrtype_rt6_to_type(const struct rt6_info *rt) +{ + u32 ret; + + if (!rt) + return XT_ADDRTYPE_UNREACHABLE; + + if (rt->rt6i_flags & RTF_REJECT) + ret = XT_ADDRTYPE_UNREACHABLE; + else + ret = 0; + + if (rt->rt6i_flags & RTF_LOCAL) + ret |= XT_ADDRTYPE_LOCAL; + if (rt->rt6i_flags & RTF_ANYCAST) + ret |= XT_ADDRTYPE_ANYCAST; + return ret; +} + +static bool match_type6(struct net *net, const struct net_device *dev, + const struct in6_addr *addr, u16 mask) +{ + int addr_type = ipv6_addr_type(addr); + + if ((mask & XT_ADDRTYPE_MULTICAST) && + !(addr_type & IPV6_ADDR_MULTICAST)) + return false; + if ((mask & XT_ADDRTYPE_UNICAST) && !(addr_type & IPV6_ADDR_UNICAST)) + return false; + if ((mask & XT_ADDRTYPE_UNSPEC) && addr_type != IPV6_ADDR_ANY) + return false; + + if ((XT_ADDRTYPE_LOCAL | XT_ADDRTYPE_ANYCAST | + XT_ADDRTYPE_UNREACHABLE) & mask) { + struct rt6_info *rt; + u32 type; + int ifindex = dev ? dev->ifindex : 0; + + rt = rt6_lookup(net, addr, NULL, ifindex, !!dev); + + type = xt_addrtype_rt6_to_type(rt); + + dst_release(&rt->dst); + return !!(mask & type); + } + return true; +} + +static bool +addrtype_mt6(struct net *net, const struct net_device *dev, + const struct sk_buff *skb, const struct xt_addrtype_info_v1 *info) +{ + const struct ipv6hdr *iph = ipv6_hdr(skb); + bool ret = true; + + if (info->source) + ret &= match_type6(net, dev, &iph->saddr, info->source) ^ + (info->flags & XT_ADDRTYPE_INVERT_SOURCE); + if (ret && info->dest) + ret &= match_type6(net, dev, &iph->daddr, info->dest) ^ + !!(info->flags & XT_ADDRTYPE_INVERT_DEST); + return ret; +} +#endif static inline bool match_type(struct net *net, const struct net_device *dev, __be32 addr, u_int16_t mask) @@ -53,7 +126,7 @@ addrtype_mt_v1(const struct sk_buff *skb, struct xt_action_param *par) { struct net *net = dev_net(par->in ? par->in : par->out); const struct xt_addrtype_info_v1 *info = par->matchinfo; - const struct iphdr *iph = ip_hdr(skb); + const struct iphdr *iph; const struct net_device *dev = NULL; bool ret = true; @@ -62,6 +135,11 @@ addrtype_mt_v1(const struct sk_buff *skb, struct xt_action_param *par) else if (info->flags & XT_ADDRTYPE_LIMIT_IFACE_OUT) dev = par->out; +#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE) + if (par->family == NFPROTO_IPV6) + return addrtype_mt6(net, dev, skb, info); +#endif + iph = ip_hdr(skb); if (info->source) ret &= match_type(net, dev, iph->saddr, info->source) ^ (info->flags & XT_ADDRTYPE_INVERT_SOURCE); @@ -98,6 +176,22 @@ static int addrtype_mt_checkentry_v1(const struct xt_mtchk_param *par) return -EINVAL; } +#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE) + if (par->family == NFPROTO_IPV6) { + if ((info->source | info->dest) & XT_ADDRTYPE_BLACKHOLE) { + pr_err("ipv6 BLACKHOLE matching not supported\n"); + return -EINVAL; + } + if ((info->source | info->dest) >= XT_ADDRTYPE_PROHIBIT) { + pr_err("ipv6 PROHIBT (THROW, NAT ..) matching not supported\n"); + return -EINVAL; + } + if ((info->source | info->dest) & XT_ADDRTYPE_BROADCAST) { + pr_err("ipv6 does not support BROADCAST matching\n"); + return -EINVAL; + } + } +#endif return 0; } @@ -111,7 +205,7 @@ static struct xt_match addrtype_mt_reg[] __read_mostly = { }, { .name = "addrtype", - .family = NFPROTO_IPV4, + .family = NFPROTO_UNSPEC, .revision = 1, .match = addrtype_mt_v1, .checkentry = addrtype_mt_checkentry_v1, -- cgit v1.2.3 From 400b871ba623b5e8263a3a43de7b45fab0103a57 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Wed, 16 Mar 2011 18:32:13 +0100 Subject: netfilter ebtables: fix xt_AUDIT to work with ebtables Even though ebtables uses xtables it still requires targets to return EBT_CONTINUE instead of XT_CONTINUE. This prevented xt_AUDIT to work as ebt module. Upon Jan's suggestion, use a separate struct xt_target for NFPROTO_BRIDGE having its own target callback returning EBT_CONTINUE instead of cloning the module. Signed-off-by: Thomas Graf Signed-off-by: Patrick McHardy --- net/netfilter/xt_AUDIT.c | 36 +++++++++++++++++++++++++++--------- 1 file changed, 27 insertions(+), 9 deletions(-) (limited to 'net/netfilter') diff --git a/net/netfilter/xt_AUDIT.c b/net/netfilter/xt_AUDIT.c index 81802d27346e..363a99ec0637 100644 --- a/net/netfilter/xt_AUDIT.c +++ b/net/netfilter/xt_AUDIT.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -168,6 +169,13 @@ errout: return XT_CONTINUE; } +static unsigned int +audit_tg_ebt(struct sk_buff *skb, const struct xt_action_param *par) +{ + audit_tg(skb, par); + return EBT_CONTINUE; +} + static int audit_tg_check(const struct xt_tgchk_param *par) { const struct xt_audit_info *info = par->targinfo; @@ -181,23 +189,33 @@ static int audit_tg_check(const struct xt_tgchk_param *par) return 0; } -static struct xt_target audit_tg_reg __read_mostly = { - .name = "AUDIT", - .family = NFPROTO_UNSPEC, - .target = audit_tg, - .targetsize = sizeof(struct xt_audit_info), - .checkentry = audit_tg_check, - .me = THIS_MODULE, +static struct xt_target audit_tg_reg[] __read_mostly = { + { + .name = "AUDIT", + .family = NFPROTO_UNSPEC, + .target = audit_tg, + .targetsize = sizeof(struct xt_audit_info), + .checkentry = audit_tg_check, + .me = THIS_MODULE, + }, + { + .name = "AUDIT", + .family = NFPROTO_BRIDGE, + .target = audit_tg_ebt, + .targetsize = sizeof(struct xt_audit_info), + .checkentry = audit_tg_check, + .me = THIS_MODULE, + }, }; static int __init audit_tg_init(void) { - return xt_register_target(&audit_tg_reg); + return xt_register_targets(audit_tg_reg, ARRAY_SIZE(audit_tg_reg)); } static void __exit audit_tg_exit(void) { - xt_unregister_target(&audit_tg_reg); + xt_unregister_targets(audit_tg_reg, ARRAY_SIZE(audit_tg_reg)); } module_init(audit_tg_init); -- cgit v1.2.3