X-Git-Url: http://drtracing.org/?a=blobdiff_plain;f=net%2Fnetfilter%2Fnf_conntrack_netlink.c;h=dd233393f6952203c2408e5d904e490d784a940f;hb=b891c5a831b13f74989dcbd7b39d04537b2a05d9;hp=94027c84be5259ee7ac882934715a5493e38c23a;hpb=0f417ce989f84cfd5418e3b316064bfbb2708196;p=deliverable%2Flinux.git diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 94027c84be52..dd233393f695 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -4,7 +4,7 @@ * (C) 2001 by Jay Schulist * (C) 2002-2006 by Harald Welte * (C) 2003 by Patrick Mchardy - * (C) 2005-2007 by Pablo Neira Ayuso + * (C) 2005-2008 by Pablo Neira Ayuso * * Initial connection tracking via netlink development funded and * generally made possible by Network Robots, Inc. (www.networkrobots.com) @@ -59,7 +59,7 @@ ctnetlink_dump_tuples_proto(struct sk_buff *skb, nest_parms = nla_nest_start(skb, CTA_TUPLE_PROTO | NLA_F_NESTED); if (!nest_parms) goto nla_put_failure; - NLA_PUT(skb, CTA_PROTO_NUM, sizeof(u_int8_t), &tuple->dst.protonum); + NLA_PUT_U8(skb, CTA_PROTO_NUM, tuple->dst.protonum); if (likely(l4proto->tuple_to_nlattr)) ret = l4proto->tuple_to_nlattr(skb, tuple); @@ -95,7 +95,7 @@ nla_put_failure: return -1; } -static inline int +static int ctnetlink_dump_tuples(struct sk_buff *skb, const struct nf_conntrack_tuple *tuple) { @@ -120,8 +120,7 @@ ctnetlink_dump_tuples(struct sk_buff *skb, static inline int ctnetlink_dump_status(struct sk_buff *skb, const struct nf_conn *ct) { - __be32 status = htonl((u_int32_t) ct->status); - NLA_PUT(skb, CTA_STATUS, sizeof(status), &status); + NLA_PUT_BE32(skb, CTA_STATUS, htonl(ct->status)); return 0; nla_put_failure: @@ -131,15 +130,12 @@ nla_put_failure: static inline int ctnetlink_dump_timeout(struct sk_buff *skb, const struct nf_conn *ct) { - long timeout_l = ct->timeout.expires - jiffies; - __be32 timeout; + long timeout = (ct->timeout.expires - jiffies) / HZ; - if (timeout_l < 0) + if (timeout < 0) timeout = 0; - else - timeout = htonl(timeout_l / HZ); - NLA_PUT(skb, CTA_TIMEOUT, sizeof(timeout), &timeout); + NLA_PUT_BE32(skb, CTA_TIMEOUT, htonl(timeout)); return 0; nla_put_failure: @@ -149,10 +145,11 @@ nla_put_failure: static inline int ctnetlink_dump_protoinfo(struct sk_buff *skb, const struct nf_conn *ct) { - struct nf_conntrack_l4proto *l4proto = nf_ct_l4proto_find_get(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num, ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum); + struct nf_conntrack_l4proto *l4proto; struct nlattr *nest_proto; int ret; + l4proto = nf_ct_l4proto_find_get(nf_ct_l3num(ct), nf_ct_protonum(ct)); if (!l4proto->to_nlattr) { nf_ct_l4proto_put(l4proto); return 0; @@ -193,7 +190,7 @@ ctnetlink_dump_helpinfo(struct sk_buff *skb, const struct nf_conn *ct) nest_helper = nla_nest_start(skb, CTA_HELP | NLA_F_NESTED); if (!nest_helper) goto nla_put_failure; - NLA_PUT(skb, CTA_HELP_NAME, strlen(helper->name), helper->name); + NLA_PUT_STRING(skb, CTA_HELP_NAME, helper->name); if (helper->to_nlattr) helper->to_nlattr(skb, ct); @@ -209,23 +206,21 @@ nla_put_failure: } #ifdef CONFIG_NF_CT_ACCT -static inline int +static int ctnetlink_dump_counters(struct sk_buff *skb, const struct nf_conn *ct, enum ip_conntrack_dir dir) { enum ctattr_type type = dir ? CTA_COUNTERS_REPLY: CTA_COUNTERS_ORIG; struct nlattr *nest_count; - __be32 tmp; nest_count = nla_nest_start(skb, type | NLA_F_NESTED); if (!nest_count) goto nla_put_failure; - tmp = htonl(ct->counters[dir].packets); - NLA_PUT(skb, CTA_COUNTERS32_PACKETS, sizeof(u_int32_t), &tmp); - - tmp = htonl(ct->counters[dir].bytes); - NLA_PUT(skb, CTA_COUNTERS32_BYTES, sizeof(u_int32_t), &tmp); + NLA_PUT_BE32(skb, CTA_COUNTERS32_PACKETS, + htonl(ct->counters[dir].packets)); + NLA_PUT_BE32(skb, CTA_COUNTERS32_BYTES, + htonl(ct->counters[dir].bytes)); nla_nest_end(skb, nest_count); @@ -242,9 +237,7 @@ nla_put_failure: static inline int ctnetlink_dump_mark(struct sk_buff *skb, const struct nf_conn *ct) { - __be32 mark = htonl(ct->mark); - - NLA_PUT(skb, CTA_MARK, sizeof(u_int32_t), &mark); + NLA_PUT_BE32(skb, CTA_MARK, htonl(ct->mark)); return 0; nla_put_failure: @@ -254,6 +247,20 @@ nla_put_failure: #define ctnetlink_dump_mark(a, b) (0) #endif +#ifdef CONFIG_NF_CONNTRACK_SECMARK +static inline int +ctnetlink_dump_secmark(struct sk_buff *skb, const struct nf_conn *ct) +{ + NLA_PUT_BE32(skb, CTA_SECMARK, htonl(ct->secmark)); + return 0; + +nla_put_failure: + return -1; +} +#else +#define ctnetlink_dump_secmark(a, b) (0) +#endif + #define master_tuple(ct) &(ct->master->tuplehash[IP_CT_DIR_ORIGINAL].tuple) static inline int @@ -278,22 +285,21 @@ nla_put_failure: } #ifdef CONFIG_NF_NAT_NEEDED -static inline int +static int dump_nat_seq_adj(struct sk_buff *skb, const struct nf_nat_seq *natseq, int type) { - __be32 tmp; struct nlattr *nest_parms; nest_parms = nla_nest_start(skb, type | NLA_F_NESTED); if (!nest_parms) goto nla_put_failure; - tmp = htonl(natseq->correction_pos); - NLA_PUT(skb, CTA_NAT_SEQ_CORRECTION_POS, sizeof(tmp), &tmp); - tmp = htonl(natseq->offset_before); - NLA_PUT(skb, CTA_NAT_SEQ_OFFSET_BEFORE, sizeof(tmp), &tmp); - tmp = htonl(natseq->offset_after); - NLA_PUT(skb, CTA_NAT_SEQ_OFFSET_AFTER, sizeof(tmp), &tmp); + NLA_PUT_BE32(skb, CTA_NAT_SEQ_CORRECTION_POS, + htonl(natseq->correction_pos)); + NLA_PUT_BE32(skb, CTA_NAT_SEQ_OFFSET_BEFORE, + htonl(natseq->offset_before)); + NLA_PUT_BE32(skb, CTA_NAT_SEQ_OFFSET_AFTER, + htonl(natseq->offset_after)); nla_nest_end(skb, nest_parms); @@ -329,8 +335,7 @@ ctnetlink_dump_nat_seq_adj(struct sk_buff *skb, const struct nf_conn *ct) static inline int ctnetlink_dump_id(struct sk_buff *skb, const struct nf_conn *ct) { - __be32 id = htonl((unsigned long)ct); - NLA_PUT(skb, CTA_ID, sizeof(u_int32_t), &id); + NLA_PUT_BE32(skb, CTA_ID, htonl((unsigned long)ct)); return 0; nla_put_failure: @@ -340,9 +345,7 @@ nla_put_failure: static inline int ctnetlink_dump_use(struct sk_buff *skb, const struct nf_conn *ct) { - __be32 use = htonl(atomic_read(&ct->ct_general.use)); - - NLA_PUT(skb, CTA_USE, sizeof(u_int32_t), &use); + NLA_PUT_BE32(skb, CTA_USE, htonl(atomic_read(&ct->ct_general.use))); return 0; nla_put_failure: @@ -366,8 +369,7 @@ ctnetlink_fill_info(struct sk_buff *skb, u32 pid, u32 seq, nfmsg = NLMSG_DATA(nlh); nlh->nlmsg_flags = (nowait && pid) ? NLM_F_MULTI : 0; - nfmsg->nfgen_family = - ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + nfmsg->nfgen_family = nf_ct_l3num(ct); nfmsg->version = NFNETLINK_V0; nfmsg->res_id = 0; @@ -392,6 +394,7 @@ ctnetlink_fill_info(struct sk_buff *skb, u32 pid, u32 seq, ctnetlink_dump_protoinfo(skb, ct) < 0 || ctnetlink_dump_helpinfo(skb, ct) < 0 || ctnetlink_dump_mark(skb, ct) < 0 || + ctnetlink_dump_secmark(skb, ct) < 0 || ctnetlink_dump_id(skb, ct) < 0 || ctnetlink_dump_use(skb, ct) < 0 || ctnetlink_dump_master(skb, ct) < 0 || @@ -451,7 +454,7 @@ static int ctnetlink_conntrack_event(struct notifier_block *this, nfmsg = NLMSG_DATA(nlh); nlh->nlmsg_flags = flags; - nfmsg->nfgen_family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + nfmsg->nfgen_family = nf_ct_l3num(ct); nfmsg->version = NFNETLINK_V0; nfmsg->res_id = 0; @@ -469,14 +472,17 @@ static int ctnetlink_conntrack_event(struct notifier_block *this, goto nla_put_failure; nla_nest_end(skb, nest_parms); + if (ctnetlink_dump_id(skb, ct) < 0) + goto nla_put_failure; + + if (ctnetlink_dump_status(skb, ct) < 0) + goto nla_put_failure; + if (events & IPCT_DESTROY) { if (ctnetlink_dump_counters(skb, ct, IP_CT_DIR_ORIGINAL) < 0 || ctnetlink_dump_counters(skb, ct, IP_CT_DIR_REPLY) < 0) goto nla_put_failure; } else { - if (ctnetlink_dump_status(skb, ct) < 0) - goto nla_put_failure; - if (ctnetlink_dump_timeout(skb, ct) < 0) goto nla_put_failure; @@ -488,9 +494,9 @@ static int ctnetlink_conntrack_event(struct notifier_block *this, && ctnetlink_dump_helpinfo(skb, ct) < 0) goto nla_put_failure; -#ifdef CONFIG_NF_CONNTRACK_MARK - if ((events & IPCT_MARK || ct->mark) - && ctnetlink_dump_mark(skb, ct) < 0) +#ifdef CONFIG_NF_CONNTRACK_SECMARK + if ((events & IPCT_SECMARK || ct->secmark) + && ctnetlink_dump_secmark(skb, ct) < 0) goto nla_put_failure; #endif @@ -508,6 +514,12 @@ static int ctnetlink_conntrack_event(struct notifier_block *this, goto nla_put_failure; } +#ifdef CONFIG_NF_CONNTRACK_MARK + if ((events & IPCT_MARK || ct->mark) + && ctnetlink_dump_mark(skb, ct) < 0) + goto nla_put_failure; +#endif + nlh->nlmsg_len = skb->tail - b; nfnetlink_send(skb, 0, group, 0); return NOTIFY_DONE; @@ -526,8 +538,6 @@ static int ctnetlink_done(struct netlink_callback *cb) return 0; } -#define L3PROTO(ct) ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num - static int ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb) { @@ -537,19 +547,19 @@ ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb) struct nfgenmsg *nfmsg = NLMSG_DATA(cb->nlh); u_int8_t l3proto = nfmsg->nfgen_family; - read_lock_bh(&nf_conntrack_lock); + rcu_read_lock(); last = (struct nf_conn *)cb->args[1]; for (; cb->args[0] < nf_conntrack_htable_size; cb->args[0]++) { restart: - hlist_for_each_entry(h, n, &nf_conntrack_hash[cb->args[0]], - hnode) { + hlist_for_each_entry_rcu(h, n, &nf_conntrack_hash[cb->args[0]], + hnode) { if (NF_CT_DIRECTION(h) != IP_CT_DIR_ORIGINAL) continue; ct = nf_ct_tuplehash_to_ctrack(h); /* Dump entries of a given L3 protocol number. * If it is not specified, ie. l3proto == 0, * then dump everything. */ - if (l3proto && L3PROTO(ct) != l3proto) + if (l3proto && nf_ct_l3num(ct) != l3proto) continue; if (cb->args[1]) { if (ct != last) @@ -560,7 +570,8 @@ restart: cb->nlh->nlmsg_seq, IPCTNL_MSG_CT_NEW, 1, ct) < 0) { - nf_conntrack_get(&ct->ct_general); + if (!atomic_inc_not_zero(&ct->ct_general.use)) + continue; cb->args[1] = (unsigned long)ct; goto out; } @@ -576,7 +587,7 @@ restart: } } out: - read_unlock_bh(&nf_conntrack_lock); + rcu_read_unlock(); if (last) nf_ct_put(last); @@ -624,7 +635,7 @@ ctnetlink_parse_tuple_proto(struct nlattr *attr, if (!tb[CTA_PROTO_NUM]) return -EINVAL; - tuple->dst.protonum = *(u_int8_t *)nla_data(tb[CTA_PROTO_NUM]); + tuple->dst.protonum = nla_get_u8(tb[CTA_PROTO_NUM]); l4proto = nf_ct_l4proto_find_get(tuple->src.l3num, tuple->dst.protonum); @@ -640,7 +651,7 @@ ctnetlink_parse_tuple_proto(struct nlattr *attr, return ret; } -static inline int +static int ctnetlink_parse_tuple(struct nlattr *cda[], struct nf_conntrack_tuple *tuple, enum ctattr_tuple type, u_int8_t l3num) { @@ -687,27 +698,18 @@ static int nfnetlink_parse_nat_proto(struct nlattr *attr, struct nf_nat_range *range) { struct nlattr *tb[CTA_PROTONAT_MAX+1]; - struct nf_nat_protocol *npt; + const struct nf_nat_protocol *npt; int err; err = nla_parse_nested(tb, CTA_PROTONAT_MAX, attr, protonat_nla_policy); if (err < 0) return err; - npt = nf_nat_proto_find_get(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum); - - if (!npt->nlattr_to_range) { - nf_nat_proto_put(npt); - return 0; - } - - /* nlattr_to_range returns 1 if it parsed, 0 if not, neg. on error */ - if (npt->nlattr_to_range(tb, range) > 0) - range->flags |= IP_NAT_RANGE_PROTO_SPECIFIED; - + npt = nf_nat_proto_find_get(nf_ct_protonum(ct)); + if (npt->nlattr_to_range) + err = npt->nlattr_to_range(tb, range); nf_nat_proto_put(npt); - - return 0; + return err; } static const struct nla_policy nat_nla_policy[CTA_NAT_MAX+1] = { @@ -729,12 +731,12 @@ nfnetlink_parse_nat(struct nlattr *nat, return err; if (tb[CTA_NAT_MINIP]) - range->min_ip = *(__be32 *)nla_data(tb[CTA_NAT_MINIP]); + range->min_ip = nla_get_be32(tb[CTA_NAT_MINIP]); if (!tb[CTA_NAT_MAXIP]) range->max_ip = range->min_ip; else - range->max_ip = *(__be32 *)nla_data(tb[CTA_NAT_MAXIP]); + range->max_ip = nla_get_be32(tb[CTA_NAT_MAXIP]); if (range->min_ip) range->flags |= IP_NAT_RANGE_MAP_IPS; @@ -804,15 +806,14 @@ ctnetlink_del_conntrack(struct sock *ctnl, struct sk_buff *skb, ct = nf_ct_tuplehash_to_ctrack(h); if (cda[CTA_ID]) { - u_int32_t id = ntohl(*(__be32 *)nla_data(cda[CTA_ID])); + u_int32_t id = ntohl(nla_get_be32(cda[CTA_ID])); if (id != (u32)(unsigned long)ct) { nf_ct_put(ct); return -ENOENT; } } - if (del_timer(&ct->timeout)) - ct->timeout.function((unsigned long)ct); + nf_ct_kill(ct); nf_ct_put(ct); return 0; @@ -880,29 +881,28 @@ out: return err; } -static inline int +static int ctnetlink_change_status(struct nf_conn *ct, struct nlattr *cda[]) { unsigned long d; - unsigned int status = ntohl(*(__be32 *)nla_data(cda[CTA_STATUS])); + unsigned int status = ntohl(nla_get_be32(cda[CTA_STATUS])); d = ct->status ^ status; if (d & (IPS_EXPECTED|IPS_CONFIRMED|IPS_DYING)) /* unchangeable */ - return -EINVAL; + return -EBUSY; if (d & IPS_SEEN_REPLY && !(status & IPS_SEEN_REPLY)) /* SEEN_REPLY bit can only be set */ - return -EINVAL; - + return -EBUSY; if (d & IPS_ASSURED && !(status & IPS_ASSURED)) /* ASSURED bit can only be set */ - return -EINVAL; + return -EBUSY; if (cda[CTA_NAT_SRC] || cda[CTA_NAT_DST]) { #ifndef CONFIG_NF_NAT_NEEDED - return -EINVAL; + return -EOPNOTSUPP; #else struct nf_nat_range range; @@ -910,19 +910,17 @@ ctnetlink_change_status(struct nf_conn *ct, struct nlattr *cda[]) if (nfnetlink_parse_nat(cda[CTA_NAT_DST], ct, &range) < 0) return -EINVAL; - if (nf_nat_initialized(ct, - HOOK2MANIP(NF_INET_PRE_ROUTING))) + if (nf_nat_initialized(ct, IP_NAT_MANIP_DST)) return -EEXIST; - nf_nat_setup_info(ct, &range, NF_INET_PRE_ROUTING); + nf_nat_setup_info(ct, &range, IP_NAT_MANIP_DST); } if (cda[CTA_NAT_SRC]) { if (nfnetlink_parse_nat(cda[CTA_NAT_SRC], ct, &range) < 0) return -EINVAL; - if (nf_nat_initialized(ct, - HOOK2MANIP(NF_INET_POST_ROUTING))) + if (nf_nat_initialized(ct, IP_NAT_MANIP_SRC)) return -EEXIST; - nf_nat_setup_info(ct, &range, NF_INET_POST_ROUTING); + nf_nat_setup_info(ct, &range, IP_NAT_MANIP_SRC); } #endif } @@ -945,7 +943,7 @@ ctnetlink_change_helper(struct nf_conn *ct, struct nlattr *cda[]) /* don't change helper of sibling connections */ if (ct->master) - return -EINVAL; + return -EBUSY; err = ctnetlink_parse_help(cda[CTA_HELP], &helpname); if (err < 0) @@ -963,7 +961,7 @@ ctnetlink_change_helper(struct nf_conn *ct, struct nlattr *cda[]) helper = __nf_conntrack_helper_find_byname(helpname); if (helper == NULL) - return -EINVAL; + return -EOPNOTSUPP; if (help) { if (help->helper == helper) @@ -986,7 +984,7 @@ ctnetlink_change_helper(struct nf_conn *ct, struct nlattr *cda[]) static inline int ctnetlink_change_timeout(struct nf_conn *ct, struct nlattr *cda[]) { - u_int32_t timeout = ntohl(*(__be32 *)nla_data(cda[CTA_TIMEOUT])); + u_int32_t timeout = ntohl(nla_get_be32(cda[CTA_TIMEOUT])); if (!del_timer(&ct->timeout)) return -ETIME; @@ -1002,14 +1000,11 @@ ctnetlink_change_protoinfo(struct nf_conn *ct, struct nlattr *cda[]) { struct nlattr *tb[CTA_PROTOINFO_MAX+1], *attr = cda[CTA_PROTOINFO]; struct nf_conntrack_l4proto *l4proto; - u_int16_t npt = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum; - u_int16_t l3num = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; int err = 0; nla_parse_nested(tb, CTA_PROTOINFO_MAX, attr, NULL); - l4proto = nf_ct_l4proto_find_get(l3num, npt); - + l4proto = nf_ct_l4proto_find_get(nf_ct_l3num(ct), nf_ct_protonum(ct)); if (l4proto->from_nlattr) err = l4proto->from_nlattr(tb, ct); nf_ct_l4proto_put(l4proto); @@ -1029,19 +1024,19 @@ change_nat_seq_adj(struct nf_nat_seq *natseq, struct nlattr *attr) return -EINVAL; natseq->correction_pos = - ntohl(*(__be32 *)nla_data(cda[CTA_NAT_SEQ_CORRECTION_POS])); + ntohl(nla_get_be32(cda[CTA_NAT_SEQ_CORRECTION_POS])); if (!cda[CTA_NAT_SEQ_OFFSET_BEFORE]) return -EINVAL; natseq->offset_before = - ntohl(*(__be32 *)nla_data(cda[CTA_NAT_SEQ_OFFSET_BEFORE])); + ntohl(nla_get_be32(cda[CTA_NAT_SEQ_OFFSET_BEFORE])); if (!cda[CTA_NAT_SEQ_OFFSET_AFTER]) return -EINVAL; natseq->offset_after = - ntohl(*(__be32 *)nla_data(cda[CTA_NAT_SEQ_OFFSET_AFTER])); + ntohl(nla_get_be32(cda[CTA_NAT_SEQ_OFFSET_AFTER])); return 0; } @@ -1108,7 +1103,7 @@ ctnetlink_change_conntrack(struct nf_conn *ct, struct nlattr *cda[]) #if defined(CONFIG_NF_CONNTRACK_MARK) if (cda[CTA_MARK]) - ct->mark = ntohl(*(__be32 *)nla_data(cda[CTA_MARK])); + ct->mark = ntohl(nla_get_be32(cda[CTA_MARK])); #endif #ifdef CONFIG_NF_NAT_NEEDED @@ -1133,13 +1128,13 @@ ctnetlink_create_conntrack(struct nlattr *cda[], struct nf_conn_help *help; struct nf_conntrack_helper *helper; - ct = nf_conntrack_alloc(otuple, rtuple); + ct = nf_conntrack_alloc(otuple, rtuple, GFP_KERNEL); if (ct == NULL || IS_ERR(ct)) return -ENOMEM; if (!cda[CTA_TIMEOUT]) goto err; - ct->timeout.expires = ntohl(*(__be32 *)nla_data(cda[CTA_TIMEOUT])); + ct->timeout.expires = ntohl(nla_get_be32(cda[CTA_TIMEOUT])); ct->timeout.expires = jiffies + ct->timeout.expires * HZ; ct->status |= IPS_CONFIRMED; @@ -1158,14 +1153,15 @@ ctnetlink_create_conntrack(struct nlattr *cda[], #if defined(CONFIG_NF_CONNTRACK_MARK) if (cda[CTA_MARK]) - ct->mark = ntohl(*(__be32 *)nla_data(cda[CTA_MARK])); + ct->mark = ntohl(nla_get_be32(cda[CTA_MARK])); #endif - helper = nf_ct_helper_find_get(rtuple); + rcu_read_lock(); + helper = __nf_ct_helper_find(rtuple); if (helper) { help = nf_ct_helper_ext_add(ct, GFP_KERNEL); if (help == NULL) { - nf_ct_helper_put(helper); + rcu_read_unlock(); err = -ENOMEM; goto err; } @@ -1181,9 +1177,7 @@ ctnetlink_create_conntrack(struct nlattr *cda[], add_timer(&ct->timeout); nf_conntrack_hash_insert(ct); - - if (helper) - nf_ct_helper_put(helper); + rcu_read_unlock(); return 0; @@ -1214,11 +1208,11 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, return err; } - write_lock_bh(&nf_conntrack_lock); + spin_lock_bh(&nf_conntrack_lock); if (cda[CTA_TUPLE_ORIG]) - h = __nf_conntrack_find(&otuple, NULL); + h = __nf_conntrack_find(&otuple); else if (cda[CTA_TUPLE_REPLY]) - h = __nf_conntrack_find(&rtuple, NULL); + h = __nf_conntrack_find(&rtuple); if (h == NULL) { struct nf_conntrack_tuple master; @@ -1231,9 +1225,9 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, CTA_TUPLE_MASTER, u3); if (err < 0) - return err; + goto out_unlock; - master_h = __nf_conntrack_find(&master, NULL); + master_h = __nf_conntrack_find(&master); if (master_h == NULL) { err = -ENOENT; goto out_unlock; @@ -1242,7 +1236,7 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, atomic_inc(&master_ct->ct_general.use); } - write_unlock_bh(&nf_conntrack_lock); + spin_unlock_bh(&nf_conntrack_lock); err = -ENOENT; if (nlh->nlmsg_flags & NLM_F_CREATE) err = ctnetlink_create_conntrack(cda, @@ -1262,12 +1256,12 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, if (!(nlh->nlmsg_flags & NLM_F_EXCL)) { /* we only allow nat config for new conntracks */ if (cda[CTA_NAT_SRC] || cda[CTA_NAT_DST]) { - err = -EINVAL; + err = -EOPNOTSUPP; goto out_unlock; } /* can't link an existing conntrack to a master */ if (cda[CTA_TUPLE_MASTER]) { - err = -EINVAL; + err = -EOPNOTSUPP; goto out_unlock; } err = ctnetlink_change_conntrack(nf_ct_tuplehash_to_ctrack(h), @@ -1275,7 +1269,7 @@ ctnetlink_new_conntrack(struct sock *ctnl, struct sk_buff *skb, } out_unlock: - write_unlock_bh(&nf_conntrack_lock); + spin_unlock_bh(&nf_conntrack_lock); return err; } @@ -1343,13 +1337,15 @@ nla_put_failure: return -1; } -static inline int +static int ctnetlink_exp_dump_expect(struct sk_buff *skb, const struct nf_conntrack_expect *exp) { struct nf_conn *master = exp->master; - __be32 timeout = htonl((exp->timeout.expires - jiffies) / HZ); - __be32 id = htonl((unsigned long)exp); + long timeout = (exp->timeout.expires - jiffies) / HZ; + + if (timeout < 0) + timeout = 0; if (ctnetlink_exp_dump_tuple(skb, &exp->tuple, CTA_EXPECT_TUPLE) < 0) goto nla_put_failure; @@ -1360,8 +1356,8 @@ ctnetlink_exp_dump_expect(struct sk_buff *skb, CTA_EXPECT_MASTER) < 0) goto nla_put_failure; - NLA_PUT(skb, CTA_EXPECT_TIMEOUT, sizeof(timeout), &timeout); - NLA_PUT(skb, CTA_EXPECT_ID, sizeof(u_int32_t), &id); + NLA_PUT_BE32(skb, CTA_EXPECT_TIMEOUT, htonl(timeout)); + NLA_PUT_BE32(skb, CTA_EXPECT_ID, htonl((unsigned long)exp)); return 0; @@ -1464,7 +1460,7 @@ ctnetlink_exp_dump_table(struct sk_buff *skb, struct netlink_callback *cb) struct hlist_node *n; u_int8_t l3proto = nfmsg->nfgen_family; - read_lock_bh(&nf_conntrack_lock); + rcu_read_lock(); last = (struct nf_conntrack_expect *)cb->args[1]; for (; cb->args[0] < nf_ct_expect_hsize; cb->args[0]++) { restart: @@ -1481,7 +1477,8 @@ restart: cb->nlh->nlmsg_seq, IPCTNL_MSG_EXP_NEW, 1, exp) < 0) { - atomic_inc(&exp->use); + if (!atomic_inc_not_zero(&exp->use)) + continue; cb->args[1] = (unsigned long)exp; goto out; } @@ -1492,7 +1489,7 @@ restart: } } out: - read_unlock_bh(&nf_conntrack_lock); + rcu_read_unlock(); if (last) nf_ct_expect_put(last); @@ -1534,7 +1531,7 @@ ctnetlink_get_expect(struct sock *ctnl, struct sk_buff *skb, return -ENOENT; if (cda[CTA_EXPECT_ID]) { - __be32 id = *(__be32 *)nla_data(cda[CTA_EXPECT_ID]); + __be32 id = nla_get_be32(cda[CTA_EXPECT_ID]); if (ntohl(id) != (u32)(unsigned long)exp) { nf_ct_expect_put(exp); return -ENOENT; @@ -1588,7 +1585,7 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, return -ENOENT; if (cda[CTA_EXPECT_ID]) { - __be32 id = *(__be32 *)nla_data(cda[CTA_EXPECT_ID]); + __be32 id = nla_get_be32(cda[CTA_EXPECT_ID]); if (ntohl(id) != (u32)(unsigned long)exp) { nf_ct_expect_put(exp); return -ENOENT; @@ -1605,11 +1602,11 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, struct nf_conn_help *m_help; /* delete all expectations for this helper */ - write_lock_bh(&nf_conntrack_lock); + spin_lock_bh(&nf_conntrack_lock); h = __nf_conntrack_helper_find_byname(name); if (!h) { - write_unlock_bh(&nf_conntrack_lock); - return -EINVAL; + spin_unlock_bh(&nf_conntrack_lock); + return -EOPNOTSUPP; } for (i = 0; i < nf_ct_expect_hsize; i++) { hlist_for_each_entry_safe(exp, n, next, @@ -1623,10 +1620,10 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, } } } - write_unlock_bh(&nf_conntrack_lock); + spin_unlock_bh(&nf_conntrack_lock); } else { /* This basically means we have to flush everything*/ - write_lock_bh(&nf_conntrack_lock); + spin_lock_bh(&nf_conntrack_lock); for (i = 0; i < nf_ct_expect_hsize; i++) { hlist_for_each_entry_safe(exp, n, next, &nf_ct_expect_hash[i], @@ -1637,7 +1634,7 @@ ctnetlink_del_expect(struct sock *ctnl, struct sk_buff *skb, } } } - write_unlock_bh(&nf_conntrack_lock); + spin_unlock_bh(&nf_conntrack_lock); } return 0; @@ -1723,11 +1720,11 @@ ctnetlink_new_expect(struct sock *ctnl, struct sk_buff *skb, if (err < 0) return err; - write_lock_bh(&nf_conntrack_lock); + spin_lock_bh(&nf_conntrack_lock); exp = __nf_ct_expect_find(&tuple); if (!exp) { - write_unlock_bh(&nf_conntrack_lock); + spin_unlock_bh(&nf_conntrack_lock); err = -ENOENT; if (nlh->nlmsg_flags & NLM_F_CREATE) err = ctnetlink_create_expect(cda, u3); @@ -1737,7 +1734,7 @@ ctnetlink_new_expect(struct sock *ctnl, struct sk_buff *skb, err = -EEXIST; if (!(nlh->nlmsg_flags & NLM_F_EXCL)) err = ctnetlink_change_expect(exp, cda); - write_unlock_bh(&nf_conntrack_lock); + spin_unlock_bh(&nf_conntrack_lock); return err; }