From c71caf4114a0e1da3451cc92fba6a152929cd4c2 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 24 Jan 2011 19:01:07 +0100 Subject: [PATCH 1/4] netfilter: ctnetlink: fix missing refcount increment during dumps In 13ee6ac netfilter: fix race in conntrack between dump_table and destroy, we recovered spinlocks to protect the dump of the conntrack table according to reports from Stephen and acknowledgments on the issue from Eric. In that patch, the refcount bump that allows to keep a reference to the current ct object was removed. However, we still decrement the refcount for that object in the output path of ctnetlink_dump_table(): if (last) nf_ct_put(last) Cc: Stephen Hemminger Signed-off-by: Pablo Neira Ayuso Acked-by: Eric Dumazet Signed-off-by: Patrick McHardy --- net/netfilter/nf_conntrack_netlink.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 93297aaceb2b..eead9db6f899 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -667,6 +667,7 @@ ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb) if (ctnetlink_fill_info(skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, IPCTNL_MSG_CT_NEW, ct) < 0) { + nf_conntrack_get(&ct->ct_general); cb->args[1] = (unsigned long)ct; goto out; } From 08b5194b5d6485d12ebf24cf6ee389fc55691122 Mon Sep 17 00:00:00 2001 From: Thomas Jacob Date: Mon, 24 Jan 2011 21:35:36 +0100 Subject: [PATCH 2/4] netfilter: xt_iprange: Incorrect xt_iprange boundary check for IPv6 iprange_ipv6_sub was substracting 2 unsigned ints and then casting the result to int to find out whether they are lt, eq or gt each other, this doesn't work if the full 32 bits of each part can be used in IPv6 addresses. Patch should remedy that without significant performance penalties. Also number of ntohl calls can be reduced this way (Jozsef Kadlecsik). Signed-off-by: Thomas Jacob Signed-off-by: Patrick McHardy --- net/netfilter/xt_iprange.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) diff --git a/net/netfilter/xt_iprange.c b/net/netfilter/xt_iprange.c index 88f7c3511c72..73c33a42f87f 100644 --- a/net/netfilter/xt_iprange.c +++ b/net/netfilter/xt_iprange.c @@ -53,15 +53,13 @@ iprange_mt4(const struct sk_buff *skb, struct xt_action_param *par) } static inline int -iprange_ipv6_sub(const struct in6_addr *a, const struct in6_addr *b) +iprange_ipv6_lt(const struct in6_addr *a, const struct in6_addr *b) { unsigned int i; - int r; for (i = 0; i < 4; ++i) { - r = ntohl(a->s6_addr32[i]) - ntohl(b->s6_addr32[i]); - if (r != 0) - return r; + if (a->s6_addr32[i] != b->s6_addr32[i]) + return ntohl(a->s6_addr32[i]) < ntohl(b->s6_addr32[i]); } return 0; @@ -75,15 +73,15 @@ iprange_mt6(const struct sk_buff *skb, struct xt_action_param *par) bool m; if (info->flags & IPRANGE_SRC) { - m = iprange_ipv6_sub(&iph->saddr, &info->src_min.in6) < 0; - m |= iprange_ipv6_sub(&iph->saddr, &info->src_max.in6) > 0; + m = iprange_ipv6_lt(&iph->saddr, &info->src_min.in6); + m |= iprange_ipv6_lt(&info->src_max.in6, &iph->saddr); m ^= !!(info->flags & IPRANGE_SRC_INV); if (m) 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 = iprange_ipv6_lt(&iph->daddr, &info->dst_min.in6); + m |= iprange_ipv6_lt(&info->dst_max.in6, &iph->daddr); m ^= !!(info->flags & IPRANGE_DST_INV); if (m) return false; From 9d0db8b6b1da9e3d4c696ef29449700c58d589db Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 1 Feb 2011 16:03:46 +0100 Subject: [PATCH 3/4] netfilter: arpt_mangle: fix return values of checkentry In 135367b "netfilter: xtables: change xt_target.checkentry return type", the type returned by checkentry was changed from boolean to int, but the return values where not adjusted. arptables: Input/output error This broke arptables with the mangle target since it returns true under success, which is interpreted by xtables as >0, thus returning EIO. Signed-off-by: Pablo Neira Ayuso Signed-off-by: Patrick McHardy --- net/ipv4/netfilter/arpt_mangle.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/ipv4/netfilter/arpt_mangle.c b/net/ipv4/netfilter/arpt_mangle.c index b8ddcc480ed9..a5e52a9f0a12 100644 --- a/net/ipv4/netfilter/arpt_mangle.c +++ b/net/ipv4/netfilter/arpt_mangle.c @@ -60,12 +60,12 @@ static int checkentry(const struct xt_tgchk_param *par) if (mangle->flags & ~ARPT_MANGLE_MASK || !(mangle->flags & ARPT_MANGLE_MASK)) - return false; + return -EINVAL; if (mangle->target != NF_DROP && mangle->target != NF_ACCEPT && mangle->target != XT_CONTINUE) - return false; - return true; + return -EINVAL; + return 0; } static struct xt_target arpt_mangle_reg __read_mostly = { From 3db7e93d3308fb882884b9f024235d6fbf542034 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Tue, 1 Feb 2011 16:06:30 +0100 Subject: [PATCH 4/4] netfilter: ecache: always set events bits, filter them later For the following rule: iptables -I PREROUTING -t raw -j CT --ctevents assured The event delivered looks like the following: [UPDATE] tcp 6 src=192.168.0.2 dst=192.168.1.2 sport=37041 dport=80 src=192.168.1.2 dst=192.168.1.100 sport=80 dport=37041 [ASSURED] Note that the TCP protocol state is not included. For that reason the CT event filtering is not very useful for conntrackd. To resolve this issue, instead of conditionally setting the CT events bits based on the ctmask, we always set them and perform the filtering in the late stage, just before the delivery. Thus, the event delivered looks like the following: [UPDATE] tcp 6 432000 ESTABLISHED src=192.168.0.2 dst=192.168.1.2 sport=37041 dport=80 src=192.168.1.2 dst=192.168.1.100 sport=80 dport=37041 [ASSURED] Signed-off-by: Pablo Neira Ayuso Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_ecache.h | 3 --- net/netfilter/nf_conntrack_ecache.c | 3 +++ 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/include/net/netfilter/nf_conntrack_ecache.h b/include/net/netfilter/nf_conntrack_ecache.h index 96ba5f7dcab6..349cefedc9f3 100644 --- a/include/net/netfilter/nf_conntrack_ecache.h +++ b/include/net/netfilter/nf_conntrack_ecache.h @@ -77,9 +77,6 @@ nf_conntrack_event_cache(enum ip_conntrack_events event, struct nf_conn *ct) if (e == NULL) return; - if (!(e->ctmask & (1 << event))) - return; - set_bit(event, &e->cache); } diff --git a/net/netfilter/nf_conntrack_ecache.c b/net/netfilter/nf_conntrack_ecache.c index 5702de35e2bb..63a1b915a7e4 100644 --- a/net/netfilter/nf_conntrack_ecache.c +++ b/net/netfilter/nf_conntrack_ecache.c @@ -63,6 +63,9 @@ void nf_ct_deliver_cached_events(struct nf_conn *ct) * this does not harm and it happens very rarely. */ unsigned long missed = e->missed; + if (!((events | missed) & e->ctmask)) + goto out_unlock; + ret = notify->fcn(events | missed, &item); if (unlikely(ret < 0 || missed)) { spin_lock_bh(&ct->lock);