netfilter: merge ctinfo into nfct pointer storage area

After this change conntrack operations (lookup, creation, matching from
ruleset) only access one instead of two sk_buff cache lines.

This works for normal conntracks because those are allocated from a slab
that guarantees hw cacheline or 8byte alignment (whatever is larger)
so the 3 bits needed for ctinfo won't overlap with nf_conn addresses.

Template allocation now does manual address alignment (see previous change)
on arches that don't have sufficent kmalloc min alignment.

Some spots intentionally use skb->_nfct instead of skb_nfct() helpers,
this is to avoid undoing the skb_nfct() use when we remove untracked
conntrack object in the future.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>

4 files changed, 12 insertions, 8 deletions

diff --git a/net/netfilter/core.c b/net/netfilter/core.c
index ce6adfae521a..a87a6f8a74d8 100644
--- a/net/netfilter/core.c
+++ b/net/netfilter/core.c
@@ -375,7 +375,7 @@ void nf_ct_attach(struct sk_buff *new, const struct sk_buff *skb)
 {
 	void (*attach)(struct sk_buff *, const struct sk_buff *);
 
-	if (skb->nfct) {
+	if (skb->_nfct) {
 		rcu_read_lock();
 		attach = rcu_dereference(ip_ct_attach);
 		if (attach)
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 768968fba7f6..47c4ea53daa6 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -1239,7 +1239,7 @@ init_conntrack(struct net *net, struct nf_conn *tmpl,
 	return &ct->tuplehash[IP_CT_DIR_ORIGINAL];
 }
 
-/* On success, returns conntrack ptr, sets skb->nfct and ctinfo */
+/* On success, returns conntrack ptr, sets skb->_nfct | ctinfo */
 static inline struct nf_conn *
 resolve_normal_ct(struct net *net, struct nf_conn *tmpl,
 		  struct sk_buff *skb,
@@ -1323,7 +1323,7 @@ nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum,
 			NF_CT_STAT_INC_ATOMIC(net, ignore);
 			return NF_ACCEPT;
 		}
-		skb->nfct = NULL;
+		skb->_nfct = 0;
 	}
 
 	/* rcu_read_lock()ed by nf_hook_thresh */
@@ -1352,7 +1352,7 @@ nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum,
 			goto out;
 		}
 		/* ICMP[v6] protocol trackers may assign one conntrack. */
-		if (skb->nfct)
+		if (skb->_nfct)
 			goto out;
 	}
 repeat:
@@ -1383,7 +1383,7 @@ repeat:
 		 * the netfilter core what to do */
 		pr_debug("nf_conntrack_in: Can't track with proto module\n");
 		nf_conntrack_put(&ct->ct_general);
-		skb->nfct = NULL;
+		skb->_nfct = 0;
 		NF_CT_STAT_INC_ATOMIC(net, invalid);
 		if (ret == -NF_DROP)
 			NF_CT_STAT_INC_ATOMIC(net, drop);
@@ -1878,7 +1878,8 @@ int nf_conntrack_init_start(void)
 	nf_conntrack_max = max_factor * nf_conntrack_htable_size;
 
 	nf_conntrack_cachep = kmem_cache_create("nf_conntrack",
-						sizeof(struct nf_conn), 0,
+						sizeof(struct nf_conn),
+						NFCT_INFOMASK + 1,
 						SLAB_DESTROY_BY_RCU | SLAB_HWCACHE_ALIGN, NULL);
 	if (!nf_conntrack_cachep)
 		goto err_cachep;
diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c
index d009ae663453..2256147dcaad 100644
--- a/net/netfilter/nf_conntrack_standalone.c
+++ b/net/netfilter/nf_conntrack_standalone.c
@@ -642,6 +642,9 @@ static int __init nf_conntrack_standalone_init(void)
 	if (ret < 0)
 		goto out_start;
 
+	BUILD_BUG_ON(SKB_NFCT_PTRMASK != NFCT_PTRMASK);
+	BUILD_BUG_ON(NFCT_INFOMASK <= IP_CT_NUMBER);
+
 #ifdef CONFIG_SYSCTL
 	nf_ct_netfilter_header =
 		register_net_sysctl(&init_net, "net", nf_ct_netfilter_table);
diff --git a/net/netfilter/xt_CT.c b/net/netfilter/xt_CT.c
index 51f00e1e1208..b008db0184b8 100644
--- a/net/netfilter/xt_CT.c
+++ b/net/netfilter/xt_CT.c
@@ -23,7 +23,7 @@
 static inline int xt_ct_target(struct sk_buff *skb, struct nf_conn *ct)
 {
 	/* Previously seen (loopback)? Ignore. */
-	if (skb->nfct != NULL)
+	if (skb->_nfct != 0)
 		return XT_CONTINUE;
 
 	/* special case the untracked ct : we want the percpu object */
@@ -409,7 +409,7 @@ static unsigned int
 notrack_tg(struct sk_buff *skb, const struct xt_action_param *par)
 {
 	/* Previously seen (loopback)? Ignore. */
-	if (skb->nfct != NULL)
+	if (skb->_nfct != 0)
 		return XT_CONTINUE;
 
 	nf_ct_set(skb, nf_ct_untracked_get(), IP_CT_NEW);