[PATCH 2/2] nf_conntrack: Don't try to find clashed expectation

[PATCH 2/2] nf_conntrack: Don't try to find clashed expectation

[Yasuyuki KOZAKAI]

The original code continues loop to find expectation in list if the master
conntrack of the found expectation is unconfirmed. But it never success
in that case, because nf_conntrack_expect_related() never insert
clashed expectation to the list.

This stops loop in that case.

Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
---
 net/netfilter/nf_conntrack_expect.c |   27 +++++++++++++++------------
 1 files changed, 15 insertions(+), 12 deletions(-)

diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c
index 588d379..7df8f9a 100644
--- a/net/netfilter/nf_conntrack_expect.c
+++ b/net/netfilter/nf_conntrack_expect.c
@@ -91,25 +91,28 @@ EXPORT_SYMBOL_GPL(nf_conntrack_expect_fi
 struct nf_conntrack_expect *
 find_expectation(const struct nf_conntrack_tuple *tuple)
 {
-	struct nf_conntrack_expect *i;
+	struct nf_conntrack_expect *exp;
+
+	exp = __nf_conntrack_expect_find(tuple);
+	if (!exp)
+		return NULL;
 
-	list_for_each_entry(i, &nf_conntrack_expect_list, list) {
 	/* If master is not in hash table yet (ie. packet hasn't left
 	   this machine yet), how can other end know about expected?
 	   Hence these are not the droids you are looking for (if
 	   master ct never got confirmed, we'd hold a reference to it
 	   and weird things would happen to future packets). */
-		if (nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask)
-		    && nf_ct_is_confirmed(i->master)) {
-			if (i->flags & NF_CT_EXPECT_PERMANENT) {
-				atomic_inc(&i->use);
-				return i;
-			} else if (del_timer(&i->timeout)) {
-				nf_ct_unlink_expect(i);
-				return i;
-			}
-		}
+	if (!nf_ct_is_confirmed(exp->master))
+		return NULL;
+
+	if (exp->flags & NF_CT_EXPECT_PERMANENT) {
+		atomic_inc(&exp->use);
+		return exp;
+	} else if (del_timer(&exp->timeout)) {
+		nf_ct_unlink_expect(exp);
+		return exp;
 	}
+
 	return NULL;
 }
 
-- 
1.4.4

Re: [PATCH 2/2] nf_conntrack: Don't try to find clashed expectation

[Patrick McHardy]

Yasuyuki KOZAKAI wrote:
> The original code continues loop to find expectation in list if the master
> conntrack of the found expectation is unconfirmed. But it never success
> in that case, because nf_conntrack_expect_related() never insert
> clashed expectation to the list.
> 
> This stops loop in that case.

Also applied, thanks.