[PATCH] netfilter fix for 3.6-rc7

[PATCH] netfilter fix for 3.6-rc7

[pablo]

From: Pablo Neira Ayuso <pablo@netfilter.org>

Hi David,

If time allows, I'd appreciate if you can take the following fix
for the xt_limit match.

As Jan indicates, random things may occur while using the xt_limit
match due to use of uninitialized memory.

You can pull this change from:

git://1984.lsi.us.es/nf master

Thanks!

Jan Engelhardt (1):
  netfilter: xt_limit: have r->cost != 0 case work

 net/netfilter/xt_limit.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

-- 
1.7.10.4

[PATCH] netfilter: xt_limit: have r->cost != 0 case work

[pablo]

From: Jan Engelhardt <jengelh@inai.de>

Commit v2.6.19-rc1~1272^2~41 tells us that r->cost != 0 can happen when
a running state is saved to userspace and then reinstated from there.

Make sure that private xt_limit area is initialized with correct values.
Otherwise, random matchings due to use of uninitialized memory.

Signed-off-by: Jan Engelhardt <jengelh@inai.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
---
 net/netfilter/xt_limit.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/netfilter/xt_limit.c b/net/netfilter/xt_limit.c
index 5c22ce8..a4c1e45 100644
--- a/net/netfilter/xt_limit.c
+++ b/net/netfilter/xt_limit.c
@@ -117,11 +117,11 @@ static int limit_mt_check(const struct xt_mtchk_param *par)
 
 	/* For SMP, we only want to use one set of state. */
 	r->master = priv;
+	/* User avg in seconds * XT_LIMIT_SCALE: convert to jiffies *
+	   128. */
+	priv->prev = jiffies;
+	priv->credit = user2credits(r->avg * r->burst); /* Credits full. */
 	if (r->cost == 0) {
-		/* User avg in seconds * XT_LIMIT_SCALE: convert to jiffies *
-		   128. */
-		priv->prev = jiffies;
-		priv->credit = user2credits(r->avg * r->burst); /* Credits full. */
 		r->credit_cap = priv->credit; /* Credits full. */
 		r->cost = user2credits(r->avg);
 	}
-- 
1.7.10.4

Re: [PATCH] netfilter fix for 3.6-rc7

[David Miller]

From: pablo@netfilter.org
Date: Wed, 26 Sep 2012 01:34:10 +0200

> If time allows, I'd appreciate if you can take the following fix
> for the xt_limit match.
> 
> As Jan indicates, random things may occur while using the xt_limit
> match due to use of uninitialized memory.
> 
> You can pull this change from:
> 
> git://1984.lsi.us.es/nf master

Pulled, thanks.