All of lore.kernel.org
 help / color / mirror / Atom feed
From: Patrick McHardy <kaber@trash.net>
To: davem@davemloft.net
Cc: netdev@vger.kernel.org, Patrick McHardy <kaber@trash.net>,
	netfilter-devel@vger.kernel.org
Subject: netfilter 03/03: ctnetlink: fix regression in expectation handling
Date: Wed,  8 Apr 2009 18:52:19 +0200 (MEST)	[thread overview]
Message-ID: <20090408165216.6702.57294.sendpatchset@x2.localnet> (raw)
In-Reply-To: <20090408165212.6702.44171.sendpatchset@x2.localnet>

commit 83731671d9e6878c0a05d309c68fb71c16d3235a
Author: Pablo Neira Ayuso <pablo@netfilter.org>
Date:   Mon Apr 6 17:47:20 2009 +0200

    netfilter: ctnetlink: fix regression in expectation handling
    
    This patch fixes a regression (introduced by myself in commit 19abb7b:
    netfilter: ctnetlink: deliver events for conntracks changed from
    userspace) that results in an expectation re-insertion since
    __nf_ct_expect_check() may return 0 for expectation timer refreshing.
    
    This patch also removes a unnecessary refcount bump that
    pretended to avoid a possible race condition with event delivery
    and expectation timers (as said, not needed since we hold a
    reference to the object since until we finish the expectation
    setup). This also merges nf_ct_expect_related_report() and
    nf_ct_expect_related() which look basically the same.
    
    Reported-by: Patrick McHardy <kaber@trash.net>
    Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
    Signed-off-by: Patrick McHardy <kaber@trash.net>

diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h
index ab17a15..a965280 100644
--- a/include/net/netfilter/nf_conntrack_expect.h
+++ b/include/net/netfilter/nf_conntrack_expect.h
@@ -99,9 +99,12 @@ void nf_ct_expect_init(struct nf_conntrack_expect *, unsigned int, u_int8_t,
 		       const union nf_inet_addr *,
 		       u_int8_t, const __be16 *, const __be16 *);
 void nf_ct_expect_put(struct nf_conntrack_expect *exp);
-int nf_ct_expect_related(struct nf_conntrack_expect *expect);
 int nf_ct_expect_related_report(struct nf_conntrack_expect *expect, 
 				u32 pid, int report);
+static inline int nf_ct_expect_related(struct nf_conntrack_expect *expect)
+{
+	return nf_ct_expect_related_report(expect, 0, 0);
+}
 
 #endif /*_NF_CONNTRACK_EXPECT_H*/
 
diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c
index 3940f99..afde8f9 100644
--- a/net/netfilter/nf_conntrack_expect.c
+++ b/net/netfilter/nf_conntrack_expect.c
@@ -372,7 +372,7 @@ static inline int __nf_ct_expect_check(struct nf_conntrack_expect *expect)
 	struct net *net = nf_ct_exp_net(expect);
 	struct hlist_node *n;
 	unsigned int h;
-	int ret = 0;
+	int ret = 1;
 
 	if (!master_help->helper) {
 		ret = -ESHUTDOWN;
@@ -412,41 +412,23 @@ out:
 	return ret;
 }
 
-int nf_ct_expect_related(struct nf_conntrack_expect *expect)
+int nf_ct_expect_related_report(struct nf_conntrack_expect *expect,
+				u32 pid, int report)
 {
 	int ret;
 
 	spin_lock_bh(&nf_conntrack_lock);
 	ret = __nf_ct_expect_check(expect);
-	if (ret < 0)
+	if (ret <= 0)
 		goto out;
 
+	ret = 0;
 	nf_ct_expect_insert(expect);
-	atomic_inc(&expect->use);
-	spin_unlock_bh(&nf_conntrack_lock);
-	nf_ct_expect_event(IPEXP_NEW, expect);
-	nf_ct_expect_put(expect);
-	return ret;
-out:
 	spin_unlock_bh(&nf_conntrack_lock);
+	nf_ct_expect_event_report(IPEXP_NEW, expect, pid, report);
 	return ret;
-}
-EXPORT_SYMBOL_GPL(nf_ct_expect_related);
-
-int nf_ct_expect_related_report(struct nf_conntrack_expect *expect, 
-				u32 pid, int report)
-{
-	int ret;
-
-	spin_lock_bh(&nf_conntrack_lock);
-	ret = __nf_ct_expect_check(expect);
-	if (ret < 0)
-		goto out;
-	nf_ct_expect_insert(expect);
 out:
 	spin_unlock_bh(&nf_conntrack_lock);
-	if (ret == 0)
-		nf_ct_expect_event_report(IPEXP_NEW, expect, pid, report);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(nf_ct_expect_related_report);

  parent reply	other threads:[~2009-04-08 16:52 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-04-08 16:52 netfilter 00/03: netfilter fixes Patrick McHardy
2009-04-08 16:52 ` netfilter 01/03: ip6tables regression fix Patrick McHardy
2009-04-08 16:52 ` netfilter 02/03: fix selection of "LED" target in netfilter Patrick McHardy
2009-04-08 16:52 ` Patrick McHardy [this message]
2009-04-08 20:43 ` netfilter 00/03: netfilter fixes David Miller

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20090408165216.6702.57294.sendpatchset@x2.localnet \
    --to=kaber@trash.net \
    --cc=davem@davemloft.net \
    --cc=netdev@vger.kernel.org \
    --cc=netfilter-devel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.