Skip to contentSkip to conversation

netfilter: bridge: rename br_parse_ip_options

PATCH Applied Bernhard Thaler 2 participants 2 messages May 30, 2015 – Jun 12, 2015

Subsystems (3)
Networking netfilter Ethernet bridge
lore
Share & tools
lore

Bernhard Thaler

author
#0 lore raw

PATCH v6 2/4 netfilter: bridge: rename br_parse_ip_options

br_parse_ip_options() does not parse any IP options, it validates IP packets as a whole and the function name is misleading.

Rename br_parse_ip_options() to br_validate_ipv4() and remove unneeded commments.

Signed-off-by Bernhard Thaler <[email protected]>

---
Patch revision history:

v6
* put this patch in front of "netfilter: bridge: forward IPv6 fragmented packets"

v5
* rebase to current davem/net-next

v4
* re-post due to errors in v3 formatting introduced by my MUA

v3
* re-assignment of iph variable needed because pskb_may_pull() can invalidate the network header
* same patch as v1 again

v2
* first patch did not contain statement removing double iph variable assignment

 net/bridge/br_netfilter.c |    9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)
net/bridge/br_netfilter.c +4 -5
index 1f30b28..962d5f8 100644
@@ -216,7 +216,7 @@ static inline void nf_bridge_pull_encap_header_rcsum(struct sk_buff *skb)
  * expected format
  */
 
-static int br_parse_ip_options(struct sk_buff *skb)
+static int br_validate_ipv4(struct sk_buff *skb)
 {
 	const struct iphdr *iph;
 	struct net_device *dev = skb->dev;

@@ -692,7 +692,7 @@ static unsigned int br_nf_pre_routing(const struct nf_hook_ops *ops,
 
 	nf_bridge_pull_encap_header_rcsum(skb);
 
-	if (br_parse_ip_options(skb))
+	if (br_validate_ipv4(skb))
 		return NF_DROP;
 
 	nf_bridge_put(skb->nf_bridge);

@@ -802,7 +802,7 @@ static unsigned int br_nf_forward_ip(const struct nf_hook_ops *ops,
 	}
 
 	if (pf == NFPROTO_IPV4) {
-		if (br_parse_ip_options(skb))
+		if (br_validate_ipv4(skb))
 			return NF_DROP;
 		IPCB(skb)->frag_max_size = nf_bridge->frag_max_size;
 	}

@@ -913,8 +913,7 @@ static int br_nf_dev_queue_xmit(struct sock *sk, struct sk_buff *skb)
 	if (skb->len + mtu_reserved > skb->dev->mtu) {
 		struct brnf_frag_data *data;
 
-		if (br_parse_ip_options(skb))
-			/* Drop invalid packet */
+		if (br_validate_ipv4(skb))
 			return NF_DROP;
 
 		IPCB(skb)->frag_max_size = nf_bridge->frag_max_size;
Signature · 1 line
1.7.10.4
maintainer Applied
#1 lore raw
Quoted 5 lines · Bernhard Thaler

br_parse_ip_options() does not parse any IP options, it validates IP packets as a whole and the function name is misleading.

Rename br_parse_ip_options() to br_validate_ipv4() and remove unneeded commments.

Applied, thanks.