Re: [PATCH 10/17] netfilter: ipv6: expand skb head in ip6_route_me_harder after oif change

On 11/21/2011 11:29 AM, Eric Dumazet wrote:

Le lundi 21 novembre 2011 à 06:46 +0100, kaber@trash.net a écrit :

quoted

From: Patrick McHardy<redacted>

Expand the skb headroom if the oif changed due to rerouting similar to
how IPv4 packets are handled.

Signed-off-by: Patrick McHardy<redacted>
---
  net/ipv6/netfilter.c |    7 +++++++
  1 files changed, 7 insertions(+), 0 deletions(-)

diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c
index 30fcee4..d39e954 100644
--- a/net/ipv6/netfilter.c
+++ b/net/ipv6/netfilter.c

@@ -14,6 +14,7 @@ int ip6_route_me_harder(struct sk_buff *skb)
  {
  	struct net *net = dev_net(skb_dst(skb)->dev);
  	const struct ipv6hdr *iph = ipv6_hdr(skb);
+	unsigned int hh_len;
  	struct dst_entry *dst;
  	struct flowi6 fl6 = {
  		.flowi6_oif = skb->sk ? skb->sk->sk_bound_dev_if : 0,

@@ -46,6 +47,12 @@ int ip6_route_me_harder(struct sk_buff *skb)
  	}
  #endif

+	/* Change in oif may mean change in hh_len. */
+	hh_len = skb_dst(skb)->dev->hard_header_len;
+	if (skb_headroom(skb)<  hh_len&&
+	    pskb_expand_head(skb, hh_len - skb_headroom(skb), 0, GFP_ATOMIC))
+		return -1;
+

Hmm, We had a problem recently on Tile arches because of unaligned
accesses.

Had you seen the bug report and proposed patch ?

Fix is to instead do :

	pskb_expand_head(skb,
			 HH_DATA_ALIGN(hh_len - skb_headroom(skb)),
			 0,
			 GFP_ATOMIC);

I'll have a look and will update the patch accordingly.

--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html