Re: [PATCH] sit: reload iphdr in ipip6_rcv

From: 严海双 <redacted>
Date: Tue, 6 Jun 2017 10:09:45 +0800

quoted

On 5 Jun 2017, at 9:03 PM, Eric Dumazet [off-list ref] wrote:

On Sun, 2017-06-04 at 14:43 +0800, Haishuang Yan wrote:

quoted

Since iptunnel_pull_header() can call pskb_may_pull(),
we must reload any pointer that was related to skb->head.

Fixes: a09a4c8dd1ec ("tunnels: Remove encapsulation offloads on decap")
Signed-off-by: Haishuang Yan <redacted>
---
net/ipv6/sit.c | 1 +
1 file changed, 1 insertion(+)

diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 61e5902..af832e7 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c

@@ -657,6 +657,7 @@ static int ipip6_rcv(struct sk_buff *skb)

		if (iptunnel_pull_header(skb, 0, htons(ETH_P_IPV6),
		    !net_eq(tunnel->net, dev_net(tunnel->dev))))
			goto out;
+		iph = ip_hdr(skb);

		err = IP_ECN_decapsulate(iph, skb);
		if (unlikely(err)) {

This seems unnecessary.

By the time ipip6_rcv() is called, we already have the guarantee the
IPv4 header is in skb linear part.

Otherwise we could not use iph->saddr and iph->daddr in the call to
ipip6_tunnel_lookup()

Therefore, the pskb_may_pull() is a nop in this particular case.

Yes, it’s right. Thanks!

Ok, I reverted, thanks everyone.