Skip to contentSkip to conversation

net/ipv4: reset mac head before call ip_tunnel_rcv()

PATCH In review Chen Minqiang 2 participants 2 messages Aug 6, 2019

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 net/ipv4: reset mac head before call ip_tunnel_rcv()

---
 net/ipv4/ipip.c | 1 +
 net/ipv6/sit.c  | 1 +
 2 files changed, 2 insertions(+)
net/ipv4/ipip.c +1 -0
index 43adfc1641ba..ba2b5fc8910f 100644
@@ -242,6 +242,7 @@ static int ipip_tunnel_rcv(struct sk_buff *skb, u8 ipproto)
 			if (!tun_dst)
 				return 0;
 		}
+		skb_reset_mac_header(skb);
 		return ip_tunnel_rcv(tunnel, skb, tpi, tun_dst, log_ecn_error);
 	}
 
net/ipv6/sit.c +1 -0
index 80610899a323..44a9674d06a6 100644
@@ -739,6 +739,7 @@ static int sit_tunnel_rcv(struct sk_buff *skb, u8 ipproto)
 			tpi = &ipip_tpi;
 		if (iptunnel_pull_header(skb, 0, tpi->proto, false))
 			goto drop;
+		skb_reset_mac_header(skb);
 		return ip_tunnel_rcv(tunnel, skb, tpi, NULL, log_ecn_error);
 	}
 
Signature · 1 line
2.17.1
maintainer
#1 lore raw

From: Chen Minqiang <[email protected]>
Date: Tue, 6 Aug 2019 18:47:31 +0800

Quoted 1 line

No commit message means I'm not even going to look at this patch and try to understand it.

You must always completely explain, in detail, what change you are making, how you are making it, and above all why you are making this change.

Is there some bug you are fixing? What is that bug and what does that bug cause to happen? Are there potential negative side effects to your fix? What are they and what was the cost/benefit analysis for that?

Where was the bug introduced? You must provide a proper Fixes: tag which shows this.

Thank you.