Re: [PATCH net v5 1/1] ip6_tunnel: snapshot encap in xmit

[Paolo Abeni <pabeni@redhat.com>]

On 9/15/26 11:20 AM, Paolo Abeni wrote:
> On 9/9/26 9:01 PM, Ren Wei wrote:
>> diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
>> index 7c9aadfe8fe3..ab217ddbeb20 100644
>> --- a/include/net/ip_tunnels.h
>> +++ b/include/net/ip_tunnels.h
>> @@ -522,6 +522,16 @@ skb_vlan_inet_prepare(struct sk_buff *skb, bool inner_proto_inherit)
>>  	return SKB_NOT_DROPPED_YET;
>>  }
>>  
>> +static inline void
>> +ip_tunnel_encap_snapshot(struct ip_tunnel_encap *dst,
>> +			 const struct ip_tunnel_encap *src)
>> +{
>> +	dst->type = READ_ONCE(src->type);
>> +	dst->flags = READ_ONCE(src->flags);
>> +	dst->sport = READ_ONCE(src->sport);
>> +	dst->dport = READ_ONCE(src->dport);
> 
> The above can still race with ip6_tnl_encap_setup() giving inconsistent
> results. As noted by Eric in v4 the solution would be convering the encap
> to RCU. I think that the additional dereference would be actually faster
> than the extra memory copy.
whoops... I almost forgot: I think such work should target net-next with
not fixes tag.

/P