Re: [PATCH net] ipv6: Skip XFRM lookup if dst_entry in socket cache is valid

2016-06-08 Thread David Miller
From: Jakub Sitnicki Date: Wed, 8 Jun 2016 15:13:34 +0200 > At present we perform an xfrm_lookup() for each UDPv6 message we > send. The lookup involves querying the flow cache (flow_cache_lookup) > and, in case of a cache miss, creating an XFRM bundle. > > If we miss the flow cache, we can end

[PATCH net] ipv6: Skip XFRM lookup if dst_entry in socket cache is valid

2016-06-08 Thread Jakub Sitnicki
At present we perform an xfrm_lookup() for each UDPv6 message we send. The lookup involves querying the flow cache (flow_cache_lookup) and, in case of a cache miss, creating an XFRM bundle. If we miss the flow cache, we can end up creating a new bundle and deriving the path MTU (xfrm_init_pmtu) fr