diff options
| author | David S. Miller <davem@davemloft.net> | 2014-01-03 19:41:39 -0500 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2014-01-03 19:41:39 -0500 |
| commit | e9bf3b075384b3b400ab4689a2014860704164b7 (patch) | |
| tree | e5e6d38f79704d282679a06d6bf00cf819a85d7a /include | |
| parent | f916ec9608af017dda3e0fa7b5c6a344fc05285e (diff) | |
| parent | 9a4aa9af447f784f0a47313c8dcb79ac63442cf7 (diff) | |
Merge branch 'tunnel_dst_caching'
Tom Herbert says:
====================
ipv4: Cache dst in tunnels
Version 3 of caching routes in tunnels.
Addressed some comments from Eric in this series.
There are two patches (variants) in the series:
1) One dst cached for each tunnel.
2) Percpu dst cache per tunnel to avoid false sharing
Testing with GRE tunnels on a 32 CPU host with bnx2x (RSS support
for GRE) shows a modest improvement in CPU utilization with these
patches running 200 TCP_RR netperf clients.
Without patches
71.22% CPU utilization
138/180/244 90/95/99% latencies
1.30465e+06 CPU/tps
18318 CPU/tps
With patches
69.84%
142/186/249 90/95/99% latencies
1.30827e+06
18732 CPU/tps
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'include')
| -rw-r--r-- | include/net/ip_tunnels.h | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h index 732f8c6ae975..9e25b1bc31da 100644 --- a/include/net/ip_tunnels.h +++ b/include/net/ip_tunnels.h @@ -38,6 +38,11 @@ struct ip_tunnel_prl_entry { struct rcu_head rcu_head; }; +struct ip_tunnel_dst { + struct dst_entry __rcu *dst; + spinlock_t lock; +}; + struct ip_tunnel { struct ip_tunnel __rcu *next; struct hlist_node hash_node; @@ -54,6 +59,8 @@ struct ip_tunnel { int hlen; /* Precalculated header length */ int mlink; + struct ip_tunnel_dst __percpu *dst_cache; + struct ip_tunnel_parm parms; /* for SIT */ |