openvswitch: optimize flow compare and mask functions

Make sure the sw_flow_key structure and valid mask boundaries are always
machine word aligned. Optimize the flow compare and mask operations
using machine word size operations. This patch improves throughput on
average by 15% when CPU is the bottleneck of forwarding packets.

This patch is inspired by ideas and code from a patch submitted by Peter
Klausler titled "replace memcmp() with specialized comparator".
However, The original patch only optimizes for architectures
support unaligned machine word access. This patch optimizes for all
architectures.

Signed-off-by: Andy Zhou <azhou@nicira.com>
Signed-off-by: Jesse Gross <jesse@nicira.com>

1 files changed, 1 insertions, 18 deletions

diff --git a/net/openvswitch/flow.h b/net/openvswitch/flow.h
index e79305184b79..b65f885ac3dc 100644
--- a/net/openvswitch/flow.h
+++ b/net/openvswitch/flow.h
@@ -125,7 +125,7 @@ struct sw_flow_key {
 			} nd;
 		} ipv6;
 	};
-};
+} __aligned(__alignof__(long));
 
 struct sw_flow {
 	struct rcu_head rcu;
@@ -149,11 +149,6 @@ struct sw_flow_key_range {
 	size_t end;
 };
 
-static inline u16 ovs_sw_flow_key_range_actual_size(const struct sw_flow_key_range *range)
-{
-	return range->end - range->start;
-}
-
 struct sw_flow_match {
 	struct sw_flow_key *key;
 	struct sw_flow_key_range range;
@@ -253,18 +248,6 @@ struct sw_flow_mask {
 	struct sw_flow_key key;
 };
 
-static inline u16
-ovs_sw_flow_mask_actual_size(const struct sw_flow_mask *mask)
-{
-	return ovs_sw_flow_key_range_actual_size(&mask->range);
-}
-
-static inline u16
-ovs_sw_flow_mask_size_roundup(const struct sw_flow_mask *mask)
-{
-	return roundup(ovs_sw_flow_mask_actual_size(mask), sizeof(u32));
-}
-
 struct sw_flow_mask *ovs_sw_flow_mask_alloc(void);
 void ovs_sw_flow_mask_add_ref(struct sw_flow_mask *);
 void ovs_sw_flow_mask_del_ref(struct sw_flow_mask *, bool deferred);