Only detecting stale connections for UDP ports in kube-proxy.

The detectStaleConnections function in kube-proxy is very expensive in
terms of CPU utilization. The results of this function are only actually
used for UDP ports. This adds a protocol attribute to ServicePortName to
make it simple to only run this function for UDP connections. For
clusters with primarily TCP connections this can improve kube-proxy
performance by 2x.

pkg/proxy/endpointslicecache.go

@@ -145,8 +145,11 @@ func (cache *EndpointSliceCache) endpointInfoByServicePort(serviceNN types.Names
 				continue
 			}
 
-			svcPortName := ServicePortName{NamespacedName: serviceNN}
-			svcPortName.Port = *port.Name
+			svcPortName := ServicePortName{
+				NamespacedName: serviceNN,
+				Port:           *port.Name,
+				Protocol:       *port.Protocol,
+			}
 
 			endpointInfoBySP[svcPortName] = cache.addEndpointsByIP(serviceNN, int(*port.Port), endpointInfoBySP[svcPortName], sliceInfo.Endpoints)
 		}