On Fri, Oct 19, 2001 at 04:39:25PM -0400, Grant Taylor wrote: > The sockets are not connected (there are ~500 peers, and polling on > a mess of sockets would be poor). The sockets have checksums turned > off (local switched ethernet and loopback only). It probably would not make much difference to turn them on -- 2.4 should do all udp csum computation in parallel to copies and during that they're mostly free. > > 6.15% udp_recvmsg # !? This one is really a mystery. Could you set the profile table shift to 1 and try to narrow it to some instruction or run it under oprofile and check for cache misses etc.? > 5.38% ip_build_xmit # inlined skb fiddling, iph csum, etc > 4.57% __generic_copy_to_user # life > 4.45% udp_rcv # no "most recent socket" cache It used to have one, but it was removed for some reason I cannot remember. Again > 4.31% ip_rcv # iph checksum check on loopback Unlikely that it is the checksum check -- it is only a few inlined instructions. > 3.97% ip_route_output_key # could cache route for unconnected sockets It is probably more cache line bouncing, where the cache wouldn't help. -Andi - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
