On Mon, Jun 16, 2003 at 04:27:50PM -0700, Simon Kirby wrote: > On Mon, Jun 16, 2003 at 04:08:56PM -0700, David S. Miller wrote: > > > It depends upon the first patch that I enclosed. > > Never mind. :) Such patches don't work very well with patch --dry. Okay, here goes 2.5.71 + this patch: 60.0049 seconds passed, avg forwarding rate: 160190.859 pps 60.0085 seconds passed, avg forwarding rate: 157118.708 pps 60.0046 seconds passed, avg forwarding rate: 157211.097 pps 60.0073 seconds passed, avg forwarding rate: 157557.710 pps ...Looks like a tad worse than with your patch, but not by much. Forwarding rate is still pretty crappy for an Opteron. Will fiddle a bit more tonight to see what I can do. Cpu type: Athlon Cpu speed was (MHz estimation) : 1394.27 Counter 0 counted RETIRED_INSNS events (Retired instructions (includes exceptions, interrupts, resyncs)) with a unit mask of 0x00 (No unit mask) count 697000 vma samples % symbol name c02c0ea0 5113 9.07075 fn_hash_lookup c0293970 3264 5.79052 ip_route_input_slow c028ef90 2734 4.85027 nf_iterate c028f280 2525 4.47949 nf_hook_slow c02924b0 2127 3.77342 rt_intern_hash c0222330 2125 3.76987 tg3_start_xmit c02becc0 1755 3.11347 fib_validate_source c0290020 1684 2.98751 pfifo_fast_dequeue c0296220 1531 2.71608 ip_rcv_finish c0135230 1449 2.57061 kmem_cache_free c0134ff0 1431 2.53867 free_block c0221710 1369 2.42868 tg3_rx c0295cb0 1350 2.39498 ip_rcv c0135170 1304 2.31337 kmem_cache_alloc c02941a0 1258 2.23176 ip_route_input c028f920 1255 2.22644 eth_header c0134e20 1148 2.03662 cache_alloc_refill c0291b70 1104 1.95856 rt_hash_code c02886a0 1082 1.91953 netif_receive_skb c01351b0 983 1.7439 __kmalloc c028b610 923 1.63745 neigh_lookup c02c0050 914 1.62149 fib_semantic_match c029a660 857 1.52037 ip_finish_output2 c028c600 829 1.47069 neigh_resolve_output c01adc80 766 1.35893 memcpy c0135270 743 1.31812 kfree c0297000 741 1.31458 ip_forward c0284620 686 1.217 alloc_skb c02b9730 666 1.18152 inet_select_addr c028fa90 663 1.1762 eth_type_trans c0128a00 649 1.15136 call_rcu c0297240 623 1.10524 ip_forward_finish c028af00 620 1.09991 dst_alloc c0288160 597 1.05911 dev_queue_xmit c028ffa0 570 1.01121 pfifo_fast_enqueue c028b030 486 0.862191 dst_destroy c0292260 485 0.860417 rt_garbage_collect c028fcb0 472 0.837355 qdisc_restart c0221350 467 0.828484 tg3_tx c028c480 463 0.821388 neigh_hh_init c02215c0 455 0.807196 tg3_recycle_rx c02d0f70 447 0.793003 ipv4_sabotage_out c0298580 443 0.785907 ip_finish_output c011f080 430 0.762844 local_bh_enable c010fc40 358 0.635112 do_gettimeofday c0284860 345 0.612049 __kfree_skb size IN: hit tot mc no_rt bcast madst masrc OUT: hit tot mc GC: tot ignored goal_miss ovrf 22910 10 158190 0 0 0 0 0 0 0 0 158190 158188 1 0 20590 10 158330 0 0 0 0 0 0 0 0 158330 158328 1 0 20515 14 158306 0 0 0 0 0 0 0 0 158306 158304 1 0 21000 4 158964 0 0 0 0 0 0 0 0 158964 158962 1 0 21631 8 159300 0 0 0 0 0 0 0 0 159300 159298 0 0 20329 13 160059 0 0 0 0 0 0 0 0 160059 160057 1 0 22995 7 157441 0 0 0 0 0 0 0 0 157441 157439 1 0 22418 9 156831 0 0 0 0 0 0 0 0 156831 156829 1 0 22417 11 157321 0 0 0 0 0 0 0 0 157321 157319 1 0 21339 6 157898 0 0 0 0 0 0 0 0 157898 157896 0 0 22562 10 157734 0 0 0 0 0 0 0 0 157734 157732 1 0 20488 12 159496 0 0 0 0 0 0 0 0 159496 159493 1 0 22527 10 157674 0 0 0 0 0 0 0 0 157674 157672 1 0 21992 7 156729 0 0 0 0 0 0 0 0 156729 156727 0 0 21372 10 157106 0 0 0 0 0 0 0 0 157106 157104 1 0 22950 10 156402 0 0 0 0 0 0 0 0 156402 156400 2 0 20471 11 157057 0 0 0 0 0 0 0 0 157057 157055 1 0 20864 13 159082 0 0 0 0 0 0 0 0 159082 159080 0 0 22416 10 157658 0 0 0 0 0 0 0 0 157658 157656 1 0 22659 8 157348 0 0 0 0 0 0 0 0 157348 157346 1 0 Simon- [ Simon Kirby ][ Network Operations ] [ sim@netnation.com ][ NetNation Communications Inc. ] [ Opinions expressed are not necessarily those of my employer. ] - : 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
