linux/include/net
Zhang Yanmin f1dd9c379c [NET]: Fix tbench regression in 2.6.25-rc1
Comparing with kernel 2.6.24, tbench result has regression with
2.6.25-rc1.

1) On 2 quad-core processor stoakley: 4%.
2) On 4 quad-core processor tigerton: more than 30%.

bisect located below patch.

b4ce92775c is first bad commit
commit b4ce92775c
Author: Herbert Xu <herbert@gondor.apana.org.au>
Date:   Tue Nov 13 21:33:32 2007 -0800

    [IPV6]: Move nfheader_len into rt6_info

    The dst member nfheader_len is only used by IPv6.  It's also currently
    creating a rather ugly alignment hole in struct dst.  Therefore this patch
    moves it from there into struct rt6_info.

Above patch changes the cache line alignment, especially member
__refcnt. I did a testing by adding 2 unsigned long pading before
lastuse, so the 3 members, lastuse/__refcnt/__use, are moved to next
cache line. The performance is recovered.

I created a patch to rearrange the members in struct dst_entry.

With Eric and Valdis Kletnieks's suggestion, I made finer arrangement.

1) Move tclassid under ops in case CONFIG_NET_CLS_ROUTE=y. So
   sizeof(dst_entry)=200 no matter if CONFIG_NET_CLS_ROUTE=y/n. I
   tested many patches on my 16-core tigerton by moving tclassid to
   different place. It looks like tclassid could also have impact on
   performance.  If moving tclassid before metrics, or just don't move
   tclassid, the performance isn't good. So I move it behind metrics.

2) Add comments before __refcnt.

On 16-core tigerton:

If CONFIG_NET_CLS_ROUTE=y, the result with below patch is about 18%
better than the one without the patch;

If CONFIG_NET_CLS_ROUTE=n, the result with below patch is about 30%
better than the one without the patch.

With 32bit 2.6.25-rc1 on 8-core stoakley, the new patch doesn't
introduce regression.

Thank Eric, Valdis, and David!

Signed-off-by: Zhang Yanmin <yanmin.zhang@intel.com>
Acked-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-03-12 22:52:37 -07:00
..
9p 9p: add support for sticky bit 2008-02-06 19:25:06 -06:00
bluetooth bluetooth: Make hci_sock_cleanup() return void 2008-03-05 18:47:03 -08:00
irda irda: replace __inline with inline 2008-03-05 18:37:16 -08:00
iucv [AF_IUCV]: postpone receival of iucv-packets 2007-10-10 16:54:51 -07:00
netfilter [NETFILTER]: nf_conntrack: replace horrible hack with ksize() 2008-03-10 16:43:41 -07:00
netns [NETFILTER]: arp_tables: per-netns arp_tables FILTER 2008-01-31 19:27:41 -08:00
sctp [SCTP]: Update AUTH structures to match declarations in draft-16. 2008-02-28 16:45:04 -05:00
tc_act [PKT_SCHED]: Add stateless NAT 2007-10-10 16:53:11 -07:00
tipc [TIPC]: Kill unused static inline (x5) 2008-02-07 18:17:13 -08:00
act_api.h [NET_SCHED]: act_api: use PTR_ERR in tcf_action_init/tcf_action_get 2008-01-28 15:11:17 -08:00
addrconf.h [IPV4]: Enable use of 240/4 address space. 2008-01-28 15:08:44 -08:00
af_rxrpc.h [AF_RXRPC]: Add an interface to the AF_RXRPC module for the AFS filesystem to use 2007-04-26 15:50:17 -07:00
af_unix.h [AF_UNIX]: Remove unused declaration of sysctl_unix_max_dgram_qlen. 2008-01-28 14:57:13 -08:00
ah.h [IPSEC]: Get rid of ipv6_{auth,esp,comp}_hdr 2007-10-10 16:55:55 -07:00
arp.h [NETFILTER]: ebtables: remove casts, use consts 2008-01-31 19:27:33 -08:00
atmclip.h [ATM]: Annotations. 2006-12-02 21:22:55 -08:00
ax25.h [AX25] ax25_ds_timer: use mod_timer instead of add_timer 2008-02-12 17:53:34 -08:00
ax88796.h ax88796: add 93cx6 eeprom support 2007-10-10 16:53:56 -07:00
cfg80211.h cfg80211/nl80211: implement station attribute retrieval 2008-01-28 14:59:52 -08:00
checksum.h [NET]: Move netfilter checksum helpers to net/core/utils.c 2008-01-28 14:55:14 -08:00
cipso_ipv4.h [NetLabel]: consolidate the struct socket/sock handling to just struct sock 2007-06-08 13:33:09 -07:00
compat.h [NET]: Introduce SIOCGSTAMPNS ioctl to get timestamps with nanosec resolution 2007-04-25 22:24:04 -07:00
datalink.h
dn_dev.h
dn_fib.h [DECNet]: Use rtnl registration interface 2007-04-25 22:27:12 -07:00
dn_neigh.h
dn_nsp.h
dn_route.h [NET]: Wrap netdevice hardware header creation. 2007-10-10 16:52:50 -07:00
dn.h [DECNET]: Another unnecessary net/tcp.h inclusion in net/dn.h 2007-07-10 23:02:12 -07:00
dsfield.h [NET]: Constify include/net/dsfield.h 2008-01-28 14:55:58 -08:00
dst.h [NET]: Fix tbench regression in 2.6.25-rc1 2008-03-12 22:52:37 -07:00
esp.h [IPSEC]: Use crypto_aead and authenc in ESP 2008-01-31 19:27:02 -08:00
fib_rules.h [NETNS]: Process FIB rule action in the context of the namespace. 2008-01-28 15:08:14 -08:00
flow.h [IPV4]: Remove unused multipath cached routing defintion in net/flow.h 2008-01-28 15:00:20 -08:00
gen_stats.h [NET_SCHED]: Convert packet schedulers from rtnetlink to new netlink API 2008-01-28 15:11:10 -08:00
genetlink.h [GENETLINK]: Dynamic multicast groups. 2007-07-18 15:47:52 -07:00
icmp.h [IPV4]: Add ICMPMsgStats MIB (RFC 4293) 2007-10-10 16:51:28 -07:00
ieee80211_crypt.h [PATCH] Update my email address from jkmaline@cc.hut.fi to j@w1.fi 2007-04-28 11:01:01 -04:00
ieee80211_radiotap.h [MAC80211]: Add get_unaligned to ieee80211_get_radiotap_len 2007-10-10 16:47:40 -07:00
ieee80211.h ieee80211: fix misannotations 2008-01-28 15:08:48 -08:00
ieee80211softmac_wx.h
ieee80211softmac.h [IEEE80211]: Fix softmac lockdep reports. 2007-10-10 16:52:22 -07:00
if_inet6.h [IPV6]: Reorg struct ifmcaddr6 to save some bytes 2008-02-03 04:28:54 -08:00
inet6_connection_sock.h [TCP]: Restore SKB socket owner setting in tcp_transmit_skb(). 2007-01-26 01:04:55 -08:00
inet6_hashtables.h [SOCK] proto: Add hashinfo member to struct proto 2008-02-03 04:28:52 -08:00
inet_common.h [INET]: Remove leftover prototypes from include/net/inet_common.h 2007-11-12 21:02:51 -08:00
inet_connection_sock.h [SOCK] proto: Add hashinfo member to struct proto 2008-02-03 04:28:52 -08:00
inet_ecn.h [INET]: Give outer DSCP directly to ip*_copy_dscp 2008-01-28 14:53:45 -08:00
inet_frag.h [NETNS][FRAGS]: Make the pernet subsystem for fragments. 2008-01-28 15:10:40 -08:00
inet_hashtables.h [INET]: Fix accidentally broken inet(6)_hash_connect's port offset calculations. 2008-02-05 03:14:44 -08:00
inet_sock.h [TCP]: Improve ipv4 established hash function. 2008-03-04 14:28:41 -08:00
inet_timewait_sock.h [INET_TIMEWAIT_SOCK]: Reorganize struct inet_timewait_sock to save some bytes 2008-02-03 04:28:54 -08:00
inetpeer.h [INET]: Use list_head-s in inetpeer.c 2007-11-12 21:27:28 -08:00
ip6_checksum.h [IPV6]: Dumb typo in generic csum_ipv6_magic() 2006-12-22 11:12:07 -08:00
ip6_fib.h [IPV6] Minor cleanup: remove unused definitions in net/ip6_fib.h 2008-02-07 18:11:49 -08:00
ip6_route.h [IPV6] Minor clenup: remove two unused definitions in net/ip6_route.h 2008-02-07 18:10:19 -08:00
ip6_tunnel.h [IP_TUNNEL]: Don't limit the number of tunnels with generic name explicitly. 2008-02-23 20:19:20 -08:00
ip_fib.h [IPV4]: Fix compile error building without CONFIG_FS_PROC 2008-02-05 02:54:16 -08:00
ip_vs.h [IPVS]: Switch to using ctl_paths. 2008-01-28 15:01:08 -08:00
ip.h [NETNS][FRAGS]: Make the mem counter per-namespace. 2008-01-28 15:10:36 -08:00
ipcomp.h [IPSEC]: Get rid of ipv6_{auth,esp,comp}_hdr 2007-10-10 16:55:55 -07:00
ipconfig.h
ipip.h [IPV4]: Add ip_local_out 2008-01-28 14:53:47 -08:00
ipv6.h [IPV6]: Fix sysctl compilation error. 2008-02-05 02:57:59 -08:00
ipx.h [SK_BUFF]: Introduce skb_transport_header(skb) 2007-04-25 22:25:31 -07:00
iw_handler.h [NL80211]: add netlink interface to cfg80211 2007-10-10 16:52:14 -07:00
lapb.h
llc_c_ac.h
llc_c_ev.h
llc_c_st.h
llc_conn.h [NET]: Make socket creation namespace safe. 2007-10-10 16:49:07 -07:00
llc_if.h
llc_pdu.h [SK_BUFF]: Introduce skb_network_header() 2007-04-25 22:24:59 -07:00
llc_s_ac.h
llc_s_ev.h
llc_s_st.h
llc_sap.h
llc.h
mac80211.h mac80211: add unified BSS configuration 2008-01-28 15:09:43 -08:00
mip6.h [IPV6] MIP6: Loadable module support for MIPv6. 2007-07-10 22:15:42 -07:00
ndisc.h [IPV6]: remove unused method declaration (net/ndisc.h). 2008-02-12 22:06:53 -08:00
neighbour.h [ARP]: neigh_parms_put(destroy) are essentially local to core/neighbour.c. 2008-01-28 15:02:11 -08:00
net_namespace.h [NET]: Make /proc/net a symlink on /proc/self/net (v3) 2008-03-07 11:08:40 -08:00
netdma.h
netevent.h [NET]: Remove unnecessary inclusion of dst.h 2008-01-28 14:53:38 -08:00
netlabel.h NetLabel: introduce a new kernel configuration API for NetLabel 2008-02-05 09:44:20 -08:00
netlink.h [NETLINK]: Add nla_append() 2008-01-28 15:11:09 -08:00
netrom.h [PATCH] mark struct file_operations const 1 2007-02-12 09:48:44 -08:00
nexthop.h
p8022.h
pkt_cls.h [NET_SCHED]: Constify struct tcf_ext_map 2008-01-31 19:28:34 -08:00
pkt_sched.h [NET_SCHED]: Convert packet schedulers from rtnetlink to new netlink API 2008-01-28 15:11:10 -08:00
protocol.h [IPV6]: make inet6_register_protosw to return an error code 2008-01-28 14:57:12 -08:00
psnap.h
raw.h [RAW]: Wrong content of the /proc/net/raw6. 2008-01-31 19:27:26 -08:00
rawv6.h [IPv6] RAW: Compact the API for the kernel 2008-01-28 14:54:29 -08:00
red.h [NET_SCHED]: turn PSCHED_GET_TIME into inline function 2007-04-25 22:27:55 -07:00
request_sock.h [INET]: Fix potential kfree on vmalloc-ed area of request_sock_queue 2007-11-15 02:57:06 -08:00
rose.h [ROSE]: Fix rose.ko oops on unload 2007-10-07 23:44:17 -07:00
route.h [IPV4] route cache: Introduce rt_genid for smooth cache invalidation 2008-01-31 19:28:27 -08:00
rtnetlink.h [NET]: Make the device list and device lookups per namespace. 2007-10-10 16:49:10 -07:00
sch_generic.h [NET_SCHED]: Convert classifiers from rtnetlink to new netlink API 2008-01-28 15:11:11 -08:00
scm.h pid namespaces: changes to show virtual ids to user 2007-10-19 11:53:40 -07:00
slhc_vj.h
snmp.h [XFRM]: Define packet dropping statistics. 2008-01-28 14:59:38 -08:00
sock.h net: fix kernel-doc warnings in header files 2008-02-18 20:52:13 -08:00
syncppp.h
tcp_states.h
tcp.h [TCP]: Uninline tcp_is_cwnd_limited 2008-01-28 15:01:48 -08:00
timewait_sock.h [PATCH] slab: remove kmem_cache_t 2006-12-07 08:39:25 -08:00
transp_v6.h [IPV6]: make the protocol initialization to return an error code 2008-01-28 14:57:13 -08:00
udp.h [NET]: prot_inuse cleanups and optimizations 2008-01-28 15:00:36 -08:00
udplite.h [UDP]: Restore missing inDatagrams increments 2008-01-28 14:56:33 -08:00
wext.h [NET]: Make the device list and device lookups per namespace. 2007-10-10 16:49:10 -07:00
wireless.h [WIRELESS] cfg80211: New wireless config infrastructure. 2007-04-25 22:29:41 -07:00
x25.h [X.25]: Adds /proc/sys/net/x25/x25_forward to control forwarding. 2007-02-08 13:34:36 -08:00
x25device.h [SK_BUFF]: Introduce skb_reset_mac_header(skb) 2007-04-25 22:24:32 -07:00
xfrm.h [IPSEC]: Fix bogus usage of u64 on input sequence number 2008-02-12 22:50:35 -08:00