linux-stable/net
Eric Dumazet 7433819a1e tcp: do not create inetpeer on SYNACK message
Another problem on SYNFLOOD/DDOS attack is the inetpeer cache getting
larger and larger, using lots of memory and cpu time.

tcp_v4_send_synack()
->inet_csk_route_req()
 ->ip_route_output_flow()
  ->rt_set_nexthop()
   ->rt_init_metrics()
    ->inet_getpeer( create = true)

This is a side effect of commit a4daad6b09 (net: Pre-COW metrics for
TCP) added in 2.6.39

Possible solution :

Instruct inet_csk_route_req() to remove FLOWI_FLAG_PRECOW_METRICS

Before patch :

# grep peer /proc/slabinfo
inet_peer_cache   4175430 4175430    192   42    2 : tunables    0    0    0 : slabdata  99415  99415      0

Samples: 41K of event 'cycles', Event count (approx.): 30716565122
+  20,24%      ksoftirqd/0  [kernel.kallsyms]           [k] inet_getpeer
+   8,19%      ksoftirqd/0  [kernel.kallsyms]           [k] peer_avl_rebalance.isra.1
+   4,81%      ksoftirqd/0  [kernel.kallsyms]           [k] sha_transform
+   3,64%      ksoftirqd/0  [kernel.kallsyms]           [k] fib_table_lookup
+   2,36%      ksoftirqd/0  [ixgbe]                     [k] ixgbe_poll
+   2,16%      ksoftirqd/0  [kernel.kallsyms]           [k] __ip_route_output_key
+   2,11%      ksoftirqd/0  [kernel.kallsyms]           [k] kernel_map_pages
+   2,11%      ksoftirqd/0  [kernel.kallsyms]           [k] ip_route_input_common
+   2,01%      ksoftirqd/0  [kernel.kallsyms]           [k] __inet_lookup_established
+   1,83%      ksoftirqd/0  [kernel.kallsyms]           [k] md5_transform
+   1,75%      ksoftirqd/0  [kernel.kallsyms]           [k] check_leaf.isra.9
+   1,49%      ksoftirqd/0  [kernel.kallsyms]           [k] ipt_do_table
+   1,46%      ksoftirqd/0  [kernel.kallsyms]           [k] hrtimer_interrupt
+   1,45%      ksoftirqd/0  [kernel.kallsyms]           [k] kmem_cache_alloc
+   1,29%      ksoftirqd/0  [kernel.kallsyms]           [k] inet_csk_search_req
+   1,29%      ksoftirqd/0  [kernel.kallsyms]           [k] __netif_receive_skb
+   1,16%      ksoftirqd/0  [kernel.kallsyms]           [k] copy_user_generic_string
+   1,15%      ksoftirqd/0  [kernel.kallsyms]           [k] kmem_cache_free
+   1,02%      ksoftirqd/0  [kernel.kallsyms]           [k] tcp_make_synack
+   0,93%      ksoftirqd/0  [kernel.kallsyms]           [k] _raw_spin_lock_bh
+   0,87%      ksoftirqd/0  [kernel.kallsyms]           [k] __call_rcu
+   0,84%      ksoftirqd/0  [kernel.kallsyms]           [k] rt_garbage_collect
+   0,84%      ksoftirqd/0  [kernel.kallsyms]           [k] fib_rules_lookup

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Hans Schillstrom <hans.schillstrom@ericsson.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-01 14:22:11 -04:00
..
9p Autogenerated GPG tag for Rusty D1ADB8F1: 15EE 8D6C AB0E 7F0C F999 BFCB D920 0E6C D1AD B8F1 2012-05-21 20:20:23 -07:00
802 tokenring: delete all remaining driver support 2012-05-15 20:23:16 -04:00
8021q Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2012-05-16 22:17:37 -04:00
appletalk net: Convert all sysctl registrations to register_net_sysctl 2012-04-20 21:22:30 -04:00
atm Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial 2012-05-22 19:22:50 -07:00
ax25 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2012-04-23 23:15:17 -04:00
batman-adv batman-adv: unset the TT_CLIENT_PENDING flag if the new local entry already exists 2012-05-14 09:05:08 +02:00
bluetooth Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2012-05-24 11:54:29 -07:00
bridge ipv6: correct the ipv6 option name - Pad0 to Pad1 2012-05-17 15:49:51 -04:00
caif Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial 2012-05-22 19:22:50 -07:00
can can: fix sparse warning for cgw_list 2012-04-16 21:08:18 +02:00
ceph Merge git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client 2012-05-30 11:17:19 -07:00
core net: sock: validate data_len before allocating skb in sock_alloc_send_pskb() 2012-05-31 18:22:45 -04:00
dcb net: dcb: add CEE notify calls 2012-04-25 19:47:17 -04:00
dccp net: include/net/sock.h cleanup 2012-05-17 04:50:21 -04:00
decnet net: Convert net_ratelimit uses to net_<level>_ratelimited 2012-05-15 13:45:03 -04:00
dns_resolver Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security 2012-05-21 20:27:36 -07:00
dsa dsa: Convert compare_ether_addr to ether_addr_equal 2012-05-09 20:49:19 -04:00
ethernet net, drivers/net: Convert compare_ether_addr_64bits to ether_addr_equal_64bits 2012-05-10 23:33:01 -04:00
ieee802154 ieee802154: interface type to be added 2012-05-16 15:17:08 -04:00
ipv4 tcp: do not create inetpeer on SYNACK message 2012-06-01 14:22:11 -04:00
ipv6 ipv6: fix incorrect ipsec fragment 2012-05-27 01:11:22 -04:00
ipx ipx: Remove spurious NULL checking in ipx_ioctl(). 2012-05-19 00:51:04 -04:00
irda net: Convert all sysctl registrations to register_net_sysctl 2012-04-20 21:22:30 -04:00
iucv Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux 2012-03-22 18:15:32 -07:00
key net: cleanup unsigned to unsigned int 2012-04-15 12:44:40 -04:00
l2tp genetlink: Build a generic netlink family module alias 2012-05-29 22:33:56 -04:00
lapb lapb: Neaten debugging 2012-05-17 18:45:20 -04:00
llc net: include/net/sock.h cleanup 2012-05-17 04:50:21 -04:00
mac80211 mac80211: fix ADDBA declined after suspend with wowlan 2012-05-29 13:45:24 -04:00
mac802154 mac802154: monitor device support 2012-05-16 15:17:08 -04:00
netfilter Merge branch 'for-3.5' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu 2012-05-22 17:37:47 -07:00
netlabel netlabel: use GFP flags from caller instead of GFP_ATOMIC 2012-03-22 19:29:57 -04:00
netlink genetlink: Build a generic netlink family module alias 2012-05-29 22:33:56 -04:00
netrom net: Convert all sysctl registrations to register_net_sysctl 2012-04-20 21:22:30 -04:00
nfc NFC: Queue I frame fragments to the LLCP sockets queue tail 2012-05-16 13:08:14 -04:00
openvswitch Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2012-05-16 22:17:37 -04:00
packet af_packet: packet_getsockopt() cleanup 2012-04-21 16:36:42 -04:00
phonet net: Convert all sysctl registrations to register_net_sysctl 2012-04-20 21:22:30 -04:00
rds rds_rdma: don't assume infiniband device is PCI 2012-05-29 17:30:07 -04:00
rfkill device.h: cleanup users outside of linux/include (C files) 2012-03-11 14:27:37 -04:00
rose net: Convert all sysctl registrations to register_net_sysctl 2012-04-20 21:22:30 -04:00
rxrpc net: cleanup unsigned to unsigned int 2012-04-15 12:44:40 -04:00
sched Merge branch 'for-3.5' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup 2012-05-22 17:40:19 -07:00
sctp Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net 2012-05-16 22:17:37 -04:00
sunrpc NFS client updates for Linux 3.5 2012-05-29 10:43:51 -07:00
tipc tipc: compress out gratuitous extra carriage returns 2012-04-30 15:53:56 -04:00
unix net: sock_diag_handler structs can be const 2012-04-25 20:46:59 -04:00
wanrouter net/wanrouter: Deprecate and schedule for removal 2012-05-24 16:22:53 -04:00
wimax net: cleanup unsigned to unsigned int 2012-04-15 12:44:40 -04:00
wireless Merge git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-next 2012-05-22 15:18:06 -04:00
x25 net: add a limit parameter to sk_add_backlog() 2012-04-23 22:28:28 -04:00
xfrm ipv6: fix incorrect ipsec fragment 2012-05-27 01:11:22 -04:00
compat.c Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security 2012-05-21 20:27:36 -07:00
Kconfig net: drop NET dependency from HAVE_BPF_JIT 2012-05-21 12:50:12 -07:00
Makefile econet: remove ancient bug ridden protocol 2012-05-18 01:35:08 -04:00
nonet.c
socket.c Merge branch 'for-3.5' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu 2012-05-22 17:37:47 -07:00
sysctl_net.c net: delete all instances of special processing for token ring 2012-05-15 20:14:35 -04:00