While doing oprofile tests I noticed two loops are not properly unrolled by gcc

Using a hand coded unrolled loop provides nice speedup : ipt_do_table
credited of 2.52 % of cpu instead of 3.29 % in tbench.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>

Kernel module providing implementation of LED netfilter target.  Each
instance of the target appears as a led-trigger device, which can be
associated with one or more LEDs in /sys/class/leds/
Signed-off-by: Adam Nielsen <a.nielsen@shikadi.net>
Acked-by: Richard Purdie <rpurdie@linux.intel.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>

get_random_bytes() is sometimes called with a hard coded size assumption
of an integer. This could not be true for next centuries. This patch
replace it with a compile time statement.
Signed-off-by: Hagen Paul Pfeifer <hagen@jauu.net>
Signed-off-by: Patrick McHardy <kaber@trash.net>

Table size is defined as unsigned, wheres the table maximum size is
defined as a signed integer. The calculation of max is 8 or 4,
multiplied the table size. Therefore the max value is aligned to
unsigned.
Signed-off-by: Hagen Paul Pfeifer <hagen@jauu.net>
Signed-off-by: Patrick McHardy <kaber@trash.net>

The reader/writer lock in ip_tables is acquired in the critical path of
processing packets and is one of the reasons just loading iptables can cause
a 20% performance loss. The rwlock serves two functions:

1) it prevents changes to table state (xt_replace) while table is in use.
   This is now handled by doing rcu on the xt_table. When table is
   replaced, the new table(s) are put in and the old one table(s) are freed
   after RCU period.

2) it provides synchronization when accesing the counter values.
   This is now handled by swapping in new table_info entries for each cpu
   then summing the old values, and putting the result back onto one
   cpu.  On a busy system it may cause sampling to occur at different
   times on each cpu, but no packet/byte counts are lost in the process.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>

Sucessfully tested on my dual quad core machine too, but iptables only (no ipv6 here)
BTW, my new "tbench 8" result is 2450 MB/s, (it was 2150 MB/s not so long ago)
Acked-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>

ip6_tables netfilter module can use an ifname_compare() helper
so that two loops are unfolded.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>

xt_physdev netfilter module can use an ifname_compare() helper
so that two loops are unfolded.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>

Concern has been expressed about the changing Kconfig options.
Provide the old options that forward-select.
Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>

1) physdev_mt() incorrectly assumes nulldevname[] is aligned on an int

2) It also uses word comparisons, while it could use long word ones.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>

Suggested by: James King <t.james.king@gmail.com>

Similarly to commit c9fd4968, merge
TTL and HL. Since HL does not depend on any IPv6-specific function,
no new module dependencies would arise.

With slight adjustments to the Kconfig help text.
Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>

x86 and powerpc can perform long word accesses in an efficient maner.
We can use this to unroll two loops in arp_packet_match(), to
perform arithmetic on long words instead of bytes. This is a win
on x86_64 for example.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>

This patch adds a logging message for invalid new icmpv6 packet.
Signed-off-by: Eric Leblond <eric@inl.fr>
Signed-off-by: Patrick McHardy <kaber@trash.net>

The initialization of the lock element is not needed
since the lock is always initialized in ebt_register_table.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>

Later patches change the locking on xt_table and the initialization of
the lock element is not needed since the lock is always initialized in
xt_table_register anyway.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>

Change to proper type on private pointer rather than anonymous void.
Keep active elements on same cache line.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>

0 is used by Hop-by-hop header and so this may cause confusion.
255 is stated as 'Reserved' by IANA.
Signed-off-by: Christoph Paasch <christoph.paasch@student.uclouvain.be>
Signed-off-by: Patrick McHardy <kaber@trash.net>

Impact: Remove the write-only field 'current_ring_status' completely.

Fix this sparse warnings:
  drivers/net/tokenring/smctr.c:4410:52: warning: cast truncates bits from constant value (100 becomes 0)
  drivers/net/tokenring/smctr.c:4415:52: warning: cast truncates bits from constant value (400 becomes 0)
  drivers/net/tokenring/smctr.c:4420:52: warning: cast truncates bits from constant value (800 becomes 0)
  drivers/net/tokenring/smctr.c:4425:52: warning: cast truncates bits from constant value (1000 becomes 0)
  drivers/net/tokenring/smctr.c:4430:52: warning: cast truncates bits from constant value (2000 becomes 0)
  drivers/net/tokenring/smctr.c:4435:52: warning: cast truncates bits from constant value (4000 becomes 0)
  drivers/net/tokenring/smctr.c:4440:52: warning: cast truncates bits from constant value (8000 becomes 0)
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Fix sparse warning:
  drivers/net/wireless/prism54/isl_ioctl.c:2997:32: warning: incorrect type in argument 2 (different signedness)
  drivers/net/wireless/prism54/oid_mgt.c:712:42: warning: incorrect type in argument 2 (different signedness)
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Fix this sparse warnings:
  drivers/net/wireless/ipw2x00/ipw2100.c:1930:43: warning: incorrect type in argument 4 (different signedness)
  drivers/net/wireless/ipw2x00/ipw2100.c:1938:76: warning: incorrect type in argument 4 (different signedness)
  drivers/net/wireless/ipw2x00/ipw2100.c:1946:67: warning: incorrect type in argument 4 (different signedness)
  drivers/net/wireless/ipw2x00/ipw2100.c:1953:72: warning: incorrect type in argument 4 (different signedness)
  drivers/net/wireless/ipw2x00/ipw2100.c:4071:73: warning: incorrect type in argument 4 (different signedness)
  drivers/net/wireless/ipw2x00/ipw2100.c:4078:43: warning: incorrect type in argument 4 (different signedness)
  drivers/net/wireless/ipw2x00/ipw2100.c:4084:67: warning: incorrect type in argument 4 (different signedness)
  drivers/net/wireless/ipw2x00/ipw2100.c:7141:73: warning: incorrect type in argument 4 (different signedness)
  drivers/net/wireless/ipw2x00/ipw2100.c:8317:48: warning: incorrect type in argument 4 (different signedness)
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Fix this compilation warning:
  drivers/net/wireless/b43/main.c: In function 'b43_print_fw_helptext':
  drivers/net/wireless/b43/main.c:1971: warning: format not a string literal and no format arguments
  drivers/net/wireless/b43/main.c:1973: warning: format not a string literal and no format arguments
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Make symbols static.

Fix this sparse warnings:
  drivers/net/wireless/ath9k/eeprom.c:1343:5: warning: symbol 'ath9k_hw_4k_get_spur_channel' was not declared. Should it be static?
  drivers/net/wireless/ath9k/eeprom.c:1372:19: warning: symbol 'eep_4k_ops' was not declared. Should it be static?
  drivers/net/wireless/ath9k/eeprom.c:2649:5: warning: symbol 'ath9k_hw_def_get_spur_channel' was not declared. Should it be static?
  drivers/net/wireless/ath9k/eeprom.c:2678:19: warning: symbol 'eep_def_ops' was not declared. Should it be static?
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Fix this sparse warnings:
  drivers/net/wireless/airo.c:2102:21: warning: incorrect type in initializer (different signedness)
  drivers/net/wireless/airo.c:2126:21: warning: incorrect type in initializer (different signedness)
  drivers/net/wireless/airo.c:2167:21: warning: incorrect type in initializer (different signedness)
  drivers/net/wireless/airo.c:2191:21: warning: incorrect type in initializer (different signedness)
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact:
  Move variable closer to usage resp.
  remove redundant variables resp.
  rename function scope variable.

Fix this sparse warnings:
  drivers/net/wireless/airo.c:3840:29: warning: symbol 'i' shadows an earlier one
  drivers/net/wireless/airo.c:3751:13: originally declared here
  drivers/net/wireless/airo.c:3847:29: warning: symbol 'i' shadows an earlier one
  drivers/net/wireless/airo.c:3751:13: originally declared here
  drivers/net/wireless/airo.c:3861:21: warning: symbol 'i' shadows an earlier one
  drivers/net/wireless/airo.c:3751:13: originally declared here
  drivers/net/wireless/wavelan.c:43:13: warning: symbol 'irq' shadows an earlier one
  drivers/net/wireless/wavelan.p.h:692:17: originally declared here
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Attribute function with __acquires(...) resp. __releases(...).

Fix this sparse warnings:
  drivers/net/wireless/strip.c:955:21: warning: context imbalance in 'strip_seq_start' - wrong count at exit
  drivers/net/wireless/strip.c:975:13: warning: context imbalance in 'strip_seq_stop' - unexpected unlock
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Rename function scope variable and while being at it fix some
coding style issues.

Fix this sparse warning:
  drivers/net/wan/sbni.c:1526:14: warning: symbol 'num' shadows an earlier one
  drivers/net/wan/sbni.c:160:13: originally declared here
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Use 'static const char[]' instead of 'static char[]'.

Fix this compilation warnings:
  drivers/net/wan/lapbether.c: In function 'lapbeth_init_driver':
  drivers/net/wan/lapbether.c:441: warning: format not a string literal and no format arguments
  drivers/net/wan/z85230.c: In function 'z85230_init_driver':
  drivers/net/wan/z85230.c:1782: warning: format not a string literal and no format arguments
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Make symbols static.

Fix this sparse warnings:
  drivers/net/usb/hso.c:1249:6: warning: symbol 'hso_unthrottle_tasklet' was not declared. Should it be static?
  drivers/net/usb/hso.c:1268:6: warning: symbol 'hso_unthrottle_workfunc' was not declared. Should it be static?
  drivers/net/usb/hso.c:1466:5: warning: symbol 'tiocmget_submit_urb' was not declared. Should it be static?
  drivers/net/usb/smsc95xx.c:62:5: warning: symbol 'turbo_mode' was not declared. Should it be static?
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Rename a local variable resp. remove a redundant variable and
while being at it use more unform loop constructs.

Fix this sparse warning:
  drivers/net/tulip/de4x5.c:3944:13: warning: symbol 'i' shadows an earlier one
  drivers/net/tulip/de4x5.c:3938:9: originally declared here
  drivers/net/tulip/media.c:72:21: warning: symbol 'i' shadows an earlier one
  drivers/net/tulip/media.c:54:13: originally declared here
  drivers/net/tulip/media.c:134:21: warning: symbol 'i' shadows an earlier one
  drivers/net/tulip/media.c:117:13: originally declared here
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Use 'static const char[]' instead of 'static char[]' and while
being at it add a KERN_INFO prefix.

Fix this warning:
  drivers/net/tulip/de4x5.c: In function 'de4x5_hw_init':
  drivers/net/tulip/de4x5.c:1268: warning: format not a string literal and no format arguments
  drivers/net/tulip/winbond-840.c: In function 'w840_init':
  drivers/net/tulip/winbond-840.c:1666: warning: format not a string literal and no format arguments
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Remove redundant variables.

Fix this sparse warnings:
  drivers/net/tokenring/lanstreamer.c:619:21: warning: symbol 'i' shadows an earlier one
  drivers/net/tokenring/lanstreamer.c:589:13: originally declared here
  drivers/net/tokenring/madgemc.c:695:34: warning: symbol 'tp' shadows an earlier one
  drivers/net/tokenring/madgemc.c:689:26: originally declared here
  drivers/net/tokenring/olympic.c:702:21: warning: symbol 'i' shadows an earlier one
  drivers/net/tokenring/olympic.c:440:13: originally declared here
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Use 'static const char[]' instead of 'static char[]'.

Fix this warnings:
  drivers/net/tokenring/smctr.c:3644: warning: format not a string literal and no format arguments
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Move function declarations to header file.

Fix this sparse warnings:
  drivers/net/skfp/cfm.c:146:6: warning: symbol 'all_selection_criteria' was not declared. Should it be static?
  drivers/net/skfp/drvfbi.c:186:6: warning: symbol 'mac1_irq' was not declared. Should it be static?
  drivers/net/skfp/drvfbi.c:284:6: warning: symbol 'read_address' was not declared. Should it be static?
  drivers/net/skfp/drvfbi.c:323:6: warning: symbol 'init_board' was not declared. Should it be static?
  drivers/net/skfp/fplustm.c:72:24: warning: symbol 'fddi_broadcast' was not declared. Should it be static?
  drivers/net/skfp/fplustm.c:679:6: warning: symbol 'mac2_irq' was not declared. Should it be static?
  drivers/net/skfp/fplustm.c:805:6: warning: symbol 'mac3_irq' was not declared. Should it be static?
  drivers/net/skfp/fplustm.c:856:5: warning: symbol 'init_fplus' was not declared. Should it be static?
  drivers/net/skfp/pcmplc.c:404:6: warning: symbol 'init_plc' was not declared. Should it be static?
  drivers/net/skfp/pcmplc.c:1592:5: warning: symbol 'pcm_status_twisted' was not declared. Should it be static?
  drivers/net/skfp/smtinit.c:68:5: warning: symbol 'init_smt' was not declared. Should it be static?
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Move variable declaration as close to their usage as possible.

Fix this sparse warning:
  drivers/net/sfc/tenxpress.c:589:22: warning: symbol 'link_ok' shadows an earlier one
  drivers/net/sfc/tenxpress.c:575:30: originally declared here
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Acked-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Use 'static const char[]' instead of 'static char[]' and while
being at it prefix the message with KERN_INFO.

Fix this warnings:
  drivers/net/pcmcia/axnet_cs.c:1716: warning: format not a string literal and no format arguments
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Make symbol static.

Fix this sparse warning:
  drivers/net/ixgbe/ixgbe_82598.c:59:5: warning: symbol 'ixgbe_get_pcie_msix_count_82598' was not declared. Should it be static?
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Acked-by: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Fix this sparse warnings:
  drivers/net/hamradio/hdlcdrv.c:274:34: warning: incorrect type in argument 2 (different signedness)
  drivers/net/hamradio/hdlcdrv.c:279:47: warning: incorrect type in argument 2 (different signedness)
  drivers/net/hamradio/hdlcdrv.c:288:39: warning: incorrect type in argument 2 (different signedness)
  drivers/net/hamradio/hdlcdrv.c:300:47: warning: incorrect type in argument 2 (different signedness)
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Rename inner scrope variable.

Fix this sparse warning:
  drivers/net/hamradio/yam.c:856:35: warning: symbol 'dev' shadows an earlier one
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Impact: Attribute functions with __acquires(...) resp. __releases(...).

Fix this sparse warnings:
  drivers/net/hamradio/bpqether.c:387:13: warning: context imbalance in 'bpq_seq_start' - wrong count at exit
  drivers/net/hamradio/bpqether.c:419:13: warning: context imbalance in 'bpq_seq_stop' - unexpected unlock
Signed-off-by: Hannes Eder <hannes@hanneseder.net>
Signed-off-by: David S. Miller <davem@davemloft.net>