linux/include
Eric Dumazet 79ffef1fe2 tcp: avoid wakeups for pure ACK
TCP prequeue mechanism purpose is to let incoming packets
being processed by the thread currently blocked in tcp_recvmsg(),
instead of behalf of the softirq handler, to better adapt flow
control on receiver host capacity to schedule the consumer.

But in typical request/answer workloads, we send request, then
block to receive the answer. And before the actual answer, TCP
stack receives the ACK packets acknowledging the request.

Processing pure ACK on behalf of the thread blocked in tcp_recvmsg()
is a waste of resources, as thread has to immediately sleep again
because it got no payload.

This patch avoids the extra context switches and scheduler overhead.

Before patch :

a:~# echo 0 >/proc/sys/net/ipv4/tcp_low_latency
a:~# perf stat ./super_netperf 300 -t TCP_RR -l 10 -H 7.7.7.84 -- -r 8k,8k
231676

 Performance counter stats for './super_netperf 300 -t TCP_RR -l 10 -H 7.7.7.84 -- -r 8k,8k':

     116251.501765 task-clock                #   11.369 CPUs utilized
         5,025,463 context-switches          #    0.043 M/sec
         1,074,511 CPU-migrations            #    0.009 M/sec
           216,923 page-faults               #    0.002 M/sec
   311,636,972,396 cycles                    #    2.681 GHz
   260,507,138,069 stalled-cycles-frontend   #   83.59% frontend cycles idle
   155,590,092,840 stalled-cycles-backend    #   49.93% backend  cycles idle
   100,101,255,411 instructions              #    0.32  insns per cycle
                                             #    2.60  stalled cycles per insn
    16,535,930,999 branches                  #  142.243 M/sec
       646,483,591 branch-misses             #    3.91% of all branches

      10.225482774 seconds time elapsed

After patch :

a:~# echo 0 >/proc/sys/net/ipv4/tcp_low_latency
a:~# perf stat ./super_netperf 300 -t TCP_RR -l 10 -H 7.7.7.84 -- -r 8k,8k
233297

 Performance counter stats for './super_netperf 300 -t TCP_RR -l 10 -H 7.7.7.84 -- -r 8k,8k':

      91084.870855 task-clock                #    8.887 CPUs utilized
         2,485,916 context-switches          #    0.027 M/sec
           815,520 CPU-migrations            #    0.009 M/sec
           216,932 page-faults               #    0.002 M/sec
   245,195,022,629 cycles                    #    2.692 GHz
   202,635,777,041 stalled-cycles-frontend   #   82.64% frontend cycles idle
   124,280,372,407 stalled-cycles-backend    #   50.69% backend  cycles idle
    83,457,289,618 instructions              #    0.34  insns per cycle
                                             #    2.43  stalled cycles per insn
    13,431,472,361 branches                  #  147.461 M/sec
       504,470,665 branch-misses             #    3.76% of all branches

      10.249594448 seconds time elapsed

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Tom Herbert <therbert@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Cc: Andi Kleen <ak@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-02-28 15:37:29 -05:00
..
acpi PCI changes for the v3.9 merge window: 2013-02-25 21:18:18 -08:00
asm-generic GPIO changes for Linux 3.9 2013-02-26 09:35:29 -08:00
clocksource arm: arch_timer: add missing inline in stub function 2013-02-11 15:16:05 -08:00
crypto
drm drm: Add HDMI infoframe helpers 2013-02-22 08:20:10 +01:00
keys
linux Merge branch 'master' of git://1984.lsi.us.es/nf 2013-02-26 17:24:26 -05:00
math-emu
media [media] media: ov7670: Add possibility to disable pixclk during hblank 2013-02-08 14:35:06 -02:00
memory
misc
net tcp: avoid wakeups for pure ACK 2013-02-28 15:37:29 -05:00
pcmcia
ras
rdma IB/core: Add "type 2" memory windows support 2013-02-21 11:51:45 -08:00
rxrpc
scsi [SCSI] remove can_power_off flag from scsi_device 2013-01-25 15:36:50 -05:00
sound ASoC: Final updates for v3.9 2013-02-16 15:48:48 +01:00
target target: Rename spc_get_write_same_sectors -> sbc_get_write_same_sectors 2013-02-23 12:46:14 -08:00
trace Merge tag 'kvm-3.9-1' of git://git.kernel.org/pub/scm/virt/kvm/kvm 2013-02-24 13:07:18 -08:00
uapi Main batch of InfiniBand/RDMA changes for 3.9: 2013-02-26 11:41:08 -08:00
video Merge branch 'drm-next' of git://people.freedesktop.org/~airlied/linux 2013-02-25 16:46:44 -08:00
xen xen: event channel arrays are xen_ulong_t and not unsigned long 2013-02-20 08:45:07 -05:00
Kbuild