netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Hendry <andrew.hendry@gmail.com>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Jesper Dangaard Brouer <jdb@comx.dk>, netdev <netdev@vger.kernel.org>
Subject: Re: Loopback performance from kernel 2.6.12 to 2.6.37
Date: Tue, 9 Nov 2010 17:30:22 +1100	[thread overview]
Message-ID: <AANLkTikAPaU_2=wS_T3V-8xFZm-G3qutJBxY8yb0QCYL@mail.gmail.com> (raw)
In-Reply-To: <1289283797.2790.84.camel@edumazet-laptop>

most my slowdown was kmemleak left on.

After fixing its is still a lot slower than your dev system
.
# time dd if=/dev/zero bs=1M count=10000 | netcat  127.0.0.1 9999
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB) copied, 25.8182 s, 406 MB/s

real	0m25.821s
user	0m1.502s
sys	0m33.463s

------------------------------------------------------------------------------------------------------------------
   PerfTop:     241 irqs/sec  kernel:56.8%  exact:  0.0% [1000Hz
cycles],  (all, 8 CPUs)
------------------------------------------------------------------------------------------------------------------

             samples  pcnt function                    DSO
             _______ _____ ___________________________
______________________________________

             1255.00  8.7% hpet_msi_next_event
/lib/modules/2.6.37-rc1+/build/vmlinux
             1081.00  7.5% copy_user_generic_string
/lib/modules/2.6.37-rc1+/build/vmlinux
              863.00  6.0% __ticket_spin_lock
/lib/modules/2.6.37-rc1+/build/vmlinux
              498.00  3.5% do_sys_poll
/lib/modules/2.6.37-rc1+/build/vmlinux
              455.00  3.2% system_call
/lib/modules/2.6.37-rc1+/build/vmlinux
              409.00  2.8% fget_light
/lib/modules/2.6.37-rc1+/build/vmlinux
              348.00  2.4% tcp_sendmsg
/lib/modules/2.6.37-rc1+/build/vmlinux
              269.00  1.9% fsnotify
/lib/modules/2.6.37-rc1+/build/vmlinux
              258.00  1.8% _raw_spin_unlock_irqrestore
/lib/modules/2.6.37-rc1+/build/vmlinux
              223.00  1.6% _raw_spin_lock_irqsave
/lib/modules/2.6.37-rc1+/build/vmlinux
              203.00  1.4% __clear_user
/lib/modules/2.6.37-rc1+/build/vmlinux
              184.00  1.3% tcp_poll
/lib/modules/2.6.37-rc1+/build/vmlinux
              178.00  1.2% vfs_write
/lib/modules/2.6.37-rc1+/build/vmlinux
              165.00  1.1% tcp_recvmsg
/lib/modules/2.6.37-rc1+/build/vmlinux
              152.00  1.1% pipe_read
/lib/modules/2.6.37-rc1+/build/vmlinux
              149.00  1.0% schedule
/lib/modules/2.6.37-rc1+/build/vmlinux
              135.00  0.9% rw_verify_area
/lib/modules/2.6.37-rc1+/build/vmlinux
              135.00  0.9% __pollwait
/lib/modules/2.6.37-rc1+/build/vmlinux
              130.00  0.9% __write
/lib/libc-2.12.1.so
              127.00  0.9% __ticket_spin_unlock
/lib/modules/2.6.37-rc1+/build/vmlinux
              126.00  0.9% __poll
/lib/libc-2.12.1.so


On Tue, Nov 9, 2010 at 5:23 PM, Eric Dumazet <eric.dumazet@gmail.com> wrote:
> Le mardi 09 novembre 2010 à 06:22 +0100, Eric Dumazet a écrit :
>> Le mardi 09 novembre 2010 à 11:05 +1100, Andrew Hendry a écrit :
>> > results on an i7 860 @ 2.80Ghz machine, no virtualization involved. 2.6.37-rc1+
>> >
>> > # time dd if=/dev/zero bs=1M count=10000 | netcat  127.0.0.1 9999
>> > 10000+0 records in
>> > 10000+0 records out
>> > 10485760000 bytes (10 GB) copied, 50.2022 s, 209 MB/s
>> >
>> > real        0m50.210s
>> > user        0m1.094s
>> > sys 0m57.589s
>>
>> Thanks !
>>
>> Could you take a pef snapshot during the test ?
>>
>> # perf record -a -g sleep 10
>> # perf report
>>
>>
>
> On my laptop
> Intel(R) Core(TM)2 Duo CPU     T8300  @ 2.40GHz
> (2.6.35-22-generic #35-Ubuntu SMP Sat Oct 16 20:45:36 UTC 2010 x86_64
> GNU/Linux) :
>
> time dd if=/dev/zero bs=1M count=10000|netcat 127.0.0.1 9999
> 10000+0 enregistrements lus
> 10000+0 enregistrements écrits
> 10485760000 octets (10 GB) copiés, 38,2691 s, 274 MB/s
>
> real    0m38.274s
> user    0m1.870s
> sys     0m38.370s
>
>
> perf top result :
>
> -------------------------------------------------------------------------------------------------
>   PerfTop:    1948 irqs/sec  kernel:90.7%  exact:  0.0% [1000Hz cycles],  (all, 2 CPUs)
> -------------------------------------------------------------------------------------------------
>
>             samples  pcnt function                    DSO
>             _______ _____ ___________________________ ___________________
>
>             1867.00 12.4% copy_user_generic_string    [kernel.kallsyms]
>             1166.00  7.7% __ticket_spin_lock          [kernel.kallsyms]
>              744.00  4.9% __clear_user                [kernel.kallsyms]
>              667.00  4.4% system_call                 [kernel.kallsyms]
>              329.00  2.2% tcp_sendmsg                 [kernel.kallsyms]
>              304.00  2.0% schedule                    [kernel.kallsyms]
>              257.00  1.7% _raw_spin_unlock_irqrestore [kernel.kallsyms]
>              231.00  1.5% fget_light                  [kernel.kallsyms]
>              216.00  1.4% do_poll                     [kernel.kallsyms]
>              203.00  1.3% __read_chk                  /lib/libc-2.12.1.so
>              202.00  1.3% __pollwait                  [kernel.kallsyms]
>              201.00  1.3% __poll                      /lib/libc-2.12.1.so
>              187.00  1.2% system_call_after_swapgs    [kernel.kallsyms]
>              176.00  1.2% __write                     /lib/libc-2.12.1.so
>              173.00  1.1% _raw_spin_lock_irqsave      [kernel.kallsyms]
>              163.00  1.1% tcp_recvmsg                 [kernel.kallsyms]
>              158.00  1.0% do_sys_poll                 [kernel.kallsyms]
>              153.00  1.0% vfs_write                   [kernel.kallsyms]
>              143.00  0.9% pipe_read                   [kernel.kallsyms]
>              141.00  0.9% fput                        [kernel.kallsyms]
>              121.00  0.8% common_file_perm            [kernel.kallsyms]
>              120.00  0.8% _cond_resched               [kernel.kallsyms]
>
>
> # vmstat 1
> procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
>  r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa
>  2  0   1456 120056  51572 2606876    0    0   158    41  254  190  9  2 88  0
>  2  0   1456 120140  51580 2606868    0    0    12     0  758 158309 11 76 13  0
>  2  0   1456 119520  51588 2606896    0    0     0   176  778 160749  8 80 12  0
>  2  0   1456 120388  51588 2606896    0    0     0     0  730 158201  9 76 16  0
>  3  0   1456 120388  51588 2606896    0    0     0     0  745 158490  8 76 16  0
>  2  0   1456 120520  51588 2606896    0    0     0     0  991 159120  9 78 13  0
>  2  0   1456 120024  51588 2606896    0    0     0     0  653 160023 10 79 11  0
>  3  0   1456 120520  51588 2606896    0    0     0     0  659 160614  8 78 14  0
>  2  0   1456 120272  51596 2606896    0    0     0    80  695 159922 10 75 14  0
>  4  0   1456 120272  51596 2606896    0    0     0     0  675 158010  7 79 14  0
>
>
> # powertop
>     PowerTOP version 1.13      (C) 2007 Intel Corporation
>
> < Detailed C-state information is not P-states (frequencies)
>                                      Turbo Mode    43.1%
>                                        2.40 Ghz    48.0%
>                                        2.00 Ghz     8.2%
>                                        1.60 Ghz     0.7%
>                                        1200 Mhz     0.1%
>
> Wakeups-from-idle per second : 542.9    interval: 10.0s
> no ACPI power usage estimate available
>
> Top causes for wakeups:
>  21.9% (196.5)   [kernel scheduler] Load balancing tick
>  21.2% (190.7)   [Rescheduling interrupts] <kernel IPI>
>  12.7% (114.0)   PS/2 keyboard/mouse/touchpad interrupt
>  12.0% (107.9)   plugin-containe
>  11.1% ( 99.3)   alsa-sink
>   6.0% ( 53.8)   firefox-bin
>   4.4% ( 39.7)   fping
>   3.9% ( 35.2)   Xorg
>   1.3% ( 11.3)   [b43] <interrupt>
>   1.1% ( 10.0)   ksoftirqd/0
>   0.4% (  4.0)D  nagios3
>   0.2% (  1.9)D  gnome-terminal
>   0.7% (  6.4)   [Thermal event interrupts] <kernel IPI>
>
>
>
>

  reply	other threads:[~2010-11-09  6:30 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <1288954189.28003.178.camel@firesoul.comx.local>
     [not found] ` <1288988955.2665.297.camel@edumazet-laptop>
     [not found]   ` <1289213926.15004.19.camel@firesoul.comx.local>
     [not found]     ` <1289214289.2820.188.camel@edumazet-laptop>
2010-11-08 15:06       ` Loopback performance from kernel 2.6.12 to 2.6.37 Eric Dumazet
2010-11-09  0:05         ` Andrew Hendry
2010-11-09  5:22           ` Eric Dumazet
2010-11-09  6:23             ` Eric Dumazet
2010-11-09  6:30               ` Andrew Hendry [this message]
2010-11-09  6:38                 ` Eric Dumazet
2010-11-09  6:42                   ` Eric Dumazet
2010-11-09 13:59         ` Jesper Dangaard Brouer
2010-11-09 14:06           ` Eric Dumazet
2010-11-09 14:16           ` Jesper Dangaard Brouer
2010-11-09 14:25             ` Eric Dumazet
2010-11-18 13:52               ` Eric Dumazet
2010-11-18 17:41                 ` Eric Dumazet
2010-11-18 17:48                   ` David Miller
2010-11-09 14:38             ` Jesper Dangaard Brouer
2010-11-10 11:24               ` Jesper Dangaard Brouer
2010-12-12 15:48                 ` Arnaldo Carvalho de Melo
2010-11-09 21:35 Xose Vazquez Perez
2010-11-10  8:49 ` Jesper Dangaard Brouer

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='AANLkTikAPaU_2=wS_T3V-8xFZm-G3qutJBxY8yb0QCYL@mail.gmail.com' \
    --to=andrew.hendry@gmail.com \
    --cc=eric.dumazet@gmail.com \
    --cc=jdb@comx.dk \
    --cc=netdev@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).