From: vx 8400 <vx8400@gmail.com>
To: ALT Linux Community general discussions <community@lists.altlinux.org>
Subject: Re: [Comm] ALTLinux LXDE Remix for School [тормоза с 2.6.32]
Date: Tue, 25 Jan 2011 22:13:47 +0300
Message-ID: <AANLkTim2YdCbdnUFAh4HzKend_ttxR2XJgipALc0CcjM@mail.gmail.com> (raw)
In-Reply-To: <20110125164854.GH21594@osdn.org.ua>
[-- Attachment #1: Type: text/plain, Size: 1005 bytes --]
Добрый вечер/утро/день всем,
25 января 2011 г. 19:48 Michael Shigorin::
> On Tue, Jan 25, 2011 at 03:48:31PM +0200, Yura Kalinichenko wrote:
>> Как показала практика, для таких машин надо использовать ядра,
>> собранные без поддержки SMP. Я здесь по этому поводу не так
>> давно отчитывался.
>
> Если хватит модулей в tmc-tc, то оно UP.
Vanilla 2.6.32.27 показывает нормальные цифры с CONFIG_SMP=y и без
(см. вывод unixbench в прицепе).
.config для vanilla:
https://sites.google.com/site/strgko/2632-sis/config-2.6.32.27-smp
https://sites.google.com/site/strgko/2632-sis/config-2.6.32.27-up
Производительность 2.6.32-std-def-alt20.M50P.1 на этой конкретной машине
при всех прочих равных в ~3 раза ниже:
https://sites.google.com/site/strgko/2632-sis/ubench-2632-sis.txt
Это не шутки с калибровкой времени, проверяется внешним секундомером.
Где искать причину?
Причину надо найти, так как 2.6.32-std-def-alt20.M50P.1
идет в дистрибутиве как раз для таких старых машин.
Thx, rgrds,
Oleg.
[-- Attachment #2: ubench-2_6_32_27_vanilla-up-smp.txt --]
[-- Type: text/plain, Size: 5217 bytes --]
BYTE UNIX Benchmarks (Version 5.1.2) BYTE UNIX Benchmarks (Version 5.1.2)
System: localhost: GNU/Linux System: localhost: GNU/Linux
OS: GNU/Linux -- 2.6.32.27 -- #1 SMP Tue Jan 25 20:51:36 MSK 2011 OS: GNU/Linux -- 2.6.32.27 -- #1 Tue Jan 25 17:40:23 MSK 2011
Machine: i686 (unknown) Machine: i686 (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8") Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Pentium II (Deschutes) (668.1 bogomips) CPU 0: Pentium II (Deschutes) (668.3 bogomips)
MMX, Physical Address Ext, SYSENTER/SYSEXIT MMX, Physical Address Ext, SYSENTER/SYSEXIT
21:17:01 up 9 min, 1 user, load average: 0.48, 0.26, 0.17; runlevel 18:04:15 up 3 min, 1 user, load average: 0.62, 0.48, 0.21; runlevel
SMP UP
------------------------------------------------------------------------ ------------------------------------------------------------------------
Benchmark Run: Втр Янв 25 2011 21:17:01 - 21:45:09 Benchmark Run: Втр Янв 25 2011 18:04:15 - 18:32:23
1 CPU in system; running 1 parallel copy of tests 1 CPU in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 1034929.6 lps (10.0 s, 7 samples) 1039277.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 257.9 MWIPS (10.0 s, 7 samples) 256.5 MWIPS (10.0 s, 7 samples)
Execl Throughput 444.7 lps (29.7 s, 2 samples) 466.5 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 32525.5 KBps (30.0 s, 2 samples) 32135.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 11586.0 KBps (30.0 s, 2 samples) 11368.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 17496.7 KBps (30.0 s, 2 samples) 18428.4 KBps (30.0 s, 2 samples)
Pipe Throughput 129848.9 lps (10.0 s, 7 samples) 134065.2 lps (10.0 s, 7 samples)
Pipe-based Context Switching 26300.0 lps (10.0 s, 7 samples) 33873.8 lps (10.0 s, 7 samples)
Process Creation 1374.0 lps (30.0 s, 2 samples) 1541.3 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 318.7 lpm (60.2 s, 2 samples) 328.9 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 39.8 lpm (60.3 s, 2 samples) 40.5 lpm (60.7 s, 2 samples)
System Call Overhead 196143.7 lps (10.0 s, 7 samples) 204375.2 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 1034929.6 88.7 116700.0 1039277.5 89.1
Double-Precision Whetstone 55.0 257.9 46.9 55.0 256.5 46.6
Execl Throughput 43.0 444.7 103.4 43.0 466.5 108.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 32525.5 82.1 3960.0 32135.5 81.2
File Copy 256 bufsize 500 maxblocks 1655.0 11586.0 70.0 1655.0 11368.4 68.7
File Copy 4096 bufsize 8000 maxblocks 5800.0 17496.7 30.2 5800.0 18428.4 31.8
Pipe Throughput 12440.0 129848.9 104.4 12440.0 134065.2 107.8
Pipe-based Context Switching 4000.0 26300.0 65.8 4000.0 33873.8 84.7
Process Creation 126.0 1374.0 109.1 126.0 1541.3 122.3
Shell Scripts (1 concurrent) 42.4 318.7 75.2 42.4 328.9 77.6
Shell Scripts (8 concurrent) 6.0 39.8 66.4 6.0 40.5 67.5
System Call Overhead 15000.0 196143.7 130.8 15000.0 204375.2 136.3
======== ========
System Benchmarks Index Score 75.9 79.5
next prev parent reply other threads:[~2011-01-25 19:13 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-01-25 8:21 vx 8400
2011-01-25 9:28 ` Anton A. Vinogradov
2011-01-25 11:47 ` vx 8400
2011-01-25 13:48 ` Yura Kalinichenko
2011-01-25 16:48 ` Michael Shigorin
2011-01-25 19:13 ` vx 8400 [this message]
2011-01-26 8:44 ` Michael Shigorin
2011-01-26 12:08 ` vx 8400
2011-01-26 13:10 ` Michael Shigorin
2011-01-27 2:53 ` vx 8400
2011-01-28 10:50 ` Michael Shigorin
2011-01-28 10:56 ` vx 8400
2011-01-28 11:01 ` Радик Юсупов
2011-01-28 11:21 ` Aleksey Novodvorsky
2011-01-28 12:15 ` vx 8400
2011-01-28 13:49 ` Michael Shigorin
2011-01-28 17:42 ` Радик Юсупов
2011-02-01 22:43 ` vx 8400
2011-02-01 22:47 ` vx 8400
2011-01-26 15:09 ` [Comm] тормоза с 2.6.32 (kernel performance) Michael Shigorin
2011-01-26 15:13 ` Michael Shigorin
2011-01-28 9:17 ` vx 8400
2011-01-26 14:23 ` [Comm] ALTLinux LXDE Remix for School [тормоза с 2.6.32] Yura Kalinichenko
2011-01-30 7:50 ` Michael A. Kangin
2011-01-30 7:52 ` Michael A. Kangin
2011-01-26 1:50 ` Владимир Гусев
2011-01-26 4:27 ` vx 8400
2011-01-26 8:19 ` Anton A. Vinogradov
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=AANLkTim2YdCbdnUFAh4HzKend_ttxR2XJgipALc0CcjM@mail.gmail.com \
--to=vx8400@gmail.com \
--cc=community@lists.altlinux.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
ALT Linux Community general discussions
This inbox may be cloned and mirrored by anyone:
git clone --mirror http://lore.altlinux.org/community/0 community/git/0.git
# If you have public-inbox 1.1+ installed, you may
# initialize and index your mirror using the following commands:
public-inbox-init -V2 community community/ http://lore.altlinux.org/community \
mandrake-russian@linuxteam.iplabs.ru community@lists.altlinux.org community@lists.altlinux.ru community@lists.altlinux.com
public-inbox-index community
Example config snippet for mirrors.
Newsgroup available over NNTP:
nntp://lore.altlinux.org/org.altlinux.lists.community
AGPL code for this site: git clone https://public-inbox.org/public-inbox.git