奇怪..

sleeve · 发表于 2002-12-8 23:13:11

刚装lfs.顺便下了2.5.50内核..测试一下...2.5.50居然比不上2.4.19..当然不排除有错.这是在我计算机测试的结果...

Script started on Sun Dec  8 23:01:53 2002
bash-2.05a# free
         total    used    free    shared buffers    cached
Mem:       386324    31696    354628       0    6112    12264
-/+ buffers/cache:    13320    373004
Swap:    530104       0    530104
bash-2.05a# uname -a
Linux sleeve 2.4.19 #2 Sat Dec 7 22:24:10 CST 2002 i686 unknown
bash-2.05a# make see
cd results && make summary percent 2>/dev/null | more
make[1]: Entering directory `/root/zfile/lmbench-3.0-a1/results'

               L M B E N C H  3 . 0 S U M M A R Y
               ------------------------------------
               (Alpha software, do not distribute)

Basic system parameters
------------------------------------------------------------------------------
Host                OS Description             Mhz  tlb  cache  mem scal
                                                   pages line par load
                                                         bytes
--------- ------------- ----------------------- ---- ----- ----- ------ ----
sleeve    Linux 2.5.50    i686-pc-linux-gnu  393 32 32 2.3500 1
sleeve    Linux 2.4.19    i686-pc-linux-gnu  393 32 32 2.3500 1

Processor, Processes - times in microseconds - smaller is better
------------------------------------------------------------------------------
Host                OS  Mhz null null    open slct sig  sig  fork exec sh
                           call  I/O stat clos TCP  inst hndl proc proc proc
--------- ------------- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ----
sleeve    Linux 2.5.50  393 0.90 1.60 10.2 13.7 57.0 2.60 10.3 753. 2855 10.K
sleeve    Linux 2.4.19  393 0.78 1.05 6.57 9.01 28.8 2.11 6.71 402. 1908 8376

Basic integer operations - times in nanoseconds - smaller is better
-------------------------------------------------------------------
Host                OS  intgr intgr  intgr  intgr  intgr
                        bit add mul div mod
--------- ------------- ------ ------ ------ ------ ------
sleeve    Linux 2.5.50 2.5500 2.5600 10.2 99.6  107.5
sleeve    Linux 2.4.19 2.5000 2.5100 9.9800 97.3  104.9

Basic float operations - times in nanoseconds - smaller is better
-----------------------------------------------------------------
Host                OS  float  float  float  float
                     add mul div bogo
--------- ------------- ------ ------ ------ ------
sleeve    Linux 2.5.50 7.6700 12.8 59.9 20.6
sleeve    Linux 2.4.19 7.5100 12.5 58.8 20.1

Basic double operations - times in nanoseconds - smaller is better
------------------------------------------------------------------
Host                OS  double double double double
                     add mul div bogo
--------- ------------- ------  ------ ------ ------
sleeve    Linux 2.5.50 7.6700 12.7 59.9 23.1
sleeve    Linux 2.4.19 7.5100 12.5 58.8 22.6

Context switching - times in microseconds - smaller is better
-------------------------------------------------------------------------
Host                OS  2p/0K 2p/16K 2p/64K 8p/16K 8p/64K 16p/16K 16p/64K
                     ctxsw  ctxsw  ctxsw ctxsw  ctxsw ctxsw ctxsw
--------- ------------- ------ ------ ------ ------ ------ ------- -------
sleeve    Linux 2.5.50 2.2900 13.0 6.3600 15.9  106.6 26.8 183.0
sleeve    Linux 2.4.19 1.5400 10.8 4.8900 11.2  112.7 16.1 178.6

*Local* Communication latencies in microseconds - smaller is better
---------------------------------------------------------------------
Host                OS 2p/0K  Pipe AF    UDP  RPC/ TCP  RPC/ TCP
                     ctxsw    UNIX       UDP       TCP conn
--------- ------------- ----- ----- ---- ----- ----- ----- ----- ----
sleeve    Linux 2.5.50 2.290  16.1 28.3             95.1    313.
sleeve    Linux 2.4.19 1.540 8.700 15.5             72.8    252.

File & VM system latencies in microseconds - smaller is better
-------------------------------------------------------------------------------
Host                OS 0K File    10K File    Mmap Prot Page 100fd
                     Create Delete Create Delete Latency Fault  Fault  selct
--------- ------------- ------ ------ ------ ------ ------- ----- ------- -----
sleeve    Linux 2.5.50  165.5  246.1  596.3  344.0  4370.0 2.020 5.72570  44.8
sleeve    Linux 2.4.19  135.4  215.2  535.3  292.7  1798.0 1.508 2.92990  17.4

*Local* Communication bandwidths in MB/s - bigger is better
-----------------------------------------------------------------------------
Host             OS  Pipe AF TCP  File Mmap  Bcopy  Bcopy  Mem Mem
                           UNIX    reread reread (libc) (hand) read write
--------- ------------- ---- ---- ---- ------ ------ ------ ------ ---- -----
sleeve    Linux 2.5.50 137. 117. 74.2  141.3  252.6 93.1 69.3 252.  91.7
sleeve    Linux 2.4.19 210. 124. 47.7  125.4  258.2 94.6 70.9 257.  93.8

Memory latencies in nanoseconds - smaller is better
(WARNING - may not be correct, check graphs)
------------------------------------------------------------------
Host                OS Mhz L1 $ L2 $ Main mem Guesses
--------- ------------- --- ---- ---- -------- -------
sleeve    Linux 2.5.50 393 7.6470 56.2    270.9
sleeve    Linux 2.4.19 393 7.5080 55.1    265.5
make[1]: Leaving directory `/root/zfile/lmbench-3.0-a1/results'
bash-2.05a# exit
Script done on Sun Dec  8 23:02:20 2002

tram · 发表于 2002-12-8 23:28:07

你的2.4.19是不是patch过的啊？

sleeve · 发表于 2002-12-8 23:41:58

还没....那个我明天在patch一个看看
2.5.50应该更快才对..你把你的测试贴看看
还有..官方的2.5.50里有preempt的支持了..
这功能以前还是在gentoo的那个补丁才支持...但左右还不知道得很具体..
你翻译个看看

tram · 发表于 2002-12-8 23:44:04

我的用不上啊，很郁闷，wuwu，我的modem驱动只认2.4.19的内核，打过patch后，就不能编译了，老的module也加载不上。只能等着驱动升级了。

tram · 发表于 2002-12-8 23:47:39

还有，如果不看数字，你的内核用起来感觉怎么样呢？是新的好些还是老的好些？

doooom · 发表于 2002-12-9 04:36:53

我见过网上的测试是，2。5的快一些的。用的和你一样的测试软件。

sleeve · 发表于 2002-12-9 14:31:50

昨天草草编译了两个内核..测试一个内核要一个小时..花了4个小时..什么也干不了..今天重新编译一遍看看..不加任何patch..选项都一样..

		自动登录	找回密码
密码			注册