Linux 烧机测试


我发现这方面的知识比较少，在Google只看到来自http://blog.jpps.tcc.edu.tw的文章。好东西要珍惜，特此转载。再一次对博主表示感谢。
 
剛組裝了一部6核心的電腦 (AMD 1055t+ASUS M4A88TD-V EVO/USB3 500GB HD)，安裝好了CentOS 5後想測試一下系統能否持續運作及温度耗電情形..


 


1。安裝 硬體監測程式 lm_sensors


本來以為使用 yum install lm_sensors 就可以了，不過卻發現 lm_sensors 原始網站(http://www.lm-sensors.org/) 己經更新到 3.x版 手机赚钱，然而CentOS 5還在 2.x, 更大的問題是kernel版本未把需要的module包進去,


上網找了好久終於找出方法來了，原來在elrepo上有人幫忙維護了CentOS可用的lm_sensors套件：


安裝方法如下:


rpm -Uvh http://elrepo.org/linux/elrepo/el5/i386/RPMS/elrepo-release-5-2.el5.elrepo.noarch.rpm


vi /etc/yum.repos.d/elrepo.repo


修改[elrepo]段中的


 enabled=1 


接著可以安裝 lm_sensors相關套件


yum install lm_sensors*


＊安裝後建議把 sensors-detect 更新


cd /usr/sbin


mv sensors-detect sensors-detect.old


wget http://dl.lm-sensors.org/lm-sensors/files/sensors-detect


chmod u+x sensors-detect


接著就可以執行sensors-detect幫我找出電腦中的sensor


例如：


Driver `it87':
  * ISA bus, address 0x290
    Chip `ITE IT8721F/IT8758E Super IO Sensors' (confidence: 9)


Driver `k10temp':
  * Chip `AMD Family 10h thermal sensors' (confidence: 9)


通常還要下載欠缺相關的module


x86 执行  yum install kmod-it87-PAE  kmod-k10temp-PAE
x64 执行  yum install kmod-it87  kmod-k10temp


安裝後再執行一次 sensors-detect，便能成功了。


查看資料 執行sensors即可。


[root@localhost sbin]# sensors
k10temp-pci-00c3
Adapter: PCI adapter
CPU1 Temp:   +39°C
 


2。測試軟體 mprime


mkdir mprime


cd mprime


至http://www.mersenne.org/freesoft.htm下載 mprime for Linux版


wget http://mersenneforum.org/gimps/mprime2511.tar.gz


tar zxvf  mprime2511.tar.gz


解開後直接執行


[root@localhost mprime]# ./mprime 


Welcome to GIMPS, the hunt for huge prime numbers.  You will be asked a 
few simple questions and then the program will contact the primenet server 
to get some work for your computer.  Good luck!


Attention OVERCLOCKERS!!  Mprime has gained a reputation as a useful 
stress testing tool for people that enjoy pushing their hardware to the 
limit.  You are more than welcome to use this software for that purpose.  
Please select the stress testing choice below to avoid interfering with 
the PrimeNet server.  Use the Options/Torture Test menu choice for your 
stress tests.  Also, read the stress.txt file.


If you want to both join GIMPS and run stress tests, then Join GIMPS and 
answer the questions.  After the server gets some work for you, stop 
mprime, then run mprime -m and choose Options/Torture Test.


Join Gimps? (Y=Yes, N=Just stress testing) (Y): N
Number of torture test threads to run (6): 
Choose a type of torture test to run.
1 = Small FFTs (maximum FPU stress, data fits in L2 cache, RAM not tested 
much).
2 = In-place large FFTs (maximum heat and power consumption, some RAM 
tested).
3 = Blend (tests some of everything, lots of RAM tested).
11,12,13 = Allows you to fine tune the above three selections.
Blend is the default.  NOTE: if you fail the blend test, but can pass the 
small FFT test then your problem is likely bad memory or a bad memory 
controller.
Type of torture test to run (3): 


Accept the answers above? (Y): 
[Main thread Nov 10 14:47] Starting workers.
[Worker #1 Nov 10 14:47] Worker starting
[Worker #1 Nov 10 14:47] Setting affinity to run worker on logical CPU #0
[Worker #2 Nov 10 14:47] Worker starting
[Worker #2 Nov 10 14:47] Setting affinity to run worker on logical CPU #1
[Worker #3 Nov 10 14:47] Worker starting
[Worker #3 Nov 10 14:47] Setting affinity to run worker on logical CPU #2
[Worker #4 Nov 10 14:47] Worker starting
[Worker #4 Nov 10 14:47] Setting affinity to run worker on logical CPU #3
[Worker #6 Nov 10 14:47] Worker starting
[Worker #6 Nov 10 14:47] Setting affinity to run worker on logical CPU #5
[Worker #5 Nov 10 14:47] Worker starting
[Worker #5 Nov 10 14:47] Setting affinity to run worker on logical CPU #4
[Worker #1 Nov 10 14:47] Beginning a continuous self-test to check your computer.
[Worker #1 Nov 10 14:47] Please read stress.txt.  Hit ^C to end this test.
[Worker #3 Nov 10 14:47] Beginning a continuous self-test to check your computer.
[Worker #3 Nov 10 14:47] Please read stress.txt.  Hit ^C to end this test.
[Worker #4 Nov 10 14:47] Beginning a continuous self-test to check your computer.
[Worker #4 Nov 10 14:47] Please read stress.txt.  Hit ^C to end this test.
[Worker #2 Nov 10 14:47] Beginning a continuous self-test to check your computer.
[Worker #2 Nov 10 14:47] Please read stress.txt.  Hit ^C to end this test.
[Worker #5 Nov 10 14:47] Beginning a continuous self-test to check your computer.
[Worker #5 Nov 10 14:47] Please read stress.txt.  Hit ^C to end this test.
[Worker #6 Nov 10 14:47] Beginning a continuous self-test to check your computer.
[Worker #6 Nov 10 14:47] Please read stress.txt.  Hit ^C to end this test.
[Worker #3 Nov 10 14:47] Test 1, 4000 Lucas-Lehmer iterations of M19922945 using FFT length 1024K.
[Worker #5 Nov 10 14:47] Test 1, 4000 Lucas-Lehmer iterations of M19922945 using FFT length 1024K.
[Worker #6 Nov 10 14:47] Test 1, 4000 Lucas-Lehmer iterations of M19922945 using FFT length 1024K.
[Worker #2 Nov 10 14:47] Test 1, 4000 Lucas-Lehmer iterations of M19922945 using FFT length 1024K.
[Worker #4 Nov 10 14:47] Test 1, 4000 Lucas-Lehmer iterations of M19922945 using FFT length 1024K.
[Worker #1 Nov 10 14:47] Test 1, 4000 Lucas-Lehmer iterations of M19922945 using FFT length 1024K.
。。。


3。top狀況


[root@localhost sbin]# top


top - 10:14:23 up 20:59,  5 users,  load average: 6.00, 6.00, 6.00
Tasks: 184 total,   1 running, 182 sleeping,   0 stopped,   1 zombie
Cpu0  :  0.3%us,  0.0%sy, 95.3%ni,  4.3%id,  0.1%wa,  0.0%hi,  0.0%si,  0.0%st
Cpu1  :  0.2%us,  0.1%sy, 95.2%ni,  4.4%id,  0.0%wa,  0.0%hi,  0.0%si,  0.0%st
Cpu2  :  0.2%us,  0.1%sy, 95.3%ni,  4.4%id,  0.0%wa,  0.0%hi,  0.0%si,  0.0%st
Cpu3  :  0.1%us,  0.0%sy, 95.4%ni,  4.5%id,  0.0%wa,  0.0%hi,  0.0%si,  0.0%st
Cpu4  :  0.2%us,  0.1%sy, 95.3%ni,  4.2%id,  0.2%wa,  0.0%hi,  0.0%si,  0.0%st
Cpu5  :  0.2%us,  0.1%sy, 95.2%ni,  4.5%id,  0.0%wa,  0.0%hi,  0.0%si,  0.0%st
Mem:  16374048k total,  2879212k used, 13494836k free,   142480k buffers
Swap:  1020116k total,        0k used,  1020116k free,   734048k cached


  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
13316 root      25  10 1789m 1.7g 1616 S 593.7 10.7   6990:26 mprime
    1 root      15   0  2160  588  508 S  0.0  0.0   0:00.76 init
    2 root      RT  -5     0    0    0 S  0.0  0.0   0:00.01 migration/0
    3 root      34  19     0    0    0 S  0.0  0.0   0:00.00 ksoftirqd/0
    4 root      RT  -5     0    0    0 S  0.0  0.0   0:00.00 watchdog/0
    5 root      RT  -5     0    0    0 S  0.0  0.0   0:00.00 migration/1
    6 root      34  19     0    0    0 S  0.0  0.0   0:00.00 ksoftirqd/1
    7 root      RT  -5     0    0    0 S  0.0  0.0   0:00.00 watchdog/1
    8 root      RT  -5     0    0    0 S  0.0  0.0   0:00.01 migration/2
    9 root      34  19     0    0    0 S  0.0  0.0   0:00.00 ksoftirqd/2
   10 root      RT  -5     0    0    0 S  0.0  0.0   0:00.00 watchdog/2
   11 root      RT  -5     0    0    0 S  0.0  0.0   0:00.01 migration/3
   12 root      34  19     0    0    0 S  0.0  0.0   0:00.00 ksoftirqd/3
   13 root      RT  -5     0    0    0 S  0.0  0.0   0:00.00 watchdog/3


4。CPU温度


利用sensors指令便可查詢目前CPU的温度 


[root@localhost sbin]# sensors
k10temp-pci-00c3
Adapter: PCI adapter
CPU1 Temp:   +40°C


燒機程式運作了12小時後, CPU的温度大約還是40°C, 看來內建的散熱是足够的.


5。燒機程式運作時消耗電力會從待機的約110 W 昇到 183W左右, 最高有觀察到193 W的數字.




# rpm --import http://elrepo.org/RPM-GPG-KEY-elrepo.org
# rpm -Uvh http://elrepo.org/elrepo-release-5-2.el5.elrepo.noarch.rpm
# yum --enablerepo=elrepo install lm_sensors
# yum --enablerepo=elrepo install kmod-coretemp-PAE
# yum --enablerepo=elrepo install kmod-it87-PAE


# modprobe coretemp
# lsmod |grep coretemp


http://www.lm-sensors.org/wiki/Devices