ubuntu10.4版本不能运行fah的修补方法（含smp2&bigadv）

eqzero

前段时间有兄弟发现安装ubuntu10.4版本后无法运行fah，找了国外论坛的资料，终于解决：

原帖：
http://foldingforum.org/viewtopic.php?f=44&t=12939

测试通过环境：
ubuntu9.04升级10.4后完成当前运行任何之后无法继续运行，提示：

1. fah6: relocation error: /lib/libnss_files.so.2: symbol __rawmemchr, version GLIBC_2.2.5 not
   
2. defined in file libc.so.6 with link time reference

复制代码

出错状况与坛内兄弟的提示一样，理论上修补方法也一致的。

修补步骤：（注意！一定要用你平时计算fah的用户名登录，虽然下面升级步骤会有错误提示，那是正常的，不要使用"sudo"命令以root用户权限运行下面指令！）

1. cd ~
   
2. wget http://darkswarm.org/whosyerdaddy-0.6.sh
   
3. chmod +x whosyerdaddy-0.6.sh
   
4. cp -a /lib ~/lib64-fah
   
5. cd ~/lib64-fah
   
6. ~/whosyerdaddy-0.6.sh

复制代码

上述指令完成后，运行：

1. LD_LIBRARY_PATH=~/lib64-fah ./fah6 -smp 2 -verbosity 9

复制代码

即可运行smp2，如果你是4核或者4线程的机器，修改-smp 后面的2为4；
如果需要运行bigadv包，运行：

1. LD_LIBRARY_PATH=~/lib64-fah ./fah -smp 8 -bigadv -verbosity 9

复制代码

即可。

下面是运行命令和返回结果的示例，如果你运行正常，应该看到与下面类似的画面：

1. fah@exp:~$ cat /etc/issue.net
   
2. Ubuntu lucid (development branch)
   
3. fah@exp:~$ cd ~
   
4. fah@exp:~$ wget http://darkswarm.org/whosyerdaddy-0.6.sh
   
5. --2010-05-01 07:27:43--  http://darkswarm.org/whosyerdaddy-0.6.sh
   
6. Resolving darkswarm.org... 85.11.66.60
   
7. Connecting to darkswarm.org|85.11.66.60|:80... connected.
   
8. HTTP request sent, awaiting response... 200 OK
   
9. Length: 4408 (4.3K) [application/x-sh]
   
10. Saving to: “whosyerdaddy-0.6.sh”
    
11. 
    
12. 100%[=======================================================================>] 4,408       --.-K/s   in 0.001s  
    
13. 
    
14. 2010-05-01 07:27:43 (3.96 MB/s) - “whosyerdaddy-0.6.sh” saved [4408/4408]
    
15. 
    
16. fah@exp:~$ chmod +x whosyerdaddy-0.6.sh
    
17. fah@exp:~$ cp -a /lib ~/lib64-fah
    
18. cp: cannot create special file `/home/fah/lib64-fah/udev/devices/net/tun': Operation not permitted
    
19. cp: cannot create special file `/home/fah/lib64-fah/udev/devices/console': Operation not permitted
    
20. cp: cannot create special file `/home/fah/lib64-fah/udev/devices/loop0': Operation not permitted
    
21. cp: cannot create special file `/home/fah/lib64-fah/udev/devices/null': Operation not permitted
    
22. cp: cannot create special file `/home/fah/lib64-fah/udev/devices/ppp': Operation not permitted
    
23. cp: cannot open `/lib/ufw/user.rules' for reading: Permission denied
    
24. cp: cannot open `/lib/ufw/user6.rules' for reading: Permission denied
    
25. fah@exp:~$ cd ~/lib64-fah
    
26. fah@exp:~/lib64-fah$ ~/whosyerdaddy-0.6.sh
    
27. whosyerdaddy 0.6, workaround for broken Folding@Home client 6.29 binaries
    
28. Copyright (c) 2010 by Kris Rusocki <[email protected]>
    
29. Licensed under GPLv2
    
30. 
    
31. Checking for awk bc cat chmod cut dd grep head od readelf sed tac tr ...done.
    
32. WARNING: no STRTAB, going with 0x30 offset %-| and single-page size. YMMV.
    
33. DYNSYM at 0x00003c28
    
34. processing symbol __rawmemchr
    
35. Found at position 99 (assuming dynsym)
    
36. Symbol processed. Perhaps.
    
37. processing symbol strchr
    
38. Found at position 1652 (assuming dynsym)
    
39. Symbol processed. Perhaps.
    
40. processing symbol strlen
    
41. Found at position 749 (assuming dynsym)
    
42. Symbol processed. Perhaps.
    
43. processing symbol strcpy
    
44. Found at position 89 (assuming dynsym)
    
45. Symbol processed. Perhaps.
    
46. Rebranding shared objects... done.
    
47. fah@exp:~/lib64-fah$ cd ~/
    
48. fah@exp:~$ cd fah
    
49. fah@exp:~/fah$ LD_LIBRARY_PATH=~/lib64-fah ./fah6 -smp 2 -verbosity 9
    
50. 
    
51. Note: Please read the license agreement (fah6 -license). Further
    
52. use of this software requires that you have read and accepted this agreement.
    
53. 
    
54. 2 cores detected
    
55. 
    
56. 
    
57. --- Opening Log file [April 7 04:59:38 UTC]
    
58. 
    
59. 
    
60. # Linux SMP Console Edition ###################################################
    
61. ###############################################################################
    
62. 
    
63.                        Folding@Home Client Version 6.29
    
64. 
    
65.                           http://folding.stanford.edu
    
66. 
    
67. ###############################################################################
    
68. ###############################################################################
    
69. 
    
70. Launch directory: /home/fah/fah
    
71. Executable: ./fah6
    
72. Arguments: -smp 2 -verbosity 9
    
73. 
    
74. [04:59:38] Configuring Folding@Home...
    
75. 
    
76. User name [Anonymous]?
    
77. Team Number [0]?
    
78. Passkey []?
    
79. Ask before fetching/sending work (no/yes) [no]?
    
80. Use proxy (yes/no) [no]?
    
81. Acceptable size of work assignment and work result packets (bigger units
    
82. may have large memory demands) -- 'small' is <5MB, 'normal' is <10MB, and
    
83. 'big' is >10MB (small/normal/big) [normal]? big
    
84. Change advanced options (yes/no) [no]? yes
    
85. Core Priority (idle/low) [idle]?
    
86. Disable highly optimized assembly code (no/yes) [no]?
    
87. Interval, in minutes, between checkpoints (3-30) [15]?
    
88. Memory, in MB, to indicate (998 available) [998]?
    
89. Set -advmethods flag always, requesting new advanced
    
90.   scientific cores and/or work units if available (no/yes) [no]? yes
    
91. Ignore any deadline information (mainly useful if
    
92. system clock frequently has errors) (no/yes) [no]?
    
93. Machine ID (1-16) [1]?
    
94. The following options require you to restart the client before they take effect
    
95. Disable CPU affinity lock (no/yes) [no]?
    
96. Additional client parameters []?
    
97. IP address to bind core to (for viewer) []?
    
98. 
    
99. [05:00:03] - Ask before connecting: No
    
100. [05:00:03] - User name: Anonymous (Team 0)
     
101. [05:00:03] - User ID not found locally
     
102. [05:00:03] + Requesting User ID from server
     
103. [05:00:03] - Getting ID from AS:
     
104. [05:00:03] Connecting to http://assign.stanford.edu:8080/
     
105. [05:00:04] Posted data.
     
106. [05:00:04] Initial: 0C35; - Received User ID = 350C3F556027DD1D
     
107. [05:00:04] - Machine ID: 1
     
108. [05:00:04]
     
109. [05:00:04] Work directory not found. Creating...
     
110. [05:00:04] Could not open work queue, generating new queue...
     
111. [05:00:04] - Preparing to get new work unit...
     
112. [05:00:04] - Autosending finished units... [April 7 05:00:04 UTC]
     
113. [05:00:04] Trying to send all finished work units
     
114. [05:00:04] + No unsent completed units remaining.
     
115. [05:00:04] - Autosend completed
     
116. [05:00:04] Cleaning up work directory
     
117. [05:00:04] + Attempting to get work packet
     
118. [05:00:04] - Will indicate memory of 998 MB
     
119. [05:00:04] - Connecting to assignment server
     
120. [05:00:04] Connecting to http://assign.stanford.edu:8080/
     
121. [05:00:04] Posted data.
     
122. [05:00:04] Initial: ED82; - Successful: assigned to (130.237.232.140).
     
123. [05:00:04] + News From Folding@Home: Welcome to Folding@Home
     
124. [05:00:04] Loaded queue successfully.
     
125. [05:00:04] Connecting to http://130.237.232.140:8080/
     
126. [05:00:05] Posted data.
     
127. [05:00:05] Initial: 0000; - Receiving payload (expected size: 1799542)
     
128. [05:00:36] - Downloaded at ~56 kB/s
     
129. [05:00:36] - Averaged speed for that direction ~56 kB/s
     
130. [05:00:36] + Received work.
     
131. [05:00:36] + Closed connections
     
132. [05:00:36]
     
133. [05:00:36] + Processing work unit
     
134. [05:00:36] Core required: FahCore_a3.exe
     
135. [05:00:36] Core not found.
     
136. [05:00:36] - Core is not present or corrupted.
     
137. [05:00:36] - Attempting to download new core...
     
138. [05:00:36] + Downloading new core: FahCore_a3.exe
     
139. [05:00:36] Downloading core (/~pande/Linux/AMD64/Core_a3.fah from www.stanford.edu)
     
140. [05:00:36] Initial: AFDE; + 10240 bytes downloaded
     
141. 。。。。。。
     
142. [05:00:38] Initial: C7AF; + 1885206 bytes downloaded
     
143. [05:00:38] Verifying core Core_a3.fah...
     
144. [05:00:38] Signature is VALID
     
145. [05:00:38]
     
146. [05:00:38] Trying to unzip core FahCore_a3.exe
     
147. [05:00:38] Decompressed FahCore_a3.exe (4415280 bytes) successfully
     
148. [05:00:38] + Core successfully engaged
     
149. [05:01:02]
     
150. [05:01:02] + Processing work unit
     
151. [05:01:02] Core required: FahCore_a3.exe
     
152. [05:01:02] Core found.
     
153. [05:01:02] Working on queue slot 01 [April 7 05:01:02 UTC]
     
154. [05:01:02] + Working ...
     
155. [05:01:02] - Calling './FahCore_a3.exe -dir work/ -nice 19 -suffix 01 -np 2 -checkpoint 15 -verbose -lifeline 10668 -version 629'
     
156. 
     
157. [05:01:03]
     
158. [05:01:03] *------------------------------*
     
159. [05:01:03] Folding@Home Gromacs SMP Core
     
160. [05:01:03] Version 2.17 (March 6, 2010)
     
161. [05:01:03]
     
162. [05:01:03] Preparing to commence simulation
     
163. [05:01:03] - Looking at optimizations...
     
164. [05:01:03] - Created dyn
     
165. [05:01:03] - Files status OK
     
166. [05:01:03] - Expanded 1799030 -> 2392545 (decompressed 132.9 percent)
     
167. [05:01:03] Called DecompressByteArray: compressed_data_size=1799030 data_size=2392545, decompressed_data_size=2392545 diff=0
     
168. [05:01:03] - Digital signature verified
     
169. [05:01:03]
     
170. [05:01:03] Project: 6015 (Run 0, Clone 222, Gen 34)
     
171. [05:01:03]
     
172. [05:01:03] Assembly optimizations on if available.
     
173. [05:01:03] Entering M.D.
     
174. Starting 2 threads
     
175. NNODES=2, MYRANK=0, HOSTNAME=thread #0
     
176. NNODES=2, MYRANK=1, HOSTNAME=thread #1
     
177. Reading file work/wudata_01.tpr, VERSION 4.0.99_development_20090605 (single precision)
     
178. Note: tpx file_version 68, software version 70
     
179. Making 1D domain decomposition 2 x 1 x 1
     
180. starting mdrun 'Protein in POPC'
     
181. 17500002 steps,  35000.0 ps (continuing from step 17000002,  34000.0 ps).
     
182. [05:01:10] Completed 0 out of 500000 steps  (0%)

复制代码

bugfix

金鹏

回复 1# eqzero

感谢兄弟分享

panzerkiller

Super Nice!

caozhonghua

先围观一下……说不定哪天要用上

金鹏

回复 1# eqzero

标题有误导－－应该是SMP2吧

eqzero

回复 6# 金鹏


    谢谢，已经修改

vennyfan

有空再玩玩UBUNTU

金鹏

回复 8# vennyfan

显卡大王准备战略转移到BIGADV和SMP2了？

sellength

特无敌了！！！明天弄谢谢版主

eqzero

james@james-890gx:~$ cd folding/
james@james-890gx:~/folding$ ./fah6

Note: Please read the license agreement (fah6 -license). Further
use of this software requires that you have read and accepted this agreement.

8 cores detected


--- Opening Log file [June 21 12:33:50 UTC]


# Linux SMP Console Edition ###################################################
###############################################################################

                       Folding@Home Client Version 6.29

                          http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: /home/james/folding
Executable: ./fah6
Arguments: -smp -advmethods -verbosity 9

[12:33:50] - Ask before connecting: No
[12:33:50] - User name: eQzERO (Team 3213)
[12:33:50] - User ID: 6E39C51E5BA86127
[12:33:50] - Machine ID: 1
[12:33:50]
[12:33:50] Loaded queue successfully.
[12:33:50]
[12:33:50] - Autosending finished units... [June 21 12:33:50 UTC]
[12:33:50] + Processing work unit
[12:33:50] Trying to send all finished work units
[12:33:50] Core required: FahCore_a3.exe
[12:33:50] + No unsent completed units remaining.
[12:33:50] - Autosend completed
[12:33:50] Core found.
[12:33:51] Working on queue slot 03 [June 21 12:33:51 UTC]
[12:33:51] + Working ...
[12:33:51] - Calling './FahCore_a3.exe -dir work/ -nice 19 -suffix 03 -np 8 -checkpoint 15 -verbose -lifeline 1764 -version 629'

[12:33:51]
[12:33:51] *------------------------------*
[12:33:51] Folding@Home Gromacs SMP Core
[12:33:51] Version 2.22 (June 10, 2010)
[12:33:51]
[12:33:51] Preparing to commence simulation
[12:33:51] - Ensuring status. Please wait.
[12:34:00] - Looking at optimizations...
[12:34:00] - Working with standard loops on this execution.
[12:34:00] - Previous termination of core was improper.
[12:34:00] - Files status OK
[12:34:01] - Expanded 1765327 -> 2253505 (decompressed 127.6 percent)
[12:34:01] Called DecompressByteArray: compressed_data_size=1765327 data_size=2253505, decompressed_data_size=2253505 diff=0
[12:34:01] - Digital signature verified
[12:34:01]
[12:34:01] Project: 6050 (Run 1, Clone 106, Gen 25)
[12:34:01]
[12:34:01] Entering M.D.
Starting 8 threads
NNODES=8, MYRANK=0, HOSTNAME=thread #0
NNODES=8, MYRANK=1, HOSTNAME=thread #1
NNODES=8, MYRANK=2, HOSTNAME=thread #2
NNODES=8, MYRANK=4, HOSTNAME=thread #4
NNODES=8, MYRANK=5, HOSTNAME=thread #5
NNODES=8, MYRANK=6, HOSTNAME=thread #6
NNODES=8, MYRANK=7, HOSTNAME=thread #7
NNODES=8, MYRANK=3, HOSTNAME=thread #3
Reading file work/wudata_03.tpr, VERSION 4.0.99_development_20090605 (single precision)

Reading checkpoint file work/wudata_03.cpt generated: Mon Jun 21 12:08:37 2010

  #nodes mismatch,
    current program: 8
    checkpoint file: 3

Gromacs binary or parallel settings not identical to previous run.
Continuation is exact, but is not guaranteed to be binary identical.

[12:34:07] Using Gromacs checkpoints
Making 1D domain decomposition 8 x 1 x 1
starting mdrun 'Mutant_scan'
13000001 steps,  26000.0 ps (continuing from step 12660630,  25321.3 ps).
[12:34:07] Resuming from checkpoint
[12:34:07] Verified work/wudata_03.log
[12:34:07] Verified work/wudata_03.trr
[12:34:07] Verified work/wudata_03.edr
[12:34:07] Completed 160629 out of 500000 steps  (32%)

NOTE: Turning on dynamic load balancing

[04:36:52] Completed 165000 out of 500000 steps  (33%)

eqzero

发现移植性还挺强的嘛！

3核的AMD x3-440移植的4C8T的i7完全正常，继续运算，自动开8核......

1.   #nodes mismatch,
   
2.     current program: 8
   
3.     checkpoint file: 3

复制代码

eqzero

还是8线程的i7强啊！4G的x3-440才5k左右的PPD，接手运算的3.66G的x3470的PPD直下17K+！

eqzero

渐入佳境了了，6050现在的TPF为2‘53’‘，PPD近19K了！

发现功耗控制不错！一破航嘉钻石版额定195W的电源，带着超频的x3470、6600显卡，摸着电源没什么温度啊，就是电源超额定了，以200瓦电源获得19K的PPD，值了！

主板：GA-P55A-UD3R
CPU：intel [email protected]
内存：2G*4@1600
显卡NV6600
硬盘：8G-CF2Sata
散热器：U12P+P12

金鹏

回复 14# eqzero

倒，是SMP2跑到19K？不会吧？我4G的860在WIN下也跑不到啊