log in |
1)
Message boards :
Number crunching :
Trial Factoring tests
(Message 6145)
Posted 13 Apr 2020 by Sergey Kovalchuk I have extended the parameter list for mfakto. (and compiled) Are you sure that you have taken into account all possible combinations of these parameters? - numbering platform / device (from one) and just a device (from zero). - "ingenious" user config and trivial "-device 0" from wrapper - or a simple conflict in the numbers of these parameters a couple of examples for a logic test ( -d 11 --device 0) ( -d g --device 0) ( -d c --device 0) can simply ignore (delete when analyzing arguments) the --device parameter in the presence of the -d parameter? |
2)
Message boards :
Number crunching :
Trial Factoring tests
(Message 5991)
Posted 2 Apr 2020 by Sergey Kovalchuk option -d don't work with 0, start from 1 -d 00-10 Select device - Error: No platform found -d 11 Select device - Get device info - Compiling kernels. |
3)
Message boards :
Number crunching :
Trial Factoring tests
(Message 5984)
Posted 1 Apr 2020 by Sergey Kovalchuk no, app_config is better wrapper receives this argument (--device N) and passes it to the application (job.xml <append_cmdline_args/>) the question is in the source of this argument - a task template on the server ($GPU_DEVICE_NUM), a sheduler, or a local client? find and neutralize :o) or - change wrapper - force it to translate into a form that the application understands. - change application - add this form of argument - switch to another platform/device selection scheme - for example opencl_platform.txt |
4)
Message boards :
Number crunching :
Trial Factoring tests
(Message 5948)
Posted 29 Mar 2020 by Sergey Kovalchuk mfakto has option "-d c" - force using all CPUs mfaktc does not have such an option, so my tip is past the target :( |
5)
Message boards :
Number crunching :
Trial Factoring tests
(Message 5936)
Posted 29 Mar 2020 by Sergey Kovalchuk <app_config> <app_version> <app_name>GPU72</app_name> <plan_class>cuda100</plan_class> # 1 <cmdline>-d c</cmdline> # 2 - try multi-GPU, 2 tasks * 2 GPU </app_version> </app_config> |
6)
Message boards :
Number crunching :
Trial Factoring
(Message 5917)
Posted 28 Mar 2020 by Sergey Kovalchuk there is no <cmdline> either in the app_version section or in workunit so that it can be added to the app_config file without any problems <app_version> <app_name>GPU72</app_name> <plan_class>opencl_ati_101</plan_class> <cmdline> -d 1 </cmdline> </app_version> that's just different tasks need different parameters, as an option to launch the second instance of BOINC for the second GPU with the option "-d 1" in this project PS in this way I was able to run some tasks on Intel GPU although the server did not recognize this GPU |
7)
Message boards :
Number crunching :
Trial Factoring
(Message 5843)
Posted 23 Mar 2020 by Sergey Kovalchuk linux + ati 22:57:23 (1328): wrapper (7.2.26012): starting
application mfakto-x64 missing job_GPU72_l64_00002.xml <zipfilename>mfakto-linux64-v2.zip</zipfilename> |
8)
Message boards :
Number crunching :
Trial Factoring
(Message 5788)
Posted 21 Mar 2020 by Sergey Kovalchuk 13:10:35 (948): mfakto-x64.exe exited; CPU time 0.000000 13:10:35 (948): app exit status: 0xc000007b No DLL? VC redist or .NET ? 21.03.2020 12:58:45 | | OpenCL: AMD/ATI GPU 0: AMD Radeon HD 6790/6850/6870 series (Barts) (driver version 1800.8 (VM), device version OpenCL 1.2 AMD-APP (1800.8), 2048MB, 2015MB available, 2976 GFLOPS peak) |
9)
Message boards :
Number crunching :
LLR Version 3.8.17 released
(Message 2370)
Posted 15 Feb 2016 by Sergey Kovalchuk 2 tasks - "Completed and validated" |
10)
Message boards :
Number crunching :
LLR Version 3.8.17 released
(Message 2364)
Posted 15 Feb 2016 by Sergey Kovalchuk Problem not solved x64 - ok
-rw-r--r-- 1 boinc boinc 43 Feb 15 10:12 inputa60_0400526.prp
-rwxr-xr-x 1 boinc boinc 188 Feb 15 10:03 job_srbase7_l64_00018.xml
-rwxr-xr-x 1 boinc boinc 188 Feb 15 10:12 job_srbase_l64_00018.xml
-rw-r--r-- 1 boinc boinc 6471371 Feb 15 10:03 srbase_v5_l64.zip
-rwxr-xr-x 1 boinc boinc 1434493 Feb 15 10:03 wrapper_26012-v2_x86_64-pc-linux-gnu
x32 - error
-rw-r--r-- 1 boinc boinc 43 Feb 15 11:02 inputa60_0401332.prp
-rwxr-xr-x 1 boinc boinc 188 Feb 15 11:02 job_srbase7_l32_00018.xml
-rwxr-xr-x 1 boinc boinc 188 Feb 15 11:02 job_srbase_l32_00018.xml
-rw-r--r-- 1 boinc boinc 5037430 Feb 15 11:02 srbase_v5_l32.zip
-rwxr-xr-x 1 boinc boinc 1261648 Feb 15 11:02 wrapper_26012-v2_i686-pc-linux-gnu
May be problem with UID or attribute for llr32 in .zip ? |
11)
Message boards :
Number crunching :
LLR Version 3.8.17 released
(Message 2362)
Posted 15 Feb 2016 by Sergey Kovalchuk No problem with old srbase_v4_l32.zip on this host. No problem with new srbase_v5_l64.zip on x64 - hosts BOINC# Same install of Ubuntu 14.04, but 64 bit. ls -la
total 6172
drwxrwx--x 2 boinc boinc 4096 Feb 15 10:02 .
drwxrwx--x 16 boinc boinc 4096 Feb 1 11:12 ..
-rwxr-xr-x 1 boinc boinc 188 Feb 15 09:22 job_srbase7_l32_00018.xml
-rwxr-xr-x 1 boinc boinc 188 Feb 15 09:58 job_srbase_l32_00018.xml
-rw-r--r-- 1 boinc boinc 5037430 Feb 15 09:22 srbase_v5_l32.zip
-rwxr-xr-x 1 boinc boinc 1261648 Feb 15 09:22 wrapper_26012-v2_i686-pc-linux-gnu
|
12)
Message boards :
Number crunching :
LLR Version 3.8.17 released
(Message 2360)
Posted 15 Feb 2016 by Sergey Kovalchuk Totally errored on linux 32 bit, ok on linux 64 bit (Ubuntu 14.04) Host |