PDA

View Full Version : Issues running 7850 & 7950 in one case


litehouse43
03-04-13, 01:32 AM
Hi all,
Just ran into a snag when I added an HD 7850 to my case which already had a 7950. It appears the 7850 is recognized but still not utilized. I had an app_info running on the 7950, could that be part of the problem? Anyone familiar with the problem/available workarounds?

Thanks!
~Jack

3/4/2013 1:29:18 AM | | No config file found - using defaults
3/4/2013 1:29:18 AM | | Starting BOINC client version 7.0.44 for windows_x86_64
3/4/2013 1:29:18 AM | | log flags: file_xfer, sched_ops, task
3/4/2013 1:29:18 AM | | Libraries: libcurl/7.25.0 OpenSSL/1.0.1 zlib/1.2.6
3/4/2013 1:29:18 AM | | Data directory: C:\ProgramData\BOINC
3/4/2013 1:29:18 AM | | Running under account Jack Home
3/4/2013 1:29:18 AM | | Processor: 12 GenuineIntel Intel(R) Core(TM) i7-3930K CPU @ 3.20GHz [Family 6 Model 45 Stepping 7]
3/4/2013 1:29:18 AM | | Processor: 256.00 KB cache
3/4/2013 1:29:18 AM | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 nx lm vmx tm2 dca popcnt aes pbe
3/4/2013 1:29:18 AM | | OS: Microsoft Windows 7: Ultimate x64 Edition, Service Pack 1, (06.01.7601.00)
3/4/2013 1:29:18 AM | | Memory: 15.95 GB physical, 31.90 GB virtual
3/4/2013 1:29:18 AM | | Disk: 931.29 GB total, 53.98 GB free
3/4/2013 1:29:18 AM | | Local time is UTC -5 hours
3/4/2013 1:29:18 AM | | CAL: ATI GPU 0: AMD Radeon HD 7900 series (Tahiti) (CAL version 1.4.1741, 3072MB, 3032MB available, 8064 GFLOPS peak)
3/4/2013 1:29:18 AM | | CAL: ATI GPU 1: (not used) AMD Radeon HD 7800 series (Pitcairn) (CAL version 1.4.1741, 2048MB, 2008MB available, 4608 GFLOPS peak)
3/4/2013 1:29:18 AM | | OpenCL: ATI GPU 0: AMD Radeon HD 7900 series (Tahiti) (driver version 1084.4 (VM), device version OpenCL 1.2 AMD-APP (1084.4), 3072MB, 3032MB available, 8064 GFLOPS peak)
3/4/2013 1:29:18 AM | | OpenCL: ATI GPU 1 (not used): AMD Radeon HD 7800 series (Pitcairn) (driver version 1084.4 (VM), device version OpenCL 1.2 AMD-APP (1084.4), 2048MB, 2008MB available, 4608 GFLOPS peak)
Al
03-04-13, 09:10 AM
Looks like you need a cc_config.xml file with the <use_all_coprocessors>1</use_all_coprocessors> line. Right now bm is just using the most capable gpu.
John P. Myers
03-04-13, 06:40 PM
Looks like you need a cc_config.xml file with the <use_all_coprocessors>1</use_all_coprocessors> line. Right now bm is just using the most capable gpu.

What Al said. BOINC always refuses to use your weakest GPU unless you specifically tell it to use it. Just add the line Al mentioned in between the <Option> tags in your cc_config file and you'll be good to go. May require a BOINC restart to make it kick in since ordering BOINC to read the config file is usually glitchy.
litehouse43
03-05-13, 12:42 AM
I actually do not have a cc_config file enabled. Any recommendations on getting one written? Are there pre-done ones I can use?
Fire$torm
03-05-13, 01:33 AM
I actually do not have a cc_config file enabled. Any recommendations on getting one written? Are there pre-done ones I can use?

The cc_config.xml file goes in your BOINC Data folder.

Example:


<cc_config>
<options>
<start_delay>10</start_delay>
<ncpus>8</ncpus>
<network_test_url>http://www.yahoo.com/</network_test_url>
<report_results_immediately>1</report_results_immediately>
</options>
</cc_config>


You can add <use_all_coprocessors>1</use_all_coprocessors> anywhere within the <options> </options> section.

Edit: More info here ---> http://boinc.berkeley.edu/wiki/Client_configuration
Slicker
03-05-13, 06:04 PM
The cc_config.xml file goes in your BOINC Data folder.

Example:


<cc_config>
<options>
<start_delay>10</start_delay>
<ncpus>8</ncpus>
<network_test_url>http://www.yahoo.com/</network_test_url>
<report_results_immediately>1</report_results_immediately>
</options>
</cc_config>


You can add <use_all_coprocessors>1</use_all_coprocessors> anywhere within the <options> </options> section.

Edit: More info here ---> http://boinc.berkeley.edu/wiki/Client_configuration


DA removed support for report_results_immediately so you will likely see a warning in the event log if you use it in a recent (7.x) version.
Mumps
03-05-13, 10:37 PM
DA removed support for report_results_immediately so you will likely see a warning in the event log if you use it in a recent (7.x) version.

Where did you see that Slicker? I went hunting on the dev forums and see what looks like a log from 7.0.52 that implies it was still valid for that version. I need to know if I stop at 7.0.28, 7.0.42 or go current and 7.0.52 with all my Linux hosts... :) (7.0.52 has survived almost an entire month as the latest version.)
Fire$torm
03-06-13, 12:15 AM
I'm not sure but I think that support for that was re-enabled. Running 7.0.31 on all my boxes and I do not see any error messages.





01 ---- ---- ---- 3/5/2013 7:35:19 PM ---- Starting BOINC client version 7.0.31 for windows_x86_64
02 ---- ---- ---- 3/5/2013 7:35:19 PM ---- log flags: file_xfer, sched_ops, task
03 ---- ---- ---- 3/5/2013 7:35:19 PM ---- Libraries: libcurl/7.25.0 OpenSSL/1.0.1 zlib/1.2.6
04 ---- ---- ---- 3/5/2013 7:35:19 PM ---- ***********************************
05 ---- ---- ---- 3/5/2013 7:35:19 PM ---- Running under account *******
06 ---- ---- ---- 3/5/2013 7:35:19 PM ---- Processor: 8 GenuineIntel Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz [Family 6 Model 26 Stepping 5]
07 ---- ---- ---- 3/5/2013 7:35:19 PM ---- Processor: 256.00 KB cache
08 ---- ---- ---- 3/5/2013 7:35:19 PM ---- Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 nx lm vmx tm2 popcnt pbe
09 ---- ---- ---- 3/5/2013 7:35:19 PM ---- OS: Microsoft Windows 7: Home Premium x64 Edition, (06.01.7600.00)
10 ---- ---- ---- 3/5/2013 7:35:19 PM ---- Memory: 11.99 GB physical, 30.30 GB virtual
11 ---- ---- ---- 3/5/2013 7:35:19 PM ---- Disk: 350.49 GB total, 263.54 GB free
12 ---- ---- ---- 3/5/2013 7:35:19 PM ---- Local time is UTC -5 hours
13 ---- ---- ---- 3/5/2013 7:35:19 PM ---- NVIDIA GPU 0: GeForce GTX 460 (driver version 306.23, CUDA version 5.0, compute capability 2.1, 768MB, 695MB available, 1142 GFLOPS peak)
14 ---- ---- ---- 3/5/2013 7:35:19 PM ---- ATI GPU 0: ATI Radeon HD 5800 series (Cypress) (CAL version 1.4.1741, 1024MB, 991MB available, 4144 GFLOPS peak)
15 ---- ---- ---- 3/5/2013 7:35:19 PM ---- OpenCL: NVIDIA GPU 0: GeForce GTX 460 (driver version 306.23, device version OpenCL 1.1 CUDA, 768MB, 695MB available)
16 ---- ---- ---- 3/5/2013 7:35:19 PM ---- OpenCL: ATI GPU 0: ATI Radeon HD 5800 series (Cypress) (driver version 1084.2 (VM), device version OpenCL 1.2 AMD-APP (1084.2), 1024MB, 991MB available)
17 ---- Poem@Home ---- 3/5/2013 7:35:19 PM ---- Found app_info.xml; using anonymous platform
18 ---- ---- ---- 3/5/2013 7:35:19 PM ---- Config: simulate 8 CPUs
19 ---- ---- ---- 3/5/2013 7:35:19 PM ---- Config: report completed tasks immediately
20 ---- ---- ---- 3/5/2013 7:35:19 PM ---- Config: GUI RPC allowed from:
Duke of Buckingham
03-06-13, 08:06 AM
That is working for me either. It is reporting tasks as long as they are ready.
Mumps
03-06-13, 10:26 PM
Where did you see that Slicker? I went hunting on the dev forums and see what looks like a log from 7.0.52 that implies it was still valid for that version. I need to know if I stop at 7.0.28, 7.0.42 or go current and 7.0.52 with all my Linux hosts... :) (7.0.52 has survived almost an entire month as the latest version.)

LOL And 7.0.54 was posted this evening! :)

EDIT: But as of yet, still no change log posted in the Dev forum.