GPU not picking up work...need help.
3 Dec 2015 17:54:34 UTC
Topic 198332
(moderation:
This is what is picked up from the logs. not sure what I am doing wrong. As it wont pick up new work.
https://einstein.phys.uwm.edu/host_sched_logs/12141/12141585
GPU not picking up work...need help.
)
The relevant message from the scheduler log is this:
CUDA compute capability required min: 100, supplied: 0
I would try to either reinstall or update the graphics driver.
Thank you Holmls small
)
Thank you Holmls
small update from the synaptic package manager, got me up and running. :)
Your welcome and good to know
)
Your welcome and good to know that it was an easy fix! =)
AM back with another one..
)
AM back with another one.. :D
in my other system... It wont find my GPU,(r9 280x)seems like I have installed the driver ok, but boinc is not picking it up.
Sat 05 Dec 2015 11:05:52 GMT | | Starting BOINC client version 7.2.42 for x86_64-pc-linux-gnu
Sat 05 Dec 2015 11:05:52 GMT | | log flags: file_xfer, sched_ops, task
Sat 05 Dec 2015 11:05:52 GMT | | Libraries: libcurl/7.35.0 OpenSSL/1.0.1f zlib/1.2.8 libidn/1.28 librtmp/2.3
Sat 05 Dec 2015 11:05:52 GMT | | Data directory: /var/lib/boinc-client
Sat 05 Dec 2015 11:05:52 GMT | | OpenCL CPU: AMD Opteron(TM) Processor 6276 (OpenCL driver vendor: Advanced Micro Devices, Inc., driver version 1729.3 (sse2,avx,fma4), device version OpenCL 1.2 AMD-APP (1729.3))
Sat 05 Dec 2015 11:05:52 GMT | | No usable GPUs found
Sat 05 Dec 2015 11:05:52 GMT | | Host name: H8QG6
Sat 05 Dec 2015 11:05:52 GMT | | Processor: 64 AuthenticAMD AMD Opteron(TM) Processor 6276 [Family 21 Model 1 Stepping 2]
Sat 05 Dec 2015 11:05:52 GMT | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid amd_dcm aperfmperf pni pclmulqdq monitor ssse3 cx16 sse4_1 sse4_2 popcnt aes xsave avx lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt lwp fma4 nodeid_msr topoext perfctr_core perfctr_nb arat cpb hw_pstate npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold vmmcall
Sat 05 Dec 2015 11:05:52 GMT | | OS: Linux: 3.16.0-38-generic
Sat 05 Dec 2015 11:05:52 GMT | | Memory: 31.40 GB physical, 31.98 GB virtual
Sat 05 Dec 2015 11:05:52 GMT | | Disk: 23.17 GB total, 15.57 GB free
Sat 05 Dec 2015 11:05:52 GMT | | Local time is UTC +0 hours
Sat 05 Dec 2015 11:05:52 GMT | | Config: GUI RPCs allowed from:
Sat 05 Dec 2015 11:05:52 GMT | Einstein@Home | URL http://einstein.phys.uwm.edu/; Computer ID 12142002; resource share 100
Sat 05 Dec 2015 11:05:52 GMT | NumberFields@home | URL http://numberfields.asu.edu/NumberFields/; Computer ID 26286; resource share 100
any help would be great
RE: any help would be
)
Short answer (common problem), boinc may not be finding the libOpenCL.so
$ cd /usr/lib
$ ls -l libOpen* # is it missing ?
lrwxrwxrwx 1 root root 23 Jun 4 2015 libOpenCL.so -> /usr/lib/libOpenCL.so.1
-rw-r--r-- 1 root root 26184 Jun 2 2015 libOpenCL.so.1
# not on my system but if missing - create a symbolic link
$ sudo ln -s /usr/lib/libOpenCL.so.1 libOpenCL.so
# restart boinc
Long answer: if this does not solve the issue please look over the boinc forum, where this issue is getting some traction at the moment. see here The GPU forum there has a lot of information as well. Good luck!
Edit: you mentioned
What was that update?
Thank you very much. That
)
Thank you very much. That seem to do the trick. Is up and running now.
did notice, that if I restart the system, I have to then restart boinc from terminal to see the GPU.
''small update from the synaptic package manager, got me up and running. :)''
Am sure I didn't have CUDA package installed.
Linux newbie here. :D
RE: Thank you very much.
)
Good news, you are now probably hitting the xhost issue. There may be several ways around the issue - this assumes you are running lightdm (ubuntu standard display manager)
See here for more details.
OK thanks for that, useful to know.
I haven't been able to get
)
I haven't been able to get any work units for several weeks. The Server Status page says that none of the Work Generator Programs are running. Yet, some people appear to be getting work units. I would really like the people who run this project to say what is going on.
RE: I haven't been able to
)
You're getting plenty of work units (for CPU that is), you just haven't managed to download a single one in weeks. I'd guess there's some antivirus/firewall/whatever interfering. Besides that, there's no application for Intel GPU on Mac and something is wrong with your CUDA setup so you're not getting GPU work units.
They are not running because there is enough work ready to send.
More than some, or we'd have heard of it.
They can't. Nor can the people who run Rosetta (hint).
(Edit: typo)
RE: I would really like the
)
Comparing the post at Rosetta here - i can't see anything abnormal.