| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Beta Testing
Forum: Beta Test Support Forum
Thread: Beta Test for Help Conquer Cancer - GPU v6.51 |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 63
|
|
| Author |
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges: 
|
FWIW I checked the resends I have received so far. Of the 10 errors and 2 invalids 10 of the crunchers were running either a 6.10.xx version or 6.12.xx version of BOINC. This is pretty much the same results I saw during the last beta. Looks to me like anything less that a 7.0.xx version of BOINC is a problem.
----------------------------------------
In 1969 I took an oath to defend and protect the U S Constitution against all enemies, both foreign and Domestic. There was no expiration date.
  |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
https://secure.worldcommunitygrid.org/ms/devi...og.do?resultId=1060132751
This WU restarted @ 0% after I suspended boinc - changed from 100% cpu to 90%cpu. WU was @ 89% when I suspended |
||
|
|
slakin
Advanced Cruncher Joined: Jul 4, 2008 Post Count: 79 Status: Offline Project Badges: 
|
I decided to actually turn Throttle on to see if I could reproduce the problem but BETA units seem to run fine ..longer elapsed times ..double when set Max CPU 50% but no other issues seen. So on my machine the issue seems to be when I set the CPU usage below 100% in preferences.
----------------------------------------Hope this helps. By the way, I successfully run BETA WU's on <core_client_version>6.10.58</core_client_version> [Edit 1 times, last edit by slakin at Sep 21, 2012 9:25:58 PM] |
||
|
|
hendermd
Cruncher United States Joined: Apr 30, 2010 Post Count: 29 Status: Offline Project Badges: 
|
My system[2600k w/AMD HD 5670] with Boinc Manager ver. 6.12.34(x64) and running 100% of processors is running without errors or WU hanging (the same result as ver 6.50). The only difference I am seeing with this test is the large increase in cache size (my 3 day limit looks to have doubled so far) which did not happen with beta 6.50.
---------------------------------------- |
||
|
|
Eurwin
Cruncher Joined: Apr 28, 2007 Post Count: 17 Status: Offline Project Badges: 
|
Hi,
----------------------------------------I was running the beta without any problem. So I also tried to reproduce the problem of a stuck work unit. I altered "max CPU time" too 80% and then at 99.415% completion (that's when the GPU part is done) , the work unit got stuck, 0% CPU usage for that thread and it stayed there. When I changed the "max CPU time" back to 100% , the work unit stayed stuck. Suspending the work unit, then resume it, and it ran nicely too it's end. These are the last lines from stderr.txt when the work unit was stuck: Average kernel time: 0.093015 Min kernel time: 0.087920 (dx=3 dy=25 sample_dist=24 ) Max kernel time: 0.099186 dx=1 dy=1 sample_dist=0 INFO: Flushing queue...Waiting for all commands to finish...Releasing kernel...Releasing program...Freeing strings...Releasing queue... So for one reason the work units get stuck when they switch from GPU to CPU AND max CPU time usage is NOT 100%. I only use 4 of my 6 cores but that's causing no problems. Cheers, Erwin PS: my system is a Phenom II X6 1090T and a AMD HD7750 running Win7 64bit with BOINC client 7.0.28 [Edit 3 times, last edit by Eurwin at Sep 22, 2012 11:35:31 AM] |
||
|
|
wplachy
Senior Cruncher Joined: Sep 4, 2007 Post Count: 423 Status: Offline |
I have a different problem. I have 4 Betas that have been "Ready to start" for the last 60 minutes. CPU WUs end and new ones start but the Betas never change.
----------------------------------------OS: Win 7 64b Client: 6.10.58 CPU: I7P920 - Not OCed GPU: ATI Radeon HD 5770 Preferences are: While computer is in use is checked; Use 100% of CPU; and Use GPU while computer in use not checked. The GPU is not in use and hasn't been since the Betas downloaded. I have tried changing the On multiprocessor systems use at most to 90%. This suspended one of the CPU WUs but did not start a Beta. I then tried checking Use GPU while computer in use but no Beta started. I then reset Preferences and did a project suspend/resume. No Betas started. I then suspended all CPU ready to start WUs. Since then 4 CPU WUs have ended and three started. There are 7 (4 core with HT) CPU WUs running and no change to the Betas, all 4 are ready to start. I then aborted one of the ready to start WU whith no change. I then shut down the Client and re-booted the PC. After restarting the client it downloaded 1 more GPU WU and it now has 4 Ready to start GPU WUs. I have nothing in the Event Log other than the preference changes, suspend/resume WU and project activity and CPU WU end, upload messages. The GPU drivers are current and this PC have been running GPU Betas successfully since the GPU testing initially started. The pattern I've seen before was that after a few minutes one of the CPU tasks would go Waiting to Run and a GPU WU would start. Once a GPU WU started running they would continue as long as any were Ready to Start. Any one have a recommendation on getting these to start? Edit: corrected CPU type, added re-boot step
Bill P
---------------------------------------- [Edit 2 times, last edit by wplachy at Sep 21, 2012 10:23:51 PM] |
||
|
|
DunaDan_2004
Cruncher Joined: Nov 30, 2011 Post Count: 3 Status: Offline Project Badges:
|
1) Are you using the boinc CPU throttle? (Use at most N% of CPU) - Yes
----------------------------------------2) Do you have a value set "While processor usage is less than <blank>" that is not 0? - 80% 3) Can you paste the last 3 -4 lines from stderr.txt? This can be found in the slot directory where the gpu workunit is running whcih is in the boinc data dir typically in ProgramData\BOINC\slots\. There are numbered directories in there and the slot where the gpu workunit is running should have the opencl binary. Estimated kernel execution time = 0.36106 [sec] Starting analysis of ../../projects/www.worldcommunitygrid.org/161aea56b2b33bc3a1cfbb3249e1c239.jp2... Extracting GLCM features... Total kernel time: 193.516449 (1026 kernel executions) Total memory transfer time: 1.980802 Average kernel time: 0.188613 Min kernel time: 0.178055 (dx=5 dy=25 sample_dist=24 ) Max kernel time: 0.209928 dx=3 dy=1 sample_dist=2 INFO: Flushing queue...Waiting for all commands to finish...Releasing kernel...Releasing program...Freeing strings...Releasing queue...Releasing context...Releasing Memory Objects INFO: GPU calculations complete. Radon blur Level 0 Stuck in 99,415 % OS: Win 7 32 Client: 6.10.58 CPU:Athlon II X4 630 GPU: ATI Radeon HD 5770 [Edit 1 times, last edit by DunaDan_2004 at Sep 21, 2012 10:10:18 PM] |
||
|
|
ICG Studio
Cruncher Joined: Jan 17, 2011 Post Count: 12 Status: Offline Project Badges:
|
Win7 64, Boinc Manager 7.031 x64, 2 radeon 1st hd5870 2nd hd5830, so far ok when cpu set to 0%.
I suspend unit then i resume gpu 0 start computing next wu in quene, then gpu1 continue computing (suspended before) wu. No issue with return to 0% progress, it was start from 38%, after feeding gpu via cpu(in my case around 25 sec) computing was slightly slower... I would say like 1min approx. ( no cpu task at all, CPU is hexacore). Kind regards everyone:) |
||
|
|
Bearcat
Master Cruncher USA Joined: Jan 6, 2007 Post Count: 2803 Status: Offline Project Badges: 
|
I have always set my gpu to 60% and never had issues until this series of betas. Currently did not receive any for my ati 6950. Not sure why though.
----------------------------------------
Crunching for humanity since 2007!
 |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Are you using the boinc CPU throttle? (Use at most N% of CPU) No
----------------------------------------Do you have a value set "While processor usage is less than <blank>" that is not 0? Yes, 50% Can you paste the last 3 -4 lines from stderr.txt? This can be found in the slot directory where the gpu workunit is running whcih is in the boinc data dir typically in ProgramData\BOINC\slots\. There are numbered directories in there and the slot where the gpu workunit is running should have the opencl binary. Estimated kernel execution time = 1.45903 [sec] Starting analysis of ../../projects/www.worldcommunitygrid.org/d02ceee12e5c285359139c0192abe7ad.jp2... Extracting GLCM features... Total kernel time: 1010.657471 (1026 kernel executions) Total memory transfer time: 1.711292 Average kernel time: 0.985046 Min kernel time: 0.846665 (dx=25 dy=3 sample_dist=24 ) Max kernel time: 1.189607 dx=1 dy=1 sample_dist=0 INFO: Flushing queue...Waiting for all commands to finish...Releasing kernel...Releasing program...Freeing strings...Releasing queue...Releasing context...Releasing Memory Objects INFO: GPU calculations complete. Radon blur Level 0 Radon blur Level 1 Radon blur Level 2 Radon blur Level 3 Radon blur Level 4 Win7 64 Nvidia GT 540m [Edit 1 times, last edit by Former Member at Sep 21, 2012 10:38:49 PM] |
||
|
|
|