| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: GPU Support Forum
Thread: BOINC 7.0.40 warning |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 109
|
|
| Author |
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Thanks Rob!
|
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges: 
|
Has anyone who jumped on 7.0.40 actually tested app_config to see if it functions to control how many GPU tasks run concurrent on a task, AND important too, how it behaves with multiple cards of different capability (without app_info to prevent conflict)? No multiple cards, but some experience (without using an app_info) I wanna share with Rob and other interested people: I simplified the app_config.xml, because the gpu-part is of no interest for cep2 and removed the max_concurrent in the hcc-app, because the number of concurrent tasks should be controlled by <gpu_usage>, if you don't run HCC CPU.: <app_config>At first every seems to working as expected. The max of 1 CEP-task and 2 GPU HCC were running. But after receiving new HCC tasks one of the running HCC tasks change the state to "Waiting to run". Even suspending all CPU tasks couldn't force this 2nd task to resume again. Before customizing the app_config I had the same experience, but wasn't sure, whether it was caused by a starting cep-task. Now I knew it too was caused by receiving new GPU-tasks for WCG. When you were running HCC GPU and CEP2 CPU tasks how did you have your profile set up to get only CEP2 CPU tasks and no HCC CPU tasks? A screenshot, if possible, would be a big help.
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.
----------------------------------------  [Edit 1 times, last edit by nanoprobe at Dec 14, 2012 6:27:28 PM] |
||
|
|
Crystal Pellet
Veteran Cruncher Joined: May 21, 2008 Post Count: 1294 Status: Offline Project Badges:
|
When you were running HCC GPU and CEP2 CPU tasks how did you have your profile set up to get only CEP2 CPU tasks and no HCC CPU tasks? A screenshot, if possible, would be a big help. No 1 profile. That's still a wish where WCG-techs are working on.I only have to switch between a CPU-profile and after loading enough CPU-tasks, back to a HCC GPU only profile.  |
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges:
|
When you were running HCC GPU and CEP2 CPU tasks how did you have your profile set up to get only CEP2 CPU tasks and no HCC CPU tasks? A screenshot, if possible, would be a big help. No 1 profile. That's still a wish where WCG-techs are working on.I only have to switch between a CPU-profile and after loading enough CPU-tasks, back to a HCC GPU only profile. I was afraid of that. Thanks. On the plus side the app_config now lets me run multiple GPU tasks on my 32 bit XP box. The app_info would not let me do that.
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.
----------------------------------------[Edit 1 times, last edit by nanoprobe at Dec 14, 2012 9:23:23 PM] |
||
|
|
ThreadRipper
Veteran Cruncher Sweden Joined: Apr 26, 2007 Post Count: 1314 Status: Offline Project Badges: 
|
Nope...7.0.42 kept erroring out WUs and I had to switch back to 7.0.38. Now seems to be runing smooth again with AMD 12.11 Beta 11 driver. However, I have once got a "the graphics driver stopped responding and has been reset" message or similar. As this happened, the running WUs in BOINC got stuck and I had to abort them.
----------------------------------------Still, a beta driver is a beta driver...but will have to wait for a while before they release a newer version of BOINC client before I try that again. Too bad since I like the new app_config file :(  Join The International Team: https://www.worldcommunitygrid.org/team/viewTeamInfo.do?teamId=CK9RP1BKX1 AMD TR2990WX @ PBO, 64GB Quad 3200MHz 14-17-17-17-1T, RX6900XT @ Stock AMD 3800X @ PBO AMD 2700X @ 4GHz |
||
|
|
OldChap
Veteran Cruncher UK Joined: Jun 5, 2009 Post Count: 978 Status: Offline Project Badges: 
|
On each rig I ran the existing work dry (important), stopped Boinc, removed app_info, loaded 7.0.42, installed app_config and re-started rig. I played with the settings a little and found that one can induce an imbalance between cards which may be of use to some folks with dissimilar GPU's in the same machine but in general I would have to say that it does exactly what it says on the tin. smooth running and feels mature.
----------------------------------------I am running gpu only 7950 and 5850 and 5870's on win7 and Vista (both 64) and using both 12.10 and 12.11 drivers. nano: in the instructions there is reference to running two instances of boinc (in different folders). Could you run one as gpu only using one profile and another that is cpu only and uses a profile that excludes HCC?  [Edit 1 times, last edit by OldChap at Dec 16, 2012 1:24:18 AM] |
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges:
|
However, I have once got a "the graphics driver stopped responding and has been reset" message or similar. As this happened, the running WUs in BOINC got stuck and I had to abort them. The driver error you're getting is a problem with how Windows handles the drivers. MS has acknowledged the problem and to fix it requires a very simple registry edit. You don't have to abort the tasks that get stuck. Just suspend all of you GPU tasks and restart the stuck ones one at a time. Then resume all your other tasks. And FWIW I had problems with the 12.11 beta 11 driver. Switched back to the 12.11 beta 4 and no more problems.
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.
----------------------------------------[Edit 2 times, last edit by nanoprobe at Dec 16, 2012 1:12:32 AM] |
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges:
|
nano: in the instructions there is reference to running two instances of boinc (in different folders). Could you run one as gpu only using one profile and another that is cpu only and uses a profile that excludes HCC? Have no idea. Sounds like it could work but I'm too lazy to try it. For me it's easier to fill up SN2S with 5 days of cache and switch the profile back to GPU and wait for a resolution from the techs.
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.
|
||
|
|
OldChap
Veteran Cruncher UK Joined: Jun 5, 2009 Post Count: 978 Status: Offline Project Badges:
|
I may have spoken a little too soon.
----------------------------------------I have two win7 rigs each running a pair of 5870's One is a ud4 b3 with a 2600K which has crunched for years it seems The other is a x68 ud4 with 3770K at stock Using app_info these would both throw a small number of errors a day usually at around the 8 second mark. Since changing to the app_config and its associated boinc version the incidence of errors has risen dramatically when there have been no other changes. So far I have seen that reducing the number of instances running on a card has had a positive effect on this problem on one machine (yet to try the other) This is then a warning to those of you that are trialling this to keep an eye out for errors ... at least on older cards. |
||
|
|
nanoprobe
Master Cruncher Classified Joined: Aug 29, 2008 Post Count: 2998 Status: Offline Project Badges:
|
How many tasks were you running on the 5870s
----------------------------------------
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.
|
||
|
|
|