1)
Message boards :
Number crunching :
Some GPU tasks failing
(Message 71741)
Posted 11 Feb 2022 by jjv Post: Don't usually bother with DDU or it's ilk. Several decades of experience of limited usefulness. I did however recently do a thorough driver cleanup due to an unrelated problem. I also went through BOINC logs and can find no mention of older GPUs or drivers in recent logs. I think you accidentally looked at a year old invalid task. i don't do a lot of invalids ;-) Fully aware of the risks involved in using bleeding edge software (or hardware for that matter). Recent driver updates have however included certain fixes that pertain to issues I'm personally experiencing. Also, since I'm a very technical person by profession and hobby I consider it my role to be a tester in these things. I dialed down the amount of WUs allowed per GPU simultaneously. Will keep monitoring the situation but no similar failures have since happened. JJ |
2)
Message boards :
Number crunching :
Some GPU tasks failing
(Message 71721)
Posted 10 Feb 2022 by jjv Post: Well, yes. It's kind of hard to fully utilize a 3090 not to mention two of them without running multiple WUs. It actually might be the case that the error pertains to the WU not fitting in the GPU memory. Which is silly since these things have 24GB each but there seem to be limitations on GPU memory utilization with BOINC. Where on earth did that 1080Ti reference come from??? Those haven't been in use for over half a year. JJ |
3)
Message boards :
Number crunching :
Some GPU tasks failing
(Message 71718)
Posted 10 Feb 2022 by jjv Post: Noticed a fair amount of errors among my GPU tasks. Some work, others don't. The logs seem to indicate a problem with device detection: Success: https://milkyway.cs.rpi.edu/milkyway/result.php?resultid=106001665 Fail: https://milkyway.cs.rpi.edu/milkyway/result.php?resultid=106001938 This is a dual GPU machine but can't figure out from the log which GPU the WU was using or trying to use. JJ |
4)
Message boards :
Number crunching :
All fixedangles WUs crashing immediately
(Message 64888)
Posted 14 Jul 2016 by jjv Post: Seems a bit better. Now only some of them fail :-) The active discussion on this issue seems to be in the "No More Work From Modfit Project" thread. http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=3973 JJ |
5)
Message boards :
Number crunching :
All fixedangles WUs crashing immediately
(Message 64879)
Posted 13 Jul 2016 by jjv Post: Same here. And I'm running win7. Checking the tasks they seem to be failing for everyone. Regardless of OS, GPU vendor or even CPU task. JJ |
6)
Message boards :
Number crunching :
GPU lacks necessary double precision extension
(Message 53028)
Posted 10 Feb 2012 by jjv Post: Hmm. So it seems. I wonder what changed. JJ |
7)
Message boards :
Number crunching :
GPU lacks necessary double precision extension
(Message 52970)
Posted 9 Feb 2012 by jjv Post: My other machine is running "stable" and is using the 6.12.34 version. It is running fine. Both machines have an nVidia GPU, but otherwise they are different softwarewise. I personally also suspect the pre-release BOINC. The last three versions have all had changes to the OpenCL detection and reporting. JJ |
8)
Message boards :
Number crunching :
GPU lacks necessary double precision extension
(Message 52962)
Posted 9 Feb 2012 by jjv Post: I haven't encountered any problems with these drivers. This particular machine is a sort of testbed where I run the bleeding edge versions of most software unless I run into problems. This isn't a serious issue as the GPU is fully utilized by other projects, but I was wondering if anyone had an idea where in the software chain the problem originates... JJ |
9)
Message boards :
Number crunching :
GPU lacks necessary double precision extension
(Message 52959)
Posted 9 Feb 2012 by jjv Post: So apparently my GTX580 has recently lost it's double precision capability... "Milkyway@Home | Message from server: GPU lacks necessary double precision extension" I'm running the latest pre-release BOINC (7.0.14 as of this writing) and I know they have been fiddling with the OpenCL detections with the recent versions. I'm also using the latest nVidia beta drivers (295.51 x64) so either could be at fault I suppose. JJ |
©2024 Astroinformatics Group