PrimeGrid
Please visit donation page to help the project cover running costs for this month

Toggle Menu

Join PrimeGrid

Returning Participants

Community

Leader Boards

Results

Other

drummers-lowrise
1) Message boards : Generalized Fermat Prime Search : GFN17Low GPU Error (Message 111911)
Posted 1241 days ago by NUCCpod_NAPTIMELABS_01
Side question, as I am yet unfamiliar with the the details concerning the creation of applications for CUDA and the like, is it not possible to check the CC level of a given card before the execution begins in earnest?
2) Message boards : Generalized Fermat Prime Search : GFN17Low GPU Error (Message 111908)
Posted 1241 days ago by NUCCpod_NAPTIMELABS_01
Purging and re-installation of the driver (340.102) did not bring us a fix sadly.
I've taken note of the error and will circle back around to further testing down the road when I have the time.

That being said, and after double checking my logs, I'm going to chalk this up #10 on the GFN notes.

At this time the OCL4-high transform (used only for n=15, n=16, n=17-low, and n=17-mega), when run on an Nvidia GPU, requires CC 2.0 (Compute Capability). This generally means at least a GTX 4xx series GPU, although there are some exceptions. If you're not sure, you can check on Nvidia's CUDA GPU list and Nvidia's Legacy CUDA GPU list. There's no restriction on ATI/AMD GPUs.


The gtx260 I used to double check the presence of the error is only 1.3 (iirc) and I'm currently unable to check on the quadro fx 4800, but given that the same error is present for both cards.... Well, seems like a pretty strong signifier.

Thanks for the help poking this

3) Message boards : Generalized Fermat Prime Search : GFN17Low GPU Error (Message 111905)
Posted 1241 days ago by NUCCpod_NAPTIMELABS_01
Interesting, I'm going to poke in a bit more and then check in on the drivers.
4) Message boards : Generalized Fermat Prime Search : GFN17Low GPU Error (Message 111900)
Posted 1241 days ago by NUCCpod_NAPTIMELABS_01
In my test case I have been watching the following task on a worker restart from scratch a few times in row, I suspended primegrid overnight and have been attempting to gather more data today.

https://www.primegrid.com/workunit.php?wuid=545875931

Is one of the tasks that is restarting and throwing this message to the event log.

Running on linux, and checking on /var/log/boincerr.log shows nothing (seemingly) useful for this message. But I did find that a new entry of "No protocol specified" is thrown there about once per second, so I'll be digging into that for a moment.
5) Message boards : Generalized Fermat Prime Search : GFN17Low GPU Error (Message 111891)
Posted 1242 days ago by NUCCpod_NAPTIMELABS_01
Sat 25 Nov 2017 02:08:41 AM PST | PrimeGrid | <![CDATA[Task genefer17low_13338015_2 exited with zero status but no 'finished' file]]>
Sat 25 Nov 2017 02:08:41 AM PST | PrimeGrid | <![CDATA[If this happens repeatedly you may need to reset the project.]]>

The task just restarts from scratch at this point...

After finding this error while doing some burn in testing on a new worker, I went back through my logs and found that several more workers had this message as well. Not just with primegrid but on other project's work units as well.

I'm curious about this message and am wondering if anyone here has some more detailed information.

[Return to PrimeGrid main page]
DNS Powered by DNSEXIT.COM
Copyright © 2005 - 2021 Rytis Slatkevičius (contact) and PrimeGrid community. Server load 0.00, 0.00, 0.00
Generated 20 Apr 2021 | 10:45:45 UTC