Why my bitfile does not work properly on a specific sbRIO-9607?

[richtian]

I have an sbRIO-9607 bitfile which has been deployed to 100+ sbRIOs and worked well on all those sbRIOs. However, with my latest sbRIO-9607 from NI, I find that the bitfile does not work on one specific 9607, based on current test. Since I need to deploy the bitfile to another several tens sbRIO-9607 from a same order, I'm afraid of meeting the same issue on other 9607s and not know what the root cause is. Anybody has the same experience or any idea on why this happens? 

Thanks,

Richie

[YannisS]

Hi Richie

have you had any luck in identifying the source of the problem? I have difficulties in deploying the application as well and came across this post.

My problem: the RT/FPGA applications work as desired in development mode, however, the deployment doesn't. I know that at least the FPGA application is not deployed. I'll put some checks in the RT as well. This situation applies to a new batch of sbRIOs.

Anyone any advice?

Thanks

Yannis

[Call.Me.Wilbur]

Hi Yannis,

How are you referencing the RIO resource from your RT VI? There are several ways of referencing a RIO device. Depending on how whether you're referencing it through an IP address, localhost, or just the RIO device name, you could see different behavior between development and deployment scenarios. I recommend referencing the RIO resource just by it's device name. That way the reference path doesn't have to loop back through the network stack. See this article for more details.

https://knowledge.ni.com/KnowledgeArticleDetails?id=kA00Z0000004BIiSAM&l=en-US

Richie,

Are you still having troubles with that 9607? Does the behavior show up on other devices from that same shipment? I'm also curious about the behavior your seeing. Do you get an error in deployment or does the application seem to never run?

Best,

Will

National Instruments

[richtian]

Hey, Yannis

Sorry for replying late. Just come back to work from a long long holiday because of Corov-19. 😞

I suspect that my problem is different from yours. I'm sure my VI is launched and bit file is deployed, while I'm getting a little different behavior from FPGA. It seems that my problem happens rarely among all my sbRIOs. So far, I've just run into 2 cases out of 50 sbRIOs. I ran into similar situation before on myRIO. I suspect it's something from the manufacturing of sbRIO. Looks like not a big deal.

For your case, I'm wondering whether the RT program is properly launched and run. If yes, probably, you'd check out the items listed by Will. If not, it should be easy to handle.

Thanks,

Richie

[richtian]

Hi, Will

Sorry for replying late. Just come back to work from a long long holiday because of Corov-19. 😞

I've checked almost all the sbRIOs from the shipment, it turns out that there are only few having the issue. The problem is that something calculated on FPGA is incorrect while I'm deploying a exact same image to all the sbRIOs. I ran into similar situation while using myRIO. There was only once case on myRIO, as I remember correctly. I'm suspecting that the bit file does not work properly on the specific fabric because of timing or something else. The good news for me is that this does not sound like a big deal at the moment.

Thanks for checking it out.

Richie

[Call.Me.Wilbur]

Hi Richie,

That's very interesting behavior. I would expect that any issues related to unit-to-unit manufacturing discrepancies would not easily affect operations on the FPGA. If you you're interested in troubleshooting the behavior of that unit further feel free to share your code and possibly some screenshots showing the differences in behavior.

Will

[richtian]

Hey, Will

Thanks for your offering and help as always. I'll bother you if I do run into the situation where many sbRIO cannot work my same lvbit.

Thanks,

Richie