AIT ARINC 429 Add-On Feedback

[Claudio_G]

Full text error is in message 51 from stripes

******AIT 429 CUSTOM DEVICE ERROR******
Source:    Owl429.lvlib:Owl429_DefaultErrorHandler.vi:437<ERR>
An API failure occurred. uapiMonInit returned -1. General error.

<b>Complete call chain:</b>
     Owl429.lvlib:Owl429_DefaultErrorHandler.vi:437
     Owl429.lvlib:Owl429_IBoard_resetWithConfig.vi:6
     AIT 429.lvlib:AIT 429 Configure All (BIN).vi:5
     AIT 429.lvlib:AIT 429 Configure Board (BIN).vi:4
     AIT 429 Custom Device.lvlib:429 RT Driver Init.vi:3
     AIT 429 Custom Device.lvlib:AIT 429 RT Driver VI.vi:2
     Custom Devices Storage.lvlib:Initialize Device (HW Interface).vi
     NI VeriStand Engine.lvlib:Initialize Inline Custom Devices.vi
     NI VeriStand Engine.lvlib:VeriStand Engine State Machine.vi
     NI VeriStand Engine.lvlib:VeriStand Engine.vi
     NI VeriStand Engine.lvlib:VeriStand Engine Wrapper (RT).vi
Code:    -2
******AIT 429 CUSTOM DEVICE ERROR******

[StephenB]

Hello,

I suggest contacting AIT. I cannot reproduce the deployment issue when I use my AIT 429 card with v4.6 and 2012. I can look into the performance issue if you have a configuration I can test.

[StephenB]

Just FYI I heard from AIT that the could not create FIFO message can be safely ignored. They will supress it in a future driver revision.    

[StephenB]

I have discovered several performance problems in later AIT drivers (v4.6.0) and later versions of NI VeriStand+LabVIEW (2012+). I have worked around the AIT issue and notified them, and made many changes to  improve performance in the custom device. Since I do not have near the complicated setup as many of the people posting here, I would like to post the next revision now as a beta. Please try this beta out and let me know if it 1) resolves performance issues and 2) continues to function exactly as previously.

Please note you must open your system defintion files, save, and close, after installing the beta so they are mutated up.

CHANGE LOG

1.8.0 BETA

• <Bug> Major performance issue seen in 2012+ HP loop duration fixed
• <Feature> Average CPU usage and write loop duration should be greatly reduced (sometimes an order of magnitude) because change detection is now used on TX values
• <Change> Loop duration status channels are now in nanoseconds
• <Bug> Errors should be reported more reliably in case of a catastrophic problem with read or write
• <Change> Loop late status channels have been removed. They did not function propely and often reported late when they were not. (Known issue 20)
• <Bug> Fixed known issue 18: the output loop duration channel was always zero
• <Bug> Fixed known issue 21: Worked around the significant performance degredation in the AIT 4.6.0 drivers (first used in 1.7.2) causing read and write loop slowness

I'm not posting 2010 here for now, as those builds take an hour. I can post it next week if requested.

Message was edited by: StephenB
I removed them from here as they are now officially posted on the main page.

[hajiao]

Hi,Stephen B

Do you have a plan to let this custom support "Block" and "Acycle" transfer type?

If not, is it allowble if I modified the source myself?

If yes, could you give me some advice on modifying this code?

I asked it because a customer request this feature.

[StephenB]

Hi Hajiao,

If you contact me internally we can discuss this more.

[hrobitaille]

We are looking to implement the timestamping on the incoming data.

In the Custom Device, we have seen a sequence to calculate the index in the buffer which depends if the board is AIT or AIM, but we have some difficulties to find the information about the buffer structure (where we can find the timestand) in the AIT documentation. 

The documents installed with the SDK are essentially generated automaticly from the source code. Is there a document like a programmer's guide which explain more than the API ?

Regards

[DMous]

Hey hrobitaille,

Have a look to the chapter 2.10.1.1 of the quick reference guide.
The timestamp is composed of the major time in seconds and the minor time in µs.

http://www.jtelec.fr/dl/ait/AIT_429_QRG.pdf

Best regards,

[hrobitaille]

Thanks for the link to the manual. It is exactly what we need.

I never see this manual before. It should be included in the SDK or available on the AIT Web site.

Regards

[StephenB]

Feel free to remove the AIM code path. The AIT driver dropped support for that card a long time ago.

If you are feeling charitable, it would be great if you could post your update back here so I can merge it into trunk