From Friday, April 19th (11:00 PM CDT) through Saturday, April 20th (2:00 PM CDT), 2024, ni.com will undergo system upgrades that may result in temporary service interruption.

We appreciate your patience as we improve our online experience.

LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

SVFA Peak Amplitude and Relative Phase should be reentrant

Solved!
Go to solution

I think I've found an obscure problem with one of the "Baseband FFT" VI's on the "Frequency Analysis" subpalette of the Sound & Vibration Measurement Suite.

 

The "SVFA Peak Amplitude and Relative Phase" is set to non-reentrant execution, but I think it should be set to preallocated clone reentrant execution.  All the other "Baseband FFT" Vi's on the "Frequency Analysis" subpalette are set to preallocated clone reentrant execution.  This is crucial, because if you use multiple instances of the same VI on a block diagram and feed each instance a different set of waveforms inputs, each instance must maintain its own internal state for channel info, sampling rate, averaging, etc.

 

I just found out the hard way after lots of debugging that this particular VI was set to non-reentrant execution.  Averaging kept automatically resetting with warning 1908.  Sure enough, I had two instances on my block diagram, each receiving waveforms from a different set of channels.  Because the VI was non-reentrant, it was executing from one instance with one set of waveform inputs, then executing from the other instance with a different set of waveform inputs, continuing back and forth, and automatically resetting averaging each time because the channels kept alternating.

 

Do I have this right?

 

 

Message 1 of 4
(2,799 Views)

Hi Jeremy,

 

I took a look at this VI and I'm inclined to agree that having it set to non-reentrant by default appears to be an error. Your understanding of reentrancy is correct. After discussing this issue with a couple colleagues, I have submitted a corrective action request to see if this can be corrected in the future.  For now I suggest continuing to set it manually.

 

 

0 Kudos
Message 2 of 4
(2,759 Views)
Solution
Accepted by topic author Jeremy@MSI

Jeremy,

 

Looks like our test coverage can use some improvement. You're absolutely correct - this VI ought to have been set to preallocated clone reentrant. The AE who originally repsonded filed CAR# 570225 to our development team and it's been fixed for the next release of Sound and Vibration Software. 

 

Every release, we write a KnowledgeBase articule documenting these fixes, for example: the 2015 bug fix list.

 

Keep an eye out for the next release's bug fix list and that CAR ID to show up on it. 

Nathan Murphy
Message 3 of 4
(2,712 Views)

Excellent, thanks!

0 Kudos
Message 4 of 4
(2,701 Views)