LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Show Buffer Allocations on Feedback Nodes

Using LV2009, when I click "Show Buffer Allocations", I see no blinking rectangles on feedback nodes, all the way back to the source of the initializer terminal. All items are checked on the SBA options page. What gives?

0 Kudos
Message 1 of 12
(4,453 Views)

Hey mechelecengr,

 

The blinking rectangles that are present when using the show buffer allocations tool represent places where your code is creating, not suprising, buffer allocations. The reason why you are able to see one of these blocks on a Feedback Node in, say, LabVIEW 8.6 and not LabVIEW 2009 can be attributed to several things. First, this functionality could be attributed to the algorithm used to detect when a buffer is being created. Second, this could be attributed to how the feedback node works in regards to how it allocates buffers.

Hope this helps.
-Ben

WaterlooLabs
0 Kudos
Message 2 of 12
(4,424 Views)

Would you do some poking around and see which is it?

 

If the tool is broken, it needs fixing. If FN's are somehow miraculously not allocating memory, let us know, so we can store all our data in feedback nodes!

 

Best regards,

Jack

0 Kudos
Message 3 of 12
(4,415 Views)

Hey Jack,

 

I'll look into whether this is an optimization of the feedback node or an alteration in the algorithm used to identify the buffer allocations through this Show Buffer Allocations tool and try to post back tomorrow.

Message Edited by BCho on 08-20-2009 02:26 PM
Hope this helps.
-Ben

WaterlooLabs
0 Kudos
Message 4 of 12
(4,396 Views)

Hey Jack,

 

Sorry for the delay. I have had a chance to look at this issue a little more and I believe your suspicions about the feedback node are correct. After running a simple feedback node example with the Desktop Execution Trace toolkit, I noticed that the buffered are allocated in a similar fashion as they are with shift registers; in that, each iteration of the loop they are in, the buffer associated with them increases (as one would expect). So, in terms of a buffer being allocated or not, there is a more memory being allocated for the feedback node just as there is with the shift register.

 

Since the shift register is recognized as being a source of a buffer allocation using the Show Buffer Allocation tool, so too should the feedback node. So, I have filed a corrective action request to revert back to recognize the feedback node as a source of buffer allocation just as the shift register is.

Hope this helps.
-Ben

WaterlooLabs
0 Kudos
Message 5 of 12
(4,346 Views)

Alright! Yes, that definitely helped, Ben. Of course, memory must be allocated for the nodes, so I suspected the Show Buffer Allocations detection routine was not properly registering FN's as buffer allocations.

 

What I REALLY wanted to hear is that Feedback Nodes became magical in LV2009, that are infinitely fast and consume zero memory. Until NI is able to implement that execution priority (you know, ...High Priority, Real-Time, Magical...), we will need to depend on the Show Buffer Allocations for FN code validation.

 

Please post the CAR#. 

 

Best regards,

Jack

Message 6 of 12
(4,331 Views)

Unfortunately, infinite execution speed is only available with the LabVIEW Magical Development System (MDS).

 

Chris M

 

(no, LabVIEW MDS doesn't really exist) 

0 Kudos
Message 7 of 12
(4,323 Views)

HI Ben,

 

I echo Jack's request for the CAR#

 

I use the show buffer alloations regularly and I really on it working correctly.

 

Please post the CAR# so that I can track this bug.

 

Thank you,

 

(another) Ben

Retired Senior Automation Systems Architect with Data Science Automation LabVIEW Champion Knight of NI and Prepper LinkedIn Profile YouTube Channel
0 Kudos
Message 8 of 12
(4,305 Views)

Hi all,

 

The CAR number is 183414.  Thanks again for the feedback!

 

 

Nick Keel 

Applications Engineering 

National Instruments

Nick Keel
Product Manager - NI VeriStand and Model Interface Toolkit
National Instruments
Message 9 of 12
(4,298 Views)
Sorry for the delay, that CAR # is 183414
Hope this helps.
-Ben

WaterlooLabs
Message 10 of 12
(4,265 Views)