05-22-2014 01:28 PM
When using LabVIEW 2013 Professional with LabVIEW Unit Test Framework Toolkit 2013 on Windows 7 Professional 32-bit, there seems to be an issue with the code coverage metric when using parallel for loops. In my test, there is an array that is fed into a parallel for loop for processing. When I run the unit test on the VI, I can't get 100% code coverage because of the for loop. If I disable parallel execution in my for loop, the unit test framework gives 100% code coverage. Does anyone know how to allow for parallel for loops while still getting 100% code coverage in the unit test framework?
05-23-2014 02:05 PM
Hi ductape1,
I am checking with our R&D team on this behavior. It may take them a few days to get back to me but I will post here as soon as I have an update.
05-23-2014 02:12 PM
It brings up a few interesting points to ponder.
Can you share how much the metric changes with a parallized loop? (does it resolve to 1 diagram or "P-1" that is uncovered?)
06-04-2014 12:40 PM
Hi ductape1,
I have filed a Corrective Action Request for this issue (CAR# 475037). There currently isn't a timeline for fixing this issue, but you can use the CAR number to check whether this issue has been fixed/
06-04-2014 05:39 PM
Catherine, thanks for the information! I'm glad to know it wasn't just me having the problem.
06-04-2014 05:41 PM
Jeff,
The metric just counts the parallelized for loop as one path, exactly the same as passing n=1, and the metric wants to count the looping of the loop as a path, so n=2 would be sufficient for 100% coverage.