Clear Filters
Clear Filters

Issue with Simulink Model Simulation using CVSIM command inside Parfor ?

7 views (last 30 days)
Hi,
I am trying to run multiple simulations with different data set on model 'ModelTest.slx' using parfor.
my code goes like this,
parfor
...
...
testObj = cvtest(model);
testObj.settings.decision = 1;
testObj.settings.condition =1;
tempData.data = cvsim(testObj);
...
..
end
I am getting below error after execution
Caused by:
Error using cvsim
Transparency violation error.
Below are the details of matlab version and model settings,
Matalb 2015b,
parpool with local profile.
Simulation Mode : Normal;
Solver :Fixed Step, discrete
Does anyone have any idea about this error. ?
Or Is there any other way we can extract the coverage information without using "CV" functions. ?

Accepted Answer

Edric Ellis
Edric Ellis on 28 Feb 2019
Caveat: I know nothing about cvtest or cvsim. I suspect that cvsim is trying to inspect the calling workspace - this is not allowed inside parfor, and causes a "transparency violation". (This slightly cryptic term simply means that all accesses to variables must be "transparent" - i.e. obvious in the text of the program.)
Often, the simplest route is to pop the body of your parfor loop into a separate function, so you'd do:
parfor idx = ...
out(idx) = myCvSim(idx,model)
end
...
function out = myCvSim(idx,model)
testObj = cvtest(model);
...
out = cvsim(testObj);
end
  3 Comments
Pat Canny
Pat Canny on 18 Mar 2019
Just as a note: beginning with R2018b, you can collect coverage when using parsim. This might simplify your workflow. In older releases, Edric's solution is the way to go.
Shivaputra Narke
Shivaputra Narke on 25 Apr 2019
Thanks for the update I will try once I get 2018b. For now, I'm still using Matlab 2015b and 2017b.

Sign in to comment.

More Answers (0)

Categories

Find more on Parallel for-Loops (parfor) in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!