I’d like to solve problems in parallel using OpenMP (using C++). However, the global parameter database does not contain the “num_threads” field. I’m using the FEniCS build from Ubuntu’s repos on a Podman image. I’m not using the official Docker image. Does anyone know if Ubuntu’s repos have FEniCS compiled with OpenMP support? Or is there another issue that may result in the “num_threads” option not being available?
Edit: To add some context, back when I used Python for this the Anaconda install would automatically use the full concurrency of the processor. Now, I just get single thread behavior under the Ubuntu repo version of FEniCS.