Currently tests that require controlling terminal are run sequentially, because at most one job can be in the foreground at a time. To run all the tests in parallel, we need to assign a terminal for each test file (by dynamically allocating a pseudo-terminal?).
magicant 
Fork
Currently tests that require controlling terminal are run sequentially, because at most one job can be in the foreground at a time. To run all the tests in parallel, we need to assign a terminal for each test file (by dynamically allocating a pseudo-terminal?).