By design, CTest WILL_FAIL test property does not handle segmentation faults and other unexpected terminations. That is, a failing test will also show as failed in CTest, when the desire may have been to show an intended failure as success via WILL_FAIL. A simple workaround for this is to make a wrapper script that calls the intentionally segfaulting executable and catch that result, and pass that to CTest, as in this example project.