When running RMC on code with loops and without an --unwind bound, RMC will often not terminate.
But, it presently produces no output at all during this infinite loop, whereas CBMC produces continuous stream of logging data that helps indicate what's gone wrong.
I believe this is because at some point we're buffering stdout from cbmc and only printing it once it's finished. We should allow these log messages to stream to the terminal, so we can see them during a non-termination problem.
When running RMC on code with loops and without an
--unwindbound, RMC will often not terminate.But, it presently produces no output at all during this infinite loop, whereas CBMC produces continuous stream of logging data that helps indicate what's gone wrong.
I believe this is because at some point we're buffering stdout from cbmc and only printing it once it's finished. We should allow these log messages to stream to the terminal, so we can see them during a non-termination problem.