Continuing from my first question, if the process continues where it left off when it was removed from running, then the time that the command calls exit will correspond to the total amount of time that process is on the CPU. That sounds too good to be true, so I'm assuming that the command file should not have a syscall interval that is greater than the time quantum.