I read a description where each of the 2 processors care for a specific subset of the 15 (or so) FCC processes. It is only in case a fault is detected that the valid processor take over all the processes.
I wrote in another thread : I wouldn't be surprised if the overload comes from the additional checks added to the MCAS process. If the MCAS trigger keeps being repeated and keeps being inhibited, if the new checks involve much I/O with wait states...