Cosmic Dave doesn't understand the first thing about computers, especially embedded systems. An embedded system is different than your typical laptop or desktop. Very different.
In the AGC one doesn't "load" a program into memory as what is done in a general-purpose computer. The program is already in read-only memory and can run from there; there is no disk. It just has to be identified to the operating system as a program that needs to be run in the regular rotation. That was the reason why programs were activated and deactivated through the various phases of the flight.
Now in truth the erasable memory was allocated and deallocated. And one of the 120x program alarms had to do with the allocation of an erasable memory block. But the allocation scheme was a combination of dynamic and static assignments, as well as shared memory. The mix of programs had to be carefully controlled ahead of time so that the static assignments wouldn't stomp on each other.
But obviously the landing and takeoff programs could use the same allocation because they would never be run simultaneously.
|