Hands-On System Programming with Linux

Why a process stack?

We're taught to write nice modular code: divide your work into subroutines, and implement them as small, easily readable, and maintainable C functions. That's great.

The CPU, though, does not really understand how to invoke a C function, how to pass parameters, store local variables, and return a result to the calling function. Our savior, the compiler, takes over, converting C code into an assembly language that is capable of making this whole function thing work.

The compiler generates assembly code to invoke a function, passes along parameters, allocates space for local variables, and finally, emits a return result back to the caller. To do this, it uses the stack! So, similar to the heap, the stack is also a dynamic ...