Stack memory is a memory usage mechanism that allows the system memory to be used as temporary data storage that behaves as a first-in-last-out buffer. One of the essential elements of stack memory operation is a register called the Stack Pointer. The stack pointer indicates where the current stack memory location is, and is adjusted automatically each time a stack operation is carried out.
In the Cortex®-M processors, the Stack Pointer is register R13 in the register bank. Physically there are two stack pointers in the Cortex-M processors, but only one of them is used at a time, depending on the current value of the CONTROL register and the state of the processor.
In common terms, storing data to the stack is called pushing (using the PUSH instruction) and restoring data from the stack is called popping (using the POP instruction). Depending on processor architecture, some processors perform storing of new data to stack memory using incremental address indexing and some use decrement address indexing. In the Cortex-M processors, the stack operation is based on a “full-descending” stack model. This means the stack pointer always points to the last filled data in the stack memory, and the stack pointer predecrements for each new data store (PUSH)
