Hands-On RTOS with Microcontrollers

Stack memory

A stack is used for function-scoped storage. Each time a function is called, information for that function (like its parameters and local variables) are placed onto a stack. When the function exits, all of the information that was placed onto the stack is removed (this is why passing pointers to local variables is a problem). In FreeRTOS, each task has its own private stack whose size is determined when the task is created.

Since stacks have such an orderly way of being accessed, it isn't possible for them to become fragmented, like a heap. It is possible, however, to overflow the stack by placing more information onto it than its size allows.

On the Cortex-M, there is also one additional stack—the main stack. The main stack ...