Is longjmp supposed to restore the stack?

Question

From what I understood, setjmp saves the current context and it's supposed to restore it when calling longjmp. However the next piece of code prints 15 (I compiled with -g and without any optimization). Did I misunderstand this construct or am I missing anything else?

#include <iostream>
#include <csetjmp>


std::jmp_buf jump_buffer;

int main()
{
    int a = 0;
    if (setjmp(jump_buffer) == 0) {
      a = 15;
      std::longjmp(jump_buffer, 42);
    }
    std::cerr << a << std::endl;
}

Disclaimer: only trying to use it for curiosity. I never heard about this construct until I recently read some paper about NASA coding guidelines that mentioned it's forbidden to use this flow of control construct

Using both c and c++ tags because the code is mixed and I would assume the actual relevant functions are more relevant to c heavy users rather than c++... :/

Answer 1

That's the expected behavior:

Upon return to the scope of setjmp, all accessible objects, floating-point status flags, and other components of the abstract machine have the same values as they had when std::longjmp was executed, except for the non-volatile local variables in setjmp's scope, whose values are indeterminate if they have been changed since the setjmp invocation.

The value of a when executing longjmp is 15, so that is a value one could expect to see (it's indeterminate in general). The jmp_buf only stores the point of execution. Not the state of every variable in the program.