Suspicious usage of below-stack-pointer memory area for local variables

Question

I have recently discovered unclear fact about memory reservation for local variables in generated by g++ code: the variables are placed at the addresses below the stack pointer.

For example, when this program (main.cpp) is compiled:

void __attribute__ ((noinline)) foo() {
    volatile int v0, v1, v2, v3;
    v0 = 0;
    v1 = 1;
    v2 = 2;
    v3 = 3;
}

int main() {
    foo();
}

by command:

g++ -Ofast -S main.cpp

with compiler

g++ (Ubuntu 4.8.5-4ubuntu2) 4.8.5 on x86_64

then this assembly is created (main.s):

        .file   "main.cpp"
        .text
        .p2align 4,,15
        .globl  _Z3foov
        .type   _Z3foov, @function
_Z3foov:
.LFB0:
        .cfi_startproc
        movl    $0, -16(%rsp)
        movl    $1, -12(%rsp)
        movl    $2, -8(%rsp)
        movl    $3, -4(%rsp)
        ret
        .cfi_endproc
.LFE0:
        .size   _Z3foov, .-_Z3foov
        .section        .text.startup,"ax",@progbits
        .p2align 4,,15
        .globl  main
        .type   main, @function
main:
.LFB1:
        .cfi_startproc
        call    _Z3foov
        xorl    %eax, %eax
        ret
        .cfi_endproc
.LFE1:
        .size   main, .-main
        .ident  "GCC: (Ubuntu 4.8.5-4ubuntu2) 4.8.5"
        .section        .note.GNU-stack,"",@progbits

As you can see, local variables of function "foo" are placed at "-16(%rsp)" ... "-4(%rsp)" addresses.

But, as far as I know, these addresses (below the stack pointer register) are reserved for flow interrupting events, for example, for signal handling. So, the below-stack-pointer memory area can be unpredictably changed in any moment. So, it can't be used for local variables storage.

Can somebody, please, clarify this issue?