I am aware, my point is about how the following is dealt with:

  fn foo(x: i32) i32 {
      return x + 100;
  }
  fn bar(x: i32) i32 {
      @setRuntimeSafety(false);
      return foo(x + 100);
  }

Particularly when you have arbitrarily deep call graphs where some explicitly enable and disable runtime safety.

You mean overflow checking? It only applies to the `x + 100` in bar. Bar can still panic if the `x + 100` in foo overflows. Why should it be any other way?