On Sun, Sep 20, 2020 at 4:59 PM Steve Nickolas wrote: > I was under the impression that there was explicitly no requirement that a > null pointer be 0, > Indeed, section 7.19 states it is *implementation-defined*. See my previous message.