2 ms·
In fact, with a sufficiently large address space, you can prove that a "random" return value for malloc() is more reliable than a correct implementation, becaus
by kabdib 13y ago
In fact, with a sufficiently large address space, you can prove that a "random" return value for malloc() is more reliable than a correct implementation, because of the likelihood that a machine failure would return a "bad" (collision) result when running the longer code path of the correct implementation.
256 bit addresses, anyone? Well, maybe 1024... :-)