Well so the kernel does zero fill pages. I get the impression no one except me cares about this (in my eyes) pretty useless waste of machine resources (and therefore my valuable time). Even worse this puts a natural limit on the Mullocator, at least when new memory needs to be acquired. In certain benchmarks, you just can't go any faster because you are starting to hit that brick wall.
In most not so long running programs the zero filling is absolutely noticable. It's effects tend to peter off, if a program settles into a malloc free pattern with a stable amount of memory used.
So there isn't much I can do about the Safari.app benchmark.