They’re not leaks, it’s just that the slop at the end of a present page is potentially 512 times more than it would be with 4K pages.
On a stock kernel, it's 511. TCMalloc's docs recommend using max_ptes_none set to 0 for this reason: https://github.com/google/tcmalloc/blob/master/docs/tuning.m...
(Disclosure: I work on TCMalloc and authored the above doc.)