[Guy Harris <guy@xxxxxxxxxxxx>]

Guy Harris wrote:

> So, no, I don't see any reason not to implement ep_realloc().

...unless it's impossible for ep_realloc() to find out the *existing* 
size of the object being reallocated (malloc()/realloc() tend to attach 
a header before the chunk of memory being allocated, so free() knows how 
much to free, but, as there's no ep_free(), I don't think ep_malloc() 
attaches such a header).