> Anthony's argument about exposing the mem layout is crucial, though.
>

Yes it is!

The patch I attached un-xors only the part for the object's handle.
The memory pointer is kept xored.

Reply via email to