Cling/Nim-style UAF prevention by strict heap segmentation is neat, but it seems incomplete as a solution for memory safety without something to deal with type confusion caused by union misuse.
-
-
What assumptions? especially if multithreaded, you need to make absolutely everything on the heap volatile for reads- even if protected by locks (locks'll be reinitialized after reallocating).
-
This also would seem to preclude any sort of variable-length heap data- do you store pointers to that as (ptr, len)? obviously won't work if *ptr is reallocated and used for something smaller. len inline with data? if you fetch len once (e.g. at start of iteration),
- 3 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.