Re: Valgrind meets UML

[Jeff Dike]

[e-mail] said:
> The main problem will be that newly allocated memory will still be
> considered initialized by its previous owner.  Also, if UML allocates
> memory using mmap, all memory will be considered to be initialized.

What I was doing was having kfree and free_pages set the freed object to
noaccess.  Presumably, that tells valgrind to consider the memory 
uninitialized.

Presumably, that will also cause errors from inside the allocator if it
touches that memory at all before it's allocated again.

				Jeff