Re: [PATCH v2] mm: Add kvfree_sensitive() for freeing sensitive data objects

[Matthew Wilcox]

On Tue, Apr 07, 2020 at 10:14:11PM +0100, David Howells wrote:
> Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
> 
> > So the _real_ prototype for 'free()'-like operations should be something like
> > 
> >     void free(const volatile killed void *ptr);
> > 
> > where that "killed" also tells the compiler that the pointer lifetime
> > is dead, so that using it afterwards is invalid. So that the compiler
> > could warn us about some of the most trivial use-after-free cases.
> 
> It might be worth asking the compiler folks to give us an __attribute__ for
> that - even if they don't do anything with it immediately.  So we might have
> something like:
> 
> 	void free(const volatile void *ptr) __attribute__((free(1)));
> 
> There are some for allocation functions, some of which we use, though I'm not
> sure we do so as consistently as we should (should inline functions like
> kcalloc() have them, for example?).

GCC recognises free() as being a __builtin.  I don't know if there's
an __attribute__ for it.

gcc/builtins.def:DEF_LIB_BUILTIN        (BUILT_IN_FREE, "free", BT_FN_VOID_PTR, ATTR_NOTHROW_LEAF_LIST)

It looks like the only two things this really does is warn you if you
try to free a pointer that gcc can prove isn't in the heap, and elide
the call if gcc can prove it's definitely NULL.  Which are both things
that a compiler should do, but aren't all that valuable.