Re: [PATCH v4 2/2] rust: zpool: add abstraction for zpool drivers

[Vitaly Wool]

<snip>
> > > +    extern "C" fn destroy_(pool: *mut c_void) {
> > > +        // SAFETY: The pointer originates from an `into_foreign` call.
> > > +        T::destroy(unsafe { T::Pool::from_foreign(pool) })
> > > +    }
> > > +    extern "C" fn malloc_(
> > > +        pool: *mut c_void,
> > > +        size: usize,
> > > +        gfp: u32,
> > > +        handle: *mut usize,
> > > +        nid: c_int,
> > > +    ) -> c_int {
> > > +        // SAFETY: The pointer originates from an `into_foreign` 
> > > call. If `pool` is passed to
> > > +        // `from_foreign`, then that happens in `_destroy` which 
> > > will not be called during this
> > > +        // method.
> > > +        let pool = unsafe { T::Pool::borrow_mut(pool) };
> >
> > Wait, can't this happen concurrently to all the other functions that 
> > borrow the
> > pool? This would be undefined behavior, no?
>
> Theoretically, yes, but since pool is actually Box<T>, it's only the 
> inner T that is mutable.
>
> Anyway, the only reason for malloc() to require a mutable reference is 
> that the backend implementation *may* use RBTree::cursor_lower_bound() 
> which requires a mutable reference of the tree.
>
> Would it be okay if I
> * change the Zpool API so that malloc takes an immutable reference
> * extend the RBTree API with a cursor_lower_bound analog which doesn't 
> require a mutable tree?

As a matter of fact, the RBTree change may be postponed and submitted 
together with zblock, it is not relevant until then.

~Vitaly