Re: [PATCH v7 2/2] rust: xarray: Add an abstraction for XArray

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



> From: Asahi Lina <lina@xxxxxxxxxxxxx>
> 
> The XArray is an abstract data type which behaves like a very large
> array of pointers. Add a Rust abstraction for this data type.
> 
> The initial implementation uses explicit locking on get operations and
> returns a guard which blocks mutation, ensuring that the referenced
> object remains alive. To avoid excessive serialization, users are
> expected to use an inner type that can be efficiently cloned (such as
> Arc<T>), and eagerly clone and drop the guard to unblock other users
> after a lookup.
> 
> Future variants may support using RCU instead to avoid mutex locking.
> 
> This abstraction also introduces a reservation mechanism, which can be
> used by alloc-capable XArrays to reserve a free slot without immediately
> filling it, and then do so at a later time. If the reservation is
> dropped without being filled, the slot is freed again for other users,
> which eliminates the need for explicit cleanup code.
> 
> Signed-off-by: Asahi Lina <lina@xxxxxxxxxxxxx>
> Co-developed-by: Maíra Canal <mcanal@xxxxxxxxxx>
> Signed-off-by: Maíra Canal <mcanal@xxxxxxxxxx>

Sorry for the delay in reviewing this.

I have one important comment (the first one), and the rest are nits that
are not so important.


> +pub struct Guard<'a, T: ForeignOwnable>(NonNull<T>, &'a XArray<T>);

This stores a pointer to a `T`, but really it's the pointer returned by
`into_foreign`, correct? So e.g. in the case of `T = Box<U>`, you have
an `NonNull<Box<U>>`, even though the pointer actually points at an
`U`, not a `Box<U>`.

I think it's better to keep this as `NonNull<c_void>`, since that's the
type used by `into_foreign`. That also lets you remove a bunch of calls
to `.cast()`.

> +/// INVARIANT: All pointers stored in the array are pointers obtained by
> +/// calling `T::into_foreign` or are NULL pointers. By using the pin-init
> +/// initialization, `self.xa` is always an initialized and valid XArray.

Nit: usually invariants listed on structs use this heading:

/// # Invariants
/// 
/// All pointers stored ...

Nit: You also do not need to mention that you use pin-init to make
`self.xa` be initialized and valid. It's enough to explain that it is
always initialized and valid - the *how* is not necessary in this type
of comment.

> +/// Represents a reserved slot in an `XArray`, which does not yet have a value but has an assigned
> +/// index and may not be allocated by any other user. If the Reservation is dropped without
> +/// being filled, the entry is marked as available again.
> +///
> +/// Users must ensure that reserved slots are not filled by other mechanisms, or otherwise their
> +/// contents may be dropped and replaced (which will print a warning).
> +pub struct Reservation<'a, T: ForeignOwnable>(&'a XArray<T>, usize, PhantomData<T>);

Nit: I don't think the PhantomData is necessary here.

> +            if !new.is_null() {
> +                // SAFETY: If `new` is not NULL, it came from the `ForeignOwnable` we got
> +                // from the caller.
> +                unsafe { T::from_foreign(new) };
> +            }

Nit: Adding a call to `drop` here makes the code more clear to me.

> +        unsafe {
> +            bindings::xa_destroy(self.xa.get());
> +        }

Nit: Moving the semicolon outside of the unsafe block formats better.

Alice





[Index of Archives]     [Linux Ext4 Filesystem]     [Union Filesystem]     [Filesystem Testing]     [Ceph Users]     [Ecryptfs]     [NTFS 3]     [AutoFS]     [Kernel Newbies]     [Share Photos]     [Security]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux Cachefs]     [Reiser Filesystem]     [Linux RAID]     [NTFS 3]     [Samba]     [Device Mapper]     [CEPH Development]

  Powered by Linux