On 12.08.2024 20:22, Danilo Krummrich wrote:
Now that all existing `Box` users were moved to the kernel `Box` type,
remove the `BoxExt` extension and all other related extensions.
I just noticed that in the recent 'rust-dev' branch we have a change
which *adds* something to BoxExt:
rust: kernel: add drop_contents to BoxExt
https://github.com/Rust-for-Linux/linux/commit/62c34da1da6c01a635ea2308cb42996d0571059e
I'm unclear how relevant that is. Just want to mention this in case it
would make sense to include that directly in this patch series to avoid
a future add-on patch ;)
Thanks,
Dirk
P.S.: It looks like anything like this at least makes the compiler happy:
diff --git a/rust/kernel/alloc/kbox.rs b/rust/kernel/alloc/kbox.rs
index d67f975502246..e91d441835d54 100644
--- a/rust/kernel/alloc/kbox.rs
+++ b/rust/kernel/alloc/kbox.rs
@@ -9,6 +9,7 @@
use core::mem::MaybeUninit;
use core::ops::{Deref, DerefMut};
use core::pin::Pin;
+use core::ptr;
use core::ptr::NonNull;
use core::result::Result;
@@ -270,6 +271,28 @@ pub fn new_uninit(flags: Flags) ->
Result<Box<MaybeUninit<T>, A>, AllocError> {
Ok(Box(ptr, PhantomData::<A>))
}
+ /// Drops the contents, but keeps the allocation.
+ ///
+ /// # Examples
+ ///
+ /// ```
+ /// use kernel::alloc::{Flags, KBox};
+ /// let value = KBox::new([0; 32], GFP_KERNEL)?;
+ /// assert_eq!(*value, [0; 32]);
+ /// let value = KBox::drop_contents(value);
+ /// // Now we can re-use `value`:
+ /// let value = KBox::write(value, [1; 32]);
+ /// assert_eq!(*value, [1; 32]);
+ /// # Ok::<(), Error>(())
+ /// ```
+ pub fn drop_contents(this: Self) -> Box<MaybeUninit<T>, A> {
+ let ptr = Box::into_raw(this);
+ // SAFETY: `ptr` is valid, because it came from `Box::into_raw`.
+ unsafe { ptr::drop_in_place(ptr) };
+ // SAFETY: `ptr` is valid, because it came from `Box::into_raw`.
+ unsafe { Box::from_raw(ptr.cast()) }
+ }
+
/// Constructs a new `Pin<Box<T, A>>`. If `T` does not implement
[`Unpin`], then `x` will be
/// pinned in memory and can't be moved.
#[inline]
--