On Fri, Oct 25, 2024 at 6:10 AM Trevor Gross <tmgross@xxxxxxxxx> wrote:
>
> On Tue, Oct 1, 2024 at 3:23 AM Alice Ryhl <aliceryhl@xxxxxxxxxx> wrote:
> >
> > This will be used by the miscdevice abstractions, as the C function
> > `misc_register` is fallible.
> >
> > Signed-off-by: Alice Ryhl <aliceryhl@xxxxxxxxxx>
> > ---
> > rust/kernel/types.rs | 16 ++++++++++++++++
> > 1 file changed, 16 insertions(+)
> >
> > diff --git a/rust/kernel/types.rs b/rust/kernel/types.rs
> > index 9e7ca066355c..070d03152937 100644
> > --- a/rust/kernel/types.rs
> > +++ b/rust/kernel/types.rs
> > @@ -299,6 +299,22 @@ pub fn ffi_init(init_func: impl FnOnce(*mut T)) -> impl PinInit<Self> {
> > }
> > }
> >
> > + /// Creates a fallible pin-initializer from the given initializer closure.
> > + ///
> > + /// The returned initializer calls the given closure with the pointer to the inner `T` of this
> > + /// `Opaque`. Since this memory is uninitialized, the closure is not allowed to read from it.
> > + ///
> > + /// This function is safe, because the `T` inside of an `Opaque` is allowed to be
> > + /// uninitialized. Additionally, access to the inner `T` requires `unsafe`, so the caller needs
> > + /// to verify at that point that the inner value is valid.
> > + pub fn try_ffi_init<E>(
> > + init_func: impl FnOnce(*mut T) -> Result<(), E>,
> > + ) -> impl PinInit<Self, E> {
> > + // SAFETY: We contain a `MaybeUninit`, so it is OK for the `init_func` to not fully
> > + // initialize the `T`.
> > + unsafe { init::pin_init_from_closure::<_, E>(move |slot| init_func(Self::raw_get(slot))) }
> > + }
>
> [1] adjusts `ffi_init` to use `try_ffi_init`. Maybe this should do the same?
>
> [1]: https://lore.kernel.org/rust-for-linux/20241022213221.2383-2-dakr@xxxxxxxxxx/
Ah, I wasn't able to find previous patches for this, but I guess there was one.
This patch has already landed in char-misc-next, so this can be a
follow-up if you want to change it.
Alice
