Re: [PATCH v12 4/5] rust: drm: gem: Introduce shmem::SGTable

[lyude@redhat.com]

On Fri, 2026-04-24 at 00:01 +0900, Alexandre Courbot wrote:
> There are 4 sites where we acquire and release the DMA resv lock,
> each
> of which require unsafe blocks and carrying the risk that we forget
> releasing the lock in the end. For this method in particular we need
> to
> jump through hoops a bit and store the return value into a temporary
> variable so we can unlock the DMA reservation.
> 
> Let's do ourselves a favor and implement a small, private guard type:
> 
>     struct DmaResvGuard<'a, T: DriverObject>(&'a Object<T>);
> 
>     impl<'a, T: DriverObject> DmaResvGuard<'a, T> {
>         fn new(object: &'a Object<T>) -> Self {
>             // SAFETY: This lock is initialized throughout the
> lifetime of `object`
>             unsafe { bindings::dma_resv_lock(object.raw_dma_resv(),
> ptr::null_mut()) };
> 
>             Self(object)
>         }
>     }
> 
>     impl<'a, T> Drop for DmaResvGuard<'a, T>
>     where
>         T: DriverObject,
>     {
>         fn drop(&mut self) {
>             // SAFETY: We are releasing the lock grabbed during the
> creation of this object.
>             unsafe { bindings::dma_resv_unlock(self.0.raw_dma_resv())
> };
>         }
>     }
> 
> There here you would just do
> 
>     let _dma_resv = DmaResvGuard::new(self);

I thought of doing this but lost track of how many times I was actually
grabbing this lock…
BTW - want me to just give you authorship on the patch for adding
DmaResvGuard? Since on my branch I've pretty much only just added two
inline annotations, it's otherwise identical to what you wrote here.