[WIP RFC v2 27/35] rust: drm/kms: Add drm_framebuffer bindings

[Daniel Almeida]

> On 30 Sep 2024, at 20:10, Lyude Paul <lyude at redhat.com> wrote:
> 
> This adds some very simple bindings for drm_framebuffer. We don't use them
> much yet, but we'll eventually be using them when rvkms eventually gets CRC
> and writeback support. Just like Connector objects, these use RcModeObject.

Yeah, for the non-KMS devs like me, a brief explanation on CRC and writeback support
would be helpful

> 
> Signed-off-by: Lyude Paul <lyude at redhat.com>
> ---
> rust/bindings/bindings_helper.h    |  1 +
> rust/kernel/drm/kms.rs             |  1 +
> rust/kernel/drm/kms/framebuffer.rs | 58 ++++++++++++++++++++++++++++++
> 3 files changed, 60 insertions(+)
> create mode 100644 rust/kernel/drm/kms/framebuffer.rs
> 
> diff --git a/rust/bindings/bindings_helper.h b/rust/bindings/bindings_helper.h
> index 7695f11f4363c..7c324003c3885 100644
> --- a/rust/bindings/bindings_helper.h
> +++ b/rust/bindings/bindings_helper.h
> @@ -15,6 +15,7 @@
> #include <drm/drm_fourcc.h>
> #include <drm/drm_fbdev_dma.h>
> #include <drm/drm_fbdev_shmem.h>
> +#include <drm/drm_framebuffer.h>
> #include <drm/drm_gem.h>
> #include <drm/drm_gem_framebuffer_helper.h>
> #include <drm/drm_gem_shmem_helper.h>
> diff --git a/rust/kernel/drm/kms.rs b/rust/kernel/drm/kms.rs
> index 4ab039d67352e..3edd90bc0025a 100644
> --- a/rust/kernel/drm/kms.rs
> +++ b/rust/kernel/drm/kms.rs
> @@ -7,6 +7,7 @@
> pub mod crtc;
> pub mod encoder;
> pub mod fbdev;
> +pub mod framebuffer;
> pub mod plane;
> 
> use crate::{
> diff --git a/rust/kernel/drm/kms/framebuffer.rs b/rust/kernel/drm/kms/framebuffer.rs
> new file mode 100644
> index 0000000000000..bbe408c187670
> --- /dev/null
> +++ b/rust/kernel/drm/kms/framebuffer.rs
> @@ -0,0 +1,58 @@
> +// SPDX-License-Identifier: GPL-2.0 OR MIT
> +//! DRM Framebuffer related bindings
> +
> +use crate::{
> +    drm::{
> +        device::Device,
> +        fourcc::*,
> +    },
> +    types::*,
> +    prelude::*,
> +};
> +use bindings;
> +use core::{
> +    marker::*,
> +    slice,
> +};
> +use super::{ModeObject, RcModeObject, KmsDriver};
> +
> +/// The main interface for [`struct drm_framebuffer`].
> +///
> +/// [`struct drm_framebuffer`]: srctree/include/drm/drm_framebuffer.h
> +#[repr(transparent)]
> +pub struct Framebuffer<T: KmsDriver> {
> +    inner: Opaque<bindings::drm_framebuffer>,
> +    _p: PhantomData<T>,
> +}
> +
> +impl<T: KmsDriver> ModeObject for Framebuffer<T> {
> +    type Driver = T;
> +
> +    fn drm_dev(&self) -> &Device<Self::Driver> {
> +        unsafe { Device::borrow((*self.inner.get()).dev) }
> +    }
> +
> +    fn raw_mode_obj(&self) -> *mut bindings::drm_mode_object {
> +        // SAFETY: We don't expose Framebuffer<T> to users before it's initialized, so `base` is
> +        // always initialized
> +        unsafe { &mut (*self.inner.get()).base }
> +    }
> +}
> +
> +// SAFETY: Framebuffers are refcounted mode objects.
> +unsafe impl<T: KmsDriver> RcModeObject for Framebuffer<T> {}
> +
> +// SAFETY: References to framebuffers are safe to be accessed from any thread
> +unsafe impl<T: KmsDriver> Send for Framebuffer<T> {}
> +// SAFETY: References to framebuffers are safe to be accessed from any thread
> +unsafe impl<T: KmsDriver> Sync for Framebuffer<T> {}
> +
> +impl<T: KmsDriver> crate::private::Sealed for Framebuffer<T> {}

A brief comment on why you’re implementing Sealed would also be good :)

> +
> +impl<T: KmsDriver> Framebuffer<T> {
> +    #[inline]
> +    pub(super) unsafe fn from_raw<'a>(ptr: *const bindings::drm_framebuffer) -> &'a Self {
> +        // SAFETY: Our data layout is identical to drm_framebuffer
> +        unsafe { &*ptr.cast() }
> +    }
> +}
> -- 
> 2.46.1
> 

This LGTM.

— Daniel