On Mon, Dec 08, 2025 at 04:44:05PM +1000, Brendan Shephard wrote:
> Use page::page_align for GEM object memory allocation to ensure the
> allocation is page aligned. This ensures that the allocation is page
> aligned with the system in cases where 4096 is not the default.
> For example on 16k or 64k aarch64 systems this allocation should be
> aligned accordingly.
> 
> This change uses the updated page_align() function from commit:
> "rust: Return Option from page_align and ensure no usize overflow"
> Which returns an Option<usize> for overflow safety.
> 
> Signed-off-by: Brendan Shephard <[email protected]>
> ---
> Changes in v2:
> - Updated to use the new page_align() Option<usize> return value from:
>   Commit: "rust: Return Option from page_align and ensure no usize overflow"
>   
> https://lore.kernel.org/rust-for-linux/[email protected]/T/#u
> 
>  drivers/gpu/drm/nova/gem.rs | 7 ++-----
>  1 file changed, 2 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/gpu/drm/nova/gem.rs b/drivers/gpu/drm/nova/gem.rs
> index 2760ba4f3450..41439f3b9361 100644
> --- a/drivers/gpu/drm/nova/gem.rs
> +++ b/drivers/gpu/drm/nova/gem.rs
> @@ -3,6 +3,7 @@
>  use kernel::{
>      drm,
>      drm::{gem, gem::BaseObject},
> +    page::page_align,
>      prelude::*,
>      sync::aref::ARef,
>  };
> @@ -27,11 +28,7 @@ fn new(_dev: &NovaDevice, _size: usize) -> impl 
> PinInit<Self, Error> {
>  impl NovaObject {
>      /// Create a new DRM GEM object.
>      pub(crate) fn new(dev: &NovaDevice, size: usize) -> 
> Result<ARef<gem::Object<Self>>> {
> -        let aligned_size = size.next_multiple_of(1 << 12);
> -
> -        if size == 0 || size > aligned_size {
> -            return Err(EINVAL);
> -        }
> +        let aligned_size = page_align(size).ok_or(EINVAL)?

Don't you lose the size == 0 check here? If intentional, please update the
changelog mentioning it.

thanks,

 - Joel




>  
>          gem::Object::new(dev, aligned_size)
>      }
> 
> base-commit: 37bb2e7217b01404e2abf9d90d8e5705a5603b52
> -- 
> 2.52.0
> 

Reply via email to