Re: [PATCH v7] rust: Return Option from page_align and ensure no usize overflow

[Brendan Shephard <bshephar@bne-home.net>]

On Thu, Dec 18, 2025 at 10:13:23PM +0900, Alexandre Courbot wrote:
> On Fri Dec 5, 2025 at 7:40 AM JST, Brendan Shephard wrote:
> > +/// Returns [`None`] on integer overflow.
> > +///
> > +/// # Examples
> > +///
> > +/// ```
> > +/// use kernel::page::{page_align, PAGE_SIZE};
> 
> Add an extra line to separate the imports from the code.
> 
> > +/// // Requested address is already aligned
> 
> Please consistently finish sentences with a `.`.
> 
> > +/// assert_eq!(page_align(0x0), Some(0x0));
> > +/// assert_eq!(page_align(PAGE_SIZE), Some(PAGE_SIZE));
> > +///
> > +/// // Requested address needs alignment up
> > +/// assert_eq!(page_align(0x1), Some(PAGE_SIZE));
> > +/// assert_eq!(page_align(PAGE_SIZE + 1), Some(2 * PAGE_SIZE));
> > +///
> > +/// // Requested address causes overflow (returns None)
> > +/// // The check asserts that `None` is returned when a value is requested within one `PAGE_SIZE` of
> > +/// // `usize::MAX`.
> 
> We can see what the check does by looking at it, so this last sentence
> feels heavy and unnecessary.
> 
> > +/// let overflow_addr = usize::MAX - (PAGE_SIZE / 2);
> > +/// assert_eq!(page_align(overflow_addr), None);
> > +///
> > +/// // Requested address of `usize::MAX` would overflow, and subsequently returns `None`.
> > +/// assert_eq!(page_align(usize::MAX), None);
> 
> This test is basically equivalent to the previous one and can be
> discarded imho.
> 
> I have checked the tests and rustdoc fwiw, and saw no issue, so this
> looks good to go after these last details are fixed.
> 
> 

Yeah, sounds good. I fixed up those and sent a new revision:
https://lore.kernel.org/rust-for-linux/20251223055647.9761-1-bshephar@bne-home.net/T/#u

Thanks again! Hope you enjoy your holidays.