Re: [PATCH 4/5] rust: hrtimer: use READ_ONCE instead of read_volatile

[Boqun Feng <boqun.feng@gmail.com>]

On Tue, Jan 06, 2026 at 04:47:35PM -0800, John Hubbard wrote:
> On 1/6/26 10:43 AM, Alice Ryhl wrote:
> > On Tue, Jan 06, 2026 at 03:23:00PM +0000, Gary Guo wrote:
> >> On Tue, 06 Jan 2026 13:37:34 +0100
> >> Andreas Hindborg <a.hindborg@kernel.org> wrote:
> >>
> >>> "FUJITA Tomonori" <fujita.tomonori@gmail.com> writes:
> >>>>
> >>>> Sorry, of course this should be:
> >>>>
> >>>> +__rust_helper ktime_t rust_helper_hrtimer_get_expires(const struct hrtimer *timer)
> >>>> +{
> >>>> +	return hrtimer_get_expires(timer);
> >>>> +}
> >>>>  
> >>>
> >>> This is a potentially racy read. As far as I recall, we determined that
> >>> using read_once is the proper way to handle the situation.
> >>>
> >>> I do not think it makes a difference that the read is done by C code.
> >>
> >> If that's the case I think the C code should be fixed by inserting the
> >> READ_ONCE?
> > 
> > I maintain my position that if this is what you recommend C code does,
> > it's confusing to not make the same recommendation for Rust abstractions
> > to the same thing.
> > 
> > After all, nothing is stopping you from calling atomic_read() in C too.
> > 
> 
> Hi Alice and everyone!
> 
> I'm having trouble fully understanding the latest reply, so maybe what
> I'm saying is actually what you just said.
> 
> Anyway, we should use READ_ONCE in both the C and Rust code. Relying
> on the compiler for that is no longer OK. We shouldn't be shy about
> fixing the C side (not that I think you have been, so far!).
> 

Agreed on most of it, except that we should be more explicit in Rust,
by using atomic_load[1] instead of READ_ONCE().

[1]: https://lore.kernel.org/rust-for-linux/aV0FxCRzXFrNLZik@tardis-2.local/

Regards,
Boqun

> thanks,
> -- 
> John Hubbard
>