[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20250603.225143.1147206358060510971.fujita.tomonori@gmail.com>
Date: Tue, 03 Jun 2025 22:51:43 +0900 (JST)
From: FUJITA Tomonori <fujita.tomonori@...il.com>
To: a.hindborg@...nel.org
Cc: fujita.tomonori@...il.com, rust-for-linux@...r.kernel.org,
boqun.feng@...il.com, frederic@...nel.org, lyude@...hat.com,
tglx@...utronix.de, anna-maria@...utronix.de, jstultz@...gle.com,
sboyd@...nel.org, ojeda@...nel.org, alex.gaynor@...il.com,
gary@...yguo.net, bjorn3_gh@...tonmail.com, benno.lossin@...ton.me,
aliceryhl@...gle.com, tmgross@...ch.edu, dakr@...nel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v1 3/5] rust: time: Add HrTimerExpires trait
On Fri, 30 May 2025 15:04:00 +0200
Andreas Hindborg <a.hindborg@...nel.org> wrote:
>> +/// Defines a new `HrTimerMode` implementation with a given expiration type and C mode.
>> +#[doc(hidden)]
>> +macro_rules! define_hrtimer_mode {
>> + (
>> + $(#[$meta:meta])*
>> + $vis:vis struct $name:ident<$clock:ident> {
>> + c = $mode:ident,
>> + expires = $expires:ty
>> + }
>> + ) => {
>> + $(#[$meta])*
>> + $vis struct $name<$clock: $crate::time::ClockSource>(
>> + ::core::marker::PhantomData<$clock>
>> + );
>
> I think a macro is too much here. The code would be easier to read
> without the macro, and the macro does not remove much code here.
>
> Could you try to do the trait implementations without the macro?
Something like the following, right? If so, I'll do in the next
version. I'm also fine with that way.
/// Timer that expires at a fixed point in time.
pub struct AbsoluteMode<C: ClockSource>(PhantomData<C>);
impl<C: ClockSource> HrTimerMode for AbsoluteMode<C> {
const C_MODE: bindings::hrtimer_mode = bindings::hrtimer_mode_HRTIMER_MODE_ABS;
type Clock = C;
type Expires = Instant<C>;
}
instead of
define_hrtimer_mode! {
/// Timer that expires at a fixed point in time.
pub struct AbsoluteMode<C> {
c = HRTIMER_MODE_ABS,
expires = Instant<C>
}
}
Powered by blists - more mailing lists