Rust for Linux List
 help / color / mirror / Atom feed
From: FUJITA Tomonori <tomo@flapping.org>
To: a.hindborg@kernel.org, boqun@kernel.org, ojeda@kernel.org
Cc: acourbot@nvidia.com, aliceryhl@google.com,
	anna-maria@linutronix.de, bjorn3_gh@protonmail.com,
	dakr@kernel.org, daniel.almeida@collabora.com,
	frederic@kernel.org, gary@garyguo.net, jstultz@google.com,
	lossin@kernel.org, lyude@redhat.com, mike@fireburn.co.uk,
	sboyd@kernel.org, tamird@kernel.org, tglx@kernel.org,
	tmgross@umich.edu, work@onurozkan.dev,
	rust-for-linux@vger.kernel.org,
	FUJITA Tomonori <fujita.tomonori@gmail.com>
Subject: [PATCH v1 1/2] rust: time: make Instant generic over the time unit
Date: Sat, 31 Oct 2026 09:33:33 +0900	[thread overview]
Message-ID: <20261031003334.4161188-2-tomo@flapping.org> (raw)
In-Reply-To: <20261031003334.4161188-1-tomo@flapping.org>

From: FUJITA Tomonori <fujita.tomonori@gmail.com>

Add a type parameter `U` to `Instant`, with `Nsec` as the default, so
that an `Instant` in seconds can be added later. The value is stored as
`U::Repr`, like `Delta<U>`.

`U` is bound by a new trait `InstantUnit` rather than `TimeUnit`. It
reads the clock in its unit, and jiffies is not a clock, so
`Instant<C, Jiffy>` must not exist.

`now()` stays one function on `Instant<C, U>`. With two inherent `now()`,
one per unit, the existing `Instant::now()` calls do not compile (E0034,
multiple applicable items). The other methods and operators are
implemented for `Instant<C, Nsec>` only. `Instant<C>` still means
`Instant<C, Nsec>`, so no user changes.

No functional change.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@gmail.com>
---
 rust/kernel/time.rs | 44 ++++++++++++++++++++++++++++++++------------
 1 file changed, 32 insertions(+), 12 deletions(-)

diff --git a/rust/kernel/time.rs b/rust/kernel/time.rs
index b6a2aab12a50..8232880bffe7 100644
--- a/rust/kernel/time.rs
+++ b/rust/kernel/time.rs
@@ -179,36 +179,43 @@ fn ktime_get() -> bindings::ktime_t {
 
 /// A specific point in time.
 ///
+/// The time is read from the clock given by the type parameter `C` (see
+/// [`ClockId`]) and stored in the unit given by `U` (see [`InstantUnit`]); its
+/// value has type `U::Repr`. `U` defaults to [`Nsec`], so a plain
+/// [`Instant<C>`] is a point in time in nanoseconds.
+///
 /// # Invariants
 ///
 /// The `inner` value is in the range from 0 to `KTIME_MAX`.
 #[repr(transparent)]
 #[derive(PartialEq, PartialOrd, Eq, Ord)]
-pub struct Instant<C: ClockId> {
-    inner: bindings::ktime_t,
+pub struct Instant<C: ClockId, U: InstantUnit = Nsec> {
+    inner: U::Repr,
     _c: PhantomData<C>,
 }
 
-impl<C: ClockId> Clone for Instant<C> {
+impl<C: ClockId, U: InstantUnit> Clone for Instant<C, U> {
     fn clone(&self) -> Self {
         *self
     }
 }
 
-impl<C: ClockId> Copy for Instant<C> {}
+impl<C: ClockId, U: InstantUnit> Copy for Instant<C, U> {}
 
-impl<C: ClockId> Instant<C> {
-    /// Get the current time from the clock source.
+impl<C: ClockId, U: InstantUnit> Instant<C, U> {
+    /// Get the current time from the clock `C`.
     #[inline]
     pub fn now() -> Self {
-        // INVARIANT: The `ClockId::ktime_get()` function returns a value in the range
-        // from 0 to `KTIME_MAX`.
+        // INVARIANT: `InstantUnit` is sealed and its only impl, `Nsec`, calls
+        // `ClockId::ktime_get()`, which returns a value in the range from 0 to `KTIME_MAX`.
         Self {
-            inner: C::ktime_get(),
+            inner: U::now::<C>(),
             _c: PhantomData,
         }
     }
+}
 
+impl<C: ClockId> Instant<C, Nsec> {
     /// Return the amount of time elapsed since the [`Instant`].
     #[inline]
     pub fn elapsed(&self) -> Delta {
@@ -243,7 +250,7 @@ pub(crate) unsafe fn from_ktime(ktime: bindings::ktime_t) -> Self {
     }
 }
 
-impl<C: ClockId> ops::Sub for Instant<C> {
+impl<C: ClockId> ops::Sub for Instant<C, Nsec> {
     type Output = Delta;
 
     // By the type invariant, it never overflows.
@@ -255,7 +262,7 @@ fn sub(self, other: Instant<C>) -> Delta {
     }
 }
 
-impl<T: ClockId> ops::Add<Delta> for Instant<T> {
+impl<T: ClockId> ops::Add<Delta> for Instant<T, Nsec> {
     type Output = Self;
 
     #[inline]
@@ -275,7 +282,7 @@ fn add(self, rhs: Delta) -> Self::Output {
     }
 }
 
-impl<T: ClockId> ops::Sub<Delta> for Instant<T> {
+impl<T: ClockId> ops::Sub<Delta> for Instant<T, Nsec> {
     type Output = Self;
 
     #[inline]
@@ -335,6 +342,19 @@ impl TimeUnit for Jiffy {
     const ZERO: Self::Repr = 0;
 }
 
+/// A trait for time units that can be used with [`Instant`].
+pub trait InstantUnit: TimeUnit {
+    /// Return the current time of the clock `C` in this unit.
+    fn now<C: ClockId>() -> Self::Repr;
+}
+
+impl InstantUnit for Nsec {
+    #[inline]
+    fn now<C: ClockId>() -> i64 {
+        C::ktime_get()
+    }
+}
+
 /// A span of time.
 ///
 /// The span is stored in the unit given by the type parameter `U` (see
-- 
2.43.0


  reply	other threads:[~2026-10-31  0:33 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-31  0:33 [PATCH v1 0/2] rust: time: seconds based Instant FUJITA Tomonori
2026-10-31  0:33 ` FUJITA Tomonori [this message]
2026-10-31  0:33 ` [PATCH v1 2/2] rust: time: add Instant in seconds FUJITA Tomonori
2026-10-31  3:56 ` [PATCH v1 0/2] rust: time: seconds based Instant Boqun Feng

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20261031003334.4161188-2-tomo@flapping.org \
    --to=tomo@flapping.org \
    --cc=a.hindborg@kernel.org \
    --cc=acourbot@nvidia.com \
    --cc=aliceryhl@google.com \
    --cc=anna-maria@linutronix.de \
    --cc=bjorn3_gh@protonmail.com \
    --cc=boqun@kernel.org \
    --cc=dakr@kernel.org \
    --cc=daniel.almeida@collabora.com \
    --cc=frederic@kernel.org \
    --cc=fujita.tomonori@gmail.com \
    --cc=gary@garyguo.net \
    --cc=jstultz@google.com \
    --cc=lossin@kernel.org \
    --cc=lyude@redhat.com \
    --cc=mike@fireburn.co.uk \
    --cc=ojeda@kernel.org \
    --cc=rust-for-linux@vger.kernel.org \
    --cc=sboyd@kernel.org \
    --cc=tamird@kernel.org \
    --cc=tglx@kernel.org \
    --cc=tmgross@umich.edu \
    --cc=work@onurozkan.dev \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox