From: Andreas Hindborg <a.hindborg@kernel.org>
To: "Benno Lossin" <lossin@kernel.org>
Cc: "Miguel Ojeda" <ojeda@kernel.org>,
"Alex Gaynor" <alex.gaynor@gmail.com>,
"Boqun Feng" <boqun.feng@gmail.com>,
"Gary Guo" <gary@garyguo.net>,
"Björn Roy Baron" <bjorn3_gh@protonmail.com>,
"Alice Ryhl" <aliceryhl@google.com>,
"Masahiro Yamada" <masahiroy@kernel.org>,
"Nathan Chancellor" <nathan@kernel.org>,
"Luis Chamberlain" <mcgrof@kernel.org>,
"Danilo Krummrich" <dakr@kernel.org>,
"Nicolas Schier" <nicolas.schier@linux.dev>,
"Trevor Gross" <tmgross@umich.edu>,
"Adam Bratschi-Kaye" <ark.email@gmail.com>,
rust-for-linux@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-kbuild@vger.kernel.org, "Petr Pavlu" <petr.pavlu@suse.com>,
"Sami Tolvanen" <samitolvanen@google.com>,
"Daniel Gomez" <da.gomez@samsung.com>,
"Simona Vetter" <simona.vetter@ffwll.ch>,
"Greg KH" <gregkh@linuxfoundation.org>,
"Fiona Behrens" <me@kloenk.dev>,
"Daniel Almeida" <daniel.almeida@collabora.com>,
linux-modules@vger.kernel.org
Subject: Re: [PATCH v14 1/7] rust: sync: add `OnceLock`
Date: Thu, 03 Jul 2025 11:03:38 +0200 [thread overview]
Message-ID: <87bjq1ve6t.fsf@kernel.org> (raw)
In-Reply-To: <DB1NVTWHU7BN.2WGPMAY9LQYNW@kernel.org> (Benno Lossin's message of "Wed, 02 Jul 2025 17:07:12 +0200")
"Benno Lossin" <lossin@kernel.org> writes:
> On Wed Jul 2, 2025 at 3:18 PM CEST, Andreas Hindborg wrote:
>> Introduce the `OnceLock` type, a container that can only be written once.
>> The container uses an internal atomic to synchronize writes to the internal
>> value.
>>
>> Signed-off-by: Andreas Hindborg <a.hindborg@kernel.org>
>> ---
>> rust/kernel/sync.rs | 1 +
>> rust/kernel/sync/once_lock.rs | 104 ++++++++++++++++++++++++++++++++++++++++++
>> 2 files changed, 105 insertions(+)
>>
>> diff --git a/rust/kernel/sync.rs b/rust/kernel/sync.rs
>> index c7c0e552bafe..f2ee07315091 100644
>> --- a/rust/kernel/sync.rs
>> +++ b/rust/kernel/sync.rs
>> @@ -15,6 +15,7 @@
>> mod condvar;
>> pub mod lock;
>> mod locked_by;
>> +pub mod once_lock;
>
> As Alice already said, we should reexport the type. And then make the
> module private, no need to have `kernel::sync::OnceLock` and
> `kernel::sync::once_lock::OnceLock`...
Will do.
>
> Also, I agree with the name change to `SetOnce` or something similar.
I'm OK with that, but please see my comments on Alice suggestion.
>
>> pub mod poll;
>> pub mod rcu;
>>
>> diff --git a/rust/kernel/sync/once_lock.rs b/rust/kernel/sync/once_lock.rs
>> new file mode 100644
>> index 000000000000..cd311bea3919
>> --- /dev/null
>> +++ b/rust/kernel/sync/once_lock.rs
>> @@ -0,0 +1,104 @@
>> +//! A container that can be initialized at most once.
>> +
>> +use super::atomic::ordering::Acquire;
>> +use super::atomic::ordering::Release;
>> +use super::atomic::Atomic;
>> +use kernel::types::Opaque;
>> +
>> +/// A container that can be populated at most once. Thread safe.
>> +///
>> +/// Once the a [`OnceLock`] is populated, it remains populated by the same object for the
>> +/// lifetime `Self`.
>> +///
>> +/// # Invariants
>> +///
>> +/// `init` tracks the state of the container:
>> +///
>> +/// - If the container is empty, `init` is `0`.
>> +/// - If the container is mutably accessed, `init` is `1`.
>
> I think we should swap the order and change the ifs to iffs:
>
> - `init == 0` iff the container is empty.
> - `init == 1` iff the container is being accessed mutably.
Right, that is better, but I will expand "iff".
>
>> +/// - If the container is populated and ready for shared access, `init` is `2`.
>
> You also need that `init` is only increased and never decreases.
> Otherwise you could read a `2` and then access the value, but `init`
> changed under your nose to `0`.
>
> Then the INVARIANT comments below also need to be updated.
OK.
>
>> +///
>> +/// # Example
>> +///
>> +/// ```
>> +/// # use kernel::sync::once_lock::OnceLock;
>> +/// let value = OnceLock::new();
>> +/// assert_eq!(None, value.as_ref());
>> +///
>> +/// let status = value.populate(42u8);
>> +/// assert_eq!(true, status);
>> +/// assert_eq!(Some(&42u8), value.as_ref());
>> +/// assert_eq!(Some(42u8), value.copy());
>> +///
>> +/// let status = value.populate(101u8);
>> +/// assert_eq!(false, status);
>> +/// assert_eq!(Some(&42u8), value.as_ref());
>> +/// assert_eq!(Some(42u8), value.copy());
>> +/// ```
>> +pub struct OnceLock<T> {
>> + init: Atomic<u32>,
>> + value: Opaque<T>,
>> +}
>> +
>> +impl<T> Default for OnceLock<T> {
>> + fn default() -> Self {
>> + Self::new()
>> + }
>> +}
>> +
>> +impl<T> OnceLock<T> {
>> + /// Create a new [`OnceLock`].
>> + ///
>> + /// The returned instance will be empty.
>> + pub const fn new() -> Self {
>> + // INVARIANT: The container is empty and we set `init` to `0`.
>> + Self {
>> + value: Opaque::uninit(),
>> + init: Atomic::new(0),
>> + }
>> + }
>> +
>> + /// Get a reference to the contained object.
>> + ///
>> + /// Returns [`None`] if this [`OnceLock`] is empty.
>> + pub fn as_ref(&self) -> Option<&T> {
>> + if self.init.load(Acquire) == 2 {
>> + // SAFETY: As determined by the load above, the object is ready for shared access.
>
> // SAFETY: By the safety requirements of `Self`, `self.init == 2` means that `self.value` contains
> // a valid value.
By the *type invariants* I guess?
>
>> + Some(unsafe { &*self.value.get() })
>> + } else {
>> + None
>> + }
>> + }
>> +
>> + /// Populate the [`OnceLock`].
>> + ///
>> + /// Returns `true` if the [`OnceLock`] was successfully populated.
>> + pub fn populate(&self, value: T) -> bool {
>> + // INVARIANT: We obtain exclusive access to the contained allocation and write 1 to
>> + // `init`.
>> + if let Ok(0) = self.init.cmpxchg(0, 1, Acquire) {
>> + // SAFETY: We obtained exclusive access to the contained object.
>> + unsafe { core::ptr::write(self.value.get(), value) };
>> + // INVARIANT: We release our exclusive access and transition the object to shared
>> + // access.
>> + self.init.store(2, Release);
>> + true
>> + } else {
>> + false
>> + }
>> + }
>> +}
>> +
>> +impl<T: Copy> OnceLock<T> {
>> + /// Get a copy of the contained object.
>> + ///
>> + /// Returns [`None`] if the [`OnceLock`] is empty.
>> + pub fn copy(&self) -> Option<T> {
>> + if self.init.load(Acquire) == 2 {
>> + // SAFETY: As determined by the load above, the object is ready for shared access.
>> + Some(unsafe { *self.value.get() })
>> + } else {
>> + None
>> + }
>
> The impl can just be:
>
> self.as_ref().copied()
Nice. I was thinking of dropping this method and just have callers do
my_once_lock.as_ref().map(|v| v.copied())
What do you think?
Best regards,
Andreas Hindborg
next prev parent reply other threads:[~2025-07-03 9:03 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-07-02 13:18 [PATCH v14 0/7] rust: extend `module!` macro with integer parameter support Andreas Hindborg
2025-07-02 13:18 ` [PATCH v14 1/7] rust: sync: add `OnceLock` Andreas Hindborg
2025-07-02 13:32 ` Alice Ryhl
2025-07-02 13:54 ` Andreas Hindborg
2025-07-02 14:50 ` Alice Ryhl
2025-07-03 7:51 ` Andreas Hindborg
2025-07-02 15:07 ` Benno Lossin
2025-07-02 15:27 ` Alice Ryhl
2025-07-02 15:40 ` Benno Lossin
2025-07-03 9:03 ` Andreas Hindborg [this message]
2025-07-03 9:42 ` Benno Lossin
2025-07-03 16:25 ` Andreas Hindborg
2025-07-03 20:41 ` Benno Lossin
2025-07-03 9:36 ` Wren Turkal
2025-07-03 16:41 ` Andreas Hindborg
2025-07-02 13:18 ` [PATCH v14 2/7] rust: str: add radix prefixed integer parsing functions Andreas Hindborg
2025-07-02 13:18 ` [PATCH v14 3/7] rust: introduce module_param module Andreas Hindborg
2025-07-02 15:21 ` Benno Lossin
2025-07-04 11:45 ` Andreas Hindborg
2025-07-06 20:00 ` Miguel Ojeda
2025-07-03 21:49 ` Danilo Krummrich
2025-07-04 7:29 ` Andreas Hindborg
2025-07-04 7:37 ` Andreas Hindborg
2025-07-04 9:59 ` Benno Lossin
2025-07-04 11:46 ` Andreas Hindborg
2025-07-02 13:18 ` [PATCH v14 4/7] rust: module: use a reference in macros::module::module Andreas Hindborg
2025-07-02 13:18 ` [PATCH v14 5/7] rust: module: update the module macro with module parameter support Andreas Hindborg
2025-07-02 15:38 ` Benno Lossin
2025-07-04 12:29 ` Andreas Hindborg
2025-07-04 12:48 ` Benno Lossin
2025-07-04 13:51 ` Andreas Hindborg
2025-07-04 14:00 ` Benno Lossin
2025-07-02 13:18 ` [PATCH v14 6/7] rust: samples: add a module parameter to the rust_minimal sample Andreas Hindborg
2025-07-02 13:18 ` [PATCH v14 7/7] modules: add rust modules files to MAINTAINERS Andreas Hindborg
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=87bjq1ve6t.fsf@kernel.org \
--to=a.hindborg@kernel.org \
--cc=alex.gaynor@gmail.com \
--cc=aliceryhl@google.com \
--cc=ark.email@gmail.com \
--cc=bjorn3_gh@protonmail.com \
--cc=boqun.feng@gmail.com \
--cc=da.gomez@samsung.com \
--cc=dakr@kernel.org \
--cc=daniel.almeida@collabora.com \
--cc=gary@garyguo.net \
--cc=gregkh@linuxfoundation.org \
--cc=linux-kbuild@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-modules@vger.kernel.org \
--cc=lossin@kernel.org \
--cc=masahiroy@kernel.org \
--cc=mcgrof@kernel.org \
--cc=me@kloenk.dev \
--cc=nathan@kernel.org \
--cc=nicolas.schier@linux.dev \
--cc=ojeda@kernel.org \
--cc=petr.pavlu@suse.com \
--cc=rust-for-linux@vger.kernel.org \
--cc=samitolvanen@google.com \
--cc=simona.vetter@ffwll.ch \
--cc=tmgross@umich.edu \
/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;
as well as URLs for NNTP newsgroup(s).