From: "Gary Guo" <gary@garyguo.net>
To: "Danilo Krummrich" <dakr@kernel.org>,
<gregkh@linuxfoundation.org>, <rafael@kernel.org>,
<acourbot@nvidia.com>, <aliceryhl@google.com>,
<david.m.ertman@intel.com>, <ira.weiny@intel.com>,
<leon@kernel.org>, <viresh.kumar@linaro.org>,
<m.wilczynski@samsung.com>, <ukleinek@kernel.org>,
<bhelgaas@google.com>, <kwilczynski@kernel.org>,
<abdiel.janulgue@gmail.com>, <robin.murphy@arm.com>,
<markus.probst@posteo.de>, <ojeda@kernel.org>, <boqun@kernel.org>,
<gary@garyguo.net>, <bjorn3_gh@protonmail.com>,
<lossin@kernel.org>, <a.hindborg@kernel.org>, <tmgross@umich.edu>,
<igor.korotin@linux.dev>, <daniel.almeida@collabora.com>
Cc: <driver-core@lists.linux.dev>, <linux-kernel@vger.kernel.org>,
<nova-gpu@lists.linux.dev>, <dri-devel@lists.freedesktop.org>,
<linux-pm@vger.kernel.org>, <linux-pwm@vger.kernel.org>,
<linux-pci@vger.kernel.org>, <rust-for-linux@vger.kernel.org>
Subject: Re: [PATCH v2 04/25] rust: driver: make Adapter trait lifetime-parameterized
Date: Thu, 07 May 2026 13:24:08 +0100 [thread overview]
Message-ID: <DICFXB4BLACF.3GTAQPRYURQCB@garyguo.net> (raw)
In-Reply-To: <20260506215113.851360-5-dakr@kernel.org>
On Wed May 6, 2026 at 10:50 PM BST, Danilo Krummrich wrote:
> Parameterize the Adapter trait with a lifetime, changing the id_info(),
> of_id_info() and acpi_id_info() methods to take &'bound Device<Bound> and
> return &'bound Self::IdInfo instead of &'static.
>
> This is needed for the ForLt conversion of bus driver traits. Once
> Driver becomes lifetime-parameterized, its IdInfo associated type may
> depend on the lifetime parameter.
How? The `IdInfo` is part of the ID table which is static.
I feel that this is really just adding lifetime for the sake of adding it.
Nothing in `Adapter` is really tied to device lifetime.
> With Adapter<'bound>, the impl can set
> IdInfo = <F::Of<'bound> as Driver<'bound>>::IdInfo and the lifetime flows
Well, I do get that you're trying to avoid writing `Driver<'static>`. This is
needed because conceptually, a `Driver` trait have a lifetime-dependent part and
a lifetime independent part.
`IdInfo` and `ID_TABLE` are the lifetime-independent part, and `probe` and
`unbind` are lifetime dependent part. We really only have this issue because we
use `Self` instead of an associated data for driver's data.
The `Driver` could be more cleanly designed if expressed like this:
trait Driver {
type IdInfo;
const ID_TABLE: IdTable<Self::IdInfo>;
type Data<'bound>;
fn probe<'bound>(
dev: &'bound Device<device::Core>,
id_info: &'bound Self::IdInfo,
) -> impl PinInit<Self::Data<'bound>, Error> + 'bound;
...
}
Then there's no `Driver<'static>` at all.
But even if we don't go with that design, I'd prefer using `Driver<'static>` to
giving lifetime parameters to types that shouldn't have one.
> through naturally, avoiding the need for transmute.
I applied the following diff and it doesn't require
additional transmute?
diff --git a/rust/kernel/driver.rs b/rust/kernel/driver.rs
index 7c5148772697..4c5bc31fd71d 100644
--- a/rust/kernel/driver.rs
+++ b/rust/kernel/driver.rs
@@ -316,7 +316,7 @@ unsafe fn acpi_of_match_device(
/// of a device and a driver.
///
/// It provides bus independent functions for device / driver interactions.
-pub trait Adapter<'bound> {
+pub trait Adapter {
/// The type holding driver private data about each device id supported by the driver.
type IdInfo: 'static;
@@ -326,7 +326,7 @@ pub trait Adapter<'bound> {
/// Returns the driver's private data from the matching entry in the [`acpi::IdTable`], if any.
///
/// If this returns `None`, it means there is no match with an entry in the [`acpi::IdTable`].
- fn acpi_id_info(dev: &'bound Device<Bound>) -> Option<&'bound Self::IdInfo> {
+ fn acpi_id_info(dev: &Device<Bound>) -> Option<&'static Self::IdInfo> {
#[cfg(not(CONFIG_ACPI))]
{
let _ = dev;
@@ -360,7 +360,7 @@ fn acpi_id_info(dev: &'bound Device<Bound>) -> Option<&'bound Self::IdInfo> {
/// Returns the driver's private data from the matching entry in the [`of::IdTable`], if any.
///
/// If this returns `None`, it means there is no match with an entry in the [`of::IdTable`].
- fn of_id_info(dev: &'bound Device<Bound>) -> Option<&'bound Self::IdInfo> {
+ fn of_id_info(dev: &Device<Bound>) -> Option<&'static Self::IdInfo> {
let table = Self::of_id_table()?;
#[cfg(not(any(CONFIG_OF, CONFIG_ACPI)))]
@@ -424,7 +424,7 @@ fn of_id_info(dev: &'bound Device<Bound>) -> Option<&'bound Self::IdInfo> {
///
/// If this returns `None`, it means that there is no match in any of the ID tables directly
/// associated with a [`device::Device`].
- fn id_info(dev: &'bound Device<Bound>) -> Option<&'bound Self::IdInfo> {
+ fn id_info(dev: &Device<Bound>) -> Option<&'static Self::IdInfo> {
let id = Self::acpi_id_info(dev);
if id.is_some() {
return id;
diff --git a/rust/kernel/i2c.rs b/rust/kernel/i2c.rs
index 5a601cba7dc0..739b97c9917d 100644
--- a/rust/kernel/i2c.rs
+++ b/rust/kernel/i2c.rs
@@ -183,7 +183,7 @@ extern "C" fn probe_callback(idev: *mut bindings::i2c_client) -> kernel::ffi::c_
from_result(|| {
let info = Self::i2c_id_info(idev)
- .or_else(|| <Self as driver::Adapter<'_>>::id_info(idev.as_ref()));
+ .or_else(|| <Self as driver::Adapter>::id_info(idev.as_ref()));
let data = <F::Of<'_> as Driver<'_>>::probe(idev, info);
idev.as_ref().set_drvdata::<F>(data)?;
@@ -243,19 +243,19 @@ fn i2c_id_info<'bound>(
}
}
-impl<'bound, F> driver::Adapter<'bound> for Adapter<F>
+impl<F> driver::Adapter for Adapter<F>
where
F: ForLt + 'static,
- F::Of<'bound>: Driver<'bound>,
+ for<'bound> F::Of<'bound>: Driver<'bound>,
{
- type IdInfo = <F::Of<'bound> as Driver<'bound>>::IdInfo;
+ type IdInfo = <F::Of<'static> as Driver<'static>>::IdInfo;
fn of_id_table() -> Option<of::IdTable<Self::IdInfo>> {
- <F::Of<'bound> as Driver<'bound>>::OF_ID_TABLE
+ <F::Of<'static> as Driver<'static>>::OF_ID_TABLE
}
fn acpi_id_table() -> Option<acpi::IdTable<Self::IdInfo>> {
- <F::Of<'bound> as Driver<'bound>>::ACPI_ID_TABLE
+ <F::Of<'static> as Driver<'static>>::ACPI_ID_TABLE
}
}
diff --git a/rust/kernel/platform.rs b/rust/kernel/platform.rs
index f59f22283c53..f4df2e474ca0 100644
--- a/rust/kernel/platform.rs
+++ b/rust/kernel/platform.rs
@@ -120,7 +120,7 @@ extern "C" fn probe_callback(pdev: *mut bindings::platform_device) -> kernel::ff
let pdev = unsafe { &*pdev.cast::<Device<device::CoreInternal>>() };
from_result(|| {
- let info = <Self as driver::Adapter<'_>>::id_info(pdev.as_ref());
+ let info = <Self as driver::Adapter>::id_info(pdev.as_ref());
let data = <F::Of<'_> as Driver<'_>>::probe(pdev, info);
pdev.as_ref().set_drvdata::<F>(data)?;
@@ -143,19 +143,19 @@ extern "C" fn remove_callback(pdev: *mut bindings::platform_device) {
}
}
-impl<'bound, F> driver::Adapter<'bound> for Adapter<F>
+impl<F> driver::Adapter for Adapter<F>
where
F: ForLt + 'static,
for<'b> F::Of<'b>: Driver<'b>,
{
- type IdInfo = <F::Of<'bound> as Driver<'bound>>::IdInfo;
+ type IdInfo = <F::Of<'static> as Driver<'static>>::IdInfo;
fn of_id_table() -> Option<of::IdTable<Self::IdInfo>> {
- <F::Of<'bound> as Driver<'bound>>::OF_ID_TABLE
+ <F::Of<'static> as Driver<'static>>::OF_ID_TABLE
}
fn acpi_id_table() -> Option<acpi::IdTable<Self::IdInfo>> {
- <F::Of<'bound> as Driver<'bound>>::ACPI_ID_TABLE
+ <F::Of<'static> as Driver<'static>>::ACPI_ID_TABLE
}
}
next prev parent reply other threads:[~2026-05-07 12:24 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-05-06 21:50 [PATCH v2 00/25] rust: device: Higher-Ranked Lifetime Types for device drivers Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 01/25] rust: driver core: drop drvdata before devres release Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 02/25] rust: types: add `ForLt` trait for higher-ranked lifetime support Danilo Krummrich
2026-05-06 22:14 ` Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 03/25] rust: device: generalize drvdata methods over ForLt Danilo Krummrich
2026-05-07 12:50 ` Alice Ryhl
2026-05-06 21:50 ` [PATCH v2 04/25] rust: driver: make Adapter trait lifetime-parameterized Danilo Krummrich
2026-05-07 12:24 ` Gary Guo [this message]
2026-05-06 21:50 ` [PATCH v2 05/25] rust: pci: implement Sync for Device<Bound> Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 06/25] rust: platform: " Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 07/25] rust: auxiliary: " Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 08/25] rust: usb: " Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 09/25] rust: device: " Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 10/25] rust: pci: make Driver trait lifetime-parameterized Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 11/25] rust: platform: " Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 12/25] rust: auxiliary: " Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 13/25] rust: auxiliary: generalize Registration over ForLt Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 14/25] samples: rust: rust_driver_auxiliary: showcase lifetime-bound registration data Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 15/25] rust: usb: make Driver trait lifetime-parameterized Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 16/25] rust: i2c: " Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 17/25] rust: pci: make Bar lifetime-parameterized Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 18/25] rust: io: make IoMem and ExclusiveIoMem lifetime-parameterized Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 19/25] samples: rust: rust_driver_pci: use HRT lifetime for Bar Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 20/25] rust: driver-core: rename 'a lifetime to 'bound Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 21/25] gpu: nova-core: " Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 22/25] gpu: nova-core: use HRT lifetime for Bar Danilo Krummrich
2026-05-06 21:50 ` [PATCH v2 23/25] gpu: nova-core: unregister sysmem flush page from Drop Danilo Krummrich
2026-05-06 21:51 ` [PATCH v2 24/25] gpu: nova-core: replace ARef<Device> with &'bound Device in SysmemFlush Danilo Krummrich
2026-05-06 21:51 ` [PATCH v2 25/25] gpu: drm: tyr: use HRT lifetime for IoMem Danilo Krummrich
2026-05-07 12:14 ` [PATCH v2 00/25] rust: device: Higher-Ranked Lifetime Types for device drivers Dirk Behme
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=DICFXB4BLACF.3GTAQPRYURQCB@garyguo.net \
--to=gary@garyguo.net \
--cc=a.hindborg@kernel.org \
--cc=abdiel.janulgue@gmail.com \
--cc=acourbot@nvidia.com \
--cc=aliceryhl@google.com \
--cc=bhelgaas@google.com \
--cc=bjorn3_gh@protonmail.com \
--cc=boqun@kernel.org \
--cc=dakr@kernel.org \
--cc=daniel.almeida@collabora.com \
--cc=david.m.ertman@intel.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=driver-core@lists.linux.dev \
--cc=gregkh@linuxfoundation.org \
--cc=igor.korotin@linux.dev \
--cc=ira.weiny@intel.com \
--cc=kwilczynski@kernel.org \
--cc=leon@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=linux-pwm@vger.kernel.org \
--cc=lossin@kernel.org \
--cc=m.wilczynski@samsung.com \
--cc=markus.probst@posteo.de \
--cc=nova-gpu@lists.linux.dev \
--cc=ojeda@kernel.org \
--cc=rafael@kernel.org \
--cc=robin.murphy@arm.com \
--cc=rust-for-linux@vger.kernel.org \
--cc=tmgross@umich.edu \
--cc=ukleinek@kernel.org \
--cc=viresh.kumar@linaro.org \
/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