[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <DFPWOMZOX6CI.27JJBKSO6IONE@kernel.org>
Date: Fri, 16 Jan 2026 10:28:51 +0100
From: "Benno Lossin" <lossin@...nel.org>
To: "Tamir Duberstein" <tamird@...il.com>
Cc: "Gary Guo" <gary@...yguo.net>, "Miguel Ojeda" <ojeda@...nel.org>, "Boqun
Feng" <boqun.feng@...il.com>, Björn Roy Baron
<bjorn3_gh@...tonmail.com>, "Andreas Hindborg" <a.hindborg@...nel.org>,
"Alice Ryhl" <aliceryhl@...gle.com>, "Trevor Gross" <tmgross@...ch.edu>,
"Danilo Krummrich" <dakr@...nel.org>, "Fiona Behrens" <me@...enk.dev>,
"Christian Schrefl" <chrisi.schrefl@...il.com>, "Alban Kurti"
<kurti@...icto.ai>, <linux-kernel@...r.kernel.org>,
<rust-for-linux@...r.kernel.org>
Subject: Re: [PATCH v3 09/15] rust: pin-init: rewrite the initializer macros
using `syn`
On Fri Jan 16, 2026 at 1:19 AM CET, Tamir Duberstein wrote:
> On Thu, Jan 15, 2026 at 5:43 PM Benno Lossin <lossin@...nel.org> wrote:
>>
>> On Thu Jan 15, 2026 at 10:45 PM CET, Tamir Duberstein wrote:
>> > On Wed, Jan 14, 2026 at 1:20 PM Benno Lossin <lossin@...nel.org> wrote:
>> >>
>> >> Rewrite the initializer macros `[pin_]init!` using `syn`. No functional
>> >> changes intended aside from improved error messages on syntactic and
>> >> semantical errors. For example if one forgets to use `<-` with an
>> >> initializer (and instead uses `:`):
>> >>
>> >> <snip>
>> >>
>> >> Tested-by: Andreas Hindborg <a.hindborg@...nel.org>
>> >> Signed-off-by: Benno Lossin <lossin@...nel.org>
>> >> ---
>> >>
>> >> <snip>
>> >>
>> >> -#[cfg(kernel)]
>> >> -pub use ::macros::paste;
>> >> -#[cfg(not(kernel))]
>> >> -pub use ::paste::paste;
>> >
>> > Now that `cfg(kernel)` is gone, can we also remove it from `rust/Makefile`
>> > and `scripts/generate_rust_analyzer.py`? That is already done for
>> > pin-init in patch 3 of this series:
>> > https://lore.kernel.org/all/20260114181934.1782470-4-lossin@kernel.org/.
>>
>> I'm still using `cfg(kernel)` in `src/lib.rs` for linking to the std
>> Box and Arc types.
>
> Are you sure? I'm not seeing it. It looks like the crate is no_std
> unless the std feature is activated and no features are activated in
> the kernel.
Yes:
$ rg 'cfg.*kernel'
src/lib.rs
260:#![cfg_attr(not(kernel), doc = "[`Arc<T>`]: alloc::alloc::sync::Arc")]
261:#![cfg_attr(not(kernel), doc = "[`Box<T>`]: alloc::alloc::boxed::Box")]
921:#[cfg_attr(not(kernel), doc = "[`Arc<T>`]: alloc::alloc::sync::Arc")]
922:#[cfg_attr(not(kernel), doc = "[`Box<T>`]: alloc::alloc::boxed::Box")]
1020:#[cfg_attr(not(kernel), doc = "[`Arc<T>`]: alloc::alloc::sync::Arc")]
1021:#[cfg_attr(not(kernel), doc = "[`Box<T>`]: alloc::alloc::boxed::Box")]
>> Also I don't want to have to reintroduce all that infrastructure when I
>> inevitably have to add new kernel workarounds.
>
> Then why remove it for pin-init?
I don't understand, what is removed where? In patch 3 the `--cfg kernel`
is kept in all locations.
Cheers,
Benno
Powered by blists - more mailing lists