[PATCH] rust: miscdevice: use `pin_init::zeroed()` for C type initialization

Atharv Dubey posted 1 patch 3 days, 5 hours ago
There is a newer version of this series
rust/kernel/miscdevice.rs | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
[PATCH] rust: miscdevice: use `pin_init::zeroed()` for C type initialization
Posted by Atharv Dubey 3 days, 5 hours ago
Replace manual zero-initialization using
`MaybeUninit::zeroed().assume_init()` with `pin_init::zeroed()`.
The `pin_init` helper provides a safer and clearer API for
zero-initializing C structs without requiring an `unsafe` block.

Link: https://github.com/Rust-for-Linux/linux/issues/1189
Signed-off-by: Atharv Dubey <atharvd440@gmail.com>
---
 rust/kernel/miscdevice.rs | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/rust/kernel/miscdevice.rs b/rust/kernel/miscdevice.rs
index d698cddcb4a5..e168e9da8f4a 100644
--- a/rust/kernel/miscdevice.rs
+++ b/rust/kernel/miscdevice.rs
@@ -20,7 +20,7 @@
     seq_file::SeqFile,
     types::{ForeignOwnable, Opaque},
 };
-use core::{marker::PhantomData, mem::MaybeUninit, pin::Pin};
+use core::{marker::PhantomData, pin::Pin};
 
 /// Options for creating a misc device.
 #[derive(Copy, Clone)]
@@ -33,7 +33,7 @@ impl MiscDeviceOptions {
     /// Create a raw `struct miscdev` ready for registration.
     pub const fn into_raw<T: MiscDevice>(self) -> bindings::miscdevice {
         // SAFETY: All zeros is valid for this C type.
-        let mut result: bindings::miscdevice = unsafe { MaybeUninit::zeroed().assume_init() };
+        let mut result: bindings::miscdevice = pin_init::zeroed();
         result.minor = bindings::MISC_DYNAMIC_MINOR as ffi::c_int;
         result.name = crate::str::as_char_ptr_in_const_context(self.name);
         result.fops = MiscdeviceVTable::<T>::build();
@@ -421,7 +421,7 @@ impl<T: MiscDevice> MiscdeviceVTable<T> {
             None
         },
         // SAFETY: All zeros is a valid value for `bindings::file_operations`.
-        ..unsafe { MaybeUninit::zeroed().assume_init() }
+        ..pin_init::zeroed()
     };
 
     const fn build() -> &'static bindings::file_operations {
-- 
2.43.0
Re: [PATCH] rust: miscdevice: use `pin_init::zeroed()` for C type initialization
Posted by Dirk Behme 2 days, 14 hours ago
On 28.11.25 17:23, Atharv Dubey wrote:
> Replace manual zero-initialization using
> `MaybeUninit::zeroed().assume_init()` with `pin_init::zeroed()`.
> The `pin_init` helper provides a safer and clearer API for
> zero-initializing C structs without requiring an `unsafe` block.
> 
> Link: https://github.com/Rust-for-Linux/linux/issues/1189
> Signed-off-by: Atharv Dubey <atharvd440@gmail.com>
> ---
>  rust/kernel/miscdevice.rs | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/rust/kernel/miscdevice.rs b/rust/kernel/miscdevice.rs
> index d698cddcb4a5..e168e9da8f4a 100644
> --- a/rust/kernel/miscdevice.rs
> +++ b/rust/kernel/miscdevice.rs
> @@ -20,7 +20,7 @@
>      seq_file::SeqFile,
>      types::{ForeignOwnable, Opaque},
>  };
> -use core::{marker::PhantomData, mem::MaybeUninit, pin::Pin};
> +use core::{marker::PhantomData, pin::Pin};
>  
>  /// Options for creating a misc device.
>  #[derive(Copy, Clone)]
> @@ -33,7 +33,7 @@ impl MiscDeviceOptions {
>      /// Create a raw `struct miscdev` ready for registration.
>      pub const fn into_raw<T: MiscDevice>(self) -> bindings::miscdevice {
>          // SAFETY: All zeros is valid for this C type.
> -        let mut result: bindings::miscdevice = unsafe { MaybeUninit::zeroed().assume_init() };
> +        let mut result: bindings::miscdevice = pin_init::zeroed();

While getting rid of the `unsafe`what's about dropping the `SAFETY`
comment as well? As done by Benno in [1]?

Cheers

Dirk

[1]
https://lore.kernel.org/all/20250814093046.2071971-4-lossin@kernel.org/


>          result.minor = bindings::MISC_DYNAMIC_MINOR as ffi::c_int;
>          result.name = crate::str::as_char_ptr_in_const_context(self.name);
>          result.fops = MiscdeviceVTable::<T>::build();
> @@ -421,7 +421,7 @@ impl<T: MiscDevice> MiscdeviceVTable<T> {
>              None
>          },
>          // SAFETY: All zeros is a valid value for `bindings::file_operations`.
> -        ..unsafe { MaybeUninit::zeroed().assume_init() }
> +        ..pin_init::zeroed()
>      };
>  
>      const fn build() -> &'static bindings::file_operations {
Re: [PATCH] rust: miscdevice: use `pin_init::zeroed()` for C type initialization
Posted by Miguel Ojeda 2 days, 5 hours ago
On Sat, Nov 29, 2025 at 7:46 AM Dirk Behme <dirk.behme@gmail.com> wrote:
>
> While getting rid of the `unsafe`what's about dropping the `SAFETY`
> comment as well? As done by Benno in [1]?

Yeah -- Atharv: please make sure to check your patches with `CLIPPY=1`
(which should detect that), `rustdoc`, `rusttest`... Some notes at:

    https://rust-for-linux.com/contributing#submit-checklist-addendum

Thanks!

Cheers,
Miguel
Re: [PATCH] rust: miscdevice: use `pin_init::zeroed()` for C type initialization
Posted by Atharv Dubey 7 hours ago
On Sat, Nov 29, 2025 at 05:33:28PM +0100, Miguel Ojeda wrote:
> On Sat, Nov 29, 2025 at 7:46 AM Dirk Behme <dirk.behme@gmail.com> wrote:
> >
> > While getting rid of the `unsafe`what's about dropping the `SAFETY`
> > comment as well? As done by Benno in [1]?
> 
> Yeah -- Atharv: please make sure to check your patches with `CLIPPY=1`
> (which should detect that), `rustdoc`, `rusttest`... Some notes at:
> 
>     https://rust-for-linux.com/contributing#submit-checklist-addendum
> 
> Thanks!
> 
> Cheers,
> Miguel 

>Sure,will keep this in mind while sending out patches.  

-Atharv
Re: [PATCH] rust: miscdevice: use `pin_init::zeroed()` for C type initialization
Posted by Miguel Ojeda 3 days, 2 hours ago
On Fri, Nov 28, 2025 at 5:23 PM Atharv Dubey <atharvd440@gmail.com> wrote:
>
> Replace manual zero-initialization using
> `MaybeUninit::zeroed().assume_init()` with `pin_init::zeroed()`.
> The `pin_init` helper provides a safer and clearer API for
> zero-initializing C structs without requiring an `unsafe` block.
>
> Link: https://github.com/Rust-for-Linux/linux/issues/1189
> Signed-off-by: Atharv Dubey <atharvd440@gmail.com>

Welcome! I am glad you ended up sending a patch :)

Minor procedural nit: I think this is missing:

Suggested-by: Benno Lossin <lossin@kernel.org>

It should be picked up automatically, no need to resend a v2 just for this.

Thanks!

Cheers,
Miguel