rust: module_param: return value by copy from `value`

PATCH In review Andreas Hindborg 6 participants 11 messages Jun 1 – Jul 10

Subsystems (2)

Rust Module support

lore

Share & tools

lore

Andreas Hindborg

author reviewer

2026-06-01 10:17 UTC #0 lore raw

PATCH v2 1/1 rust: module_param: return value by copy from `value`

For Copy parameter types it is more ergonomic to retrieve the parameter value by copy than through a shared reference. Change ModuleParamAccess::value to return T by copy when T: Copy, and rename the previous reference-returning accessor to value_ref. Update the in-tree caller in rust_minimal.

Signed-off-by Andreas Hindborg <[email protected]>

---
This change was suggested during review of the rust null block series [1].

[1] https://lore.kernel.org/r/[email protected]
---
Changes in v2:
- Move value into existing impl block (Gary).
- Add inline attributes (Sashiko).
- Link to v1: https://msgid.link/[email protected]

---
 rust/kernel/module_param.rs  | 18 +++++++++++++++++-
 samples/rust/rust_minimal.rs |  2 +-
 2 files changed, 18 insertions(+), 2 deletions(-)

rust/kernel/module_param.rs +17 -1

index 6a8a7a875643..a33302d1b8a3 100644
@@ -131,10 +131,26 @@ pub const fn new(default: T) -> Self {
         }
     }
 
+    /// Get a copy of the parameter value.
+    ///
+    /// Returns the value supplied at module load time, or the default value
+    /// if the parameter has not been set.
+    #[inline]
+    pub fn value(&self) -> T
+    where
+        T: Copy,
+    {
+        self.value.copy().unwrap_or(self.default)
+    }
+
     /// Get a shared reference to the parameter value.
+    ///
+    /// Returns a reference to the value supplied at module load time, or a
+    /// reference to the default value if the parameter has not been set.
     // Note: When sysfs access to parameters are enabled, we have to pass in a
     // held lock guard here.
-    pub fn value(&self) -> &T {
+    #[inline]
+    pub fn value_ref(&self) -> &T {
         self.value.as_ref().unwrap_or(&self.default)
     }

samples/rust/rust_minimal.rs +1 -1

index 8eb9583571d7..60d03df6cd80 100644
@@ -28,7 +28,7 @@ fn init(_module: &'static ThisModule) -> Result<Self> {
         pr_info!("Am I built-in? {}\n", !cfg!(MODULE));
         pr_info!(
             "test_parameter: {}\n",
-            *module_parameters::test_parameter.value()
+            module_parameters::test_parameter.value()
         );
 
         let mut numbers = KVec::new();

---
base-commit: 7fd2df204f342fc17d1a0bfcd474b24232fb0f32
change-id: 20260529-modules-value-ref-e95a7ab94fdb

Best regards,

Signature · 1 line

Andreas Hindborg <[email protected]>

[email protected]

AI review

2026-06-01 10:28 UTC #1 lore raw

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] Casting a pointer of ModuleParamAccess<T> to SetOnce<T> invokes undefined behavior due to missing #[repr(C)] on ModuleParamAccess. --

commit 9dc35dfddf4037b5b314ed7586fa775219811371
Author: Andreas Hindborg <[email protected]>

rust: module_param: return value by copy from value

This commit updates parameter access to be more ergonomic for Copy types by returning the parameter value by copy rather than through a shared reference. It changes ModuleParamAccess::value to return T by copy when
T: Copy, and renames the previous reference-returning accessor to value_ref. The in-tree caller in rust_minimal is updated to reflect this change.

rust/kernel/module_param.rs

Quoted 4 lines

rust/kernel/module_param.rs +0 -0
index 6a8a7a875643a..a33302d1b8a36 100644

[ ... ]

[Severity: High]
This isn't a bug introduced by this patch, but casting the pointer to SetOnce<T> appears to rely on an unspecified memory layout.

In rust/kernel/module_param.rs:ModuleParamAccess:

pub struct ModuleParamAccess<T> {
value: SetOnce<T>,
default: T,
}

Since ModuleParamAccess lacks a #[repr(C)] annotation, the compiler is free to reorder its fields. If the default field is placed before value to minimize padding, would the cast in set_param() overwrite the default field instead?

rust/kernel/module_param.rs:set_param() {
    ...
    // SAFETY: By function safety requirements, this access is safe.
    let container = unsafe {
        &*((*param).__bindgen_anon_1.arg.cast::<SetOnce<T>>())
    };

    container
        .populate(new_value)
    ...
}

Should ModuleParamAccess be marked with #[repr(C)], or should the pointer be cast back to ModuleParamAccess<T> before accessing the value field?

Quoted 6 lines

rust/kernel/module_param.rs +2 -0

@@ -131,10 +131,26 @@ pub const fn new(default: T) -> Self {
         }
     }
 
+    /// Get a copy of the parameter value.
+    ///

[ ... ]

Signature · 1 line

Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=1

Gary Guo

reviewer Reviewed-by on v2 1/1

2026-06-01 20:15 UTC #2 lore raw

In reply to Andreas Hindborg

Quoted 8 lines · Andreas Hindborg

For Copy parameter types it is more ergonomic to retrieve the parameter value by copy than through a shared reference. Change ModuleParamAccess::value to return T by copy when T: Copy, and rename the previous reference-returning accessor to value_ref. Update the in-tree caller in rust_minimal.
Suggested-by Alice Ryhl <[email protected]>
Signed-off-by Andreas Hindborg <[email protected]>

Reviewed-by Gary Guo <[email protected]>

Quoted 13 lines

---
This change was suggested during review of the rust null block series [1].
[1] https://lore.kernel.org/r/[email protected]
---
Changes in v2:
- Move value into existing impl block (Gary).
- Add inline attributes (Sashiko).
- Link to v1: https://msgid.link/[email protected]
---
 rust/kernel/module_param.rs  | 18 +++++++++++++++++-
 samples/rust/rust_minimal.rs |  2 +-
 2 files changed, 18 insertions(+), 2 deletions(-)

Petr Pavlu

maintainer Reviewed-by on v2 1/1

2026-07-09 11:48 UTC #3 lore raw

In reply to Andreas Hindborg

Quoted 8 lines · Andreas Hindborg

For Copy parameter types it is more ergonomic to retrieve the parameter value by copy than through a shared reference. Change ModuleParamAccess::value to return T by copy when T: Copy, and rename the previous reference-returning accessor to value_ref. Update the in-tree caller in rust_minimal.
Suggested-by Alice Ryhl <[email protected]>
Signed-off-by Andreas Hindborg <[email protected]>

Reviewed-by Petr Pavlu <[email protected]>

@Miguel, please let me know if I should take this patch on modules-next, or if you'd prefer for it to go through the Rust tree.

Signature · 2 lines

Thanks,
Petr

Miguel Ojeda

2026-07-09 14:13 UTC #4 lore raw

Quoted 3 lines · Petr Pavlu

@Miguel, please let me know if I should take this patch on modules-next, or if you'd prefer for it to go through the Rust tree.

Please feel free to take it through modules-next, of course.

Thanks!

Cheers,
Miguel

Petr Pavlu

maintainer

2026-07-10 08:13 UTC #5 lore raw

Quoted 6 lines · Miguel Ojeda

Quoted 3 lines · Petr Pavlu
@Miguel, please let me know if I should take this patch on modules-next, or if you'd prefer for it to go through the Rust tree.
Please feel free to take it through modules-next, of course.

Thanks for the confirmation. Queued now on modules-next for v7.3-rc1.

-- Petr

Alice Ryhl

reviewer

2026-07-10 14:26 UTC #6 lore raw

Quoted 9 lines · Petr Pavlu

Quoted 6 lines · Miguel Ojeda
Quoted 3 lines · Petr Pavlu
@Miguel, please let me know if I should take this patch on modules-next, or if you'd prefer for it to go through the Rust tree.
Please feel free to take it through modules-next, of course.
Thanks for the confirmation. Queued now on modules-next for v7.3-rc1.

I'm seeing this error on x86 with latest nightly rust compiler:

rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramaEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramhEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramhEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramiEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramiEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramjEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramjEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramlEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramlEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_parammEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_parammEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramsEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramsEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramtEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramtEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramyEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramyEB4_() falls through to next function _RINvNvNtCs65fqAZscXCv_6kernel6devres16register_foreign8callbackINtNtCsbtRzKR1rrUg_4core3pin3PinINtNtNtB6_5alloc4kbox3BoxNtNtNtB6_3pci3irq21IrqVectorRegistrationNtNtB1A_9allocator7KmallocEEEB6_() rust/kernel.o: error: objtool: _RNvMNtCs65fqAZscXCv_6kernel6deviceNtB2_6Device10get_device() falls through to next function _RNvMNtCs65fqAZscXCv_6kernel6devresINtB2_6DevresNtNtB4_3i2c12RegistrationE19devres_node_releaseB4_()
rust/kernel.o: error: objtool: _RNvMNtCs65fqAZscXCv_6kernel6devresINtB2_6DevresNtNtB4_3i2c12RegistrationE19devres_node_releaseB4_() falls through to next function _RNvMNtCs65fqAZscXCv_6kernel6devresINtB2_6DevresNtNtB4_3i2c12RegistrationE21devres_node_free_nodeB4_()

I understand this is probably due to a new function in Rust stdlib that is no_return, so probably not caused by this patch even if this is triggering it.

Alice

Miguel Ojeda

2026-07-10 14:42 UTC #7 lore raw

Quoted 4 lines · Alice Ryhl

I understand this is probably due to a new function in Rust stdlib that is no_return, so probably not caused by this patch even if this is triggering it.

I can take a look and perhaps push a patch via rust-fixes, which would make it fixed by the time this one lands in the merge window.

Cheers,
Miguel

Petr Pavlu

maintainer

2026-07-10 15:58 UTC #8 lore raw

In reply to Alice Ryhl

Quoted 25 lines · Alice Ryhl

Quoted 9 lines · Petr Pavlu
Quoted 6 lines · Miguel Ojeda
Quoted 3 lines · Petr Pavlu
@Miguel, please let me know if I should take this patch on modules-next, or if you'd prefer for it to go through the Rust tree.
Please feel free to take it through modules-next, of course.
Thanks for the confirmation. Queued now on modules-next for v7.3-rc1.
I'm seeing this error on x86 with latest nightly rust compiler:
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramaEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramhEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramhEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramiEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramiEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramjEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramjEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramlEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramlEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_parammEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_parammEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramsEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramsEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramtEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramtEB4_() falls through to next function _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramyEB4_()
rust/kernel.o: error: objtool: _RINvNtCs65fqAZscXCv_6kernel12module_param9set_paramyEB4_() falls through to next function _RINvNvNtCs65fqAZscXCv_6kernel6devres16register_foreign8callbackINtNtCsbtRzKR1rrUg_4core3pin3PinINtNtNtB6_5alloc4kbox3BoxNtNtNtB6_3pci3irq21IrqVectorRegistrationNtNtB1A_9allocator7KmallocEEEB6_() rust/kernel.o: error: objtool: _RNvMNtCs65fqAZscXCv_6kernel6deviceNtB2_6Device10get_device() falls through to next function _RNvMNtCs65fqAZscXCv_6kernel6devresINtB2_6DevresNtNtB4_3i2c12RegistrationE19devres_node_releaseB4_()
rust/kernel.o: error: objtool: _RNvMNtCs65fqAZscXCv_6kernel6devresINtB2_6DevresNtNtB4_3i2c12RegistrationE19devres_node_releaseB4_() falls through to next function _RNvMNtCs65fqAZscXCv_6kernel6devresINtB2_6DevresNtNtB4_3i2c12RegistrationE21devres_node_free_nodeB4_()

Quoted 3 lines

I understand this is probably due to a new function in Rust stdlib that is no_return, so probably not caused by this patch even if this is triggering it.

I can't immediately reproduce this issue. I tried with linux-next (next-20260710), rustc 1.99.0-nightly (af3d95584 2026-07-09) and x86_64_defconfig with Rust enabled. What am I missing? Could you send me your config and the produced rust/kernel.o off-list?

Signature · 2 lines

Thanks,
Petr

Miguel Ojeda

2026-07-10 17:00 UTC #9 lore raw

Quoted 5 lines · Petr Pavlu

I can't immediately reproduce this issue. I tried with linux-next (next-20260710), rustc 1.99.0-nightly (af3d95584 2026-07-09) and x86_64_defconfig with Rust enabled. What am I missing? Could you send me your config and the produced rust/kernel.o off-list?

I can reproduce it independently of this patch.

It looks like panic_null_reference_constructed, added the other day -- I will send the patch.

Cheers,
Miguel

Miguel Ojeda

2026-07-10 17:34 UTC #10 lore raw

Quoted 3 lines · Miguel Ojeda

It looks like panic_null_reference_constructed, added the other day -- I will send the patch.

https://lore.kernel.org/rust-for-linux/[email protected]/

Cheers,
Miguel