linux.git/rust/kernel, branch v6.12.80

rust: pin-init: internal: init: document load-bearing fact of field accessors

2026-04-02T11:09:50+00:00

commit 580cc37b1de4fcd9997c48d7080e744533f09f36 upstream.

The functions `[Pin]Init::__[pinned_]init` and `ptr::write` called from
the `init!` macro require the passed pointer to be aligned. This fact is
ensured by the creation of field accessors to previously initialized
fields.

Since we missed this very important fact from the beginning [1],
document it in the code.

Link: https://rust-for-linux.zulipchat.com/#narrow/channel/561532-pin-init/topic/initialized.20field.20accessor.20detection/with/576210658 [1]
Fixes: 90e53c5e70a6 ("rust: add pin-init API core")
Cc:  # 6.6.y, 6.12.y: 42415d163e5d: rust: pin-init: add references to previously initialized fields
Cc:  # 6.6.y, 6.12.y, 6.18.y, 6.19.y
Signed-off-by: Benno Lossin 
Reviewed-by: Gary Guo 
Link: https://patch.msgid.link/20260302140424.4097655-2-lossin@kernel.org
[ Updated Cc: stable@ tags as discussed. - Miguel ]
Signed-off-by: Miguel Ojeda 
[ Moved changes to the declarative macro, because 6.19.y and earlier do not
  have `syn`. Also duplicated the comment for all field accessor creations.
  - Benno ]
Signed-off-by: Benno Lossin 
Signed-off-by: Greg Kroah-Hartman

rust: pin-init: add references to previously initialized fields

2026-04-02T11:09:50+00:00

commit 42415d163e5df6db799c7de6262d707e402c2c7e upstream.

After initializing a field in an initializer macro, create a variable
holding a reference that points at that field. The type is either
`Pin<&mut T>` or `&mut T` depending on the field's structural pinning
kind.

[ Applied fixes to devres and rust_driver_pci sample - Benno]
Reviewed-by: Danilo Krummrich 
Signed-off-by: Benno Lossin 
[ Removed the devres changes, because devres is not present in 6.12.y and
  earlier. Also adjusted paths in the macro to account for the fact that
  pin-init is part of the kernel crate in 6.12.y and earlier. - Benno ]
Signed-off-by: Benno Lossin 
Signed-off-by: Greg Kroah-Hartman

rust: kunit: fix warning when !CONFIG_PRINTK

2026-03-13T16:20:40+00:00

[ Upstream commit 7dd34dfc8dfa92a7244242098110388367996ac3 ]

If `CONFIG_PRINTK` is not set, then the following warnings are issued
during build:

  warning: unused variable: `args`
    --> ../rust/kernel/kunit.rs:16:12
    |
  16 | pub fn err(args: fmt::Arguments<'_>) {
    |            ^^^^ help: if this is intentional, prefix it with an underscore: `_args`
    |
    = note: `#[warn(unused_variables)]` (part of `#[warn(unused)]`) on by default

  warning: unused variable: `args`
    --> ../rust/kernel/kunit.rs:32:13
    |
  32 | pub fn info(args: fmt::Arguments<'_>) {
    |             ^^^^ help: if this is intentional, prefix it with an underscore: `_args`

Fix this by adding a no-op assignment using `args` when `CONFIG_PRINTK`
is not set.

Fixes: a66d733da801 ("rust: support running Rust documentation tests as KUnit ones")
Signed-off-by: Alexandre Courbot 
Reviewed-by: Alice Ryhl 
Reviewed-by: David Gow 
Signed-off-by: Shuah Khan 
Signed-off-by: Sasha Levin

rust: rbtree: fix documentation typo in CursorMut peek_next method

2026-02-06T15:55:44+00:00

commit 45f6aed8a835ee2bdd0a5d5ee626a91fe285014f upstream.

The peek_next method's doc comment incorrectly stated it accesses the
"previous" node when it actually accesses the next node.

Fix the documentation to accurately reflect the method's behavior.

Fixes: 98c14e40e07a ("rust: rbtree: add cursor")
Reviewed-by: Alice Ryhl 
Signed-off-by: Hang Shu 
Reported-by: Miguel Ojeda 
Closes: https://github.com/Rust-for-Linux/linux/issues/1205
Cc: stable@vger.kernel.org
Reviewed-by: Gary Guo 
Link: https://patch.msgid.link/20251107093921.3379954-1-m18080292938@163.com
[ Reworded slightly. - Miguel ]
Signed-off-by: Miguel Ojeda 
Signed-off-by: Greg Kroah-Hartman

rust: block: fix `srctree/` links

2025-10-12T10:57:16+00:00

commit 208d7f788e84e80992d7b1c82ff17b620eb1371e upstream.

This `srctree/` link pointed to a file with an underscore, but the header
used a dash instead.

Thus fix it.

This cleans a future warning that will check our `srctree/` links.

Cc: stable@vger.kernel.org
Fixes: 3253aba3408a ("rust: block: introduce `kernel::block::mq` module")
Reviewed-by: Daniel Almeida 
Signed-off-by: Miguel Ojeda 
Signed-off-by: Greg Kroah-Hartman

rust: alloc: fix `rusttest` by providing `Cmalloc::aligned_layout` too

2025-08-28T14:31:13+00:00

[ Upstream commit 0f580d5d3d9d9cd0953695cd32e43aac3a946338 ]

Commit fde578c86281 ("rust: alloc: replace aligned_size() with
Kmalloc::aligned_layout()") provides a public `aligned_layout` function
in `Kamlloc`, but not in `Cmalloc`, and thus uses of it will trigger an
error in `rusttest`.

Such a user appeared in the following commit 22ab0641b939 ("rust: drm:
ensure kmalloc() compatible Layout"):

    error[E0599]: no function or associated item named `aligned_layout` found for struct `alloc::allocator_test::Cmalloc` in the current scope
       --> rust/kernel/drm/device.rs:100:31
        |
    100 |         let layout = Kmalloc::aligned_layout(Layout::new::());
        |                               ^^^^^^^^^^^^^^ function or associated item not found in `Cmalloc`
        |
       ::: rust/kernel/alloc/allocator_test.rs:19:1
        |
    19  | pub struct Cmalloc;
        | ------------------ function or associated item `aligned_layout` not found for this struct

Thus add an equivalent one for `Cmalloc`.

Fixes: fde578c86281 ("rust: alloc: replace aligned_size() with Kmalloc::aligned_layout()")
Signed-off-by: Miguel Ojeda 
Link: https://lore.kernel.org/r/20250816204215.2719559-1-ojeda@kernel.org
Signed-off-by: Danilo Krummrich 
Signed-off-by: Sasha Levin

rust: alloc: replace aligned_size() with Kmalloc::aligned_layout()

2025-08-28T14:31:12+00:00

[ Upstream commit fde578c86281f27b182680c7642836a0dbbd0be7 ]

aligned_size() dates back to when Rust did support kmalloc() only, but
is now used in ReallocFunc::call() and hence for all allocators.

However, the additional padding applied by aligned_size() is only
required by the kmalloc() allocator backend.

Hence, replace aligned_size() with Kmalloc::aligned_layout() and use it
for the affected allocators, i.e. kmalloc() and kvmalloc(), only.

While at it, make Kmalloc::aligned_layout() public, such that Rust
abstractions, which have to call subsystem specific kmalloc() based
allocation primitives directly, can make use of it.

Fixes: 8a799831fc63 ("rust: alloc: implement `ReallocFunc`")
Reviewed-by: Alice Ryhl 
Link: https://lore.kernel.org/r/20250731154919.4132-2-dakr@kernel.org
[ Remove `const` from Kmalloc::aligned_layout(). - Danilo ]
Signed-off-by: Danilo Krummrich 
Signed-off-by: Sasha Levin

rust: use `#[used(compiler)]` to fix build and `modpost` with Rust >= 1.89.0

2025-07-24T06:56:37+00:00

commit 7498159226772d66f150dd406be462d75964a366 upstream.

Starting with Rust 1.89.0 (expected 2025-08-07), the Rust compiler fails
to build the `rusttest` target due to undefined references such as:

    kernel...-cgu.0:(.text....+0x116): undefined reference to
    `rust_helper_kunit_get_current_test'

Moreover, tooling like `modpost` gets confused:

    WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/gpu/drm/nova/nova.o
    ERROR: modpost: missing MODULE_LICENSE() in drivers/gpu/nova-core/nova_core.o

The reason behind both issues is that the Rust compiler will now [1]
treat `#[used]` as `#[used(linker)]` instead of `#[used(compiler)]`
for our targets. This means that the retain section flag (`R`,
`SHF_GNU_RETAIN`) will be used and that they will be marked as `unique`
too, with different IDs. In turn, that means we end up with undefined
references that did not get discarded in `rusttest` and that multiple
`.modinfo` sections are generated, which confuse tooling like `modpost`
because they only expect one.

Thus start using `#[used(compiler)]` to keep the previous behavior
and to be explicit about what we want. Sadly, it is an unstable feature
(`used_with_arg`) [2] -- we will talk to upstream Rust about it. The good
news is that it has been available for a long time (Rust >= 1.60) [3].

The changes should also be fine for previous Rust versions, since they
behave the same way as before [4].

Alternatively, we could use `#[no_mangle]` or `#[export_name = ...]`
since those still behave like `#[used(compiler)]`, but of course it is
not really what we want to express, and it requires other changes to
avoid symbol conflicts.

Cc: David Wood 
Cc: Wesley Wiser 
Cc: stable@vger.kernel.org # Needed in 6.12.y and later (Rust is pinned in older LTSs).
Link: https://github.com/rust-lang/rust/pull/140872 [1]
Link: https://github.com/rust-lang/rust/issues/93798 [2]
Link: https://github.com/rust-lang/rust/pull/91504 [3]
Link: https://godbolt.org/z/sxzWTMfzW [4]
Reviewed-by: Alice Ryhl 
Acked-by: Björn Roy Baron 
Link: https://lore.kernel.org/r/20250712160103.1244945-3-ojeda@kernel.org
Signed-off-by: Miguel Ojeda 
Signed-off-by: Greg Kroah-Hartman

rust: init: allow `dead_code` warnings for Rust >= 1.89.0

2025-07-17T16:37:14+00:00

Starting with Rust 1.89.0 (expected 2025-08-07), the Rust compiler
may warn:

    error: trait `MustNotImplDrop` is never used
       --> rust/kernel/init/macros.rs:927:15
        |
    927 |         trait MustNotImplDrop {}
        |               ^^^^^^^^^^^^^^^
        |
       ::: rust/kernel/sync/arc.rs:133:1
        |
    133 | #[pin_data]
        | ----------- in this procedural macro expansion
        |
        = note: `-D dead-code` implied by `-D warnings`
        = help: to override `-D warnings` add `#[allow(dead_code)]`
        = note: this error originates in the macro `$crate::__pin_data`
                which comes from the expansion of the attribute macro
                `pin_data` (in Nightly builds, run with
                -Z macro-backtrace for more info)

Thus `allow` it to clean it up.

This does not happen in mainline nor 6.15.y, because there the macro was
moved out of the `kernel` crate, and `dead_code` warnings are not
emitted if the macro is foreign to the crate. Thus this patch is
directly sent to stable and intended for 6.12.y only.

Similarly, it is not needed in previous LTSs, because there the Rust
version is pinned.

Acked-by: Benno Lossin 
Signed-off-by: Miguel Ojeda 
Signed-off-by: Greg Kroah-Hartman

rust: alloc: add missing invariant in Vec::set_len()

2025-06-19T13:32:09+00:00

[ Upstream commit fb1bf1067de979c89ae33589e0466d6ce0dde204 ]

When setting a new length, we have to justify that the set length
represents the exact number of elements stored in the vector.

Reviewed-by: Benno Lossin 
Reported-by: Alice Ryhl 
Closes: https://lore.kernel.org/rust-for-linux/20250311-iov-iter-v1-4-f6c9134ea824@google.com
Fixes: 2aac4cd7dae3 ("rust: alloc: implement kernel `Vec` type")
Link: https://lore.kernel.org/r/20250315154436.65065-2-dakr@kernel.org
Signed-off-by: Danilo Krummrich 
Signed-off-by: Sasha Levin