The C code defines 2 new workqueues: system_percpu_wq and system_dfl_wq,
respectively the futures replacement for system_wq and system_unbound_wq.

This change introduce system_percpu(), that use the new system_percpu_wq.

In order to enqueue on a specific CPU, two new functions have been added
to the Queue implementation:

  enqueue_cpu()		- that receive a u32 CPU id as 2nd argument
  enqueue_delayed_cpu()	- that receive a u32 CPU id as 3rd argument

system_wq (and so workqueue::system()) will be removed in a future
release cycle and should not be used.

Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
---
 rust/kernel/sync/completion.rs |  2 +-
 rust/kernel/workqueue.rs       | 96 +++++++++++++++++++++++++++++++---
 2 files changed, 91 insertions(+), 7 deletions(-)

diff --git a/rust/kernel/sync/completion.rs b/rust/kernel/sync/completion.rs
index c50012a940a3..ee6aa8f18507 100644
--- a/rust/kernel/sync/completion.rs
+++ b/rust/kernel/sync/completion.rs
@@ -38,7 +38,7 @@
 ///             done <- Completion::new(),
 ///         }), GFP_KERNEL)?;
 ///
-///         let _ = workqueue::system().enqueue(this.clone());
+///         let _ = workqueue::system_percpu().enqueue(this.clone());
 ///
 ///         Ok(this)
 ///     }
diff --git a/rust/kernel/workqueue.rs b/rust/kernel/workqueue.rs
index adbecdb2b3a9..fd28fc74d5b7 100644
--- a/rust/kernel/workqueue.rs
+++ b/rust/kernel/workqueue.rs
@@ -67,7 +67,7 @@
 //! /// This method will enqueue the struct for execution on the system workqueue, where its value
 //! /// will be printed.
 //! fn print_later(val: Arc<MyStruct>) {
-//!     let _ = workqueue::system().enqueue(val);
+//!     let _ = workqueue::system_percpu().enqueue(val);
 //! }
 //! # print_later(MyStruct::new(42).unwrap());
 //! ```
@@ -121,11 +121,11 @@
 //! }
 //!
 //! fn print_1_later(val: Arc<MyStruct>) {
-//!     let _ = workqueue::system().enqueue::<Arc<MyStruct>, 1>(val);
+//!     let _ = workqueue::system_percpu().enqueue::<Arc<MyStruct>, 1>(val);
 //! }
 //!
 //! fn print_2_later(val: Arc<MyStruct>) {
-//!     let _ = workqueue::system().enqueue::<Arc<MyStruct>, 2>(val);
+//!     let _ = workqueue::system_percpu().enqueue::<Arc<MyStruct>, 2>(val);
 //! }
 //! # print_1_later(MyStruct::new(24, 25).unwrap());
 //! # print_2_later(MyStruct::new(41, 42).unwrap());
@@ -171,13 +171,13 @@
 //! /// This method will enqueue the struct for execution on the system workqueue, where its value
 //! /// will be printed 12 jiffies later.
 //! fn print_later(val: Arc<MyStruct>) {
-//!     let _ = workqueue::system().enqueue_delayed(val, 12);
+//!     let _ = workqueue::system_percpu().enqueue_delayed(val, 12);
 //! }
 //!
 //! /// It is also possible to use the ordinary `enqueue` method together with `DelayedWork`. This
 //! /// is equivalent to calling `enqueue_delayed` with a delay of zero.
 //! fn print_now(val: Arc<MyStruct>) {
-//!     let _ = workqueue::system().enqueue(val);
+//!     let _ = workqueue::system_percpu().enqueue(val);
 //! }
 //! # print_later(MyStruct::new(42).unwrap());
 //! # print_now(MyStruct::new(42).unwrap());
@@ -292,6 +292,39 @@ pub fn enqueue<W, const ID: u64>(&self, w: W) -> W::EnqueueOutput
         }
     }
 
+    /// Enqueues a work item on a specific CPU.
+    ///
+    /// This may fail if the work item is already enqueued in a workqueue.
+    ///
+    /// The work item will be submitted on cpu_id.
+    pub fn enqueue_cpu<W, const ID: u64>(&self, w: W, cpu_id: u32) -> W::EnqueueOutput
+    where
+        W: RawWorkItem<ID> + Send + 'static,
+    {
+        let queue_ptr = self.0.get();
+
+        // SAFETY: We only return `false` if the `work_struct` is already in a workqueue. The other
+        // `__enqueue` requirements are not relevant since `W` is `Send` and static.
+        //
+        // The call to `bindings::queue_work_on` will dereference the provided raw pointer, which
+        // is ok because `__enqueue` guarantees that the pointer is valid for the duration of this
+        // closure.
+        //
+        // Furthermore, if the C workqueue code accesses the pointer after this call to
+        // `__enqueue`, then the work item was successfully enqueued, and `bindings::queue_work_on`
+        // will have returned true. In this case, `__enqueue` promises that the raw pointer will
+        // stay valid until we call the function pointer in the `work_struct`, so the access is ok.
+        unsafe {
+            w.__enqueue(move |work_ptr| {
+                bindings::queue_work_on(
+                    cpu_id as ffi::c_int,
+                    queue_ptr,
+                    work_ptr,
+                )
+            })
+        }
+    }
+
     /// Enqueues a delayed work item.
     ///
     /// This may fail if the work item is already enqueued in a workqueue.
@@ -328,6 +361,42 @@ pub fn enqueue_delayed<W, const ID: u64>(&self, w: W, delay: Jiffies) -> W::Enqu
         }
     }
 
+    /// Enqueues a delayed work item on a specific CPU.
+    ///
+    /// This may fail if the work item is already enqueued in a workqueue.
+    ///
+    /// The work item will be submitted on cpu_id.
+    pub fn enqueue_delayed_cpu<W, const ID: u64>(&self, w: W, delay: Jiffies, cpu_id: u32) -> W::EnqueueOutput
+    where
+        W: RawDelayedWorkItem<ID> + Send + 'static,
+    {
+        let queue_ptr = self.0.get();
+
+        // SAFETY: We only return `false` if the `work_struct` is already in a workqueue. The other
+        // `__enqueue` requirements are not relevant since `W` is `Send` and static.
+        //
+        // The call to `bindings::queue_delayed_work_on` will dereference the provided raw pointer,
+        // which is ok because `__enqueue` guarantees that the pointer is valid for the duration of
+        // this closure, and the safety requirements of `RawDelayedWorkItem` expands this
+        // requirement to apply to the entire `delayed_work`.
+        //
+        // Furthermore, if the C workqueue code accesses the pointer after this call to
+        // `__enqueue`, then the work item was successfully enqueued, and
+        // `bindings::queue_delayed_work_on` will have returned true. In this case, `__enqueue`
+        // promises that the raw pointer will stay valid until we call the function pointer in the
+        // `work_struct`, so the access is ok.
+        unsafe {
+            w.__enqueue(move |work_ptr| {
+                bindings::queue_delayed_work_on(
+                    cpu_id as ffi::c_int,
+                    queue_ptr,
+                    container_of!(work_ptr, bindings::delayed_work, work),
+                    delay,
+                )
+            })
+        }
+    }
+
     /// Tries to spawn the given function or closure as a work item.
     ///
     /// This method can fail because it allocates memory to store the work item.
@@ -934,17 +1003,32 @@ unsafe impl<T, const ID: u64> RawDelayedWorkItem<ID> for Pin<KBox<T>>
 {
 }
 
-/// Returns the system work queue (`system_wq`).
+/// Returns the per-cpu system work queue (`system_wq`).
 ///
 /// It is the one used by `schedule[_delayed]_work[_on]()`. Multi-CPU multi-threaded. There are
 /// users which expect relatively short queue flush time.
 ///
 /// Callers shouldn't queue work items which can run for too long.
+///
+/// Note: `system_wq` will be removed in a future release cycle. Use `system_percpu_wq` instead.
 pub fn system() -> &'static Queue {
     // SAFETY: `system_wq` is a C global, always available.
     unsafe { Queue::from_raw(bindings::system_wq) }
 }
 
+/// Returns the per-cpu system work queue (`system_percpu_wq`).
+///
+/// It is the one used by `schedule[_delayed]_work[_on]()`. Multi-CPU multi-threaded. There are
+/// users which expect relatively short queue flush time.
+///
+/// Callers shouldn't queue work items which can run for too long.
+///
+/// Note: `system_percpu_wq` will replace `system_wq` in a future release cycle.
+pub fn system_percpu() -> &'static Queue {
+    // SAFETY: `system_percpu_wq` is a C global, always available.
+    unsafe { Queue::from_raw(bindings::system_percpu_wq) }
+}
+
 /// Returns the system high-priority work queue (`system_highpri_wq`).
 ///
 /// It is similar to the one returned by [`system`] but for work items which require higher
-- 
2.52.0