Resguardo inicial

linux-7.2.3/rust/kernel/block/mq.rs

@@ -0,0 +1,103 @@
+// SPDX-License-Identifier: GPL-2.0
+
+//! This module provides types for implementing block drivers that interface the
+//! blk-mq subsystem.
+//!
+//! To implement a block device driver, a Rust module must do the following:
+//!
+//! - Implement [`Operations`] for a type `T`.
+//! - Create a [`TagSet<T>`].
+//! - Create a [`GenDisk<T>`], via the [`GenDiskBuilder`].
+//! - Add the disk to the system by calling [`GenDiskBuilder::build`] passing in
+//!   the `TagSet` reference.
+//!
+//! The types available in this module that have direct C counterparts are:
+//!
+//! - The [`TagSet`] type that abstracts the C type `struct tag_set`.
+//! - The [`GenDisk`] type that abstracts the C type `struct gendisk`.
+//! - The [`Request`] type that abstracts the C type `struct request`.
+//!
+//! The kernel will interface with the block device driver by calling the method
+//! implementations of the `Operations` trait.
+//!
+//! IO requests are passed to the driver as [`kernel::sync::aref::ARef<Request>`]
+//! instances. The `Request` type is a wrapper around the C `struct request`.
+//! The driver must mark end of processing by calling one of the
+//! `Request::end`, methods. Failure to do so can lead to deadlock or timeout
+//! errors. Please note that the C function `blk_mq_start_request` is implicitly
+//! called when the request is queued with the driver.
+//!
+//! The `TagSet` is responsible for creating and maintaining a mapping between
+//! `Request`s and integer ids as well as carrying a pointer to the vtable
+//! generated by `Operations`. This mapping is useful for associating
+//! completions from hardware with the correct `Request` instance. The `TagSet`
+//! determines the maximum queue depth by setting the number of `Request`
+//! instances available to the driver, and it determines the number of queues to
+//! instantiate for the driver. If possible, a driver should allocate one queue
+//! per core, to keep queue data local to a core.
+//!
+//! One `TagSet` instance can be shared between multiple `GenDisk` instances.
+//! This can be useful when implementing drivers where one piece of hardware
+//! with one set of IO resources are represented to the user as multiple disks.
+//!
+//! One significant difference between block device drivers implemented with
+//! these Rust abstractions and drivers implemented in C, is that the Rust
+//! drivers have to own a reference count on the `Request` type when the IO is
+//! in flight. This is to ensure that the C `struct request` instances backing
+//! the Rust `Request` instances are live while the Rust driver holds a
+//! reference to the `Request`. In addition, the conversion of an integer tag to
+//! a `Request` via the `TagSet` would not be sound without this bookkeeping.
+//!
+//! [`GenDisk`]: gen_disk::GenDisk
+//! [`GenDisk<T>`]: gen_disk::GenDisk
+//! [`GenDiskBuilder`]: gen_disk::GenDiskBuilder
+//! [`GenDiskBuilder::build`]: gen_disk::GenDiskBuilder::build
+//!
+//! # Examples
+//!
+//! ```rust
+//! use kernel::{
+//!     alloc::flags,
+//!     block::mq::*,
+//!     new_mutex,
+//!     prelude::*,
+//!     sync::{aref::ARef, Arc, Mutex},
+//! };
+//!
+//! struct MyBlkDevice;
+//!
+//! #[vtable]
+//! impl Operations for MyBlkDevice {
+//!     type QueueData = ();
+//!
+//!     fn queue_rq(_queue_data: (), rq: ARef<Request<Self>>, _is_last: bool) -> Result {
+//!         Request::end_ok(rq);
+//!         Ok(())
+//!     }
+//!
+//!     fn commit_rqs(_queue_data: ()) {}
+//!
+//!     fn complete(rq: ARef<Request<Self>>) {
+//!         Request::end_ok(rq)
+//!             .map_err(|_e| kernel::error::code::EIO)
+//!             .expect("Fatal error - expected to be able to end request");
+//!     }
+//! }
+//!
+//! let tagset: Arc<TagSet<MyBlkDevice>> =
+//!     Arc::pin_init(TagSet::new(1, 256, 1), flags::GFP_KERNEL)?;
+//! let mut disk = gen_disk::GenDiskBuilder::new()
+//!     .capacity_sectors(4096)
+//!     .build(fmt!("myblk"), tagset, ())?;
+//!
+//! # Ok::<(), kernel::error::Error>(())
+//! ```
+
+pub mod gen_disk;
+mod operations;
+mod request;
+mod tag_set;
+
+pub use operations::Operations;
+pub use request::Request;
+pub use tag_set::TagSet;

linux-7.2.3/rust/kernel/block/mq/gen_disk.rs

@@ -0,0 +1,241 @@
+// SPDX-License-Identifier: GPL-2.0
+
+//! Generic disk abstraction.
+//!
+//! C header: [`include/linux/blkdev.h`](srctree/include/linux/blkdev.h)
+//! C header: [`include/linux/blk-mq.h`](srctree/include/linux/blk-mq.h)
+
+use crate::{
+    bindings,
+    block::mq::{Operations, TagSet},
+    error::{self, from_err_ptr, Result},
+    fmt::{self, Write},
+    prelude::*,
+    static_lock_class,
+    str::NullTerminatedFormatter,
+    sync::Arc,
+    types::{ForeignOwnable, ScopeGuard},
+};
+
+/// A builder for [`GenDisk`].
+///
+/// Use this struct to configure and add new [`GenDisk`] to the VFS.
+pub struct GenDiskBuilder {
+    rotational: bool,
+    logical_block_size: u32,
+    physical_block_size: u32,
+    capacity_sectors: u64,
+}
+
+impl Default for GenDiskBuilder {
+    fn default() -> Self {
+        Self {
+            rotational: false,
+            logical_block_size: bindings::PAGE_SIZE as u32,
+            physical_block_size: bindings::PAGE_SIZE as u32,
+            capacity_sectors: 0,
+        }
+    }
+}
+
+impl GenDiskBuilder {
+    /// Create a new instance.
+    pub fn new() -> Self {
+        Self::default()
+    }
+
+    /// Set the rotational media attribute for the device to be built.
+    pub fn rotational(mut self, rotational: bool) -> Self {
+        self.rotational = rotational;
+        self
+    }
+
+    /// Validate block size by verifying that it is between 512 and `PAGE_SIZE`,
+    /// and that it is a power of two.
+    pub fn validate_block_size(size: u32) -> Result {
+        if !(512..=bindings::PAGE_SIZE as u32).contains(&size) || !size.is_power_of_two() {
+            Err(error::code::EINVAL)
+        } else {
+            Ok(())
+        }
+    }
+
+    /// Set the logical block size of the device to be built.
+    ///
+    /// This method will check that block size is a power of two and between 512
+    /// and 4096. If not, an error is returned and the block size is not set.
+    ///
+    /// This is the smallest unit the storage device can address. It is
+    /// typically 4096 bytes.
+    pub fn logical_block_size(mut self, block_size: u32) -> Result<Self> {
+        Self::validate_block_size(block_size)?;
+        self.logical_block_size = block_size;
+        Ok(self)
+    }
+
+    /// Set the physical block size of the device to be built.
+    ///
+    /// This method will check that block size is a power of two and between 512
+    /// and 4096. If not, an error is returned and the block size is not set.
+    ///
+    /// This is the smallest unit a physical storage device can write
+    /// atomically. It is usually the same as the logical block size but may be
+    /// bigger. One example is SATA drives with 4096 byte physical block size
+    /// that expose a 512 byte logical block size to the operating system.
+    pub fn physical_block_size(mut self, block_size: u32) -> Result<Self> {
+        Self::validate_block_size(block_size)?;
+        self.physical_block_size = block_size;
+        Ok(self)
+    }
+
+    /// Set the capacity of the device to be built, in sectors (512 bytes).
+    pub fn capacity_sectors(mut self, capacity: u64) -> Self {
+        self.capacity_sectors = capacity;
+        self
+    }
+
+    /// Build a new `GenDisk` and add it to the VFS.
+    pub fn build<T: Operations>(
+        self,
+        name: fmt::Arguments<'_>,
+        tagset: Arc<TagSet<T>>,
+        queue_data: T::QueueData,
+    ) -> Result<GenDisk<T>> {
+        let data = queue_data.into_foreign();
+        let recover_data = ScopeGuard::new(|| {
+            // SAFETY: T::QueueData was created by the call to `into_foreign()` above
+            drop(unsafe { T::QueueData::from_foreign(data) });
+        });
+
+        let mut lim: bindings::queue_limits = pin_init::zeroed();
+
+        lim.logical_block_size = self.logical_block_size;
+        lim.physical_block_size = self.physical_block_size;
+        if self.rotational {
+            lim.features = bindings::BLK_FEAT_ROTATIONAL;
+        }
+
+        // SAFETY: `tagset.raw_tag_set()` points to a valid and initialized tag set
+        let gendisk = from_err_ptr(unsafe {
+            bindings::__blk_mq_alloc_disk(
+                tagset.raw_tag_set(),
+                &mut lim,
+                data,
+                static_lock_class!().as_ptr(),
+            )
+        })?;
+
+        const TABLE: bindings::block_device_operations = bindings::block_device_operations {
+            submit_bio: None,
+            open: None,
+            release: None,
+            ioctl: None,
+            compat_ioctl: None,
+            check_events: None,
+            unlock_native_capacity: None,
+            getgeo: None,
+            set_read_only: None,
+            swap_slot_free_notify: None,
+            report_zones: None,
+            devnode: None,
+            alternative_gpt_sector: None,
+            get_unique_id: None,
+            // TODO: Set to `THIS_MODULE`.
+            owner: core::ptr::null_mut(),
+            pr_ops: core::ptr::null_mut(),
+            free_disk: None,
+            poll_bio: None,
+        };
+
+        // SAFETY: `gendisk` is a valid pointer as we initialized it above
+        unsafe { (*gendisk).fops = &TABLE };
+
+        let cleanup_failure = ScopeGuard::new_with_data((gendisk, data), |(gendisk, data)| {
+            // SAFETY: `gendisk` came from `__blk_mq_alloc_disk()` above and
+            // has not been added to the VFS on this cleanup path.
+            unsafe { bindings::put_disk(gendisk) };
+            // SAFETY: `data` came from `into_foreign()` above and has not been
+            // converted back on this cleanup path.
+            drop(unsafe { T::QueueData::from_foreign(data) });
+        });
+
+        // The failure guard now owns both pieces of cleanup; the early guard
+        // must not run on this path anymore.
+        recover_data.dismiss();
+
+        let mut writer = NullTerminatedFormatter::new(
+            // SAFETY: `gendisk` points to a valid and initialized instance. We
+            // have exclusive access, since the disk is not added to the VFS
+            // yet.
+            unsafe { &mut (*gendisk).disk_name },
+        )
+        .ok_or(EINVAL)?;
+        writer.write_fmt(name)?;
+
+        // SAFETY: `gendisk` points to a valid and initialized instance of
+        // `struct gendisk`. `set_capacity` takes a lock to synchronize this
+        // operation, so we will not race.
+        unsafe { bindings::set_capacity(gendisk, self.capacity_sectors) };
+
+        crate::error::to_result(
+            // SAFETY: `gendisk` points to a valid and initialized instance of
+            // `struct gendisk`.
+            unsafe {
+                bindings::device_add_disk(core::ptr::null_mut(), gendisk, core::ptr::null_mut())
+            },
+        )?;
+
+        cleanup_failure.dismiss();
+
+        // INVARIANT: `gendisk` was initialized above.
+        // INVARIANT: `gendisk` was added to the VFS via `device_add_disk` above.
+        // INVARIANT: `gendisk.queue.queue_data` is set to `data` in the call to
+        // `__blk_mq_alloc_disk` above.
+        Ok(GenDisk {
+            _tagset: tagset,
+            gendisk,
+        })
+    }
+}
+
+/// A generic block device.
+///
+/// # Invariants
+///
+///  - `gendisk` must always point to an initialized and valid `struct gendisk`.
+///  - `gendisk` was added to the VFS through a call to
+///    `bindings::device_add_disk`.
+///  - `self.gendisk.queue.queuedata` is initialized by a call to `ForeignOwnable::into_foreign`.
+pub struct GenDisk<T: Operations> {
+    _tagset: Arc<TagSet<T>>,
+    gendisk: *mut bindings::gendisk,
+}
+
+// SAFETY: `GenDisk` is an owned pointer to a `struct gendisk` and an `Arc` to a
+// `TagSet` It is safe to send this to other threads as long as T is Send.
+unsafe impl<T: Operations + Send> Send for GenDisk<T> {}
+
+impl<T: Operations> Drop for GenDisk<T> {
+    fn drop(&mut self) {
+        // SAFETY: By type invariant of `Self`, `self.gendisk` points to a valid
+        // and initialized instance of `struct gendisk`, and, `queuedata` was
+        // initialized with the result of a call to
+        // `ForeignOwnable::into_foreign`.
+        let queue_data = unsafe { (*(*self.gendisk).queue).queuedata };
+
+        // SAFETY: By type invariant, `self.gendisk` points to a valid and
+        // initialized instance of `struct gendisk`, and it was previously added
+        // to the VFS.
+        unsafe { bindings::del_gendisk(self.gendisk) };
+
+        // SAFETY: By type invariant, `self.gendisk` was added to the VFS, so
+        // `put_disk()` must follow `del_gendisk()` to drop the final gendisk
+        // reference and trigger the remaining release path.
+        unsafe { bindings::put_disk(self.gendisk) };
+
+        // SAFETY: `queue.queuedata` was created by `GenDiskBuilder::build` with
+        // a call to `ForeignOwnable::into_foreign` to create `queuedata`.
+        // `ForeignOwnable::from_foreign` is only called here.
+        drop(unsafe { T::QueueData::from_foreign(queue_data) });
+    }
+}

linux-7.2.3/rust/kernel/block/mq/operations.rs

@@ -0,0 +1,286 @@
+// SPDX-License-Identifier: GPL-2.0
+
+//! This module provides an interface for blk-mq drivers to implement.
+//!
+//! C header: [`include/linux/blk-mq.h`](srctree/include/linux/blk-mq.h)
+
+use crate::{
+    bindings,
+    block::mq::{request::RequestDataWrapper, Request},
+    error::{from_result, Result},
+    prelude::*,
+    sync::{aref::ARef, Refcount},
+    types::ForeignOwnable,
+};
+use core::marker::PhantomData;
+
+type ForeignBorrowed<'a, T> = <T as ForeignOwnable>::Borrowed<'a>;
+
+/// Implement this trait to interface blk-mq as block devices.
+///
+/// To implement a block device driver, implement this trait as described in the
+/// [module level documentation]. The kernel will use the implementation of the
+/// functions defined in this trait to interface a block device driver. Note:
+/// There is no need for an exit_request() implementation, because the `drop`
+/// implementation of the [`Request`] type will be invoked by automatically by
+/// the C/Rust glue logic.
+///
+/// [module level documentation]: kernel::block::mq
+#[macros::vtable]
+pub trait Operations: Sized {
+    /// Data associated with the `struct request_queue` that is allocated for
+    /// the `GenDisk` associated with this `Operations` implementation.
+    type QueueData: ForeignOwnable;
+
+    /// Called by the kernel to queue a request with the driver. If `is_last` is
+    /// `false`, the driver is allowed to defer committing the request.
+    fn queue_rq(
+        queue_data: ForeignBorrowed<'_, Self::QueueData>,
+        rq: ARef<Request<Self>>,
+        is_last: bool,
+    ) -> Result;
+
+    /// Called by the kernel to indicate that queued requests should be submitted.
+    fn commit_rqs(queue_data: ForeignBorrowed<'_, Self::QueueData>);
+
+    /// Called by the kernel when the request is completed.
+    fn complete(rq: ARef<Request<Self>>);
+
+    /// Called by the kernel to poll the device for completed requests. Only
+    /// used for poll queues.
+    fn poll() -> bool {
+        build_error!(crate::error::VTABLE_DEFAULT_ERROR)
+    }
+}
+
+/// A vtable for blk-mq to interact with a block device driver.
+///
+/// A `bindings::blk_mq_ops` vtable is constructed from pointers to the `extern
+/// "C"` functions of this struct, exposed through the `OperationsVTable::VTABLE`.
+///
+/// For general documentation of these methods, see the kernel source
+/// documentation related to `struct blk_mq_operations` in
+/// [`include/linux/blk-mq.h`].
+///
+/// [`include/linux/blk-mq.h`]: srctree/include/linux/blk-mq.h
+pub(crate) struct OperationsVTable<T: Operations>(PhantomData<T>);
+
+impl<T: Operations> OperationsVTable<T> {
+    /// This function is called by the C kernel. A pointer to this function is
+    /// installed in the `blk_mq_ops` vtable for the driver.
+    ///
+    /// # Safety
+    ///
+    /// - The caller of this function must ensure that the pointee of `bd` is
+    ///   valid for reads for the duration of this function.
+    /// - This function must be called for an initialized and live `hctx`. That
+    ///   is, `Self::init_hctx_callback` was called and
+    ///   `Self::exit_hctx_callback()` was not yet called.
+    /// - `(*bd).rq` must point to an initialized and live `bindings:request`.
+    ///   That is, `Self::init_request_callback` was called but
+    ///   `Self::exit_request_callback` was not yet called for the request.
+    /// - `(*bd).rq` must be owned by the driver. That is, the block layer must
+    ///   promise to not access the request until the driver calls
+    ///   `bindings::blk_mq_end_request` for the request.
+    unsafe extern "C" fn queue_rq_callback(
+        hctx: *mut bindings::blk_mq_hw_ctx,
+        bd: *const bindings::blk_mq_queue_data,
+    ) -> bindings::blk_status_t {
+        // SAFETY: `bd.rq` is valid as required by the safety requirement for
+        // this function.
+        let request = unsafe { &*(*bd).rq.cast::<Request<T>>() };
+
+        // One refcount for the ARef, one for being in flight
+        request.wrapper_ref().refcount().set(2);
+
+        // SAFETY:
+        //  - We own a refcount that we took above. We pass that to `ARef`.
+        //  - By the safety requirements of this function, `request` is a valid
+        //    `struct request` and the private data is properly initialized.
+        //  - `rq` will be alive until `blk_mq_end_request` is called and is
+        //    reference counted by `ARef` until then.
+        let rq = unsafe { Request::aref_from_raw((*bd).rq) };
+
+        // SAFETY: `hctx` is valid as required by this function.
+        let queue_data = unsafe { (*(*hctx).queue).queuedata };
+
+        // SAFETY: `queue.queuedata` was created by `GenDiskBuilder::build` with
+        // a call to `ForeignOwnable::into_foreign` to create `queuedata`.
+        // `ForeignOwnable::from_foreign` is only called when the tagset is
+        // dropped, which happens after we are dropped.
+        let queue_data = unsafe { T::QueueData::borrow(queue_data) };
+
+        // SAFETY: We have exclusive access and we just set the refcount above.
+        unsafe { Request::start_unchecked(&rq) };
+
+        let ret = T::queue_rq(
+            queue_data,
+            rq,
+            // SAFETY: `bd` is valid as required by the safety requirement for
+            // this function.
+            unsafe { (*bd).last },
+        );
+
+        if let Err(e) = ret {
+            e.to_blk_status()
+        } else {
+            bindings::BLK_STS_OK as bindings::blk_status_t
+        }
+    }
+
+    /// This function is called by the C kernel. A pointer to this function is
+    /// installed in the `blk_mq_ops` vtable for the driver.
+    ///
+    /// # Safety
+    ///
+    /// This function may only be called by blk-mq C infrastructure. The caller
+    /// must ensure that `hctx` is valid.
+    unsafe extern "C" fn commit_rqs_callback(hctx: *mut bindings::blk_mq_hw_ctx) {
+        // SAFETY: `hctx` is valid as required by this function.
+        let queue_data = unsafe { (*(*hctx).queue).queuedata };
+
+        // SAFETY: `queue.queuedata` was created by `GenDisk::try_new()` with a
+        // call to `ForeignOwnable::into_foreign()` to create `queuedata`.
+        // `ForeignOwnable::from_foreign()` is only called when the tagset is
+        // dropped, which happens after we are dropped.
+        let queue_data = unsafe { T::QueueData::borrow(queue_data) };
+        T::commit_rqs(queue_data)
+    }
+
+    /// This function is called by the C kernel. A pointer to this function is
+    /// installed in the `blk_mq_ops` vtable for the driver.
+    ///
+    /// # Safety
+    ///
+    /// This function may only be called by blk-mq C infrastructure. `rq` must
+    /// point to a valid request that has been marked as completed. The pointee
+    /// of `rq` must be valid for write for the duration of this function.
+    unsafe extern "C" fn complete_callback(rq: *mut bindings::request) {
+        // SAFETY: This function can only be dispatched through
+        // `Request::complete`. We leaked a refcount then which we pick back up
+        // now.
+        let aref = unsafe { Request::aref_from_raw(rq) };
+        T::complete(aref);
+    }
+
+    /// This function is called by the C kernel. A pointer to this function is
+    /// installed in the `blk_mq_ops` vtable for the driver.
+    ///
+    /// # Safety
+    ///
+    /// This function may only be called by blk-mq C infrastructure.
+    unsafe extern "C" fn poll_callback(
+        _hctx: *mut bindings::blk_mq_hw_ctx,
+        _iob: *mut bindings::io_comp_batch,
+    ) -> crate::ffi::c_int {
+        T::poll().into()
+    }
+
+    /// This function is called by the C kernel. A pointer to this function is
+    /// installed in the `blk_mq_ops` vtable for the driver.
+    ///
+    /// # Safety
+    ///
+    /// This function may only be called by blk-mq C infrastructure. This
+    /// function may only be called once before `exit_hctx_callback` is called
+    /// for the same context.
+    unsafe extern "C" fn init_hctx_callback(
+        _hctx: *mut bindings::blk_mq_hw_ctx,
+        _tagset_data: *mut crate::ffi::c_void,
+        _hctx_idx: crate::ffi::c_uint,
+    ) -> crate::ffi::c_int {
+        from_result(|| Ok(0))
+    }
+
+    /// This function is called by the C kernel. A pointer to this function is
+    /// installed in the `blk_mq_ops` vtable for the driver.
+    ///
+    /// # Safety
+    ///
+    /// This function may only be called by blk-mq C infrastructure.
+    unsafe extern "C" fn exit_hctx_callback(
+        _hctx: *mut bindings::blk_mq_hw_ctx,
+        _hctx_idx: crate::ffi::c_uint,
+    ) {
+    }
+
+    /// This function is called by the C kernel. A pointer to this function is
+    /// installed in the `blk_mq_ops` vtable for the driver.
+    ///
+    /// # Safety
+    ///
+    /// - This function may only be called by blk-mq C infrastructure.
+    /// - `_set` must point to an initialized `TagSet<T>`.
+    /// - `rq` must point to an initialized `bindings::request`.
+    /// - The allocation pointed to by `rq` must be at the size of `Request`
+    ///   plus the size of `RequestDataWrapper`.
+    unsafe extern "C" fn init_request_callback(
+        _set: *mut bindings::blk_mq_tag_set,
+        rq: *mut bindings::request,
+        _hctx_idx: crate::ffi::c_uint,
+        _numa_node: crate::ffi::c_int,
+    ) -> crate::ffi::c_int {
+        from_result(|| {
+            // SAFETY: By the safety requirements of this function, `rq` points
+            // to a valid allocation.
+            let pdu = unsafe { Request::wrapper_ptr(rq.cast::<Request<T>>()) };
+
+            // SAFETY: The refcount field is allocated but not initialized, so
+            // it is valid for writes.
+            unsafe { RequestDataWrapper::refcount_ptr(pdu.as_ptr()).write(Refcount::new(0)) };
+
+            Ok(0)
+        })
+    }
+
+    /// This function is called by the C kernel. A pointer to this function is
+    /// installed in the `blk_mq_ops` vtable for the driver.
+    ///
+    /// # Safety
+    ///
+    /// - This function may only be called by blk-mq C infrastructure.
+    /// - `_set` must point to an initialized `TagSet<T>`.
+    /// - `rq` must point to an initialized and valid `Request`.
+    unsafe extern "C" fn exit_request_callback(
+        _set: *mut bindings::blk_mq_tag_set,
+        rq: *mut bindings::request,
+        _hctx_idx: crate::ffi::c_uint,
+    ) {
+        // SAFETY: The tagset invariants guarantee that all requests are allocated with extra memory
+        // for the request data.
+        let pdu = unsafe { bindings::blk_mq_rq_to_pdu(rq) }.cast::<RequestDataWrapper>();
+
+        // SAFETY: `pdu` is valid for read and write and is properly initialised.
+        unsafe { core::ptr::drop_in_place(pdu) };
+    }
+
+    const VTABLE: bindings::blk_mq_ops = bindings::blk_mq_ops {
+        queue_rq: Some(Self::queue_rq_callback),
+        queue_rqs: None,
+        commit_rqs: Some(Self::commit_rqs_callback),
+        get_budget: None,
+        put_budget: None,
+        set_rq_budget_token: None,
+        get_rq_budget_token: None,
+        timeout: None,
+        poll: if T::HAS_POLL {
+            Some(Self::poll_callback)
+        } else {
+            None
+        },
+        complete: Some(Self::complete_callback),
+        init_hctx: Some(Self::init_hctx_callback),
+        exit_hctx: Some(Self::exit_hctx_callback),
+        init_request: Some(Self::init_request_callback),
+        exit_request: Some(Self::exit_request_callback),
+        cleanup_rq: None,
+        busy: None,
+        map_queues: None,
+        #[cfg(CONFIG_BLK_DEBUG_FS)]
+        show_rq: None,
+    };
+
+    pub(crate) const fn build() -> &'static bindings::blk_mq_ops {
+        &Self::VTABLE
+    }
+}

linux-7.2.3/rust/kernel/block/mq/request.rs

@@ -0,0 +1,257 @@
+// SPDX-License-Identifier: GPL-2.0
+
+//! This module provides a wrapper for the C `struct request` type.
+//!
+//! C header: [`include/linux/blk-mq.h`](srctree/include/linux/blk-mq.h)
+
+use crate::{
+    bindings,
+    block::mq::Operations,
+    error::Result,
+    sync::{
+        aref::{ARef, AlwaysRefCounted},
+        atomic::Relaxed,
+        Refcount,
+    },
+    types::Opaque,
+};
+use core::{marker::PhantomData, ptr::NonNull};
+
+/// A wrapper around a blk-mq [`struct request`]. This represents an IO request.
+///
+/// # Implementation details
+///
+/// There are four states for a request that the Rust bindings care about:
+///
+/// 1. Request is owned by block layer (refcount 0).
+/// 2. Request is owned by driver but with zero [`ARef`]s in existence
+///    (refcount 1).
+/// 3. Request is owned by driver with exactly one [`ARef`] in existence
+///    (refcount 2).
+/// 4. Request is owned by driver with more than one [`ARef`] in existence
+///    (refcount > 2).
+///
+///
+/// We need to track 1 and 2 to ensure we fail tag to request conversions for
+/// requests that are not owned by the driver.
+///
+/// We need to track 3 and 4 to ensure that it is safe to end the request and hand
+/// back ownership to the block layer.
+///
+/// Note that the driver can still obtain new `ARef` even if there is no `ARef`s in existence by
+/// using `tag_to_rq`, hence the need to distinguish B and C.
+///
+/// The states are tracked through the private `refcount` field of
+/// `RequestDataWrapper`. This structure lives in the private data area of the C
+/// [`struct request`].
+///
+/// # Invariants
+///
+/// * `self.0` is a valid [`struct request`] created by the C portion of the
+///   kernel.
+/// * The private data area associated with this request must be an initialized
+///   and valid `RequestDataWrapper<T>`.
+/// * `self` is reference counted by atomic modification of
+///   `self.wrapper_ref().refcount()`.
+///
+/// [`struct request`]: srctree/include/linux/blk-mq.h
+///
+#[repr(transparent)]
+pub struct Request<T>(Opaque<bindings::request>, PhantomData<T>);
+
+impl<T: Operations> Request<T> {
+    /// Create an [`ARef<Request>`] from a [`struct request`] pointer.
+    ///
+    /// # Safety
+    ///
+    /// * The caller must own a refcount on `ptr` that is transferred to the
+    ///   returned [`ARef`].
+    /// * The type invariants for [`Request`] must hold for the pointee of `ptr`.
+    ///
+    /// [`struct request`]: srctree/include/linux/blk-mq.h
+    pub(crate) unsafe fn aref_from_raw(ptr: *mut bindings::request) -> ARef<Self> {
+        // INVARIANT: By the safety requirements of this function, invariants are upheld.
+        // SAFETY: By the safety requirement of this function, we own a
+        // reference count that we can pass to `ARef`.
+        unsafe { ARef::from_raw(NonNull::new_unchecked(ptr.cast())) }
+    }
+
+    /// Notify the block layer that a request is going to be processed now.
+    ///
+    /// The block layer uses this hook to do proper initializations such as
+    /// starting the timeout timer. It is a requirement that block device
+    /// drivers call this function when starting to process a request.
+    ///
+    /// # Safety
+    ///
+    /// The caller must have exclusive ownership of `self`, that is
+    /// `self.wrapper_ref().refcount() == 2`.
+    pub(crate) unsafe fn start_unchecked(this: &ARef<Self>) {
+        // SAFETY: By type invariant, `self.0` is a valid `struct request` and
+        // we have exclusive access.
+        unsafe { bindings::blk_mq_start_request(this.0.get()) };
+    }
+
+    /// Try to take exclusive ownership of `this` by dropping the refcount to 0.
+    /// This fails if `this` is not the only [`ARef`] pointing to the underlying
+    /// [`Request`].
+    ///
+    /// If the operation is successful, [`Ok`] is returned with a pointer to the
+    /// C [`struct request`]. If the operation fails, `this` is returned in the
+    /// [`Err`] variant.
+    ///
+    /// [`struct request`]: srctree/include/linux/blk-mq.h
+    fn try_set_end(this: ARef<Self>) -> Result<*mut bindings::request, ARef<Self>> {
+        // To hand back the ownership, we need the current refcount to be 2.
+        // Since we can race with `TagSet::tag_to_rq`, this needs to atomically reduce
+        // refcount to 0. `Refcount` does not provide a way to do this, so use the underlying
+        // atomics directly.
+        if let Err(_old) = this
+            .wrapper_ref()
+            .refcount()
+            .as_atomic()
+            .cmpxchg(2, 0, Relaxed)
+        {
+            return Err(this);
+        }
+
+        let request_ptr = this.0.get();
+        core::mem::forget(this);
+
+        Ok(request_ptr)
+    }
+
+    /// Notify the block layer that the request has been completed without errors.
+    ///
+    /// This function will return [`Err`] if `this` is not the only [`ARef`]
+    /// referencing the request.
+    pub fn end_ok(this: ARef<Self>) -> Result<(), ARef<Self>> {
+        let request_ptr = Self::try_set_end(this)?;
+
+        // SAFETY: By type invariant, `this.0` was a valid `struct request`. The
+        // success of the call to `try_set_end` guarantees that there are no
+        // `ARef`s pointing to this request. Therefore it is safe to hand it
+        // back to the block layer.
+        unsafe {
+            bindings::blk_mq_end_request(
+                request_ptr,
+                bindings::BLK_STS_OK as bindings::blk_status_t,
+            )
+        };
+
+        Ok(())
+    }
+
+    /// Complete the request by scheduling `Operations::complete` for
+    /// execution.
+    ///
+    /// The function may be scheduled locally, via SoftIRQ or remotely via IPMI.
+    /// See `blk_mq_complete_request_remote` in [`blk-mq.c`] for details.
+    ///
+    /// [`blk-mq.c`]: srctree/block/blk-mq.c
+    pub fn complete(this: ARef<Self>) {
+        let ptr = ARef::into_raw(this).cast::<bindings::request>().as_ptr();
+        // SAFETY: By type invariant, `self.0` is a valid `struct request`
+        if !unsafe { bindings::blk_mq_complete_request_remote(ptr) } {
+            // SAFETY: We released a refcount above that we can reclaim here.
+            let this = unsafe { Request::aref_from_raw(ptr) };
+            T::complete(this);
+        }
+    }
+
+    /// Return a pointer to the [`RequestDataWrapper`] stored in the private area
+    /// of the request structure.
+    ///
+    /// # Safety
+    ///
+    /// - `this` must point to a valid allocation of size at least size of
+    ///   [`Self`] plus size of [`RequestDataWrapper`].
+    pub(crate) unsafe fn wrapper_ptr(this: *mut Self) -> NonNull<RequestDataWrapper> {
+        let request_ptr = this.cast::<bindings::request>();
+        // SAFETY: By safety requirements for this function, `this` is a
+        // valid allocation.
+        let wrapper_ptr =
+            unsafe { bindings::blk_mq_rq_to_pdu(request_ptr).cast::<RequestDataWrapper>() };
+        // SAFETY: By C API contract, `wrapper_ptr` points to a valid allocation
+        // and is not null.
+        unsafe { NonNull::new_unchecked(wrapper_ptr) }
+    }
+
+    /// Return a reference to the [`RequestDataWrapper`] stored in the private
+    /// area of the request structure.
+    pub(crate) fn wrapper_ref(&self) -> &RequestDataWrapper {
+        // SAFETY: By type invariant, `self.0` is a valid allocation. Further,
+        // the private data associated with this request is initialized and
+        // valid. The existence of `&self` guarantees that the private data is
+        // valid as a shared reference.
+        unsafe { Self::wrapper_ptr(core::ptr::from_ref(self).cast_mut()).as_ref() }
+    }
+}
+
+/// A wrapper around data stored in the private area of the C [`struct request`].
+///
+/// [`struct request`]: srctree/include/linux/blk-mq.h
+pub(crate) struct RequestDataWrapper {
+    /// The Rust request refcount has the following states:
+    ///
+    /// - 0: The request is owned by C block layer.
+    /// - 1: The request is owned by Rust abstractions but there are no [`ARef`] references to it.
+    /// - 2+: There are [`ARef`] references to the request.
+    refcount: Refcount,
+}
+
+impl RequestDataWrapper {
+    /// Return a reference to the refcount of the request that is embedding
+    /// `self`.
+    pub(crate) fn refcount(&self) -> &Refcount {
+        &self.refcount
+    }
+
+    /// Return a pointer to the refcount of the request that is embedding the
+    /// pointee of `this`.
+    ///
+    /// # Safety
+    ///
+    /// - `this` must point to a live allocation of at least the size of `Self`.
+    pub(crate) unsafe fn refcount_ptr(this: *mut Self) -> *mut Refcount {
+        // SAFETY: Because of the safety requirements of this function, the
+        // field projection is safe.
+        unsafe { &raw mut (*this).refcount }
+    }
+}
+
+// SAFETY: Exclusive access is thread-safe for `Request`. `Request` has no `&mut
+// self` methods and `&self` methods that mutate `self` are internally
+// synchronized.
+unsafe impl<T: Operations> Send for Request<T> {}
+
+// SAFETY: Shared access is thread-safe for `Request`. `&self` methods that
+// mutate `self` are internally synchronized`
+unsafe impl<T: Operations> Sync for Request<T> {}
+
+// SAFETY: All instances of `Request<T>` are reference counted. This
+// implementation of `AlwaysRefCounted` ensure that increments to the ref count
+// keeps the object alive in memory at least until a matching reference count
+// decrement is executed.
+unsafe impl<T: Operations> AlwaysRefCounted for Request<T> {
+    fn inc_ref(&self) {
+        self.wrapper_ref().refcount().inc();
+    }
+
+    unsafe fn dec_ref(obj: core::ptr::NonNull<Self>) {
+        // SAFETY: The type invariants of `ARef` guarantee that `obj` is valid
+        // for read.
+        let wrapper_ptr = unsafe { Self::wrapper_ptr(obj.as_ptr()).as_ptr() };
+        // SAFETY: The type invariant of `Request` guarantees that the private
+        // data area is initialized and valid.
+        let refcount = unsafe { &*RequestDataWrapper::refcount_ptr(wrapper_ptr) };
+
+        #[cfg_attr(not(CONFIG_DEBUG_MISC), allow(unused_variables))]
+        let is_zero = refcount.dec_and_test();
+
+        #[cfg(CONFIG_DEBUG_MISC)]
+        if is_zero {
+            panic!("Request reached refcount zero in Rust abstractions");
+        }
+    }
+}

linux-7.2.3/rust/kernel/block/mq/tag_set.rs

@@ -0,0 +1,85 @@
+// SPDX-License-Identifier: GPL-2.0
+
+//! This module provides the `TagSet` struct to wrap the C `struct blk_mq_tag_set`.
+//!
+//! C header: [`include/linux/blk-mq.h`](srctree/include/linux/blk-mq.h)
+
+use core::pin::Pin;
+
+use crate::{
+    bindings,
+    block::mq::{operations::OperationsVTable, request::RequestDataWrapper, Operations},
+    error::{self, Result},
+    prelude::try_pin_init,
+    types::Opaque,
+};
+use core::{convert::TryInto, marker::PhantomData};
+use pin_init::{pin_data, pinned_drop, PinInit};
+
+/// A wrapper for the C `struct blk_mq_tag_set`.
+///
+/// `struct blk_mq_tag_set` contains a `struct list_head` and so must be pinned.
+///
+/// # Invariants
+///
+/// - `inner` is initialized and valid.
+#[pin_data(PinnedDrop)]
+#[repr(transparent)]
+pub struct TagSet<T: Operations> {
+    #[pin]
+    inner: Opaque<bindings::blk_mq_tag_set>,
+    _p: PhantomData<T>,
+}
+
+impl<T: Operations> TagSet<T> {
+    /// Try to create a new tag set
+    pub fn new(
+        nr_hw_queues: u32,
+        num_tags: u32,
+        num_maps: u32,
+    ) -> impl PinInit<Self, error::Error> {
+        let tag_set: bindings::blk_mq_tag_set = pin_init::zeroed();
+        let tag_set: Result<_> = core::mem::size_of::<RequestDataWrapper>()
+            .try_into()
+            .map(|cmd_size| {
+                bindings::blk_mq_tag_set {
+                    ops: OperationsVTable::<T>::build(),
+                    nr_hw_queues,
+                    timeout: 0, // 0 means default which is 30Hz in C
+                    numa_node: bindings::NUMA_NO_NODE,
+                    queue_depth: num_tags,
+                    cmd_size,
+                    flags: 0,
+                    driver_data: core::ptr::null_mut::<crate::ffi::c_void>(),
+                    nr_maps: num_maps,
+                    ..tag_set
+                }
+            })
+            .map(Opaque::new)
+            .map_err(|e| e.into());
+
+        try_pin_init!(TagSet {
+            inner <- tag_set.pin_chain(|tag_set| {
+                // SAFETY: we do not move out of `tag_set`.
+                let tag_set: &mut Opaque<_> = unsafe { Pin::get_unchecked_mut(tag_set) };
+                // SAFETY: `tag_set` is a reference to an initialized `blk_mq_tag_set`.
+                error::to_result( unsafe { bindings::blk_mq_alloc_tag_set(tag_set.get())})
+            }),
+            _p: PhantomData,
+        })
+    }
+
+    /// Return the pointer to the wrapped `struct blk_mq_tag_set`
+    pub(crate) fn raw_tag_set(&self) -> *mut bindings::blk_mq_tag_set {
+        self.inner.get()
+    }
+}
+
+#[pinned_drop]
+impl<T: Operations> PinnedDrop for TagSet<T> {
+    fn drop(self: Pin<&mut Self>) {
+        // SAFETY: By type invariant `inner` is valid and has been properly
+        // initialized during construction.
+        unsafe { bindings::blk_mq_free_tag_set(self.inner.get()) };
+    }
+}