public inbox for linux-mm@kvack.org
 help / color / mirror / Atom feed
From: Alice Ryhl <aliceryhl@google.com>
To: Andreas Hindborg <a.hindborg@kernel.org>
Cc: "Boqun Feng" <boqun.feng@gmail.com>,
	"Jens Axboe" <axboe@kernel.dk>, "Miguel Ojeda" <ojeda@kernel.org>,
	"Gary Guo" <gary@garyguo.net>,
	"Björn Roy Baron" <bjorn3_gh@protonmail.com>,
	"Benno Lossin" <lossin@kernel.org>,
	"Trevor Gross" <tmgross@umich.edu>,
	"Danilo Krummrich" <dakr@kernel.org>,
	"FUJITA Tomonori" <fujita.tomonori@gmail.com>,
	"Frederic Weisbecker" <frederic@kernel.org>,
	"Lyude Paul" <lyude@redhat.com>,
	"Thomas Gleixner" <tglx@kernel.org>,
	"Anna-Maria Behnsen" <anna-maria@linutronix.de>,
	"John Stultz" <jstultz@google.com>,
	"Stephen Boyd" <sboyd@kernel.org>,
	"Lorenzo Stoakes" <lorenzo.stoakes@oracle.com>,
	"Liam R. Howlett" <Liam.Howlett@oracle.com>,
	linux-block@vger.kernel.org, rust-for-linux@vger.kernel.org,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH 09/79] block: rust: introduce `kernel::block::bio` module
Date: Mon, 16 Mar 2026 10:29:17 +0000	[thread overview]
Message-ID: <abfbfV6-FXnVT9Ud@google.com> (raw)
In-Reply-To: <20260216-rnull-v6-19-rc5-send-v1-9-de9a7af4b469@kernel.org>

On Mon, Feb 16, 2026 at 12:34:56AM +0100, Andreas Hindborg wrote:
> Add Rust abstractions for working with `struct bio`, the core IO command
> descriptor for the block layer.
> 
> The `Bio` type wraps `struct bio` and provides safe access to the IO
> vector describing the data buffers associated with the IO command. The
> data buffers are represented as a vector of `Segment`s, where each
> segment is a contiguous region of physical memory backed by `Page`.
> 
> The `BioSegmentIterator` provides iteration over segments in a single
> bio, while `BioIterator` allows traversing a chain of bios. The
> `Segment` type offers methods for copying data to and from pages, as
> well as zeroing page contents, which are the fundamental operations
> needed by block device drivers to process IO requests.
> 
> The `Request` type is extended with methods to access the bio chain
> associated with a request, allowing drivers to iterate over all data
> buffers that need to be processed.
> 
> Signed-off-by: Andreas Hindborg <a.hindborg@kernel.org>
> ---
>  rust/helpers/blk.c              |   8 +
>  rust/kernel/block.rs            |   1 +
>  rust/kernel/block/bio.rs        | 143 +++++++++++++++
>  rust/kernel/block/bio/vec.rs    | 389 ++++++++++++++++++++++++++++++++++++++++
>  rust/kernel/block/mq/request.rs |  46 +++++
>  rust/kernel/lib.rs              |   2 +
>  rust/kernel/page.rs             |   2 +-
>  7 files changed, 590 insertions(+), 1 deletion(-)
> 
> diff --git a/rust/helpers/blk.c b/rust/helpers/blk.c
> index cc9f4e6a2d234..53beba8c7782d 100644
> --- a/rust/helpers/blk.c
> +++ b/rust/helpers/blk.c
> @@ -1,5 +1,6 @@
>  // SPDX-License-Identifier: GPL-2.0
>  
> +#include <linux/bio.h>
>  #include <linux/blk-mq.h>
>  #include <linux/blkdev.h>
>  
> @@ -12,3 +13,10 @@ struct request *rust_helper_blk_mq_rq_from_pdu(void *pdu)
>  {
>  	return blk_mq_rq_from_pdu(pdu);
>  }
> +
> +void rust_helper_bio_advance_iter_single(const struct bio *bio,
> +					 struct bvec_iter *iter,
> +					 unsigned int bytes)
> +{
> +	bio_advance_iter_single(bio, iter, bytes);
> +}

__rust_helper

> diff --git a/rust/kernel/block/bio.rs b/rust/kernel/block/bio.rs
> new file mode 100644
> index 0000000000000..94062ea5281e6
> --- /dev/null
> +++ b/rust/kernel/block/bio.rs
> @@ -0,0 +1,143 @@
> +// SPDX-License-Identifier: GPL-2.0
> +
> +//! Types for working with the bio layer.
> +//!
> +//! C header: [`include/linux/blk_types.h`](../../include/linux/blk_types.h)

srctree/

> +/// A block device IO descriptor (`struct bio`).
> +///
> +/// A `Bio` is the main unit of IO for the block layer. It describes an IO command and associated
> +/// data buffers.
> +///
> +/// The data buffers associated with a `Bio` are represented by a vector of [`Segment`]s. These
> +/// segments represent physically contiguous regions of memory. The memory is represented by
> +/// [`Page`] descriptors internally.
> +///
> +/// The vector of [`Segment`]s can be iterated by obtaining a [`SegmentIterator`].
> +///
> +/// # Invariants
> +///
> +/// Instances of this type is always reference counted. A call to
> +/// `bindings::bio_get()` ensures that the instance is valid for read at least
> +/// until a matching call to `bindings :bio_put()`.

Refcounted? None of these methods are called anywhere, and you do not
implement AlwaysRefcounted.

> +#[repr(transparent)]
> +pub struct Bio(Opaque<bindings::bio>);
> +
> +impl Bio {
> +    /// Returns an iterator over segments in this `Bio`. Does not consider
> +    /// segments of other bios in this bio chain.
> +    #[inline(always)]
> +    pub fn segment_iter(&mut self) -> BioSegmentIterator<'_> {

Not `self: Pin<&mut Self>` here?

> +    /// Create an instance of `Bio` from a raw pointer.
> +    ///
> +    /// # Safety
> +    ///
> +    /// Caller must ensure that the `ptr` is valid for use as a reference to
> +    /// `Bio` for the duration of `'a`.
> +    #[inline(always)]
> +    pub(crate) unsafe fn from_raw<'a>(ptr: *mut bindings::bio) -> Option<&'a Self> {
> +        Some(
> +            // SAFETY: by the safety requirement of this funciton, `ptr` is
> +            // valid for read for the duration of the returned lifetime
> +            unsafe { &*NonNull::new(ptr)?.as_ptr().cast::<Bio>() },
> +        )
> +    }
> +
> +    /// Create an instance of `Bio` from a raw pointer.
> +    ///
> +    /// # Safety
> +    ///
> +    /// Caller must ensure that the `ptr` is valid for use as a unique reference
> +    /// to `Bio` for the duration of `'a`.
> +    #[inline(always)]
> +    pub(crate) unsafe fn from_raw_mut<'a>(ptr: *mut bindings::bio) -> Option<&'a mut Self> {
> +        Some(
> +            // SAFETY: by the safety requirement of this funciton, `ptr` is
> +            // valid for read for the duration of the returned lifetime
> +            unsafe { &mut *NonNull::new(ptr)?.as_ptr().cast::<Bio>() },

Why the Option? I imagine every caller has a non-null pointer?

> +        )
> +    }
> +}
> +
> +impl core::fmt::Display for Bio {

We have our own fmt trait now, right?

> +    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
> +        write!(
> +            f,
> +            "Bio({:?}, vcnt: {}, idx: {}, size: 0x{:x}, completed: 0x{:x})",
> +            self.0.get(),
> +            self.io_vec_count(),
> +            self.raw_iter().bi_idx,
> +            self.raw_iter().bi_size,
> +            self.raw_iter().bi_bvec_done

This reads the entire `bi_iter` field three separate times. A local
variable may be a good idea.

> +/// An iterator over `Segment`
> +///
> +/// # Invariants
> +///
> +/// If `iter.bi_size` > 0, `iter` must always index a valid `bio_vec` in `bio.io_vec()`.
> +pub struct BioSegmentIterator<'a> {
> +    bio: &'a mut Bio,
> +    iter: bindings::bvec_iter,
> +}
> +
> +impl<'a> BioSegmentIterator<'a> {
> +    /// Creeate a new segemnt iterator for iterating the segments of `bio`. The

typo

> +impl<'a> core::iter::Iterator for BioSegmentIterator<'a> {
> +    type Item = Segment<'a>;
> +
> +    #[inline(always)]
> +    fn next(&mut self) -> Option<Self::Item> {
> +        if self.iter.bi_size == 0 {
> +            return None;
> +        }
> +
> +        // SAFETY: We checked that `self.iter.bi_size` > 0 above.
> +        let bio_vec_ret = unsafe { self.io_vec() };
> +
> +        // SAFETY: By existence of reference `&bio`, `bio.0` contains a valid
> +        // `struct bio`. By type invariant of `BioSegmentItarator` `self.iter`
> +        // indexes into a valid `bio_vec` entry. By C API contracit, `bv_len`
> +        // does not exceed the size of the bio.
> +        unsafe {
> +            bindings::bio_advance_iter_single(
> +                self.bio.0.get(),
> +                core::ptr::from_mut(&mut self.iter),

Creating this BioSegmentItarator copies the bvec_iter from the Bio, and
then here you modify the copy. Is that the intent? Is the C type such
that copying it is always okay?

Also, is the C type such that moves are ok? It's playsible that the
answer is yes - the same applies in rust/kernel/iov.rs but it could be
clearer in e.g. "Invariants" that this is the case.

Nit: core::ptr::from_mut(&mut self.iter) -> &raw mut self.iter

> +    /// Get a mutable reference to the first [`Bio`] in this request.
> +    #[inline(always)]
> +    pub fn bio_mut(&mut self) -> Option<&mut Bio> {
> +        // SAFETY: By type invariant of `Self`, `self.0` is valid and the deref
> +        // is safe.
> +        let ptr = unsafe { (*self.0.get()).bio };
> +        // SAFETY: By C API contract, if `bio` is not null it will have a
> +        // positive refcount at least for the duration of the lifetime of
> +        // `&self`.
> +        unsafe { Bio::from_raw_mut(ptr) }

Surely &mut requires refcount == 1, not just positive refcount?

Alice


  reply	other threads:[~2026-03-16 10:29 UTC|newest]

Thread overview: 113+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-02-15 23:34 [PATCH 00/79] block: rnull: complete the rust null block driver Andreas Hindborg
2026-02-15 23:34 ` [PATCH 01/79] block: rnull: adopt new formatting guidelines Andreas Hindborg
2026-03-16  9:14   ` Alice Ryhl
2026-02-15 23:34 ` [PATCH 02/79] block: rnull: add module parameters Andreas Hindborg
2026-03-16  9:18   ` Alice Ryhl
2026-02-15 23:34 ` [PATCH 03/79] block: rnull: add macros to define configfs attributes Andreas Hindborg
2026-03-16  9:23   ` Alice Ryhl
2026-02-15 23:34 ` [PATCH 04/79] block: rust: fix generation of bindings to `BLK_STS_.*` Andreas Hindborg
2026-03-16  9:27   ` Alice Ryhl
2026-03-16 11:58     ` Alice Ryhl
2026-02-15 23:34 ` [PATCH 05/79] block: rust: change `queue_rq` request type to `Owned` Andreas Hindborg
2026-03-16 10:01   ` Alice Ryhl
2026-03-23 12:08     ` Andreas Hindborg
2026-03-24 13:27       ` Alice Ryhl
2026-02-15 23:34 ` [PATCH 06/79] block: rust: add `Request` private data support Andreas Hindborg
2026-03-16 10:10   ` Alice Ryhl
2026-02-15 23:34 ` [PATCH 07/79] block: rust: allow `hrtimer::Timer` in `RequestData` Andreas Hindborg
2026-02-15 23:34 ` [PATCH 08/79] block: rnull: add timer completion mode Andreas Hindborg
2026-02-15 23:34 ` [PATCH 09/79] block: rust: introduce `kernel::block::bio` module Andreas Hindborg
2026-03-16 10:29   ` Alice Ryhl [this message]
2026-02-15 23:34 ` [PATCH 10/79] block: rust: add `command` getter to `Request` Andreas Hindborg
2026-03-16 10:30   ` Alice Ryhl
2026-02-15 23:34 ` [PATCH 11/79] block: rust: mq: use GFP_KERNEL from prelude Andreas Hindborg
2026-03-16 10:30   ` Alice Ryhl
2026-02-15 23:34 ` [PATCH 12/79] block: rust: add `TagSet` flags Andreas Hindborg
2026-03-16 10:34   ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 13/79] block: rnull: add memory backing Andreas Hindborg
2026-03-16 10:38   ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 14/79] block: rnull: add submit queue count config option Andreas Hindborg
2026-03-16 10:40   ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 15/79] block: rnull: add `use_per_node_hctx` " Andreas Hindborg
2026-03-16 10:43   ` Alice Ryhl
2026-03-16 11:01     ` Gary Guo
2026-03-16 13:57     ` Andreas Hindborg
2026-03-16 14:02       ` Miguel Ojeda
2026-03-16 14:05       ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 16/79] block: rust: allow specifying home node when constructing `TagSet` Andreas Hindborg
2026-03-16 10:46   ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 17/79] block: rnull: allow specifying the home numa node Andreas Hindborg
2026-02-15 23:35 ` [PATCH 18/79] block: rust: add Request::sectors() method Andreas Hindborg
2026-03-16 10:48   ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 19/79] block: rust: mq: add max_hw_discard_sectors support to GenDiskBuilder Andreas Hindborg
2026-03-16 10:50   ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 20/79] block: rnull: add discard support Andreas Hindborg
2026-03-16 10:52   ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 21/79] block: rust: add `NoDefaultScheduler` flag for `TagSet` Andreas Hindborg
2026-03-16 10:53   ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 22/79] block: rnull: add no_sched module parameter and configfs attribute Andreas Hindborg
2026-03-16 10:54   ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 23/79] block: rust: change sector type from usize to u64 Andreas Hindborg
2026-03-16 10:56   ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 24/79] block: rust: add `BadBlocks` for bad block tracking Andreas Hindborg
2026-03-16 11:55   ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 25/79] block: rust: mq: add Request::end() method for custom status codes Andreas Hindborg
2026-03-16 11:57   ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 26/79] block: rnull: add badblocks support Andreas Hindborg
2026-03-16 12:02   ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 27/79] block: rnull: add badblocks_once support Andreas Hindborg
2026-02-15 23:35 ` [PATCH 28/79] block: rnull: add partial I/O support for bad blocks Andreas Hindborg
2026-03-16 12:14   ` Alice Ryhl
2026-02-15 23:35 ` [PATCH 29/79] block: rust: add `TagSet` private data support Andreas Hindborg
2026-02-15 23:35 ` [PATCH 30/79] block: rust: add `hctx` " Andreas Hindborg
2026-02-15 23:35 ` [PATCH 31/79] block: rnull: add volatile cache emulation Andreas Hindborg
2026-02-15 23:35 ` [PATCH 32/79] block: rust: implement `Sync` for `GenDisk` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 33/79] block: rust: add a back reference feature to `GenDisk` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 34/79] block: rust: introduce an idle type state for `Request` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 35/79] block: rust: add a request queue abstraction Andreas Hindborg
2026-02-15 23:35 ` [PATCH 36/79] block: rust: add a method to get the request queue for a request Andreas Hindborg
2026-02-15 23:35 ` [PATCH 37/79] block: rust: introduce `kernel::block::error` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 38/79] block: rust: require `queue_rq` to return a `BlkResult` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 39/79] block: rust: add `GenDisk::queue_data` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 40/79] block: rnull: add bandwidth limiting Andreas Hindborg
2026-02-15 23:35 ` [PATCH 41/79] block: rnull: add blocking queue mode Andreas Hindborg
2026-02-15 23:35 ` [PATCH 42/79] block: rnull: add shared tags Andreas Hindborg
2026-02-15 23:35 ` [PATCH 43/79] block: rnull: add queue depth config option Andreas Hindborg
2026-02-15 23:35 ` [PATCH 44/79] block: rust: add an abstraction for `bindings::req_op` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 45/79] block: rust: add a method to set the target sector of a request Andreas Hindborg
2026-02-15 23:35 ` [PATCH 46/79] block: rust: move gendisk vtable construction to separate function Andreas Hindborg
2026-02-15 23:35 ` [PATCH 47/79] block: rust: add zoned block device support Andreas Hindborg
2026-02-15 23:35 ` [PATCH 48/79] block: rnull: add zoned storage support Andreas Hindborg
2026-02-15 23:35 ` [PATCH 49/79] block: rust: add `map_queues` support Andreas Hindborg
2026-02-15 23:35 ` [PATCH 50/79] block: rust: add an abstraction for `struct blk_mq_queue_map` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 51/79] block: rust: add polled completion support Andreas Hindborg
2026-02-15 23:35 ` [PATCH 52/79] block: rust: add accessors to `TagSet` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 53/79] block: rnull: add polled completion support Andreas Hindborg
2026-02-15 23:35 ` [PATCH 54/79] block: rnull: add REQ_OP_FLUSH support Andreas Hindborg
2026-02-15 23:35 ` [PATCH 55/79] block: rust: add request flags abstraction Andreas Hindborg
2026-02-15 23:35 ` [PATCH 56/79] block: rust: add abstraction for block queue feature flags Andreas Hindborg
2026-02-15 23:35 ` [PATCH 57/79] block: rust: allow setting write cache and FUA flags for `GenDisk` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 58/79] block: rust: add `Segment::copy_to_page_limit` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 59/79] block: rnull: add fua support Andreas Hindborg
2026-02-15 23:35 ` [PATCH 60/79] block: fix arg type in `blk_mq_update_nr_hw_queues` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 61/79] block: rust: add `GenDisk::tag_set` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 62/79] block: rust: add `TagSet::update_hw_queue_count` Andreas Hindborg
2026-02-16 23:59   ` Ken Kurematsu
2026-02-17  9:54     ` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 63/79] block: rnull: add an option to change the number of hardware queues Andreas Hindborg
2026-02-15 23:35 ` [PATCH 64/79] block: rust: add an abstraction for `struct rq_list` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 65/79] block: rust: add `queue_rqs` vtable hook Andreas Hindborg
2026-02-15 23:35 ` [PATCH 66/79] block: rnull: support queue_rqs Andreas Hindborg
2026-02-15 23:35 ` [PATCH 67/79] block: rust: remove the `is_poll` parameter from `queue_rq` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 68/79] block: rust: add a debug assert for refcounts Andreas Hindborg
2026-02-15 23:35 ` [PATCH 69/79] block: rust: add `TagSet::tag_to_rq` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 70/79] block: rust: add `Request::queue_index` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 71/79] block: rust: add `Request::requeue` Andreas Hindborg
2026-02-15 23:35 ` [PATCH 72/79] block: rust: add `request_timeout` hook Andreas Hindborg
2026-02-15 23:36 ` [PATCH 73/79] block: rnull: add fault injection support Andreas Hindborg
2026-02-15 23:36 ` [PATCH 74/79] block: rust: add max_sectors option to `GenDiskBuilder` Andreas Hindborg
2026-02-15 23:36 ` [PATCH 75/79] block: rnull: allow configuration of the maximum IO size Andreas Hindborg
2026-02-15 23:36 ` [PATCH 76/79] block: rust: add `virt_boundary_mask` option to `GenDiskBuilder` Andreas Hindborg
2026-02-15 23:36 ` [PATCH 77/79] block: rnull: add `virt_boundary` option Andreas Hindborg
2026-02-15 23:36 ` [PATCH 78/79] block: rnull: add `shared_tag_bitmap` config option Andreas Hindborg
2026-02-15 23:36 ` [PATCH 79/79] block: rnull: add zone offline and readonly configfs files Andreas Hindborg

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=abfbfV6-FXnVT9Ud@google.com \
    --to=aliceryhl@google.com \
    --cc=Liam.Howlett@oracle.com \
    --cc=a.hindborg@kernel.org \
    --cc=anna-maria@linutronix.de \
    --cc=axboe@kernel.dk \
    --cc=bjorn3_gh@protonmail.com \
    --cc=boqun.feng@gmail.com \
    --cc=dakr@kernel.org \
    --cc=frederic@kernel.org \
    --cc=fujita.tomonori@gmail.com \
    --cc=gary@garyguo.net \
    --cc=jstultz@google.com \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=lorenzo.stoakes@oracle.com \
    --cc=lossin@kernel.org \
    --cc=lyude@redhat.com \
    --cc=ojeda@kernel.org \
    --cc=rust-for-linux@vger.kernel.org \
    --cc=sboyd@kernel.org \
    --cc=tglx@kernel.org \
    --cc=tmgross@umich.edu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox