From: Danilo Krummrich <dakr@kernel.org>
To: Alice Ryhl <aliceryhl@google.com>
Cc: ojeda@kernel.org, alex.gaynor@gmail.com, wedsonaf@gmail.com,
boqun.feng@gmail.com, gary@garyguo.net, bjorn3_gh@protonmail.com,
benno.lossin@proton.me, a.hindborg@samsung.com,
akpm@linux-foundation.org, daniel.almeida@collabora.com,
faith.ekstrand@collabora.com, boris.brezillon@collabora.com,
lina@asahilina.net, mcanal@igalia.com, zhiw@nvidia.com,
acurrid@nvidia.com, cjia@nvidia.com, jhubbard@nvidia.com,
airlied@redhat.com, ajanulgu@redhat.com, lyude@redhat.com,
linux-kernel@vger.kernel.org, rust-for-linux@vger.kernel.org,
linux-mm@kvack.org
Subject: Re: [PATCH v3 01/25] rust: alloc: add `Allocator` trait
Date: Thu, 1 Aug 2024 14:26:46 +0200 [thread overview]
Message-ID: <Zqt_BuyoUk12m61Q@pollux> (raw)
In-Reply-To: <CAH5fLgj5xf4QdDU7yWrqUjcmJw4Rqe0-UzesHLhfQRvPqAv8eA@mail.gmail.com>
On Thu, Aug 01, 2024 at 10:19:41AM +0200, Alice Ryhl wrote:
> On Thu, Aug 1, 2024 at 2:07 AM Danilo Krummrich <dakr@kernel.org> wrote:
> >
> > Add a kernel specific `Allocator` trait, that in contrast to the one in
> > Rust's core library doesn't require unstable features and supports GFP
> > flags.
> >
> > Subsequent patches add the following trait implementors: `Kmalloc`,
> > `Vmalloc` and `KVmalloc`.
> >
> > Signed-off-by: Danilo Krummrich <dakr@kernel.org>
>
> It could be nice to mention that the design without `self` parameters
> is necessary for compatibility with #[derive(SmartPointer)].
Agreed, will do.
>
> > rust/kernel/alloc.rs | 73 ++++++++++++++++++++++++++++++++++++++++++++
> > 1 file changed, 73 insertions(+)
> >
> > diff --git a/rust/kernel/alloc.rs b/rust/kernel/alloc.rs
> > index 1966bd407017..b79dd2c49277 100644
> > --- a/rust/kernel/alloc.rs
> > +++ b/rust/kernel/alloc.rs
> > @@ -11,6 +11,7 @@
> > /// Indicates an allocation error.
> > #[derive(Copy, Clone, PartialEq, Eq, Debug)]
> > pub struct AllocError;
> > +use core::{alloc::Layout, ptr::NonNull};
> >
> > /// Flags to be used when allocating memory.
> > ///
> > @@ -86,3 +87,75 @@ pub mod flags {
> > /// small allocations.
> > pub const GFP_NOWAIT: Flags = Flags(bindings::GFP_NOWAIT);
> > }
> > +
> > +/// The kernel's [`Allocator`] trait.
> > +///
> > +/// An implementation of [`Allocator`] can allocate, re-allocate and free memory buffer described
> > +/// via [`Layout`].
> > +///
> > +/// [`Allocator`] is designed to be implemented as a ZST; [`Allocator`] functions do not operate on
> > +/// an object instance.
> > +///
> > +/// # Safety
> > +///
> > +/// Memory returned from an allocator must point to a valid memory buffer and remain valid until
> > +/// it is explicitly freed.
> > +///
> > +/// Any pointer to a memory buffer which is currently allocated must be valid to be passed to any
> > +/// other [`Allocator`] function. The same applies for a NULL pointer.
> > +///
> > +/// If `realloc` is called with:
> > +/// - a size of zero, the given memory allocation, if any, must be freed
> > +/// - a NULL pointer, a new memory allocation must be created
> > +pub unsafe trait Allocator {
> > + /// Allocate memory based on `layout` and `flags`.
> > + ///
> > + /// On success, returns a buffer represented as `NonNull<[u8]>` that satisfies the layout
> > + /// constraints (i.e. minimum size and alignment as specified by `layout`).
> > + ///
> > + /// This function is equivalent to `realloc` when called with a NULL pointer.
> > + fn alloc(layout: Layout, flags: Flags) -> Result<NonNull<[u8]>, AllocError> {
> > + // SAFETY: Passing a NULL pointer to `realloc` is valid by it's safety requirements and asks
> > + // for a new memory allocation.
> > + unsafe { Self::realloc(None, layout, flags) }
> > + }
> > +
> > + /// Re-allocate an existing memory allocation to satisfy the requested `layout`. If the
> > + /// requested size is zero, `realloc` behaves equivalent to `free`.
> > + ///
> > + /// If the requested size is larger than the size of the existing allocation, a successful call
> > + /// to `realloc` guarantees that the new or grown buffer has at least `Layout::size` bytes, but
> > + /// may also be larger.
> > + ///
> > + /// If the requested size is smaller than the size of the existing allocation, `realloc` may or
> > + /// may not shrink the buffer; this is implementation specific to the allocator.
> > + ///
> > + /// On allocation failure, the existing buffer, if any, remains valid.
> > + ///
> > + /// The buffer is represented as `NonNull<[u8]>`.
> > + ///
> > + /// # Safety
> > + ///
> > + /// `ptr` must point to an existing and valid memory allocation created by this allocator
> > + /// instance.
> > + ///
> > + /// Additionally, `ptr` is allowed to be a NULL pointer; in this case a new memory allocation is
> > + /// created.
> > + unsafe fn realloc(
> > + ptr: Option<NonNull<u8>>,
> > + layout: Layout,
> > + flags: Flags,
> > + ) -> Result<NonNull<[u8]>, AllocError>;
>
> Is it intentional that this allows you to change the alignment of an
> allocation? If so, that could use a note in the docs.
Yes, it's intentional and yes it really misses a note in the documentation.
The idea is to allow a change of alignment as long as the new alignment is
smaller than the old alignment.
In terms of safety, it is the callers responsibility to ensure constant
alignment throughout re-allocations (if required).
>
> > + /// Free an existing memory allocation.
> > + ///
> > + /// # Safety
> > + ///
> > + /// `ptr` must point to an existing and valid memory allocation created by this `Allocator`
> > + /// instance.
> > + unsafe fn free(ptr: NonNull<u8>) {
> > + // SAFETY: `ptr` is guaranteed to be previously allocated with this `Allocator` or NULL.
> > + // Calling `realloc` with a buffer size of zero, frees the buffer `ptr` points to.
> > + let _ = unsafe { Self::realloc(Some(ptr), Layout::new::<()>(), Flags(0)) };
> > + }
>
> At the very least, the provided implementation of `free` changes the
> alignment when it calls `realloc`.
Yes, I think that's fine though. Hopefully no one attempts to use the memory
anymore once `free` is being called.
>
> Alice
>
next prev parent reply other threads:[~2024-08-01 12:27 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-08-01 0:01 [PATCH v3 00/25] Generic `Allocator` support for Rust Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 01/25] rust: alloc: add `Allocator` trait Danilo Krummrich
2024-08-01 8:19 ` Alice Ryhl
2024-08-01 12:26 ` Danilo Krummrich [this message]
2024-08-01 14:25 ` Alice Ryhl
2024-08-01 15:09 ` Danilo Krummrich
2024-08-04 6:21 ` Boqun Feng
2024-08-04 12:29 ` Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 02/25] rust: alloc: separate `aligned_size` from `krealloc_aligned` Danilo Krummrich
2024-08-01 8:21 ` Alice Ryhl
2024-08-01 0:02 ` [PATCH v3 03/25] rust: alloc: rename `KernelAllocator` to `Kmalloc` Danilo Krummrich
2024-08-01 8:21 ` Alice Ryhl
2024-08-01 0:02 ` [PATCH v3 04/25] rust: alloc: implement `Allocator` for `Kmalloc` Danilo Krummrich
2024-08-01 8:28 ` Alice Ryhl
2024-08-01 12:30 ` Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 05/25] rust: alloc: add module `allocator_test` Danilo Krummrich
2024-08-01 8:41 ` Alice Ryhl
2024-08-01 0:02 ` [PATCH v3 06/25] rust: alloc: implement `Vmalloc` allocator Danilo Krummrich
2024-08-01 8:43 ` Alice Ryhl
2024-08-04 6:44 ` Boqun Feng
2024-08-04 12:41 ` Danilo Krummrich
2024-08-04 15:16 ` Danilo Krummrich
2024-08-04 17:39 ` Danilo Krummrich
2024-08-04 23:57 ` Boqun Feng
2024-08-05 0:54 ` Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 07/25] rust: alloc: implement `KVmalloc` allocator Danilo Krummrich
2024-08-01 8:43 ` Alice Ryhl
2024-08-01 12:31 ` Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 08/25] rust: types: implement `Unique<T>` Danilo Krummrich
2024-08-04 6:54 ` Boqun Feng
2024-08-01 0:02 ` [PATCH v3 09/25] rust: alloc: implement kernel `Box` Danilo Krummrich
2024-08-01 8:55 ` Alice Ryhl
2024-08-01 12:45 ` Danilo Krummrich
2024-08-01 12:48 ` Alice Ryhl
2024-08-01 0:02 ` [PATCH v3 10/25] rust: treewide: switch to our kernel `Box` type Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 11/25] rust: alloc: remove `BoxExt` extension Danilo Krummrich
2024-08-01 14:53 ` Alice Ryhl
2024-08-01 0:02 ` [PATCH v3 12/25] rust: alloc: add `Box` to prelude Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 13/25] rust: alloc: import kernel `Box` type in types.rs Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 14/25] rust: alloc: import kernel `Box` type in init.rs Danilo Krummrich
2024-08-01 14:55 ` Alice Ryhl
2024-08-01 0:02 ` [PATCH v3 15/25] rust: alloc: implement kernel `Vec` type Danilo Krummrich
2024-08-01 15:05 ` Alice Ryhl
2024-08-01 15:27 ` Danilo Krummrich
2024-08-01 15:31 ` Alice Ryhl
2024-08-01 15:46 ` Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 16/25] rust: alloc: implement `IntoIterator` for `Vec` Danilo Krummrich
2024-08-01 15:07 ` Alice Ryhl
2024-08-01 15:30 ` Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 17/25] rust: alloc: implement `collect` for `IntoIter` Danilo Krummrich
2024-08-01 15:10 ` Alice Ryhl
2024-08-01 15:37 ` Danilo Krummrich
2024-08-02 7:08 ` Alice Ryhl
2024-08-02 12:02 ` Danilo Krummrich
2024-08-02 12:08 ` Alice Ryhl
2024-08-01 0:02 ` [PATCH v3 18/25] rust: treewide: switch to the kernel `Vec` type Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 19/25] rust: alloc: remove `VecExt` extension Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 20/25] rust: alloc: add `Vec` to prelude Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 21/25] rust: alloc: remove `GlobalAlloc` and `krealloc_aligned` Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 22/25] rust: error: use `core::alloc::LayoutError` Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 23/25] rust: str: test: replace `alloc::format` Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 24/25] rust: alloc: update module comment of alloc.rs Danilo Krummrich
2024-08-01 0:02 ` [PATCH v3 25/25] kbuild: rust: remove the `alloc` crate Danilo Krummrich
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=Zqt_BuyoUk12m61Q@pollux \
--to=dakr@kernel.org \
--cc=a.hindborg@samsung.com \
--cc=acurrid@nvidia.com \
--cc=airlied@redhat.com \
--cc=ajanulgu@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=alex.gaynor@gmail.com \
--cc=aliceryhl@google.com \
--cc=benno.lossin@proton.me \
--cc=bjorn3_gh@protonmail.com \
--cc=boqun.feng@gmail.com \
--cc=boris.brezillon@collabora.com \
--cc=cjia@nvidia.com \
--cc=daniel.almeida@collabora.com \
--cc=faith.ekstrand@collabora.com \
--cc=gary@garyguo.net \
--cc=jhubbard@nvidia.com \
--cc=lina@asahilina.net \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lyude@redhat.com \
--cc=mcanal@igalia.com \
--cc=ojeda@kernel.org \
--cc=rust-for-linux@vger.kernel.org \
--cc=wedsonaf@gmail.com \
--cc=zhiw@nvidia.com \
/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;
as well as URLs for NNTP newsgroup(s).