linux-kernel - Re: [RFC PATCH v1 1/5] rust: file: add bindings for `struct file`

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <94a50510-c539-49d5-bc36-241f1f003172@gmail.com>
Date:   Tue, 8 Aug 2023 23:59:04 -0300
From:   Martin Rodriguez Reboredo <yakoyoku@...il.com>
To:     Alice Ryhl <aliceryhl@...gle.com>, rust-for-linux@...r.kernel.org,
        linux-fsdevel@...r.kernel.org, Miguel Ojeda <ojeda@...nel.org>,
        Alexander Viro <viro@...iv.linux.org.uk>,
        Christian Brauner <brauner@...nel.org>
Cc:     Wedson Almeida Filho <wedsonaf@...il.com>,
        Alex Gaynor <alex.gaynor@...il.com>,
        Boqun Feng <boqun.feng@...il.com>, Gary Guo <gary@...yguo.net>,
        Björn Roy Baron <bjorn3_gh@...tonmail.com>,
        Benno Lossin <benno.lossin@...ton.me>,
        linux-kernel@...r.kernel.org, patches@...ts.linux.dev,
        Wedson Almeida Filho <walmeida@...rosoft.com>,
        Daniel Xu <dxu@...uu.xyz>
Subject: Re: [RFC PATCH v1 1/5] rust: file: add bindings for `struct file`

On 7/20/23 12:28, Alice Ryhl wrote:
> From: Wedson Almeida Filho <walmeida@...rosoft.com>
> 
> Using these bindings it becomes possible to access files from drivers
> written in Rust. This patch only adds support for accessing the flags,
> and for managing the refcount of the file.
> 
> Signed-off-by: Wedson Almeida Filho <walmeida@...rosoft.com>
> Co-Developed-by: Daniel Xu <dxu@...uu.xyz>
> Signed-off-by: Daniel Xu <dxu@...uu.xyz>
> Co-Developed-by: Alice Ryhl <aliceryhl@...gle.com>
> Signed-off-by: Alice Ryhl <aliceryhl@...gle.com>
> ---
> In this patch, I am defining an error type called `BadFdError`. I'd like
> your thoughts on doing it this way vs just using the normal `Error`
> type.
> 
> Pros:
>   * The type system makes it clear that the function can only fail with
>     EBADF, and that no other errors are possible.
>   * Since the compiler knows that `ARef<Self>` cannot be null and that
>     `BadFdError` has only one possible value, the return type of
>     `File::from_fd` is represented as a pointer with null being an error.
> 
> Cons:
>   * Defining additional error types involves boilerplate.
>   * The return type becomes a tagged union, making it larger than a
>     pointer.

These two are kinda passable, as a `impl_null_ptr_err` macro can be opted
to implement error types from nulls. Also if we consider that
`File::from_fd` isn't going to be called a gorillion times a second or a
recursive call is done, then I'd say that the tagged union won't bring
any other problems, except that a compiler optim is skipped.

>   * The question mark operator will only utilize the `From` trait once,
>     which prevents you from using the question mark operator on
>     `BadFdError` in methods that return some third error type that the
>     kernel `Error` is convertible into.

Although, I want this to be mentioned in the doc of `BadFdError` as this
cannot be overlooked when said usage of `?` is done.

> 
>   rust/bindings/bindings_helper.h |   2 +
>   rust/helpers.c                  |   7 ++
>   rust/kernel/file.rs             | 176 ++++++++++++++++++++++++++++++++
>   rust/kernel/lib.rs              |   1 +
>   4 files changed, 186 insertions(+)
>   create mode 100644 rust/kernel/file.rs
> 
> [...]
> diff --git a/rust/bindings/bindings_helper.h b/rust/bindings/bindings_helper.h
> [...]
> diff --git a/rust/helpers.c b/rust/helpers.c
> [...]

This one is making my head spin, shouldn't they go first?