From: Andreas Hindborg <a.hindborg@kernel.org>
To: "Benno Lossin" <lossin@kernel.org>
Cc: "Oliver Mangold" <oliver.mangold@pm.me>,
"Miguel Ojeda" <ojeda@kernel.org>,
"Alex Gaynor" <alex.gaynor@gmail.com>,
"Boqun Feng" <boqun.feng@gmail.com>,
"Gary Guo" <gary@garyguo.net>,
"Björn Roy Baron" <bjorn3_gh@protonmail.com>,
"Alice Ryhl" <aliceryhl@google.com>,
"Trevor Gross" <tmgross@umich.edu>,
"Asahi Lina" <lina+kernel@asahilina.net>,
rust-for-linux@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v11 4/4] rust: Add `OwnableRefCounted`
Date: Mon, 07 Jul 2025 15:21:44 +0200 [thread overview]
Message-ID: <87zfdgp253.fsf@kernel.org> (raw)
In-Reply-To: <DB5SRE5S7AR3.QUF369W9JYGJ@kernel.org> (Benno Lossin's message of "Mon, 07 Jul 2025 13:47:10 +0200")
"Benno Lossin" <lossin@kernel.org> writes:
> On Mon Jul 7, 2025 at 1:12 PM CEST, Andreas Hindborg wrote:
>> "Benno Lossin" <lossin@kernel.org> writes:
>>> Ah right, I forgot about this. What was the refcount characteristics of
>>> this again?
>>>
>>> * 1 = in flight, owned by C
>>> * 2 = in flight, owned by Rust
>>> * >2 = in flight, owned by Rust + additional references used by Rust
>>> code
>>>
>>> Correct? Maybe @Andreas can check.
>>
>> We have been a bit back and forth on this. This is how we would like it
>> going forward:
>>
>>
>> /// There are three states for a request that the Rust bindings care about:
>> ///
>> /// - 0: The request is owned by C block layer or is uniquely referenced (by [`Owned<_>`]).
>> /// - 1: The request is owned by Rust abstractions but is not referenced.
>> /// - 2+: There is one or more [`ARef`] instances referencing the request.
>
> Huh, now I'm more confused... Could you go into the details again?
Well, there is not much to it. We found out we can alias "unique" and
"owned by C".
We initialize the refcount to 0 when we initialize the request
structure. This happens at queue creation time.
When C block layer hands over a request for processing to a Rust driver,
we `debug_assert!` that the refcount is 0. We unsafely invent an
`Owned<Request<_>>` and pass that to the driver.
The driver has the option of `into_shared` to obtain an
`ARef<Request<_>>`. We use this for remote completion and timer
completion in rnull.
In most drivers, when the driver hands off the request to the hardware,
the driver will stop accounting for the request. The `Owned<Request<_>>`
is consumed when issuing to the driver, or the driver could simply drop
it. Refcount goes to 1. An ID is passed along to hardware.
When a completion comes back from hardware, it carries the ID. We use
the C block layer `tag_to_rq` machinery to turn this ID back into an
`Owned<Request<_>>`. In that process, we check that `refcount == 1` and
if so, we set refcount to 0 and invent an `Owned<Request<_>>`.
There was simply no need to have two separate values for "owned by C"
and "owned by Rust, not referenced".
Best regards,
Andreas Hindborg
next prev parent reply other threads:[~2025-07-07 13:21 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <OYpTDi4YYXiWvLG3nO_8_WKsgOl9KOpun9l3a34m0jza6nmEWDCLTldSwCfZ2PRRprjXqGmrgSL2JN8rPOQH8Q==@protonmail.internalid>
2025-06-18 12:27 ` [PATCH v11 0/4] New trait OwnableRefCounted for ARef<->Owned conversion Oliver Mangold
2025-06-18 12:27 ` [PATCH v11 1/4] rust: types: Add Ownable/Owned types Oliver Mangold
2025-07-02 11:03 ` Benno Lossin
2025-07-07 6:58 ` Oliver Mangold
2025-07-07 9:23 ` Benno Lossin
2025-07-08 9:56 ` Oliver Mangold
2025-07-08 10:16 ` Miguel Ojeda
2025-07-08 13:06 ` Benno Lossin
2025-07-08 18:30 ` Andreas Hindborg
2025-07-08 19:18 ` Benno Lossin
2025-07-09 8:53 ` Andreas Hindborg
2025-07-09 9:11 ` Benno Lossin
2025-07-08 13:22 ` Andreas Hindborg
2025-07-08 14:53 ` Benno Lossin
2025-07-08 15:00 ` Benno Lossin
2025-07-07 12:26 ` Miguel Ojeda
2025-08-18 12:46 ` Andreas Hindborg
2025-08-18 13:04 ` Oliver Mangold
2025-08-18 22:27 ` Benno Lossin
2025-08-19 6:04 ` Oliver Mangold
2025-08-19 8:26 ` Benno Lossin
2025-08-19 8:45 ` Oliver Mangold
2025-08-19 9:00 ` Andreas Hindborg
2025-08-19 17:15 ` Benno Lossin
2025-08-20 10:48 ` Andreas Hindborg
2025-08-19 8:53 ` Andreas Hindborg
2025-08-19 17:13 ` Benno Lossin
2025-08-19 18:28 ` Andreas Hindborg
2025-08-20 6:02 ` Oliver Mangold
2025-08-20 7:41 ` Benno Lossin
2025-08-20 7:43 ` Oliver Mangold
2025-08-20 10:51 ` Andreas Hindborg
2025-06-18 12:27 ` [PATCH v11 2/4] rust: Split `AlwaysRefCounted` into two traits Oliver Mangold
2025-06-19 3:15 ` kernel test robot
2025-07-02 11:23 ` Benno Lossin
2025-07-07 7:42 ` Oliver Mangold
2025-07-07 9:27 ` Benno Lossin
2025-06-18 12:27 ` [PATCH v11 3/4] rust: Add missing SAFETY documentation for `ARef` example Oliver Mangold
2025-06-18 12:27 ` [PATCH v11 4/4] rust: Add `OwnableRefCounted` Oliver Mangold
2025-07-02 13:24 ` Benno Lossin
2025-07-07 8:07 ` Oliver Mangold
2025-07-07 9:33 ` Benno Lossin
2025-07-07 11:12 ` Andreas Hindborg
2025-07-07 11:47 ` Benno Lossin
2025-07-07 13:21 ` Andreas Hindborg [this message]
2025-07-07 15:39 ` Benno Lossin
2025-07-08 13:15 ` Andreas Hindborg
2025-07-08 14:50 ` Benno Lossin
2025-07-08 15:35 ` Andreas Hindborg
2025-07-08 9:36 ` Oliver Mangold
2025-07-08 13:42 ` Benno Lossin
2025-08-05 17:23 ` [PATCH v11 0/4] New trait OwnableRefCounted for ARef<->Owned conversion Danilo Krummrich
2025-08-06 5:56 ` Oliver Mangold
2025-08-15 10:12 ` Andreas Hindborg
2025-08-18 5:59 ` Oliver Mangold
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=87zfdgp253.fsf@kernel.org \
--to=a.hindborg@kernel.org \
--cc=alex.gaynor@gmail.com \
--cc=aliceryhl@google.com \
--cc=bjorn3_gh@protonmail.com \
--cc=boqun.feng@gmail.com \
--cc=gary@garyguo.net \
--cc=lina+kernel@asahilina.net \
--cc=linux-kernel@vger.kernel.org \
--cc=lossin@kernel.org \
--cc=ojeda@kernel.org \
--cc=oliver.mangold@pm.me \
--cc=rust-for-linux@vger.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;
as well as URLs for NNTP newsgroup(s).