rust-for-linux.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: FUJITA Tomonori <fujita.tomonori@gmail.com>
To: rust-for-linux@vger.kernel.org
Cc: andrew@lunn.ch, tmgross@umich.edu, miguel.ojeda.sandonis@gmail.com
Subject: [RFC PATCH v2 0/3] Rust abstractions for network PHY drivers
Date: Sun, 24 Sep 2023 15:48:59 +0900	[thread overview]
Message-ID: <20230924064902.1339662-1-fujita.tomonori@gmail.com> (raw)

This patchset adds Rust abstractions for network PHY drivers. It
doesn't fully cover the C APIs for PHY drivers but I think that it's
already useful. I implement two PHY drivers (Asix AX88772A PHYs and
Realtek Generic FE-GE). Seems they work well with real hardware.

Thanks for lots of feedback on v1! I've tried to address them but
sorry if I missed some.

Here are major changes from v1:

- uses rustified enum for C's phy_state enum

This guarantees that C and Rust never diverge. If the C code is
changed without updating the Rust code, compiling the Rust code fails.

Note that C allows using a value that isn't defined in phy_state enum
for phy_state. It leads to undefined behaviour in Rust side. But this
enum is used as state machine in the C side. Such should not
happen. If it happens, it's a bug on the C side.

- adds the Asix driver to drivers/net/phy/ instead of samples/rust/

I've implemented all the features of drivers/net/phy/ax88796b.c. The
features of C and Rust versions are equivalent. You can choose either
on kernel configuration. I suppose that how to handle drivers in Rust
will be discussed in KS or somewhere. Until then, you can use this
to make reviewing the abstractions easier.

- uses static allocation for phy_driver

v1 dynamically allocates memory for phy_driver. v2 uses static
allocation; the registration of drivers is much simpler. I added a
macro that handles registering drivers, building the device id table,
etc.

- cleaning up the interface for PHYLIB

Uses u16 for register and value, keeps genphy_ function names,
dropping lockless_ functions without MDIO bus locked, etc. u8 could be
used for C22 registers but I like to use the same type for C22 and C45.


v1:
https://lwn.net/ml/rust-for-linux/20230913133609.1668758-1-fujita.tomonori@gmail.com/

FUJITA Tomonori (3):
  rust: core abstractions for network PHY drivers
  MAINTAINERS: add Rust PHY abstractions file to the ETHERNET PHY
    LIBRARY
  net: phy: add Rust Asix PHY driver

 MAINTAINERS                      |   2 +
 drivers/net/phy/Kconfig          |  16 +
 drivers/net/phy/Makefile         |   8 +-
 drivers/net/phy/ax88796b_rust.rs | 147 ++++++
 rust/Makefile                    |   1 +
 rust/bindings/bindings_helper.h  |   3 +
 rust/kernel/lib.rs               |   3 +
 rust/kernel/net.rs               |   6 +
 rust/kernel/net/phy.rs           | 760 +++++++++++++++++++++++++++++++
 rust/uapi/uapi_helper.h          |   1 +
 10 files changed, 946 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/phy/ax88796b_rust.rs
 create mode 100644 rust/kernel/net.rs
 create mode 100644 rust/kernel/net/phy.rs


base-commit: 0bb80ecc33a8fb5a682236443c1e740d5c917d1d
-- 
2.34.1


             reply	other threads:[~2023-09-24  6:49 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-09-24  6:48 FUJITA Tomonori [this message]
2023-09-24  6:49 ` [RFC PATCH v2 1/3] rust: core abstractions for network PHY drivers FUJITA Tomonori
2023-09-24 12:56   ` Wedson Almeida Filho
2023-09-24 13:39     ` Benno Lossin
2023-09-24 13:51       ` Wedson Almeida Filho
2023-09-25 11:30       ` FUJITA Tomonori
2023-09-25 13:14         ` Andrew Lunn
2023-09-24 15:42     ` Andrew Lunn
2023-09-25  1:13       ` FUJITA Tomonori
2023-09-25 13:25         ` Andrew Lunn
2023-09-25  6:47     ` Alice Ryhl
2023-09-26  1:19     ` FUJITA Tomonori
2023-09-26  2:50       ` Andrew Lunn
2023-09-24 13:19   ` Benno Lossin
2023-09-25 10:24     ` FUJITA Tomonori
2023-09-25 15:41       ` Miguel Ojeda
2023-09-26 13:46         ` FUJITA Tomonori
2023-09-27 10:49           ` Miguel Ojeda
2023-09-27 11:19             ` FUJITA Tomonori
2023-10-09 12:28               ` Miguel Ojeda
2023-09-24 17:00   ` Andrew Lunn
2023-09-24 18:03     ` Miguel Ojeda
2023-09-25 13:28       ` Andrew Lunn
2023-09-25 13:43       ` Andrew Lunn
2023-09-25 15:42         ` Miguel Ojeda
2023-09-25 16:53           ` Andrew Lunn
2023-09-25 17:26             ` Miguel Ojeda
2023-09-25 18:32   ` Andrew Lunn
2023-09-25 19:15     ` Miguel Ojeda
2023-09-26  6:05   ` Trevor Gross
2023-09-26 12:11     ` Andrew Lunn
2023-09-27  3:26     ` FUJITA Tomonori
2023-09-26  6:54   ` Trevor Gross
2023-09-27  3:39     ` FUJITA Tomonori
2023-09-27 12:21       ` Andrew Lunn
2023-09-24  6:49 ` [RFC PATCH v2 2/3] MAINTAINERS: add Rust PHY abstractions file to the ETHERNET PHY LIBRARY FUJITA Tomonori
2023-09-24  6:49 ` [RFC PATCH v2 3/3] net: phy: add Rust Asix PHY driver FUJITA Tomonori
2023-09-24  8:05   ` Miguel Ojeda
2023-09-24  9:38     ` FUJITA Tomonori
2023-09-24 10:10       ` Miguel Ojeda
2023-09-24 11:00         ` FUJITA Tomonori
2023-09-24 13:33   ` Benno Lossin
2023-09-25  2:31     ` FUJITA Tomonori
2023-09-26  6:20   ` Trevor Gross
2023-09-26  7:07     ` FUJITA Tomonori
     [not found]       ` <CALNs47uYnQC+AXbJuk8d5506D25SDhZ-ZKuhimFkZnYOhhdfCg@mail.gmail.com>
2023-09-26 12:36         ` Andrew Lunn
2023-09-27  1:18         ` FUJITA Tomonori

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=20230924064902.1339662-1-fujita.tomonori@gmail.com \
    --to=fujita.tomonori@gmail.com \
    --cc=andrew@lunn.ch \
    --cc=miguel.ojeda.sandonis@gmail.com \
    --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).