netdev - Re: [PATCH net-next v1 3/4] rust: net::phy support Firmware API

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]

Date: Thu, 18 Apr 2024 22:10:28 +0900 (JST)
From: FUJITA Tomonori <fujita.tomonori@...il.com>
To: dakr@...hat.com
Cc: fujita.tomonori@...il.com, gregkh@...uxfoundation.org, andrew@...n.ch,
 rust-for-linux@...r.kernel.org, tmgross@...ch.edu, mcgrof@...nel.org,
 netdev@...r.kernel.org, russ.weight@...ux.dev, wedsonaf@...il.com
Subject: Re: [PATCH net-next v1 3/4] rust: net::phy support Firmware API

Hi,

On Mon, 15 Apr 2024 17:45:46 +0200
Danilo Krummrich <dakr@...hat.com> wrote:

> On 4/15/24 12:47, FUJITA Tomonori wrote:
>> This patch adds support to the following basic Firmware API:
>> - request_firmware
>> - release_firmware
>> Signed-off-by: FUJITA Tomonori <fujita.tomonori@...il.com>
>> CC: Luis Chamberlain <mcgrof@...nel.org>
>> CC: Russ Weight <russ.weight@...ux.dev>
>> ---
>>   drivers/net/phy/Kconfig         |  1 +
>>   rust/bindings/bindings_helper.h |  1 +
>>   rust/kernel/net/phy.rs          | 45 +++++++++++++++++++++++++++++++++
>>   3 files changed, 47 insertions(+)
> 
> As Greg already mentioned, this shouldn't be implemented specifically
> for struct
> phy_device, but rather for a generic struct device.

Yeah, I have a version of creating rust/kernel/firmware.rs locally but
I wanted to know if a temporary solution could be accepted.


> In order to use them from your PHY driver, I think all you need to do
> is to implement
> AsRef<> for your phy::Device:
> 
> impl AsRef<device::Device> for Device {
>     fn as_ref(&self) -> &device::Device {
>         // SAFETY: By the type invariants, we know that `self.ptr` is non-null
>         and valid.
>         unsafe { device::Device::from_raw(&mut (*self.ptr).mdio.dev) }
>     }
> }

My implementation uses RawDevice trait in old rust branch (Wedson
implemented, I suppose):

https://github.com/Rust-for-Linux/linux/blob/18b7491480025420896e0c8b73c98475c3806c6f/rust/kernel/device.rs#L37

pub unsafe trait RawDevice {
    /// Returns the raw `struct device` related to `self`.
    fn raw_device(&self) -> *mut bindings::device;


Which is better?