netdev - Re: [PATCH net-next v1 2/4] rust: net::phy support C45 helpers

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: <903a21b5-38d1-4d7c-8eb0-610b629c9856@proton.me>
Date: Thu, 18 Apr 2024 12:47:01 +0000
From: Benno Lossin <benno.lossin@...ton.me>
To: Andrew Lunn <andrew@...n.ch>
Cc: FUJITA Tomonori <fujita.tomonori@...il.com>, netdev@...r.kernel.org, rust-for-linux@...r.kernel.org, tmgross@...ch.edu
Subject: Re: [PATCH net-next v1 2/4] rust: net::phy support C45 helpers

On 17.04.24 15:34, Andrew Lunn wrote:
>> If the driver shouldn't be concerned with how the access gets handled,
>> why do we even have a naming problem?
> 
> History.
> 
> The current C code does not cleanly separate register spaces from
> access mechanisms.
> 
> C22 register space is simple, you can only access it using C22 bus
> protocol. However C45 register space can be accessed in two ways,
> either using C45 bus protocol, or using C45 over C22. The driver
> should not care, it just wants to read/write a C45 register.  But the
> current core mixes the two concepts of C45 register space and access
> mechanisms. There have been a few attempts to clean this up, but
> nothing landed yet.
> 
> Now this driver is somewhat special. The PHY itself only implements
> one of the two access mechanisms, C45 bus protocol. So this driver
> could side-step this mess and define access functions which go
> straight to C45 bus protocol. However, some day a non-special
> device/driver will come along, and we will need the generic access
> functions, which leave the core to decide on C45 bus protocol or C45
> over C22. Ideally these generic functions should have the natural name
> for accessing C45 registers, and the special case in this driver
> should use a different name.

Thanks for the explanation. What about having the following register
representing types:
- `C22` accesses a C22 register
- `C45` accesses a C45 register using whatever method phylib decides
- `C45Bus` accesses a C45 register over the C45 bus protocol (or
   `C45Direct`)

Or are you opposed to the idea of accessing any type of register via
`dev.read_register::<RegType>(..)`?

-- 
Cheers,
Benno