[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <49975111-781B-4DF8-B63B-4E78ACD9BF8B@arinc9.com>
Date: Fri, 17 Dec 2021 13:57:38 +0300
From: Arınç ÜNAL <arinc.unal@...nc9.com>
To: Alvin Šipraga <ALSI@...g-olufsen.dk>
Cc: "luizluca@...il.com" <luizluca@...il.com>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"linus.walleij@...aro.org" <linus.walleij@...aro.org>,
"andrew@...n.ch" <andrew@...n.ch>,
"vivien.didelot@...il.com" <vivien.didelot@...il.com>,
"f.fainelli@...il.com" <f.fainelli@...il.com>,
"olteanv@...il.com" <olteanv@...il.com>
Subject: Re: [PATCH net-next 07/13] net: dsa: rtl8365mb: rename rtl8365mb to rtl8367c
Hey Alvin.
> On 17 Dec 2021, at 02:42, Alvin Šipraga <ALSI@...g-olufsen.dk> wrote:
>
> Hi Luiz,
>
>> On 12/16/21 21:13, luizluca@...il.com wrote:
>> From: Luiz Angelo Daros de Luca <luizluca@...il.com>
>>
>> rtl8365mb refers to a single device supported by the driver.
>> The rtl8367c does not refer to any real device, but it is the
>> driver version name used by Realtek.
>>
>> Tested-by: Arınç ÜNAL <arinc.unal@...nc9.com>
>> Signed-off-by: Luiz Angelo Daros de Luca <luizluca@...il.com>
>> ---
>> drivers/net/dsa/realtek/Kconfig | 9 +-
>> drivers/net/dsa/realtek/Makefile | 2 +-
>> drivers/net/dsa/realtek/realtek-smi.c | 4 +-
>> drivers/net/dsa/realtek/realtek.h | 2 +-
>> drivers/net/dsa/realtek/rtl8367c.c | 1321 ++++++++++++-------------
>> drivers/net/phy/realtek.c | 2 +-
>> 6 files changed, 666 insertions(+), 674 deletions(-)
>>
>
> Is the rename really necessary? My logic in naming it rtl8365mb was
> simply that it was the first hardware to be supported by the driver,
> which was more meaningful than Realtek's fictitious rtl8367c. This seems
> to be common practice in the kernel, and bulk renames don't normally
> bring much value.
>
> I think the vendor's naming scheme is confusing at best, so it's better
> to stick to real things in the kernel.
Here's my 2 cents on this. We do know that rtl8367c is the family name of a bunch of different switch models. Yes, it’s a huge chunk of change with no real benefits but since we’re already necessarily changing function names of other parts of the code (e.g. smi -> priv), I believe changing the subdriver to the most accurate name that represents these switch models along with that would be acceptable and for the better.
Cheers.
Arınç
>
> Alvin
Powered by blists - more mailing lists