[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20231120134818.e2k673xsjec5scy5@skbuf>
Date: Mon, 20 Nov 2023 15:48:18 +0200
From: Vladimir Oltean <olteanv@...il.com>
To: Krzysztof Kozlowski <krzk@...nel.org>
Cc: Luiz Angelo Daros de Luca <luizluca@...il.com>, netdev@...r.kernel.org,
linus.walleij@...aro.org, alsi@...g-olufsen.dk, andrew@...n.ch,
f.fainelli@...il.com, davem@...emloft.net, edumazet@...gle.com,
kuba@...nel.org, pabeni@...hat.com, arinc.unal@...nc9.com
Subject: Re: [net-next 2/2] net: dsa: realtek: load switch variants on demand
Hi Krzysztof,
On Mon, Nov 20, 2023 at 10:20:13AM +0100, Krzysztof Kozlowski wrote:
> No, why do you need it? You should not need MODULE_ALIAS() in normal
> cases. If you need it, usually it means your device ID table is wrong
> (e.g. misses either entries or MODULE_DEVICE_TABLE()). MODULE_ALIAS() is
> not a substitute for incomplete ID table.
>
> Entire abstraction/macro is pointless and make the code less readable.
Are you saying that the line
MODULE_DEVICE_TABLE(of, realtek_common_of_match);
should be put in all of realtek-mdio.c, realtek-smi.c, rtl8365mb.c and
rtl8366rb.c, but not in realtek-common.c?
There are 5 kernel modules involved, 2 for interfaces and 2 for switches.
Even if the same OF device ID table could be used to load multiple
modules, I'm not sure
(a) how to avoid loading the interface driver which will not be used
(SMI if it's a MDIO-connected switch, or MDIO if it's an SMI
connected switch)
(b) how to ensure that the drivers are loaded in the right order, i.e.
the switch drivers are loaded before the interface drivers
Powered by blists - more mailing lists