RE: [PATCH net-next V3 3/4] net: lan743x: Migrate phylib to phylink
From: Ronnie.Kunin
Date: Thu Aug 08 2024 - 16:24:46 EST
> -----Original Message-----
> From: Russell King <linux@xxxxxxxxxxxxxxx>
> Sent: Friday, August 2, 2024 4:37 AM
> To: Ronnie Kunin - C21729 <Ronnie.Kunin@xxxxxxxxxxxxx>
> Cc: Raju Lakkaraju - I30499 <Raju.Lakkaraju@xxxxxxxxxxxxx>; netdev@xxxxxxxxxxxxxxx;
> davem@xxxxxxxxxxxxx; kuba@xxxxxxxxxx; andrew@xxxxxxx; horms@xxxxxxxxxx; hkallweit1@xxxxxxxxx;
> richardcochran@xxxxxxxxx; rdunlap@xxxxxxxxxxxxx; Bryan Whitehead - C21958
> <Bryan.Whitehead@xxxxxxxxxxxxx>; edumazet@xxxxxxxxxx; pabeni@xxxxxxxxxx; linux-
> kernel@xxxxxxxxxxxxxxx; UNGLinuxDriver <UNGLinuxDriver@xxxxxxxxxxxxx>
> Subject: Re: [PATCH net-next V3 3/4] net: lan743x: Migrate phylib to phylink
> ...
> > In the longer term, it would probably make sense for phylink to provide a mechanism where a MAC
> >driver can tell phylink to switch to using a fixed-link with certain parameters.
> ...
> I think it's time to implement the suggestion in the last paragraph rather than using fixed-phy.
>
We looked into an alternate way to migrate our lan743x driver from phylib to phylink continuing to support our existing hardware out in the field, without using the phylib's fixed-phy approach that you opposed to, but without modifying the phylib framework either.
While investigating how to implement it we came across this which Raju borrowed ideas from: https://lore.kernel.org/linux-arm-kernel/YtGPO5SkMZfN8b%2Fs@xxxxxxxxxxxxxxxxxxxxx/ . He is in the process of testing/cleaning it up and expects to submit it early next week.