RE: [PATCH v3 net] net: phy: micrel: Dynamically control external clock of KSZ PHY

[Wei Fang]

> -----Original Message-----
> From: Jakub Kicinski <kuba@xxxxxxxxxx>
> Sent: 2024年12月11日 9:59
> To: Wei Fang <wei.fang@xxxxxxx>
> Cc: andrew@xxxxxxx; hkallweit1@xxxxxxxxx; linux@xxxxxxxxxxxxxxx;
> davem@xxxxxxxxxxxxx; edumazet@xxxxxxxxxx; pabeni@xxxxxxxxxx;
> florian.fainelli@xxxxxxxxxxxx; heiko.stuebner@xxxxxxxxx; fank.li@xxxxxxx;
> netdev@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; imx@xxxxxxxxxxxxxxx
> Subject: Re: [PATCH v3 net] net: phy: micrel: Dynamically control external clock
> of KSZ PHY
> 
> On Wed, 11 Dec 2024 01:49:50 +0000 Wei Fang wrote:
> > > I may be missing something but if you don't need to disable the
> > > generic clock you can put the disable into the if () block for rmii-ref ?
> >
> > For my case, it's fine to disable rmii-ref because this clock source
> > is always enabled in FEC driver. But the commit 99ac4cbcc2a5 ("net:
> > phy: micrel: allow usage of generic ethernet-phy clock") was applied a
> > year ago, so I raised a concern in V2 [1], if a new platform only
> > enables rmii-ref in the PHY driver, disabling rmii-ref after getting
> > the clock rate will cause problem, which will cause RMII to not work.
> > I'm not sure if any platform actually does this, if so the following changes will
> be a more serious problem.
> 
> Put more of this explanation into the commit message and resend.
> If it convinces Andrew we can apply.

Okay, thanks