On Thu, Nov 29, 2018 at 05:12:45PM +0100, Miquel Raynal wrote:
> Hello,
>
> Miquel Raynal <miquel.raynal@xxxxxxxxxxx> wrote on Thu, 22 Nov 2018
> 22:04:16 +0100:
> > +static struct phy *mvebu_a3700_comphy_xlate(struct device *dev,
> > + struct of_phandle_args *args)
> > +{
> > + struct mvebu_a3700_comphy_lane *lane;
> > + struct phy *phy;
> > +
> > + if (WARN_ON(args->args[0] >= MVEBU_A3700_COMPHY_PORTS))
> > + return ERR_PTR(-EINVAL);
> > +
> > + phy = of_phy_simple_xlate(dev, args);
> > + if (IS_ERR(phy))
> > + return phy;
> > +
> > + lane = phy_get_drvdata(phy);
> > + if (lane->port >= 0)
> > + return ERR_PTR(-EBUSY);
>
> This is not valid. It works only the first time the consumer gets
> a PHY for this lane. If the consumer put the PHY (module is unloaded)
> and then gets the PHY again (module is re-loaded a second time), the
> port is already set to != -1 and this will exit with an error and
> prevent the module to probe.
>
> > +
> > + lane->port = args->args[0];
>
> I will remove the above check and transform it into a valid "port"
> value right here.
Please note that mvebu_comphy_xlate() in
drivers/phy/marvell/phy-mvebu-cp110-comphy.c does exactly the same.
--
RMK's Patch system: http://www.armlinux.org.uk/developer/patches/
FTTC broadband for 0.8mile line in suburbia: sync at 12.1Mbps down 622kbps up
According to speedtest.net: 11.9Mbps down 500kbps up
