Re: [PATCH 2/2] net: phy: relax error checking when creating sysfs link netdev->phydev

[Greg Kroah-Hartman]

On Wed, Mar 14, 2018 at 05:26:24PM -0500, Grygorii Strashko wrote:
> Some ethernet drivers (like TI CPSW) may connect and manage >1 Net PHYs per
> one netdevice, as result such drivers will produce warning during system
> boot and fail to connect second phy to netdevice when PHYLIB framework
> will try to create sysfs link netdev->phydev for second PHY
> in phy_attach_direct(), because sysfs link with the same name has been
> created already for the first PHY. As result, second CPSW external
> port will became unusable.
> 
> Fix it by relaxing error checking when PHYLIB framework is creating sysfs
> link netdev->phydev in phy_attach_direct(), suppressing warning by using
> sysfs_create_link_nowarn() and adding debug message instead.
> 
> Cc: Florian Fainelli <f.fainelli@xxxxxxxxx>
> Fixes: a3995460491d ("net: phy: Relax error checking on sysfs_create_link()")
> Signed-off-by: Grygorii Strashko <grygorii.strashko@xxxxxx>
> ---
>  drivers/net/phy/phy_device.c | 15 +++++++++++----
>  1 file changed, 11 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
> index 478405e..fe16f58 100644
> --- a/drivers/net/phy/phy_device.c
> +++ b/drivers/net/phy/phy_device.c
> @@ -1012,10 +1012,17 @@ int phy_attach_direct(struct net_device *dev, struct phy_device *phydev,
>  	err = sysfs_create_link(&phydev->mdio.dev.kobj, &dev->dev.kobj,
>  				"attached_dev");
>  	if (!err) {
> -		err = sysfs_create_link(&dev->dev.kobj, &phydev->mdio.dev.kobj,
> -					"phydev");
> -		if (err)
> -			goto error;
> +		err = sysfs_create_link_nowarn(&dev->dev.kobj,
> +					       &phydev->mdio.dev.kobj,
> +					       "phydev");
> +		if (err) {
> +			dev_err(&dev->dev, "could not add device link to %s err %d\n",
> +				kobject_name(&phydev->mdio.dev.kobj),
> +				err);

dev_err() is not a "debugging" message :)

What is a user going to do with this new error?  If it's not important
at all, why care about it?

> +			/* non-fatal - some net drivers can use one netdevice
> +			 * with more then one phy
> +			 */

What about devices that do not have more than one phy and this call
fails for?  Shouldn't you check for that?

thanks,

greg k-h