Re: [PATCH net-next v2 01/10] net: freescale: ucc_geth: Drop support for the "interface" DT property

[Maxime Chevallier <maxime.chevallier@bootlin.com>]

Hello Simon,

On Fri, 15 Nov 2024 12:19:14 +0000
Simon Horman <horms@kernel.org> wrote:

[...]
> > @@ -3627,18 +3588,17 @@ static int ucc_geth_probe(struct platform_device* ofdev)
> >  	/* Find the TBI PHY node.  If it's not there, we don't support SGMII */
> >  	ug_info->tbi_node = of_parse_phandle(np, "tbi-handle", 0);
> >  
> > -	/* get the phy interface type, or default to MII */
> > -	prop = of_get_property(np, "phy-connection-type", NULL);
> > -	if (!prop) {
> > -		/* handle interface property present in old trees */
> > -		prop = of_get_property(ug_info->phy_node, "interface", NULL);
> > -		if (prop != NULL) {
> > -			phy_interface = enet_to_phy_interface[*prop];
> > -			max_speed = enet_to_speed[*prop];
> > -		} else
> > -			phy_interface = PHY_INTERFACE_MODE_MII;
> > -	} else {
> > -		phy_interface = to_phy_interface((const char *)prop);
> > +	prop = of_get_property(ug_info->phy_node, "interface", NULL);
> > +	if (prop) {
> > +		dev_err(&ofdev->dev,
> > +			"Device-tree property 'interface' is no longer supported. Please use 'phy-connection-type' instead.");
> > +		goto err_put_tbi;  
> 
> Hi Maxime,
> 
> This goto will result in err being returned by this function.
> But here err is 0. Should it be set to an error value instead?
> 
> Flagged by Smatch.

arg yes you're right. I'll address this in the next iteration.

Thanks,

Maxime