On Mon, Dec 26, 2022 at 04:14:23PM +0900, Yoshihiro Shimoda wrote: > Set phy_dev->host_interfaces by pl->link_interface in > phylink_fwnode_phy_connect() for a non-sfp PHY. > > Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@xxxxxxxxxxx> > --- > drivers/net/phy/phylink.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c > index 09cc65c0da93..1958d6cc9ef9 100644 > --- a/drivers/net/phy/phylink.c > +++ b/drivers/net/phy/phylink.c > @@ -1809,6 +1809,7 @@ int phylink_fwnode_phy_connect(struct phylink *pl, > pl->link_interface = phy_dev->interface; > pl->link_config.interface = pl->link_interface; > } > + __set_bit(pl->link_interface, phy_dev->host_interfaces); This is probably going to break Macchiatobin platforms, since we declare that the link mode there is 10GBASE-R, we'll end up with host_interfaces containing just this mode. This will cause the 88x3310 driver to select a rate matching interface mode, which the mvpp2 MAC can't support. If we want to fill host_interfaces in, then it needs to be filled in properly - and by that I mean with all the host interface modes that can be electrically supported - otherwise platforms will break. So, sorry, but NAK on this change. -- RMK's Patch system: https://www.armlinux.org.uk/developer/patches/ FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
