Re: [PATCH net-next] net: phylink: allow PHYs to be attached in 802.3z inband mode

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

Hi Russell,

On 27/03/2026 13:56, Russell King (Oracle) wrote:
> Now that we have proper decision making for inband mode support which
> makes it a "best efforts" feature based on the capabilities of the PHY
> and PCS, we can relax whether we expect and permit a PHY to be
> attached. This is especially true for the 2500BASE-X case which some
> PHYs use without inband on their host side interface for 2.5G speeds,
> but use inband for slower speeds switching to SGMII on their host side
> interface.
> 
> We already have such a case for some qcom-ethqos setups, although
> qcom-ethqos overrides phylink's inband settings by accessing the PCS
> directly at the moment. This should allow qcom-ethqos to transition to
> defaulting to inband when 2500BASE-X or SGMII is specified in its DTS.
> 
> Allow PHYs to be attached when inband mode has been specified, which
> will be necessary to allow inband mode to be used on qcom-ethqos.
> 
> Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
> ---
>  drivers/net/phy/phylink.c | 8 ++------
>  1 file changed, 2 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
> index 087ac63f9193..1c178038e6f1 100644
> --- a/drivers/net/phy/phylink.c
> +++ b/drivers/net/phy/phylink.c
> @@ -1965,9 +1965,7 @@ EXPORT_SYMBOL_GPL(phylink_destroy);
>   */
>  bool phylink_expects_phy(struct phylink *pl)
>  {
> -	if (pl->cfg_link_an_mode == MLO_AN_FIXED ||
> -	    (pl->cfg_link_an_mode == MLO_AN_INBAND &&
> -	     phy_interface_mode_is_8023z(pl->link_interface)))
> +	if (pl->cfg_link_an_mode == MLO_AN_FIXED)
>  		return false;
>  	return true;
>  }
> @@ -2206,9 +2204,7 @@ static int phylink_attach_phy(struct phylink *pl, struct phy_device *phy,
>  {
>  	u32 flags = 0;
>  
> -	if (WARN_ON(pl->cfg_link_an_mode == MLO_AN_FIXED ||
> -		    (pl->cfg_link_an_mode == MLO_AN_INBAND &&
> -		     phy_interface_mode_is_8023z(interface) && !pl->sfp_bus)))
> +	if (WARN_ON(pl->cfg_link_an_mode == MLO_AN_FIXED))
>  		return -EINVAL;
>  
>  	if (pl->phydev)

So I've tested that on my socfpga setup, and I now get this when trying to
bring an interface attached to an SFP up :

  socfpga-dwmac ff702000.ethernet eth1: no phy found

This wasn't showing before because our DT has :

  &gmac1 {
    status = "okay";
    phy-mode = "1000base-x";
    managed = "in-band-status";
    sfp = <&sfp>;
  };

It used to kinda work because the problem was hidden with the use of "1000base-x"
as the PHY mode.

Now with this patch, using "sgmii" or "1000base-x" as the phy-mode doesn't
do any difference, as expected.

The real issue IMO is rather how stmmac_init_phy() behaves, it's guarded by:

  if (!phylink_expects_phy(priv->phylink))
          return 0;

and then tries to find a PHY, failing if there's none. We should probably add
an extra check there to say "if we have a PCS, then it's OK not to have a PHY" ?

Thanks,

Maxime