[PATCH net v3] dpaa_eth: support all modes with rate adapting PHYs

[PATCH net v3] dpaa_eth: support all modes with rate adapting PHYs

[Madalin Bucur]

Stop removing modes that are not supported on the system interface
when the connected PHY is capable of rate adaptation. This addresses
an issue with the LS1046ARDB board 10G interface no longer working
with an 1G link partner after autonegotiation support was added
for the Aquantia PHY on board in

commit 09c4c57f7bc4 ("net: phy: aquantia: add support for auto-negotiation configuration")

Before this commit the values advertised by the PHY were not
influenced by the dpaa_eth driver removal of system-side unsupported
modes as the aqr_config_aneg() was basically a no-op. After this
commit, the modes removed by the dpaa_eth driver were no longer
advertised thus autonegotiation with 1G link partners failed.

Reported-by: Mian Yousaf Kaukab <ykaukab@suse.de>
Signed-off-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
---

change in v3: no longer add an API for checking the capability,
  rely on PHY vendor to determine if more modes may be available
  through rate adaptation so stop removing them

 drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
index 09dbcd819d84..fd93d542f497 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
@@ -2453,6 +2453,9 @@ static void dpaa_adjust_link(struct net_device *net_dev)
 	mac_dev->adjust_link(mac_dev);
 }
 
+/* The Aquantia PHYs are capable of performing rate adaptation */
+#define PHY_VEND_AQUANTIA	0x03a1b400
+
 static int dpaa_phy_init(struct net_device *net_dev)
 {
 	__ETHTOOL_DECLARE_LINK_MODE_MASK(mask) = { 0, };
@@ -2471,9 +2474,14 @@ static int dpaa_phy_init(struct net_device *net_dev)
 		return -ENODEV;
 	}
 
-	/* Remove any features not supported by the controller */
-	ethtool_convert_legacy_u32_to_link_mode(mask, mac_dev->if_support);
-	linkmode_and(phy_dev->supported, phy_dev->supported, mask);
+	/* Unless the PHY is capable of rate adaptation */
+	if (mac_dev->phy_if != PHY_INTERFACE_MODE_XGMII ||
+	    ((phy_dev->drv->phy_id & GENMASK(31, 10)) != PHY_VEND_AQUANTIA)) {
+		/* remove any features not supported by the controller */
+		ethtool_convert_legacy_u32_to_link_mode(mask,
+							mac_dev->if_support);
+		linkmode_and(phy_dev->supported, phy_dev->supported, mask);
+	}
 
 	phy_support_asym_pause(phy_dev);
 
-- 
2.1.0

Re: [PATCH net v3] dpaa_eth: support all modes with rate adapting PHYs

[David Miller]

From: Madalin Bucur <madalin.bucur@oss.nxp.com>
Date: Tue,  4 Feb 2020 12:08:58 +0200

> Stop removing modes that are not supported on the system interface
> when the connected PHY is capable of rate adaptation. This addresses
> an issue with the LS1046ARDB board 10G interface no longer working
> with an 1G link partner after autonegotiation support was added
> for the Aquantia PHY on board in
> 
> commit 09c4c57f7bc4 ("net: phy: aquantia: add support for auto-negotiation configuration")
> 
> Before this commit the values advertised by the PHY were not
> influenced by the dpaa_eth driver removal of system-side unsupported
> modes as the aqr_config_aneg() was basically a no-op. After this
> commit, the modes removed by the dpaa_eth driver were no longer
> advertised thus autonegotiation with 1G link partners failed.
> 
> Reported-by: Mian Yousaf Kaukab <ykaukab@suse.de>
> Signed-off-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
> ---
> 
> change in v3: no longer add an API for checking the capability,
>   rely on PHY vendor to determine if more modes may be available
>   through rate adaptation so stop removing them

There was quite a discussion about the approach that should be used here
in the short term.

Therefore I'd like some reviews and ACKs from interested parties such as
Andrew, Florian, and Heiner.

Thanks.

Re: [PATCH net v3] dpaa_eth: support all modes with rate adapting PHYs

[Andrew Lunn]

On Tue, Feb 04, 2020 at 12:08:58PM +0200, Madalin Bucur wrote:
> Stop removing modes that are not supported on the system interface
> when the connected PHY is capable of rate adaptation. This addresses
> an issue with the LS1046ARDB board 10G interface no longer working
> with an 1G link partner after autonegotiation support was added
> for the Aquantia PHY on board in
> 
> commit 09c4c57f7bc4 ("net: phy: aquantia: add support for auto-negotiation configuration")
> 
> Before this commit the values advertised by the PHY were not
> influenced by the dpaa_eth driver removal of system-side unsupported
> modes as the aqr_config_aneg() was basically a no-op. After this
> commit, the modes removed by the dpaa_eth driver were no longer
> advertised thus autonegotiation with 1G link partners failed.
> 
> Reported-by: Mian Yousaf Kaukab <ykaukab@suse.de>
> Signed-off-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
> ---
> 
> change in v3: no longer add an API for checking the capability,
>   rely on PHY vendor to determine if more modes may be available
>   through rate adaptation so stop removing them
> 
>  drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 14 +++++++++++---
>  1 file changed, 11 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
> index 09dbcd819d84..fd93d542f497 100644
> --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
> +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
> @@ -2453,6 +2453,9 @@ static void dpaa_adjust_link(struct net_device *net_dev)
>  	mac_dev->adjust_link(mac_dev);
>  }
>  
> +/* The Aquantia PHYs are capable of performing rate adaptation */
> +#define PHY_VEND_AQUANTIA	0x03a1b400
> +
>  static int dpaa_phy_init(struct net_device *net_dev)
>  {
>  	__ETHTOOL_DECLARE_LINK_MODE_MASK(mask) = { 0, };
> @@ -2471,9 +2474,14 @@ static int dpaa_phy_init(struct net_device *net_dev)
>  		return -ENODEV;
>  	}
>  
> -	/* Remove any features not supported by the controller */
> -	ethtool_convert_legacy_u32_to_link_mode(mask, mac_dev->if_support);
> -	linkmode_and(phy_dev->supported, phy_dev->supported, mask);
> +	/* Unless the PHY is capable of rate adaptation */
> +	if (mac_dev->phy_if != PHY_INTERFACE_MODE_XGMII ||
> +	    ((phy_dev->drv->phy_id & GENMASK(31, 10)) != PHY_VEND_AQUANTIA)) {
> +		/* remove any features not supported by the controller */
> +		ethtool_convert_legacy_u32_to_link_mode(mask,
> +							mac_dev->if_support);
> +		linkmode_and(phy_dev->supported, phy_dev->supported, mask);
> +	}
>  
>  	phy_support_asym_pause(phy_dev);

This is reasonable until we figure out a clean way to do this.

Reviewed-by: Andrew Lunn <andrew@lunn.ch>

    Andrew

Re: [PATCH net v3] dpaa_eth: support all modes with rate adapting PHYs

[David Miller]

From: Madalin Bucur <madalin.bucur@oss.nxp.com>
Date: Tue,  4 Feb 2020 12:08:58 +0200

> Stop removing modes that are not supported on the system interface
> when the connected PHY is capable of rate adaptation. This addresses
> an issue with the LS1046ARDB board 10G interface no longer working
> with an 1G link partner after autonegotiation support was added
> for the Aquantia PHY on board in
> 
> commit 09c4c57f7bc4 ("net: phy: aquantia: add support for auto-negotiation configuration")
> 
> Before this commit the values advertised by the PHY were not
> influenced by the dpaa_eth driver removal of system-side unsupported
> modes as the aqr_config_aneg() was basically a no-op. After this
> commit, the modes removed by the dpaa_eth driver were no longer
> advertised thus autonegotiation with 1G link partners failed.
> 
> Reported-by: Mian Yousaf Kaukab <ykaukab@suse.de>
> Signed-off-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
> ---
> 
> change in v3: no longer add an API for checking the capability,
>   rely on PHY vendor to determine if more modes may be available
>   through rate adaptation so stop removing them

Applied and queued up for -stable, thank you.