[PATCH net] net: phylink: actually fix ksettings_set() ethtool call

[PATCH net] net: phylink: actually fix ksettings_set() ethtool call

[Russell King (Oracle)]

Raju Lakkaraju reported that the below commit caused a regression
with Lan743x drivers and a 2.5G SFP. Sadly, this is because the commit
was utterly wrong. Let's fix this properly by not moving the
linkmode_and(), but instead copying the link ksettings and then
modifying the advertising mask before passing the modified link
ksettings to phylib.

Fixes: df0acdc59b09 ("net: phylink: fix ksettings_set() ethtool call")
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
---
 drivers/net/phy/phylink.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index e237949deee6..cf4e51e48cdd 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -2225,11 +2225,13 @@ int phylink_ethtool_ksettings_set(struct phylink *pl,
 
 	ASSERT_RTNL();
 
-	/* Mask out unsupported advertisements */
-	linkmode_and(config.advertising, kset->link_modes.advertising,
-		     pl->supported);
-
 	if (pl->phydev) {
+		struct ethtool_link_ksettings phy_kset = *kset;
+
+		linkmode_and(phy_kset.link_modes.advertising,
+			     phy_kset.link_modes.advertising,
+			     pl->supported);
+
 		/* We can rely on phylib for this update; we also do not need
 		 * to update the pl->link_config settings:
 		 * - the configuration returned via ksettings_get() will come
@@ -2252,6 +2254,9 @@ int phylink_ethtool_ksettings_set(struct phylink *pl,
 	}
 
 	config = pl->link_config;
+	/* Mask out unsupported advertisements */
+	linkmode_and(config.advertising, kset->link_modes.advertising,
+		     pl->supported);
 
 	/* FIXME: should we reject autoneg if phy/mac does not support it? */
 	switch (kset->base.autoneg) {
-- 
2.30.2

Re: [PATCH net] net: phylink: actually fix ksettings_set() ethtool call

[Russell King (Oracle)]

Please ignore this, wrong version of the patch. :/ More coffee needed.

On Thu, Jun 01, 2023 at 10:02:07AM +0100, Russell King (Oracle) wrote:
> Raju Lakkaraju reported that the below commit caused a regression
> with Lan743x drivers and a 2.5G SFP. Sadly, this is because the commit
> was utterly wrong. Let's fix this properly by not moving the
> linkmode_and(), but instead copying the link ksettings and then
> modifying the advertising mask before passing the modified link
> ksettings to phylib.
> 
> Fixes: df0acdc59b09 ("net: phylink: fix ksettings_set() ethtool call")
> Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
> ---
>  drivers/net/phy/phylink.c | 13 +++++++++----
>  1 file changed, 9 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
> index e237949deee6..cf4e51e48cdd 100644
> --- a/drivers/net/phy/phylink.c
> +++ b/drivers/net/phy/phylink.c
> @@ -2225,11 +2225,13 @@ int phylink_ethtool_ksettings_set(struct phylink *pl,
>  
>  	ASSERT_RTNL();
>  
> -	/* Mask out unsupported advertisements */
> -	linkmode_and(config.advertising, kset->link_modes.advertising,
> -		     pl->supported);
> -
>  	if (pl->phydev) {
> +		struct ethtool_link_ksettings phy_kset = *kset;
> +
> +		linkmode_and(phy_kset.link_modes.advertising,
> +			     phy_kset.link_modes.advertising,
> +			     pl->supported);
> +
>  		/* We can rely on phylib for this update; we also do not need
>  		 * to update the pl->link_config settings:
>  		 * - the configuration returned via ksettings_get() will come
> @@ -2252,6 +2254,9 @@ int phylink_ethtool_ksettings_set(struct phylink *pl,
>  	}
>  
>  	config = pl->link_config;
> +	/* Mask out unsupported advertisements */
> +	linkmode_and(config.advertising, kset->link_modes.advertising,
> +		     pl->supported);
>  
>  	/* FIXME: should we reject autoneg if phy/mac does not support it? */
>  	switch (kset->base.autoneg) {
> -- 
> 2.30.2
> 
> 

-- 
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last!

[PATCH net] net: phylink: actually fix ksettings_set() ethtool call

[Russell King (Oracle)]

Raju Lakkaraju reported that the below commit caused a regression
with Lan743x drivers and a 2.5G SFP. Sadly, this is because the commit
was utterly wrong. Let's fix this properly by not moving the
linkmode_and(), but instead copying the link ksettings and then
modifying the advertising mask before passing the modified link
ksettings to phylib.

Fixes: df0acdc59b09 ("net: phylink: fix ksettings_set() ethtool call")
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
---
Raju,

Given the number of cockups I've made with this so far, it would be a
really good idea if you can explicitly test this patch and provide a
tested-by. Also it would be good to have a reported-by as well.

Thanks.

 drivers/net/phy/phylink.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index e237949deee6..b4831110003c 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -2225,11 +2225,13 @@ int phylink_ethtool_ksettings_set(struct phylink *pl,
 
 	ASSERT_RTNL();
 
-	/* Mask out unsupported advertisements */
-	linkmode_and(config.advertising, kset->link_modes.advertising,
-		     pl->supported);
-
 	if (pl->phydev) {
+		struct ethtool_link_ksettings phy_kset = *kset;
+
+		linkmode_and(phy_kset.link_modes.advertising,
+			     phy_kset.link_modes.advertising,
+			     pl->supported);
+
 		/* We can rely on phylib for this update; we also do not need
 		 * to update the pl->link_config settings:
 		 * - the configuration returned via ksettings_get() will come
@@ -2248,10 +2250,13 @@ int phylink_ethtool_ksettings_set(struct phylink *pl,
 		 *   the presence of a PHY, this should not be changed as that
 		 *   should be determined from the media side advertisement.
 		 */
-		return phy_ethtool_ksettings_set(pl->phydev, kset);
+		return phy_ethtool_ksettings_set(pl->phydev, &phy_kset);
 	}
 
 	config = pl->link_config;
+	/* Mask out unsupported advertisements */
+	linkmode_and(config.advertising, kset->link_modes.advertising,
+		     pl->supported);
 
 	/* FIXME: should we reject autoneg if phy/mac does not support it? */
 	switch (kset->base.autoneg) {
-- 
2.30.2

Re: [PATCH net] net: phylink: actually fix ksettings_set() ethtool call

[patchwork-bot+netdevbpf]

Hello:

This patch was applied to netdev/net.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Thu, 01 Jun 2023 10:12:06 +0100 you wrote:
> Raju Lakkaraju reported that the below commit caused a regression
> with Lan743x drivers and a 2.5G SFP. Sadly, this is because the commit
> was utterly wrong. Let's fix this properly by not moving the
> linkmode_and(), but instead copying the link ksettings and then
> modifying the advertising mask before passing the modified link
> ksettings to phylib.
> 
> [...]

Here is the summary with links:
  - [net] net: phylink: actually fix ksettings_set() ethtool call
    https://git.kernel.org/netdev/net/c/03c44a21d033

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

Re: [PATCH net] net: phylink: actually fix ksettings_set() ethtool call

[Raju Lakkaraju]

The 06/01/2023 10:12, Russell King (Oracle) wrote:
> EXTERNAL EMAIL: Do not click links or open attachments unless you know the content is safe
> 
> Raju Lakkaraju reported that the below commit caused a regression
> with Lan743x drivers and a 2.5G SFP. Sadly, this is because the commit
> was utterly wrong. Let's fix this properly by not moving the
> linkmode_and(), but instead copying the link ksettings and then
> modifying the advertising mask before passing the modified link
> ksettings to phylib.
> 
> Fixes: df0acdc59b09 ("net: phylink: fix ksettings_set() ethtool call")
> Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
> ---
> Raju,
> 
> Given the number of cockups I've made with this so far, it would be a
> really good idea if you can explicitly test this patch and provide a
> tested-by. Also it would be good to have a reported-by as well.

Tested this patch with 1G Speed Cu SFP (Axcen Photonics - AXGT-R1T4-05I1) with
different speeds (1G/100M/10M bps) changes.. Working as expected.

Tested-by: Raju Lakkaraju <Raju.Lakkaraju@microchip.com>
Reported-by: Raju Lakkaraju <Raju.Lakkaraju@microchip.com>

Thanks,
Raju
> 
> Thanks.
> 
>  drivers/net/phy/phylink.c | 15 ++++++++++-----
>  1 file changed, 10 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
> index e237949deee6..b4831110003c 100644
> --- a/drivers/net/phy/phylink.c
> +++ b/drivers/net/phy/phylink.c
> @@ -2225,11 +2225,13 @@ int phylink_ethtool_ksettings_set(struct phylink *pl,
> 
>         ASSERT_RTNL();
> 
> -       /* Mask out unsupported advertisements */
> -       linkmode_and(config.advertising, kset->link_modes.advertising,
> -                    pl->supported);
> -
>         if (pl->phydev) {
> +               struct ethtool_link_ksettings phy_kset = *kset;
> +
> +               linkmode_and(phy_kset.link_modes.advertising,
> +                            phy_kset.link_modes.advertising,
> +                            pl->supported);
> +
>                 /* We can rely on phylib for this update; we also do not need
>                  * to update the pl->link_config settings:
>                  * - the configuration returned via ksettings_get() will come
> @@ -2248,10 +2250,13 @@ int phylink_ethtool_ksettings_set(struct phylink *pl,
>                  *   the presence of a PHY, this should not be changed as that
>                  *   should be determined from the media side advertisement.
>                  */
> -               return phy_ethtool_ksettings_set(pl->phydev, kset);
> +               return phy_ethtool_ksettings_set(pl->phydev, &phy_kset);
>         }
> 
>         config = pl->link_config;
> +       /* Mask out unsupported advertisements */
> +       linkmode_and(config.advertising, kset->link_modes.advertising,
> +                    pl->supported);
> 
>         /* FIXME: should we reject autoneg if phy/mac does not support it? */
>         switch (kset->base.autoneg) {
> --
> 2.30.2
>