[PATCH v2] Revert "net: phy: marvell: avoid pause mode on SGMII-to-Copper for 88e151x"

[PATCH v2] Revert "net: phy: marvell: avoid pause mode on SGMII-to-Copper for 88e151x"

[Russell King]

This reverts commit 6623c0fba10ef45b64ca213ad5dec926f37fa9a0.

The original diagnosis was incorrect: it appears that the NIC had
PHY polling mode enabled, which meant that it overwrote the PHYs
advertisement register during negotiation.

Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
---
v2: respun on top of net-next, but I've not yet been able to boot
 test this - which is unlikely to be for a while yet.

The extraneous whitespace change comes from the reversion of the
original patch - I added a "u32 pause" and blank line there.  The
patch converting to link modes removed the "u32 pause" but left
the blank line, causing a coding style issue.  It seems only right
that a reversion of the original commit fixes that too.

 drivers/net/phy/marvell.c | 16 ----------------
 1 file changed, 16 deletions(-)

diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index 90f44ba8aca7..3ccba37bd6dd 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -842,7 +842,6 @@ static int m88e1510_config_init(struct phy_device *phydev)
 
 	/* SGMII-to-Copper mode initialization */
 	if (phydev->interface == PHY_INTERFACE_MODE_SGMII) {
-
 		/* Select page 18 */
 		err = marvell_set_page(phydev, 18);
 		if (err < 0)
@@ -865,21 +864,6 @@ static int m88e1510_config_init(struct phy_device *phydev)
 		err = marvell_set_page(phydev, MII_MARVELL_COPPER_PAGE);
 		if (err < 0)
 			return err;
-
-		/* There appears to be a bug in the 88e1512 when used in
-		 * SGMII to copper mode, where the AN advertisement register
-		 * clears the pause bits each time a negotiation occurs.
-		 * This means we can never be truely sure what was advertised,
-		 * so disable Pause support.
-		 */
-		linkmode_clear_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT,
-				   phydev->supported);
-		linkmode_clear_bit(ETHTOOL_LINK_MODE_Pause_BIT,
-				   phydev->supported);
-		linkmode_clear_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT,
-				   phydev->advertising);
-		linkmode_clear_bit(ETHTOOL_LINK_MODE_Pause_BIT,
-				   phydev->advertising);
 	}
 
 	return m88e1318_config_init(phydev);
-- 
2.7.4

Re: [PATCH v2] Revert "net: phy: marvell: avoid pause mode on SGMII-to-Copper for 88e151x"

[liuyonglong]

On 2019/2/1 0:59, Russell King wrote:
> This reverts commit 6623c0fba10ef45b64ca213ad5dec926f37fa9a0.
> 
> The original diagnosis was incorrect: it appears that the NIC had
> PHY polling mode enabled, which meant that it overwrote the PHYs
> advertisement register during negotiation.
> 
> Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>

Tested-by: Yonglong Liu <liuyonglong@huawei.com>

Had done boot testing, and the networking is fine too.

> ---
> v2: respun on top of net-next, but I've not yet been able to boot
>  test this - which is unlikely to be for a while yet.
> 
> The extraneous whitespace change comes from the reversion of the
> original patch - I added a "u32 pause" and blank line there.  The
> patch converting to link modes removed the "u32 pause" but left
> the blank line, causing a coding style issue.  It seems only right
> that a reversion of the original commit fixes that too.
> 
>  drivers/net/phy/marvell.c | 16 ----------------
>  1 file changed, 16 deletions(-)
> 
> diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
> index 90f44ba8aca7..3ccba37bd6dd 100644
> --- a/drivers/net/phy/marvell.c
> +++ b/drivers/net/phy/marvell.c
> @@ -842,7 +842,6 @@ static int m88e1510_config_init(struct phy_device *phydev)
>  
>  	/* SGMII-to-Copper mode initialization */
>  	if (phydev->interface == PHY_INTERFACE_MODE_SGMII) {
> -
>  		/* Select page 18 */
>  		err = marvell_set_page(phydev, 18);
>  		if (err < 0)
> @@ -865,21 +864,6 @@ static int m88e1510_config_init(struct phy_device *phydev)
>  		err = marvell_set_page(phydev, MII_MARVELL_COPPER_PAGE);
>  		if (err < 0)
>  			return err;
> -
> -		/* There appears to be a bug in the 88e1512 when used in
> -		 * SGMII to copper mode, where the AN advertisement register
> -		 * clears the pause bits each time a negotiation occurs.
> -		 * This means we can never be truely sure what was advertised,
> -		 * so disable Pause support.
> -		 */
> -		linkmode_clear_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT,
> -				   phydev->supported);
> -		linkmode_clear_bit(ETHTOOL_LINK_MODE_Pause_BIT,
> -				   phydev->supported);
> -		linkmode_clear_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT,
> -				   phydev->advertising);
> -		linkmode_clear_bit(ETHTOOL_LINK_MODE_Pause_BIT,
> -				   phydev->advertising);
>  	}
>  
>  	return m88e1318_config_init(phydev);
> 

Re: [PATCH v2] Revert "net: phy: marvell: avoid pause mode on SGMII-to-Copper for 88e151x"

[David Miller]

From: Russell King <rmk+kernel@armlinux.org.uk>
Date: Thu, 31 Jan 2019 16:59:46 +0000

> This reverts commit 6623c0fba10ef45b64ca213ad5dec926f37fa9a0.
> 
> The original diagnosis was incorrect: it appears that the NIC had
> PHY polling mode enabled, which meant that it overwrote the PHYs
> advertisement register during negotiation.
> 
> Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
> ---
> v2: respun on top of net-next, but I've not yet been able to boot
>  test this - which is unlikely to be for a while yet.

This looks like a fix so I have applied it to 'net' and queued it up for
-stable.