[v1] phy: mscc: Fix PTP for vsc8574 and VSC8572

[PATCH net] phy: mscc: Fix PTP for vsc8574 and VSC8572

Posted by Horatiu Vultur 2 weeks, 3 days ago

When trying to enable PTP on vsc8574 and vsc8572 it is not working even
if the function vsc8584_ptp_init it says that it has support for PHY
timestamping. It is not working because there is no PTP device.
So, to fix this make sure to create a PTP device also for this PHYs as
they have the same PTP IP as the other vsc PHYs.

Fixes: 774626fa440e ("net: phy: mscc: Add PTP support for 2 more VSC PHYs")
Signed-off-by: Horatiu Vultur <horatiu.vultur@microchip.com>
---
 drivers/net/phy/mscc/mscc_main.c | 20 +++++++++++++++++---
 1 file changed, 17 insertions(+), 3 deletions(-)

diff --git a/drivers/net/phy/mscc/mscc_main.c b/drivers/net/phy/mscc/mscc_main.c
index ef0ef1570d392..89b5cd96e8720 100644
--- a/drivers/net/phy/mscc/mscc_main.c
+++ b/drivers/net/phy/mscc/mscc_main.c
@@ -2259,6 +2259,7 @@ static int vsc8574_probe(struct phy_device *phydev)
 	u32 default_mode[4] = {VSC8531_LINK_1000_ACTIVITY,
 	   VSC8531_LINK_100_ACTIVITY, VSC8531_LINK_ACTIVITY,
 	   VSC8531_DUPLEX_COLLISION};
+	int ret;
 
 	vsc8531 = devm_kzalloc(&phydev->mdio.dev, sizeof(*vsc8531), GFP_KERNEL);
 	if (!vsc8531)
@@ -2267,8 +2268,11 @@ static int vsc8574_probe(struct phy_device *phydev)
 	phydev->priv = vsc8531;
 
 	vsc8584_get_base_addr(phydev);
-	devm_phy_package_join(&phydev->mdio.dev, phydev,
-			      vsc8531->base_addr, 0);
+	ret = devm_phy_package_join(&phydev->mdio.dev, phydev,
+				    vsc8531->base_addr,
+				    sizeof(struct vsc85xx_shared_private));
+	if (ret)
+		return ret;
 
 	vsc8531->nleds = 4;
 	vsc8531->supp_led_modes = VSC8584_SUPP_LED_MODES;
@@ -2279,6 +2283,16 @@ static int vsc8574_probe(struct phy_device *phydev)
 	if (!vsc8531->stats)
 		return -ENOMEM;
 
+	if (phy_package_probe_once(phydev)) {
+		ret = vsc8584_ptp_probe_once(phydev);
+		if (ret)
+			return ret;
+	}
+
+	ret = vsc8584_ptp_probe(phydev);
+	if (ret)
+		return ret;
+
 	return vsc85xx_dt_led_modes_get(phydev, default_mode);
 }
 
@@ -2648,7 +2662,7 @@ static struct phy_driver vsc85xx_driver[] = {
 	.config_aneg    = &vsc85xx_config_aneg,
 	.aneg_done	= &genphy_aneg_done,
 	.read_status	= &vsc85xx_read_status,
-	.handle_interrupt = vsc85xx_handle_interrupt,
+	.handle_interrupt = vsc8584_handle_interrupt,
 	.config_intr    = &vsc85xx_config_intr,
 	.suspend	= &genphy_suspend,
 	.resume		= &genphy_resume,
-- 
2.34.1

Re: [PATCH net] phy: mscc: Fix PTP for vsc8574 and VSC8572

Posted by Paolo Abeni 2 weeks, 2 days ago

On 9/15/25 10:01 AM, Horatiu Vultur wrote:
> When trying to enable PTP on vsc8574 and vsc8572 it is not working even
> if the function vsc8584_ptp_init it says that it has support for PHY
> timestamping. It is not working because there is no PTP device.
> So, to fix this make sure to create a PTP device also for this PHYs as
> they have the same PTP IP as the other vsc PHYs.

[...]

> diff --git a/drivers/net/phy/mscc/mscc_main.c b/drivers/net/phy/mscc/mscc_main.c
> index ef0ef1570d392..89b5cd96e8720 100644
> --- a/drivers/net/phy/mscc/mscc_main.c
> +++ b/drivers/net/phy/mscc/mscc_main.c
> @@ -2259,6 +2259,7 @@ static int vsc8574_probe(struct phy_device *phydev)
>  	u32 default_mode[4] = {VSC8531_LINK_1000_ACTIVITY,
>  	   VSC8531_LINK_100_ACTIVITY, VSC8531_LINK_ACTIVITY,
>  	   VSC8531_DUPLEX_COLLISION};
> +	int ret;
>  
>  	vsc8531 = devm_kzalloc(&phydev->mdio.dev, sizeof(*vsc8531), GFP_KERNEL);
>  	if (!vsc8531)

vsc8574_probe() is also used by 8504 and 8552, is the side effect intended?

/P

Re: [PATCH net] phy: mscc: Fix PTP for vsc8574 and VSC8572

Posted by Horatiu Vultur 2 weeks, 1 day ago

The 09/16/2025 15:32, Paolo Abeni wrote:

Hi Paolo,

> 
> On 9/15/25 10:01 AM, Horatiu Vultur wrote:
> > When trying to enable PTP on vsc8574 and vsc8572 it is not working even
> > if the function vsc8584_ptp_init it says that it has support for PHY
> > timestamping. It is not working because there is no PTP device.
> > So, to fix this make sure to create a PTP device also for this PHYs as
> > they have the same PTP IP as the other vsc PHYs.
> 
> [...]
> 
> > diff --git a/drivers/net/phy/mscc/mscc_main.c b/drivers/net/phy/mscc/mscc_main.c
> > index ef0ef1570d392..89b5cd96e8720 100644
> > --- a/drivers/net/phy/mscc/mscc_main.c
> > +++ b/drivers/net/phy/mscc/mscc_main.c
> > @@ -2259,6 +2259,7 @@ static int vsc8574_probe(struct phy_device *phydev)
> >       u32 default_mode[4] = {VSC8531_LINK_1000_ACTIVITY,
> >          VSC8531_LINK_100_ACTIVITY, VSC8531_LINK_ACTIVITY,
> >          VSC8531_DUPLEX_COLLISION};
> > +     int ret;
> >
> >       vsc8531 = devm_kzalloc(&phydev->mdio.dev, sizeof(*vsc8531), GFP_KERNEL);
> >       if (!vsc8531)
> 
> vsc8574_probe() is also used by 8504 and 8552, is the side effect intended?

No, that side effect is not intended because I have checked the vsc8552
and vsc8504 datasheet and I couldn't find anything about timestamping.
So this is a mistake.
My first thought was to use vsc8514_probe for these devices but this
will introduce another side effect as the statistics will be different.
So, I will need to create another probe function for these 2 devices.

> 
> /P
> 

-- 
/Horatiu