From: Dimitri Fedrau <dimitri.fedrau@liebherr.com>

The GPIO2 pin on the DP83822 can be configured as clock output. Add support
for configuration via DT.

Signed-off-by: Dimitri Fedrau <dimitri.fedrau@liebherr.com>
---
 drivers/net/phy/dp83822.c | 44 ++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 44 insertions(+)

diff --git a/drivers/net/phy/dp83822.c b/drivers/net/phy/dp83822.c
index cf8b6d0bfaa9812eee98c612c0d4259d87da7572..2abb066c3b8fe701c551c278ba31a08f9cb3fc15 100644
--- a/drivers/net/phy/dp83822.c
+++ b/drivers/net/phy/dp83822.c
@@ -14,6 +14,8 @@
 #include <linux/netdevice.h>
 #include <linux/bitfield.h>
 
+#include <dt-bindings/net/ti-dp83822.h>
+
 #define DP83822_PHY_ID	        0x2000a240
 #define DP83825S_PHY_ID		0x2000a140
 #define DP83825I_PHY_ID		0x2000a150
@@ -33,6 +35,7 @@
 #define MII_DP83822_RCSR	0x17
 #define MII_DP83822_RESET_CTRL	0x1f
 #define MII_DP83822_GENCFG	0x465
+#define MII_DP83822_IOCTRL2	0x463
 #define MII_DP83822_SOR1	0x467
 
 /* DP83826 specific registers */
@@ -106,6 +109,11 @@
 #define DP83822_RX_CLK_SHIFT	BIT(12)
 #define DP83822_TX_CLK_SHIFT	BIT(11)
 
+/* IOCTRL2 bits */
+#define DP83822_IOCTRL2_GPIO2_CLK_SRC		GENMASK(6, 4)
+#define DP83822_IOCTRL2_GPIO2_CTRL		GENMASK(2, 0)
+#define DP83822_IOCTRL2_GPIO2_CTRL_CLK_REF	GENMASK(1, 0)
+
 /* SOR1 mode */
 #define DP83822_STRAP_MODE1	0
 #define DP83822_STRAP_MODE2	BIT(0)
@@ -141,6 +149,8 @@ struct dp83822_private {
 	u8 cfg_dac_minus;
 	u8 cfg_dac_plus;
 	struct ethtool_wolinfo wol;
+	bool set_gpio2_clk_out;
+	u32 gpio2_clk_out;
 };
 
 static int dp83822_config_wol(struct phy_device *phydev,
@@ -415,6 +425,15 @@ static int dp83822_config_init(struct phy_device *phydev)
 	int err = 0;
 	int bmcr;
 
+	if (dp83822->set_gpio2_clk_out)
+		phy_modify_mmd(phydev, DP83822_DEVADDR, MII_DP83822_IOCTRL2,
+			       DP83822_IOCTRL2_GPIO2_CTRL |
+			       DP83822_IOCTRL2_GPIO2_CLK_SRC,
+			       FIELD_PREP(DP83822_IOCTRL2_GPIO2_CTRL,
+					  DP83822_IOCTRL2_GPIO2_CTRL_CLK_REF) |
+			       FIELD_PREP(DP83822_IOCTRL2_GPIO2_CLK_SRC,
+					  dp83822->gpio2_clk_out));
+
 	if (phy_interface_is_rgmii(phydev)) {
 		rx_int_delay = phy_get_internal_delay(phydev, dev, NULL, 0,
 						      true);
@@ -613,6 +632,7 @@ static int dp83822_of_init(struct phy_device *phydev)
 {
 	struct dp83822_private *dp83822 = phydev->priv;
 	struct device *dev = &phydev->mdio.dev;
+	int ret;
 
 	/* Signal detection for the PHY is only enabled if the FX_EN and the
 	 * SD_EN pins are strapped. Signal detection can only enabled if FX_EN
@@ -625,6 +645,30 @@ static int dp83822_of_init(struct phy_device *phydev)
 		dp83822->fx_enabled = device_property_present(dev,
 							      "ti,fiber-mode");
 
+	ret = of_property_read_u32(dev->of_node, "ti,gpio2-clk-out",
+				   &dp83822->gpio2_clk_out);
+	if (!ret) {
+		dp83822->set_gpio2_clk_out = true;
+		switch (dp83822->gpio2_clk_out) {
+		case DP83822_CLK_SRC_MAC_IF:
+			break;
+		case DP83822_CLK_SRC_XI:
+			break;
+		case DP83822_CLK_SRC_INT_REF:
+			break;
+		case DP83822_CLK_SRC_RMII_MASTER_MODE_REF:
+			break;
+		case DP83822_CLK_SRC_FREE_RUNNING:
+			break;
+		case DP83822_CLK_SRC_RECOVERED:
+			break;
+		default:
+			phydev_err(phydev, "ti,gpio2-clk-out value %u not valid\n",
+				   dp83822->gpio2_clk_out);
+			return -EINVAL;
+		}
+	}
+
 	return 0;
 }
 

-- 
2.39.5

>  #define MII_DP83822_RCSR	0x17
>  #define MII_DP83822_RESET_CTRL	0x1f
>  #define MII_DP83822_GENCFG	0x465
> +#define MII_DP83822_IOCTRL2	0x463
>  #define MII_DP83822_SOR1	0x467

These are sorted, so the MII_DP83822_IOCTRL2 should go before
MII_DP83822_GENCFG.

> +	if (dp83822->set_gpio2_clk_out)
> +		phy_modify_mmd(phydev, DP83822_DEVADDR, MII_DP83822_IOCTRL2,

I would of preferred MDIO_MMD_VEND2 rather than DP83822_DEVADDR, but
having just this one instance correct would look a bit odd.

> +	ret = of_property_read_u32(dev->of_node, "ti,gpio2-clk-out",
> +				   &dp83822->gpio2_clk_out);
> +	if (!ret) {
> +		dp83822->set_gpio2_clk_out = true;
> +		switch (dp83822->gpio2_clk_out) {
> +		case DP83822_CLK_SRC_MAC_IF:
> +			break;
> +		case DP83822_CLK_SRC_XI:
> +			break;
> +		case DP83822_CLK_SRC_INT_REF:
> +			break;
> +		case DP83822_CLK_SRC_RMII_MASTER_MODE_REF:
> +			break;
> +		case DP83822_CLK_SRC_FREE_RUNNING:
> +			break;
> +		case DP83822_CLK_SRC_RECOVERED:
> +			break;

You can list multiple case statements together, and have one break at
the end.

I would personally also only have:

> +		dp83822->set_gpio2_clk_out = true;

if validation passes, not that it really matters because:


> +		default:
> +			phydev_err(phydev, "ti,gpio2-clk-out value %u not valid\n",
> +				   dp83822->gpio2_clk_out);
> +			return -EINVAL;
> +		}
> +	}


    Andrew

---
pw-bot: cr

Am Mon, Dec 09, 2024 at 02:17:04PM +0100 schrieb Andrew Lunn:
> >  #define MII_DP83822_RCSR	0x17
> >  #define MII_DP83822_RESET_CTRL	0x1f
> >  #define MII_DP83822_GENCFG	0x465
> > +#define MII_DP83822_IOCTRL2	0x463
> >  #define MII_DP83822_SOR1	0x467
> 
> These are sorted, so the MII_DP83822_IOCTRL2 should go before
> MII_DP83822_GENCFG.
>
Will fix it.

> > +	if (dp83822->set_gpio2_clk_out)
> > +		phy_modify_mmd(phydev, DP83822_DEVADDR, MII_DP83822_IOCTRL2,
> 
> I would of preferred MDIO_MMD_VEND2 rather than DP83822_DEVADDR, but
> having just this one instance correct would look a bit odd.
>
Is it worth fixing it in a separate patch, replacing all DP83822_DEVADDR
with MDIO_MMD_VEND2 ?

> > +	ret = of_property_read_u32(dev->of_node, "ti,gpio2-clk-out",
> > +				   &dp83822->gpio2_clk_out);
> > +	if (!ret) {
> > +		dp83822->set_gpio2_clk_out = true;
> > +		switch (dp83822->gpio2_clk_out) {
> > +		case DP83822_CLK_SRC_MAC_IF:
> > +			break;
> > +		case DP83822_CLK_SRC_XI:
> > +			break;
> > +		case DP83822_CLK_SRC_INT_REF:
> > +			break;
> > +		case DP83822_CLK_SRC_RMII_MASTER_MODE_REF:
> > +			break;
> > +		case DP83822_CLK_SRC_FREE_RUNNING:
> > +			break;
> > +		case DP83822_CLK_SRC_RECOVERED:
> > +			break;
> 
> You can list multiple case statements together, and have one break at
> the end.
> 
Will fix it.

> I would personally also only have:
> 
> > +		dp83822->set_gpio2_clk_out = true;
> 
> if validation passes, not that it really matters because:
> 
> 
> > +		default:
> > +			phydev_err(phydev, "ti,gpio2-clk-out value %u not valid\n",
> > +				   dp83822->gpio2_clk_out);
> > +			return -EINVAL;
> > +		}
> > +	}
> 
Ok.

	Dimitri

> > I would of preferred MDIO_MMD_VEND2 rather than DP83822_DEVADDR, but
> > having just this one instance correct would look a bit odd.
> >
> Is it worth fixing it in a separate patch, replacing all DP83822_DEVADDR
> with MDIO_MMD_VEND2 ?

You could do. As a reviewer, i like to be able to quickly see, these
are vendor registers, i don't need to check if they are part of 802.3,
and should have standard names, and be pulled out into a library for
others to share. So if you want to add a patch, i'm happy with that.

	Andrew