Re: [PATCH v2 1/2] net: phy: adin: enable configuration of the LP Termination Register

On Tue, 2025-12-23 at 10:36 +0100, Andrew Lunn wrote:

quoted

+static int adin_config_zptm100(struct phy_device *phydev)
+{
+	struct device *dev = &phydev->mdio.dev;
+	int reg;
+	int rc;
+
+	if (!(device_property_read_bool(dev, "adi,low-cmode-impedance")))
+		return 0;
+
+	/* set to 0 to configure for lowest common-mode impedance */
+	rc = phy_write_mmd(phydev, MDIO_MMD_VEND1, ADIN1300_B_100_ZPTM_DIMRX, 0x0);
+	if (rc < 0)
+		return rc;
+
+	reg = phy_read_mmd(phydev, MDIO_MMD_VEND1, ADIN1300_B_100_ZPTM_DIMRX);
+	if (reg < 0)
+		return reg;
+
+	if (!(reg & ADIN1300_B_100_ZPTM_EN_DIMRX)) {
+		phydev_err(phydev, "Failed to set lowest common-mode impedance.\n");
+		return -EINVAL;
+	}

Under what condition do you think this could happen? Do you think
there are variants of the hardware which do not have this register?

	Andrew

I think he's just reading back the register to make sure the value was really updated...
If we were going to that for every write our lives would be miserable :).

I looked at both adin1200 and adin1300 and they support this in the same way so the above
should just be:

return phy_write_mmd()...

- Nuno Sá