On Tue, Nov 01, 2016 at 05:26:48PM +0100, Andrew Lunn wrote:

quoted

quoted

quoted

+	if (of_property_read_bool(np, "brcm,enet-phy-lane-swap")) {
+		/* Lane Swap - Undocumented register...magic! */
+		ret = bcm_phy_write_exp(phydev, MII_BCM54XX_EXP_SEL_ER + 0x9,
+					0x11B);
+		if (ret < 0)
+			return ret;
+	}
+

I wounder if this property could be made generic? What exactly are you
swapping? Rx and Tx lanes? Maybe we should add it to phy.txt?

Are you envisioning adding a DT check (similar to the
of_property_read_bool above, only with a more generic string) in
phy_device_create(), which will then set a PHY device flag?  This flag
would then be checked for in the PHY driver and the appropriate action
taken (in this case the bcm_phy_write_exp above).

I would keep the parsing of the property in the driver. But if we
think other PHYs could also support this feature, it would be good to
avoid having "brcm,enet-phy-lane-swap", "marvell,enet-phy-lane-swap",
"davicom,enet-phy-lane-swap", etc. It would be better to have one well
defined property documented in phy.txt which any PHY is free to
implement.

Okay, I understand what you are saying now.  I will assume that if
nothing exists today aside from this Broadcom errata, something in the
future will happen.  So, I agree that making it generic is a good idea.

I'll make the generic string be "enet-phy-lane-swap" (without the
previous "brcm"), and add the flag to phy.txt to document it.

Thanks,
Jon

	Andrew