Revert "usb: typec: mux: Remove requirement for the "orientation-switch" device property"

This reverts commit acad3e9c.

The device property that can be used to identify the device
class/type of the remote port parent when device graph is
used is always needed after all. Without it there is no real
way to know is the requested connection actually described
in the device graph or not.

If the connection is described in the device graph but the
device instance is still missing for what ever reason, the
code defers probe for now.

Adding a comment to the code to explain this.
Reviewed-by: Li Jun <jun.li@nxp.com>
Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Link: https://lore.kernel.org/r/20210602112253.70200-1-heikki.krogerus@linux.intel.comSigned-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/usb/typec/mux.c

@@ -30,6 +30,22 @@ static void *typec_switch_match(struct fwnode_handle *fwnode, const char *id,
 {
 	struct device *dev;
 
+	/*
+	 * Device graph (OF graph) does not give any means to identify the
+	 * device type or the device class of the remote port parent that @fwnode
+	 * represents, so in order to identify the type or the class of @fwnode
+	 * an additional device property is needed. With typec switches the
+	 * property is named "orientation-switch" (@id). The value of the device
+	 * property is ignored.
+	 */
+	if (id && !fwnode_property_present(fwnode, id))
+		return NULL;
+
+	/*
+	 * At this point we are sure that @fwnode is a typec switch in all
+	 * cases. If the switch hasn't yet been registered for some reason, the
+	 * function "defers probe" for now.
+	 */
 	dev = class_find_device(&typec_mux_class, NULL, fwnode,
 				switch_fwnode_match);