On Tue, Mar 29, 2016 at 7:30 PM, Alan Tull [off-list ref] wrote:

On Fri, Mar 25, 2016 at 9:31 PM, qiujiang [off-list ref] wrote:

quoted

-               if (pp->idx == 0 &&
-                   of_property_read_bool(port_np, "interrupt-controller")) {
-                       pp->irq = irq_of_parse_and_map(port_np, 0);
+               if (dev->of_node && pp->idx == 0 &&
+                       fwnode_property_read_bool(fwnode,
+                                                 "interrupt-controller")) {
+                       pp->irq = irq_of_parse_and_map(to_of_node(fwnode), 0);
                        if (!pp->irq) {
                                dev_warn(dev, "no irq for bank %s\n",
-                                        port_np->full_name);
+                                        to_of_node(fwnode)->full_name);
                        }
                }

                pp->irq_shared  = false;
                pp->gpio_base   = -1;
-               pp->name        = port_np->full_name;
+               if (dev->of_node)
+                       pp->name = to_of_node(fwnode)->full_name;

Hi Jiang,

I tested lightly on a CycloneV and it worked fine (with device tree).

One suggestion for both patches: you could remove name from struct
dwapb_port_property and get rid of pp->name and nobody would miss it.
All it is used for is some dev_err's so the device info gets printed
anyway.  For example (if I leave the irq out of the DT)

gpio-dwapb ff708000.gpio: no irq for bank /soc/gpio@ff708000/gpio-controller@0

is redundant.  The only additional info here from the name is the port
index.  That could be added to the messages without having to get the
name through the two property/of methods.

Good suggestion! That'll make patches and code cleaner.

Perhaps separate prepended  patch?

-- 
With Best Regards,
Andy Shevchenko