On Wed Aug 5, 2026 at 4:37 AM CEST, Marek Vasut wrote:
> On 8/4/26 12:05 PM, Carlo Caione wrote:
>
> [...]
>
>> +++ b/doc/device-tree-bindings/usb/mediatek,mtu3.txt
>
> You could easily swap in Linux
> Documentation/devicetree/bindings/usb/mediatek,mtu3.yaml and
> Documentation/devicetree/bindings/usb/mediatek,mtk-xhci.yaml instead ,
> and drop the mtu3.txt . But please do replace the binding documents in a
> separate patch.
Shouldn't we just drop both
doc/device-tree-bindings/usb/mediatek,mtk-xhci.txt and
doc/device-tree-bindings/usb/mediatek,mtu3.txt and automatically fall
back onto the corresponding ones in dts/upstream/Bindings/usb?
>> +++ b/drivers/usb/mtu3/Kconfig
>> @@ -22,6 +22,7 @@ choice
>> config USB_MTU3_HOST
>> bool "Host only mode"
>> depends on USB_XHCI_HCD
>> + select SPL_OF_TRANSLATE if SPL_USB_HOST
>
> Don't you need SPL variants of the MTU3 controller Kconfig symbols instead ?
Fair point, I'll add those to V6.
>> help
>> Select this when you want to use MTU3 in host mode only,
>> thereby the gadget feature will be regressed.
>
> [...]
>
>> +static int mtu3_bind_gadget(struct udevice *parent)
>> +{
>> + struct udevice *dev;
>> + int ret;
>> +
>> + /* Node-less device: name it after the controller for diagnostics. */
>> + ret = device_bind_driver(parent, "mtu3-peripheral",
>> + ofnode_get_name(dev_ofnode(parent)), &dev);
>> + if (ret)
>> + dev_err(parent, "failed to bind peripheral mode\n");
>
> Please also include the value of "ret" in printed error strings, it
> helps identify the issue faster. Please fix this for all such cases in
> this patch.
It makes sense.
Thanks!
--
Carlo Caione
