> >>> +#ifdef CONFIG_ACPI
> >>> +static const struct acpi_device_id rt1015p_acpi_match[] = {
> >>> + { "ALCP1015", 0},
> >>
> >> This isn't much better, ALC is not a vendor ID as defined by
> >> https://uefi.org/PNP_ACPI_Registry
> >>
> >> I've ask for guidance from Intel ACPI folks on this, please stay tuned.
> >>
> >
> > Hi Pierre,
> >
> > Is there any update from Intel ACPI team regarding to this ID format issue?
>
> Sorry about the delay.
>
> It seems that the options are limited. The HID needs to be made of two parts,
> vendor ID and part ID.
>
> For the vendor ID, you may use the PNP ID (RTL) or the PCI ID (10EC).
>
> For the part ID, you are limited to 4 hex-digits. I didn't get any positive answers
> on how to support '1015P', the only solution seems to be to create a new part
> number, or detect at run-time which of the '1015' skews is actually used.
>
Actually this driver is also used for rt1015 IC, the only difference is that rt1015.c is for i2c mode
and rt1015p is for non-i2c mode of rt1015 IC, so I think it's ok to use the same part ID (different vendor ID) for two drivers.
How do you think if we use "RTL1015" instead?
