RE: [PATCH 13/14] soundwire: mipi-disco: add new properties from 2.0 spec

[Liao, Bard]

> -----Original Message-----
> From: Charles Keepax <ckeepax@xxxxxxxxxxxxxxxxxxxxx>
> Sent: Monday, September 9, 2024 11:06 PM
> To: Bard Liao <yung-chuan.liao@xxxxxxxxxxxxxxx>
> Cc: linux-sound@xxxxxxxxxxxxxxx; vkoul@xxxxxxxxxx; vinod.koul@xxxxxxxxxx;
> linux-kernel@xxxxxxxxxxxxxxx; pierre-louis.bossart@xxxxxxxxxxxxxxx; Liao, Bard
> <bard.liao@xxxxxxxxx>
> Subject: Re: [PATCH 13/14] soundwire: mipi-disco: add new properties from
> 2.0 spec
> 
> On Tue, Aug 27, 2024 at 09:07:06PM +0800, Bard Liao wrote:
> > diff --git a/drivers/soundwire/mipi_disco.c
> b/drivers/soundwire/mipi_disco.c
> > index d6eb63bf1252..2215c53f95de 100644
> > --- a/drivers/soundwire/mipi_disco.c
> > +++ b/drivers/soundwire/mipi_disco.c
> > @@ -398,6 +398,19 @@ int sdw_slave_read_prop(struct sdw_slave *slave)
> >  	device_property_read_u32(dev, "mipi-sdw-sink-port-list",
> >  				 &prop->sink_ports);
> >
> > +	device_property_read_u32(dev, "mipi-sdw-sdca-interrupt-register-
> list",
> > +				 &prop->sdca_interrupt_register_list);
> > +
> > +	/*
> > +	 * The specification defines the property value as boolean, but
> > +	 * the value can be defined as zero. This is not aligned the
> > +	 * implementation of device_property_read_bool() which only checks
> > +	 * the presence of the property.
> > +	 * Let's use read_u8 to work-around this conceptual disconnect.
> > +	 */
> > +	device_property_read_u8(dev, "mipi-sdw-commit-register-
> supported",
> > +				&prop->commit_register_supported);
> 
> Would this not be a case for the new helper added earlier in the
> series? Or is this some third type of boolean?

Thanks Charles. Yes, we should use mipi_device_property_read_bool().
Will send a v2 to fix.

> 
> Thanks,
> Charles