[PATCH] Input: tl6040-vibra: Device Tree support

Dmitry Torokhov dmitry.torokhov at gmail.com
Tue May 8 01:48:26 EST 2012


On Mon, May 07, 2012 at 01:49:29PM +0300, Peter Ujfalusi wrote:
> Hi Dmitry,
> 
> On 05/07/2012 10:02 AM, Dmitry Torokhov wrote:
> > Hi Peter,
> > 
> > On Fri, May 04, 2012 at 03:05:13PM +0300, Peter Ujfalusi wrote:
> >> +	if (pdata) {
> >> +		info->vibldrv_res = pdata->vibldrv_res;
> >> +		info->vibrdrv_res = pdata->vibrdrv_res;
> >> +		info->viblmotor_res = pdata->viblmotor_res;
> >> +		info->vibrmotor_res = pdata->vibrmotor_res;
> >> +		vddvibl_uV = pdata->vddvibl_uV;
> >> +		vddvibr_uV = pdata->vddvibr_uV;
> >> +	} else {
> >> +		of_property_read_u32(node, "vibldrv_res", &info->vibldrv_res);
> >> +		of_property_read_u32(node, "vibrdrv_res", &info->vibrdrv_res);
> >> +		of_property_read_u32(node, "viblmotor_res",
> >> +				     &info->viblmotor_res);
> >> +		of_property_read_u32(node, "vibrmotor_res",
> >> +				     &info->vibrmotor_res);
> > 
> > Since these 4 appear to be mandatory properties don't we need to
> > validate they are actually present in DT?
> 
> If the property does not exist in the DT blob the variable will not be
> updated - it will remain 0.
> I have kept the validity check for the resistance values. This will
> catch the case when something is missing from the DT blob.
> 
> I just did not wanted to complicate the code with additional checks
> since the end result would be the same.

Ah, indeed, I missed that we still validate the values.

Applied, thanks Peter.

-- 
Dmitry


More information about the devicetree-discuss mailing list