On Fri, Dec 13, 2013 at 11:31:39PM +0100, Éric Piel wrote: > On 13-12-13 21:54, Sebastian Reichel wrote: > >The lis302 has already a DT binding described in [0], > >which descibes misc. hardware properties. The current > >binding does not support all values needed to convert > >the Nokia N900's platform data to DT. > > > >This patchset introduces support for describing inverted > >axis, configuration of second wakeup unit and wakeup > >threshold support. > > All the three patches related to the lis3lv02d driver look fine to me. > Here is my signed-off for patches 2, 3, and 4: > Signed-off-by: Éric Piel <eric.piel@xxxxxxxxxxxxxxxx> > > I've got 2 side questions: > * How do you write a negative value in a DTS file? Is this just the > very standard way (i.e., -), or do you need to use a specific > technique (such as 2-completemnt on 32 bits)? If it's not the > standard way, it should probably be documented somewhere. One can simply use -, but one has to put the value into (). This is because DT supports simple mathematical operations for the values, e.g. 5-10 and must have some way to distinguish between (5) (-10) and (5-10). Here is an example for N900 axis configuration from DT file: st,axis-x = <1>; /* LIS3_DEV_X */ st,axis-y = <(-2)>; /* LIS3_INV_DEV_Y */ st,axis-z = <(-3)>; /* LIS3_INV_DEV_Z */ I had some trouble to find this out myself, so probably its a good idea to document this somewhere. > * Who wants to carry these patches? Can they all go via the > devicetree branch? Or maybe Arnd or Greg would prefer to pick them > via the misc driver branch? Both ways are fine for me... I don't care, but it would be nice to see this in 3.14. -- Sebastian
Attachment:
signature.asc
Description: Digital signature
