Re: [PATCH v3 2/5] dt-bindings: input/touchscreen: Add compatible for IST3038B

[Conor Dooley]

On Sat, Dec 09, 2023 at 10:05:27AM +0100, Karel Balej wrote:
> On Mon Dec 4, 2023 at 1:52 PM CET, Conor Dooley wrote:
> > On Mon, Dec 04, 2023 at 02:40:44PM +0200, Markuss Broks wrote:
> > > On 12/3/23 13:20, Conor Dooley wrote:
> > > > On Sat, Dec 02, 2023 at 01:48:33PM +0100, Karel Balej wrote:
> > > > > From: Markuss Broks <markuss.broks@xxxxxxxxx>
> > > > > 
> > > > > Imagis IST3038B is a variant (firmware?) of Imagis IST3038 IC,
> > > > > add the compatible for it to the IST3038C bindings.
> > > > This one is better, but would be well served by mentioning what
> > > > specifically is different (register addresses or firmware commands?)
> > > 
> > > I don't think anyone knows this other than Imagis itself. I would guess it's
> > > different hardware, since register addresses are indeed different, but on
> > > the other hand, there is a possibility that firmware on the MCU could be
> > > responding to those commands. I suppose "... IST3038B is a hardware variant
> > > of ... IST3038" would be more correct.
> >
> > Only Imagis might know the specifics, but you (plural) have made driver
> > changes so you know what is different in terms of the programming model.
> > I'm just asking for you to mention how the programming model varies in
> > the commit message. Otherwise I can't know whether you should have added
> > a fallback compatible, without going and reading your driver change. The
> > commit message for the bindings should stand on its own merit in that
> > regard.
> > "Variant" alone does not suffice, as many variants of devices have a
> > compatible programming model, be that for a subset of features or
> > complete compatibility.
> >
> > > The reason why I think it could be firmware-defined is because we have a lot
> > > of variants (30xxA, 30xxB, 30xxC, plain 30xx), and the numbers usually mean
> > > feature level/completeness, e.g. some don't support the touch pressure or
> > > touchkeys, and we don't know what A/B/C/none means.
> >
> > Ultimately whether it is due to firmware or the hardware isn't
> > particular important, just mention what is incompatibly different.
> 
> I propose to update the commit description as such:
> 
> 	Imagis IST3038B is a variant (firmware?) of Imagis IST3038 IC
> 	differing from IST3038C in its register interface. Add the
> 	compatible for it to the IST3038C bindings.


SGTM. You can add
Acked-by: Conor Dooley <conor.dooley@xxxxxxxxxxxxx>
with that commit message update.

Thanks,
Conor.

Attachment: signature.asc
Description: PGP signature