Re: [PATCH 1/5] dt-bindings: can: tcan4x5x: Add tcan4552 and tcan4553 variants

[Simon Horman]

On Wed, Mar 15, 2023 at 04:58:33PM +0100, Markus Schneider-Pargmann wrote:
> On Wed, Mar 15, 2023 at 02:14:27PM +0100, Krzysztof Kozlowski wrote:
> > On 15/03/2023 12:25, Marc Kleine-Budde wrote:
> > > On 14.03.2023 21:01:10, Krzysztof Kozlowski wrote:
> > >> On 14/03/2023 16:11, Markus Schneider-Pargmann wrote:
> > >>> These two new chips do not have state or wake pins.
> > >>>
> > >>> Signed-off-by: Markus Schneider-Pargmann <msp@xxxxxxxxxxxx>
> > >>> ---
> > >>>  .../devicetree/bindings/net/can/tcan4x5x.txt          | 11 ++++++++---
> > >>>  1 file changed, 8 insertions(+), 3 deletions(-)
> > >>>
> > >>> diff --git a/Documentation/devicetree/bindings/net/can/tcan4x5x.txt b/Documentation/devicetree/bindings/net/can/tcan4x5x.txt
> > >>> index e3501bfa22e9..38a2b5369b44 100644
> > >>> --- a/Documentation/devicetree/bindings/net/can/tcan4x5x.txt
> > >>> +++ b/Documentation/devicetree/bindings/net/can/tcan4x5x.txt
> > >>> @@ -4,7 +4,10 @@ Texas Instruments TCAN4x5x CAN Controller
> > >>>  This file provides device node information for the TCAN4x5x interface contains.
> > >>>  
> > >>>  Required properties:
> > >>> -	- compatible: "ti,tcan4x5x"
> > >>> +	- compatible:
> > >>> +		"ti,tcan4x5x" or
> > >>> +		"ti,tcan4552" or
> > >>> +		"ti,tcan4553"
> > >>
> > >> Awesome, they nicely fit into wildcard... Would be useful to deprecate
> > >> the wildcard at some point and switch to proper compatibles in such
> > >> case, because now they became confusing.
> > > 
> > > I plead for DT stability!
> > > 
> > > As I understand correctly, the exact version of the chip (4550, 4552, or
> > > 4553) can be detected via the ID2 register.
> > 
> > So maybe there is no need for this patch at all? Or the new compatibles
> > should be made compatible with generic fallback?
> 
> I can use the value being read from the ID2 register to get the version.
> This at least holds the correct value for tcan4550, 4552 and 4553. But
> the state and wake gpios can't be used in case of a 4552 or 4553.
> 
> So yes, it is possible to do it without the new compatibles but the
> changes for state and wake gpios need to stay.
> 
> What do you two prefer?

FWIIW, I think it is good to have the extra compat strings,
even if the driver only uses the fallback string for now.
This would allow the driver to take into account HW differences that come
to light later, without needing to update the bindings.