diff mbox series

[V2,2/4] dt-bindings: arm: ti: am642/am654: Allow for SoC only compatibles

Message ID 20210925201430.11678-3-nm@ti.com (mailing list archive)
State New, archived
Headers show
Series arm64: dts: ti: k3-j721e/j7200: Introduce EVM compatible | expand

Commit Message

Nishanth Menon Sept. 25, 2021, 8:14 p.m. UTC
Maintain consistency in K3 SoCs by allowing AM654 and AM642 platforms
just state SoC compatibles without specific board specific compatibles
aligned with what we have done for J721E/J7200 platforms as well.

Signed-off-by: Nishanth Menon <nm@ti.com>
---

Motivation to introduce in V2:
* Retain consistency with both "soc only" and "board specific" as two
  schemes we support across k3.

V1:  did'nt exist

 .../devicetree/bindings/arm/ti/k3.yaml        | 23 +++++++++++--------
 1 file changed, 14 insertions(+), 9 deletions(-)

Comments

Suman Anna Sept. 27, 2021, 2:26 p.m. UTC | #1
On 9/25/21 3:14 PM, Nishanth Menon wrote:
> Maintain consistency in K3 SoCs by allowing AM654 and AM642 platforms
> just state SoC compatibles without specific board specific compatibles
> aligned with what we have done for J721E/J7200 platforms as well.
> 
> Signed-off-by: Nishanth Menon <nm@ti.com>

Acked-by: Suman Anna <s-anna@ti.com>

> ---
> 
> Motivation to introduce in V2:
> * Retain consistency with both "soc only" and "board specific" as two
>   schemes we support across k3.

Thanks for the updates and making it consistent across the K3 platforms.

regards
Suman

> 
> V1:  did'nt exist
> 
>  .../devicetree/bindings/arm/ti/k3.yaml        | 23 +++++++++++--------
>  1 file changed, 14 insertions(+), 9 deletions(-)
> 
> diff --git a/Documentation/devicetree/bindings/arm/ti/k3.yaml b/Documentation/devicetree/bindings/arm/ti/k3.yaml
> index cefb06424a4a..30c3f5c70014 100644
> --- a/Documentation/devicetree/bindings/arm/ti/k3.yaml
> +++ b/Documentation/devicetree/bindings/arm/ti/k3.yaml
> @@ -20,12 +20,15 @@ properties:
>      oneOf:
>  
>        - description: K3 AM654 SoC
> -        items:
> -          - enum:
> -              - ti,am654-evm
> -              - siemens,iot2050-basic
> -              - siemens,iot2050-advanced
> +        oneOf:
>            - const: ti,am654
> +          - items:
> +              - enum:
> +                  - ti,am654-evm
> +                  - siemens,iot2050-basic
> +                  - siemens,iot2050-advanced
> +              - const: ti,am654
> +
>  
>        - description: K3 J721E SoC
>          oneOf:
> @@ -44,11 +47,13 @@ properties:
>                - const: ti,j7200
>  
>        - description: K3 AM642 SoC
> -        items:
> -          - enum:
> -              - ti,am642-evm
> -              - ti,am642-sk
> +        oneOf:
>            - const: ti,am642
> +          - items:
> +              - enum:
> +                  - ti,am642-evm
> +                  - ti,am642-sk
> +              - const: ti,am642
>  
>  additionalProperties: true
>  
>
Rob Herring (Arm) Oct. 4, 2021, 5:54 p.m. UTC | #2
On Sat, Sep 25, 2021 at 03:14:28PM -0500, Nishanth Menon wrote:
> Maintain consistency in K3 SoCs by allowing AM654 and AM642 platforms
> just state SoC compatibles without specific board specific compatibles
> aligned with what we have done for J721E/J7200 platforms as well.

This is the wrong direction IMO. Why do you want this other than 
alignment?

> 
> Signed-off-by: Nishanth Menon <nm@ti.com>
> ---
> 
> Motivation to introduce in V2:
> * Retain consistency with both "soc only" and "board specific" as two
>   schemes we support across k3.
> 
> V1:  did'nt exist
> 
>  .../devicetree/bindings/arm/ti/k3.yaml        | 23 +++++++++++--------
>  1 file changed, 14 insertions(+), 9 deletions(-)
> 
> diff --git a/Documentation/devicetree/bindings/arm/ti/k3.yaml b/Documentation/devicetree/bindings/arm/ti/k3.yaml
> index cefb06424a4a..30c3f5c70014 100644
> --- a/Documentation/devicetree/bindings/arm/ti/k3.yaml
> +++ b/Documentation/devicetree/bindings/arm/ti/k3.yaml
> @@ -20,12 +20,15 @@ properties:
>      oneOf:
>  
>        - description: K3 AM654 SoC
> -        items:
> -          - enum:
> -              - ti,am654-evm
> -              - siemens,iot2050-basic
> -              - siemens,iot2050-advanced
> +        oneOf:
>            - const: ti,am654
> +          - items:
> +              - enum:
> +                  - ti,am654-evm
> +                  - siemens,iot2050-basic
> +                  - siemens,iot2050-advanced
> +              - const: ti,am654
> +
>  
>        - description: K3 J721E SoC
>          oneOf:
> @@ -44,11 +47,13 @@ properties:
>                - const: ti,j7200
>  
>        - description: K3 AM642 SoC
> -        items:
> -          - enum:
> -              - ti,am642-evm
> -              - ti,am642-sk
> +        oneOf:
>            - const: ti,am642
> +          - items:
> +              - enum:
> +                  - ti,am642-evm
> +                  - ti,am642-sk
> +              - const: ti,am642
>  
>  additionalProperties: true
>  
> -- 
> 2.32.0
> 
>
Nishanth Menon Oct. 4, 2021, 6:59 p.m. UTC | #3
On 12:54-20211004, Rob Herring wrote:
> On Sat, Sep 25, 2021 at 03:14:28PM -0500, Nishanth Menon wrote:
> > Maintain consistency in K3 SoCs by allowing AM654 and AM642 platforms
> > just state SoC compatibles without specific board specific compatibles
> > aligned with what we have done for J721E/J7200 platforms as well.
> 
> This is the wrong direction IMO. Why do you want this other than 
> alignment?

Many downstream boards tend not to have an specific compatible at least
during initial phase and I would like folks to start using checks to
make sure that the easy to catch issues via match against bindings are
already handled.

I am curious as to why you think this is wrong - because we permit an
alternative option that allows the board files to be less specific?

[...]
Nishanth Menon Oct. 5, 2021, 6:31 p.m. UTC | #4
On 13:59-20211004, Nishanth Menon wrote:
> On 12:54-20211004, Rob Herring wrote:
> > On Sat, Sep 25, 2021 at 03:14:28PM -0500, Nishanth Menon wrote:
> > > Maintain consistency in K3 SoCs by allowing AM654 and AM642 platforms
> > > just state SoC compatibles without specific board specific compatibles
> > > aligned with what we have done for J721E/J7200 platforms as well.
> > 
> > This is the wrong direction IMO. Why do you want this other than 
> > alignment?
> 
> Many downstream boards tend not to have an specific compatible at least
> during initial phase and I would like folks to start using checks to
> make sure that the easy to catch issues via match against bindings are
> already handled.
> 
> I am curious as to why you think this is wrong - because we permit an
> alternative option that allows the board files to be less specific?


Thinking again, I get the rationale. We are attempting to be specific,
and this patch reverses the direction. Agreed. Will drop applying this
patch. Also, for future SoCs, will insist on being specific compatible.
diff mbox series

Patch

diff --git a/Documentation/devicetree/bindings/arm/ti/k3.yaml b/Documentation/devicetree/bindings/arm/ti/k3.yaml
index cefb06424a4a..30c3f5c70014 100644
--- a/Documentation/devicetree/bindings/arm/ti/k3.yaml
+++ b/Documentation/devicetree/bindings/arm/ti/k3.yaml
@@ -20,12 +20,15 @@  properties:
     oneOf:
 
       - description: K3 AM654 SoC
-        items:
-          - enum:
-              - ti,am654-evm
-              - siemens,iot2050-basic
-              - siemens,iot2050-advanced
+        oneOf:
           - const: ti,am654
+          - items:
+              - enum:
+                  - ti,am654-evm
+                  - siemens,iot2050-basic
+                  - siemens,iot2050-advanced
+              - const: ti,am654
+
 
       - description: K3 J721E SoC
         oneOf:
@@ -44,11 +47,13 @@  properties:
               - const: ti,j7200
 
       - description: K3 AM642 SoC
-        items:
-          - enum:
-              - ti,am642-evm
-              - ti,am642-sk
+        oneOf:
           - const: ti,am642
+          - items:
+              - enum:
+                  - ti,am642-evm
+                  - ti,am642-sk
+              - const: ti,am642
 
 additionalProperties: true