[PATCH v1] dt-bindings: arm: stm32: lxa,stm32mp157c-mc1: add extra SiP compatible

[PATCH v1] dt-bindings: arm: stm32: lxa,stm32mp157c-mc1: add extra SiP compatible

[Ahmad Fatoum]

The Linux Automation MC-1 is built around a SIP with CPU, RAM, PMIC,
Oscillator and EEPROM. Add a further compatible identifying the SiP,
so boot firmware can match against it to apply fixups if necessary.

To avoid intermittent dtbs_check breakage in the sole upstream device
tree that uses that SiP, patch it here as well.

Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
---
I know that bindings and device tree patches should be separate. Does
this apply here as well? Should I split the dts change into a follow-up
commit? Is it ok that dtbs_check will report an intermittent breakage?
---
 Documentation/devicetree/bindings/arm/stm32/stm32.yaml | 8 +++++++-
 arch/arm/boot/dts/stm32mp157c-lxa-mc1.dts              | 2 +-
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/Documentation/devicetree/bindings/arm/stm32/stm32.yaml b/Documentation/devicetree/bindings/arm/stm32/stm32.yaml
index 009b424e456e..9127094f0208 100644
--- a/Documentation/devicetree/bindings/arm/stm32/stm32.yaml
+++ b/Documentation/devicetree/bindings/arm/stm32/stm32.yaml
@@ -40,7 +40,6 @@ properties:
       - items:
           - enum:
               - arrow,stm32mp157a-avenger96 # Avenger96
-              - lxa,stm32mp157c-mc1
               - shiratech,stm32mp157a-iot-box # IoT Box
               - shiratech,stm32mp157a-stinger96 # Stinger96
               - st,stm32mp157c-ed1
@@ -52,6 +51,13 @@ properties:
           - const: st,stm32mp157c-ev1
           - const: st,stm32mp157c-ed1
           - const: st,stm32mp157
+      - description: Octavo OSD32MP15x System-in-Package based boards
+        items:
+          - enum:
+              - lxa,stm32mp157c-mc1 # Linux Automation MC-1
+          - const: oct,stm32mp15xx-osd32
+          - enum:
+              - st,stm32mp157
       - description: Odyssey STM32MP1 SoM based Boards
         items:
           - enum:
diff --git a/arch/arm/boot/dts/stm32mp157c-lxa-mc1.dts b/arch/arm/boot/dts/stm32mp157c-lxa-mc1.dts
index 1e5333fd437f..cda8e871f999 100644
--- a/arch/arm/boot/dts/stm32mp157c-lxa-mc1.dts
+++ b/arch/arm/boot/dts/stm32mp157c-lxa-mc1.dts
@@ -15,7 +15,7 @@
 
 / {
 	model = "Linux Automation MC-1 board";
-	compatible = "lxa,stm32mp157c-mc1", "st,stm32mp157";
+	compatible = "lxa,stm32mp157c-mc1", "oct,stm32mp15xx-osd32", "st,stm32mp157";
 
 	aliases {
 		ethernet0 = &ethernet0;
-- 
2.28.0

Re: [PATCH v1] dt-bindings: arm: stm32: lxa,stm32mp157c-mc1: add extra SiP compatible

[Rob Herring]

On Wed, Nov 04, 2020 at 12:39:31PM +0100, Ahmad Fatoum wrote:
> The Linux Automation MC-1 is built around a SIP with CPU, RAM, PMIC,
> Oscillator and EEPROM. Add a further compatible identifying the SiP,
> so boot firmware can match against it to apply fixups if necessary.
> 
> To avoid intermittent dtbs_check breakage in the sole upstream device
> tree that uses that SiP, patch it here as well.
> 
> Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
> ---
> I know that bindings and device tree patches should be separate. Does
> this apply here as well? Should I split the dts change into a follow-up
> commit? 

Yes.

> Is it ok that dtbs_check will report an intermittent breakage?

If the binding comes first, it won't break.

But generally, 'dtbs_check' being warning free is not yet a requirement. 
That will probably first have to be per platform.

> ---
>  Documentation/devicetree/bindings/arm/stm32/stm32.yaml | 8 +++++++-
>  arch/arm/boot/dts/stm32mp157c-lxa-mc1.dts              | 2 +-
>  2 files changed, 8 insertions(+), 2 deletions(-)
> 
> diff --git a/Documentation/devicetree/bindings/arm/stm32/stm32.yaml b/Documentation/devicetree/bindings/arm/stm32/stm32.yaml
> index 009b424e456e..9127094f0208 100644
> --- a/Documentation/devicetree/bindings/arm/stm32/stm32.yaml
> +++ b/Documentation/devicetree/bindings/arm/stm32/stm32.yaml
> @@ -40,7 +40,6 @@ properties:
>        - items:
>            - enum:
>                - arrow,stm32mp157a-avenger96 # Avenger96
> -              - lxa,stm32mp157c-mc1
>                - shiratech,stm32mp157a-iot-box # IoT Box
>                - shiratech,stm32mp157a-stinger96 # Stinger96
>                - st,stm32mp157c-ed1
> @@ -52,6 +51,13 @@ properties:
>            - const: st,stm32mp157c-ev1
>            - const: st,stm32mp157c-ed1
>            - const: st,stm32mp157
> +      - description: Octavo OSD32MP15x System-in-Package based boards
> +        items:
> +          - enum:
> +              - lxa,stm32mp157c-mc1 # Linux Automation MC-1
> +          - const: oct,stm32mp15xx-osd32

'oct' is not docuemnted in vendor-prefixes.yaml.

> +          - enum:
> +              - st,stm32mp157
>        - description: Odyssey STM32MP1 SoM based Boards
>          items:
>            - enum:
> diff --git a/arch/arm/boot/dts/stm32mp157c-lxa-mc1.dts b/arch/arm/boot/dts/stm32mp157c-lxa-mc1.dts
> index 1e5333fd437f..cda8e871f999 100644
> --- a/arch/arm/boot/dts/stm32mp157c-lxa-mc1.dts
> +++ b/arch/arm/boot/dts/stm32mp157c-lxa-mc1.dts
> @@ -15,7 +15,7 @@
>  
>  / {
>  	model = "Linux Automation MC-1 board";
> -	compatible = "lxa,stm32mp157c-mc1", "st,stm32mp157";
> +	compatible = "lxa,stm32mp157c-mc1", "oct,stm32mp15xx-osd32", "st,stm32mp157";
>  
>  	aliases {
>  		ethernet0 = &ethernet0;
> -- 
> 2.28.0
> 

Re: [PATCH v1] dt-bindings: arm: stm32: lxa,stm32mp157c-mc1: add extra SiP compatible

[Ahmad Fatoum]

Hello Rob,

On 11/9/20 5:37 PM, Rob Herring wrote:
> On Wed, Nov 04, 2020 at 12:39:31PM +0100, Ahmad Fatoum wrote:
>> I know that bindings and device tree patches should be separate. Does
>> this apply here as well? Should I split the dts change into a follow-up
>> commit? 
> 
> Yes.
> 
>> Is it ok that dtbs_check will report an intermittent breakage?
> 
> If the binding comes first, it won't break.
> 
> But generally, 'dtbs_check' being warning free is not yet a requirement. 
> That will probably first have to be per platform.

here the old binding is deleted, so between the patches, there will be
a dtbs_check warning, which is why I asked. I've now split it up with
binding first.

>> +          - const: oct,stm32mp15xx-osd32
> 
> 'oct' is not docuemnted in vendor-prefixes.yaml.

Huh, I only checked whether it's in use, not if it's documented.
I just sent out a v2 with your points addressed.

Thanks for review,
Ahmad

-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |