[PATCH] dt-bindings: interconnect: qcom: Do not require reg for sc8180x virt NoCs

[PATCH] dt-bindings: interconnect: qcom: Do not require reg for sc8180x virt NoCs

[djakov]

From: Georgi Djakov <djakov@kernel.org>

The virtual interconnect providers do not have their own IO address space,
but this is not documented in the DT schema and the following warnings are
reported by dtbs_check:

sc8180x-lenovo-flex-5g.dtb: interconnect-camnoc-virt: 'reg' is a required property
sc8180x-lenovo-flex-5g.dtb: interconnect-mc-virt: 'reg' is a required property
sc8180x-lenovo-flex-5g.dtb: interconnect-qup-virt: 'reg' is a required property
sc8180x-primus.dtb: interconnect-camnoc-virt: 'reg' is a required property
sc8180x-primus.dtb: interconnect-mc-virt: 'reg' is a required property
sc8180x-primus.dtb: interconnect-qup-virt: 'reg' is a required property

Fix this by adding them to the list of compatibles that do not require
the reg property.

Signed-off-by: Georgi Djakov <djakov@kernel.org>
---
 Documentation/devicetree/bindings/interconnect/qcom,rpmh.yaml | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/interconnect/qcom,rpmh.yaml b/Documentation/devicetree/bindings/interconnect/qcom,rpmh.yaml
index 9318b845ec35..70f5b2670085 100644
--- a/Documentation/devicetree/bindings/interconnect/qcom,rpmh.yaml
+++ b/Documentation/devicetree/bindings/interconnect/qcom,rpmh.yaml
@@ -113,6 +113,9 @@ allOf:
         properties:
           compatible:
             enum:
+              - qcom,sc8180x-camnoc-virt
+              - qcom,sc8180x-mc-virt
+              - qcom,sc8180x-qup-virt
               - qcom,sdx65-mc-virt
               - qcom,sm8250-qup-virt
     then:

Re: [PATCH] dt-bindings: interconnect: qcom: Do not require reg for sc8180x virt NoCs

[Krzysztof Kozlowski]

On 30/07/2024 16:10, djakov@kernel.org wrote:
> From: Georgi Djakov <djakov@kernel.org>
> 
> The virtual interconnect providers do not have their own IO address space,
> but this is not documented in the DT schema and the following warnings are
> reported by dtbs_check:
> 
> sc8180x-lenovo-flex-5g.dtb: interconnect-camnoc-virt: 'reg' is a required property
> sc8180x-lenovo-flex-5g.dtb: interconnect-mc-virt: 'reg' is a required property
> sc8180x-lenovo-flex-5g.dtb: interconnect-qup-virt: 'reg' is a required property
> sc8180x-primus.dtb: interconnect-camnoc-virt: 'reg' is a required property
> sc8180x-primus.dtb: interconnect-mc-virt: 'reg' is a required property
> sc8180x-primus.dtb: interconnect-qup-virt: 'reg' is a required property
> 
> Fix this by adding them to the list of compatibles that do not require
> the reg property.

So I guess we are giving up on
https://lore.kernel.org/all/20230530162454.51708-4-vkoul@kernel.org/
?

Best regards,
Krzysztof

Re: [PATCH] dt-bindings: interconnect: qcom: Do not require reg for sc8180x virt NoCs

[Georgi Djakov]

On 30.07.24 17:32, Krzysztof Kozlowski wrote:
> On 30/07/2024 16:10, djakov@kernel.org wrote:
>> From: Georgi Djakov <djakov@kernel.org>
>>
>> The virtual interconnect providers do not have their own IO address space,
>> but this is not documented in the DT schema and the following warnings are
>> reported by dtbs_check:
>>
>> sc8180x-lenovo-flex-5g.dtb: interconnect-camnoc-virt: 'reg' is a required property
>> sc8180x-lenovo-flex-5g.dtb: interconnect-mc-virt: 'reg' is a required property
>> sc8180x-lenovo-flex-5g.dtb: interconnect-qup-virt: 'reg' is a required property
>> sc8180x-primus.dtb: interconnect-camnoc-virt: 'reg' is a required property
>> sc8180x-primus.dtb: interconnect-mc-virt: 'reg' is a required property
>> sc8180x-primus.dtb: interconnect-qup-virt: 'reg' is a required property
>>
>> Fix this by adding them to the list of compatibles that do not require
>> the reg property.
> 
> So I guess we are giving up on
> https://lore.kernel.org/all/20230530162454.51708-4-vkoul@kernel.org/
> ?

Thanks for the pointer! That approach is fine too, but i was expecting
a re-send and then later completely forgot about it. I have a slight
preference towards my patch, because it is more compact, but i can also
revive Vinod's patch if you think that it would be a better pattern to
follow in the long term.

BR,
Georgi

Re: [PATCH] dt-bindings: interconnect: qcom: Do not require reg for sc8180x virt NoCs

[Krzysztof Kozlowski]

On 24/08/2024 01:39, Georgi Djakov wrote:
> On 30.07.24 17:32, Krzysztof Kozlowski wrote:
>> On 30/07/2024 16:10, djakov@kernel.org wrote:
>>> From: Georgi Djakov <djakov@kernel.org>
>>>
>>> The virtual interconnect providers do not have their own IO address space,
>>> but this is not documented in the DT schema and the following warnings are
>>> reported by dtbs_check:
>>>
>>> sc8180x-lenovo-flex-5g.dtb: interconnect-camnoc-virt: 'reg' is a required property
>>> sc8180x-lenovo-flex-5g.dtb: interconnect-mc-virt: 'reg' is a required property
>>> sc8180x-lenovo-flex-5g.dtb: interconnect-qup-virt: 'reg' is a required property
>>> sc8180x-primus.dtb: interconnect-camnoc-virt: 'reg' is a required property
>>> sc8180x-primus.dtb: interconnect-mc-virt: 'reg' is a required property
>>> sc8180x-primus.dtb: interconnect-qup-virt: 'reg' is a required property
>>>
>>> Fix this by adding them to the list of compatibles that do not require
>>> the reg property.
>>
>> So I guess we are giving up on
>> https://lore.kernel.org/all/20230530162454.51708-4-vkoul@kernel.org/
>> ?
> 
> Thanks for the pointer! That approach is fine too, but i was expecting
> a re-send and then later completely forgot about it. I have a slight
> preference towards my patch, because it is more compact, but i can also
> revive Vinod's patch if you think that it would be a better pattern to
> follow in the long term.

Vinod sent his patch more than a year ago, so I think we are indeed
giving up on this :)

Best regards,
Krzysztof