The SBI firmware might want to communicate to the AON firmware too.
Add a mbox-name item to allow to allocate a mailbox for SBI.
Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
---
.../devicetree/bindings/firmware/thead,th1520-aon.yaml | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/Documentation/devicetree/bindings/firmware/thead,th1520-aon.yaml b/Documentation/devicetree/bindings/firmware/thead,th1520-aon.yaml
index 3365124c7fd47..555465f4aab4e 100644
--- a/Documentation/devicetree/bindings/firmware/thead,th1520-aon.yaml
+++ b/Documentation/devicetree/bindings/firmware/thead,th1520-aon.yaml
@@ -26,11 +26,12 @@ properties:
const: thead,th1520-aon
mboxes:
- maxItems: 1
+ maxItems: 2
mbox-names:
items:
- const: aon
+ - const: aon-for-sbi
resets:
maxItems: 1
@@ -54,7 +55,7 @@ examples:
- |
aon: aon {
compatible = "thead,th1520-aon";
- mboxes = <&mbox_910t 1>;
- mbox-names = "aon";
+ mboxes = <&mbox_910t 1>, <&mbox_910r 1>;
+ mbox-names = "aon", "aon-for-sbi";
#power-domain-cells = <1>;
};
--
2.50.1
On 14/08/2025 09:07, Icenowy Zheng wrote:
> The SBI firmware might want to communicate to the AON firmware too.
>
> Add a mbox-name item to allow to allocate a mailbox for SBI.
>
> Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
> ---
> .../devicetree/bindings/firmware/thead,th1520-aon.yaml | 7 ++++---
> 1 file changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/Documentation/devicetree/bindings/firmware/thead,th1520-aon.yaml b/Documentation/devicetree/bindings/firmware/thead,th1520-aon.yaml
> index 3365124c7fd47..555465f4aab4e 100644
> --- a/Documentation/devicetree/bindings/firmware/thead,th1520-aon.yaml
> +++ b/Documentation/devicetree/bindings/firmware/thead,th1520-aon.yaml
> @@ -26,11 +26,12 @@ properties:
> const: thead,th1520-aon
>
> mboxes:
> - maxItems: 1
> + maxItems: 2
ABI break without explanation why ("allow" is not a reason to affect
ABI) and its impact.
Best regards,
Krzysztof
在 2025-08-14星期四的 09:18 +0200,Krzysztof Kozlowski写道:
> On 14/08/2025 09:07, Icenowy Zheng wrote:
> > The SBI firmware might want to communicate to the AON firmware too.
> >
> > Add a mbox-name item to allow to allocate a mailbox for SBI.
> >
> > Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
> > ---
> > .../devicetree/bindings/firmware/thead,th1520-aon.yaml | 7
> > ++++---
> > 1 file changed, 4 insertions(+), 3 deletions(-)
> >
> > diff --git
> > a/Documentation/devicetree/bindings/firmware/thead,th1520-aon.yaml
> > b/Documentation/devicetree/bindings/firmware/thead,th1520-aon.yaml
> > index 3365124c7fd47..555465f4aab4e 100644
> > --- a/Documentation/devicetree/bindings/firmware/thead,th1520-
> > aon.yaml
> > +++ b/Documentation/devicetree/bindings/firmware/thead,th1520-
> > aon.yaml
> > @@ -26,11 +26,12 @@ properties:
> > const: thead,th1520-aon
> >
> > mboxes:
> > - maxItems: 1
> > + maxItems: 2
>
>
> ABI break without explanation why ("allow" is not a reason to affect
> ABI) and its impact.
Is adding items an ABI break?
Or should I explicitly say "minItems: 1" here?
>
>
> Best regards,
> Krzysztof
On 14/08/2025 09:30, Icenowy Zheng wrote:
> 在 2025-08-14星期四的 09:18 +0200,Krzysztof Kozlowski写道:
>> On 14/08/2025 09:07, Icenowy Zheng wrote:
>>> The SBI firmware might want to communicate to the AON firmware too.
>>>
>>> Add a mbox-name item to allow to allocate a mailbox for SBI.
>>>
>>> Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
>>> ---
>>> .../devicetree/bindings/firmware/thead,th1520-aon.yaml | 7
>>> ++++---
>>> 1 file changed, 4 insertions(+), 3 deletions(-)
>>>
>>> diff --git
>>> a/Documentation/devicetree/bindings/firmware/thead,th1520-aon.yaml
>>> b/Documentation/devicetree/bindings/firmware/thead,th1520-aon.yaml
>>> index 3365124c7fd47..555465f4aab4e 100644
>>> --- a/Documentation/devicetree/bindings/firmware/thead,th1520-
>>> aon.yaml
>>> +++ b/Documentation/devicetree/bindings/firmware/thead,th1520-
>>> aon.yaml
>>> @@ -26,11 +26,12 @@ properties:
>>> const: thead,th1520-aon
>>>
>>> mboxes:
>>> - maxItems: 1
>>> + maxItems: 2
>>
>>
>> ABI break without explanation why ("allow" is not a reason to affect
>> ABI) and its impact.
>
> Is adding items an ABI break?
Adding required items is ABI break. You can easily test it. Apply patch
#1 and test your DTS. Apply patch #2 and test your DTS. New warnings
appear, so that's a proof of ABI impact.
>
> Or should I explicitly say "minItems: 1" here?
Yes, but you should clearly explain the impact. Is it working? Not
working? Are you fixing something?
Best regards,
Krzysztof
在 2025-08-14星期四的 09:49 +0200,Krzysztof Kozlowski写道:
> On 14/08/2025 09:30, Icenowy Zheng wrote:
> > 在 2025-08-14星期四的 09:18 +0200,Krzysztof Kozlowski写道:
> > > On 14/08/2025 09:07, Icenowy Zheng wrote:
> > > > The SBI firmware might want to communicate to the AON firmware
> > > > too.
> > > >
> > > > Add a mbox-name item to allow to allocate a mailbox for SBI.
> > > >
> > > > Signed-off-by: Icenowy Zheng <uwu@icenowy.me>
> > > > ---
> > > > .../devicetree/bindings/firmware/thead,th1520-aon.yaml | 7
> > > > ++++---
> > > > 1 file changed, 4 insertions(+), 3 deletions(-)
> > > >
> > > > diff --git
> > > > a/Documentation/devicetree/bindings/firmware/thead,th1520-
> > > > aon.yaml
> > > > b/Documentation/devicetree/bindings/firmware/thead,th1520-
> > > > aon.yaml
> > > > index 3365124c7fd47..555465f4aab4e 100644
> > > > --- a/Documentation/devicetree/bindings/firmware/thead,th1520-
> > > > aon.yaml
> > > > +++ b/Documentation/devicetree/bindings/firmware/thead,th1520-
> > > > aon.yaml
> > > > @@ -26,11 +26,12 @@ properties:
> > > > const: thead,th1520-aon
> > > >
> > > > mboxes:
> > > > - maxItems: 1
> > > > + maxItems: 2
> > >
> > >
> > > ABI break without explanation why ("allow" is not a reason to
> > > affect
> > > ABI) and its impact.
> >
> > Is adding items an ABI break?
>
> Adding required items is ABI break. You can easily test it. Apply
> patch
> #1 and test your DTS. Apply patch #2 and test your DTS. New warnings
> appear, so that's a proof of ABI impact.
Ah sorry I don't mean that item is required.
>
>
> >
> > Or should I explicitly say "minItems: 1" here?
>
> Yes, but you should clearly explain the impact. Is it working? Not
> working? Are you fixing something?
The jsonschema draft says "Omitting this keyword has the same behavior
as a value of 0." for minItems. [1]
[1]
https://json-schema.org/draft/2020-12/json-schema-validation#section-6.4.2-3
>
> Best regards,
> Krzysztof
On 14/08/2025 10:04, Icenowy Zheng wrote: >> >>> >>> Or should I explicitly say "minItems: 1" here? >> >> Yes, but you should clearly explain the impact. Is it working? Not >> working? Are you fixing something? > > The jsonschema draft says "Omitting this keyword has the same behavior > as a value of 0." for minItems. [1] No, omitting this means it is implied by maxItems, see fixups.py:130. Best regards, Krzysztof
© 2016 - 2026 Red Hat, Inc.