Re: [PATCH v2 1/2] dt-bindings: interrupt-controller: apple,aic2: Add AICv3

[Sven Peter]

On 05.02.26 23:19, Janne Grunau wrote:
> On Thu, Feb 05, 2026 at 07:01:51PM +0100, Sven Peter wrote:
> > Hi,
> >
> > On 05.02.26 13:50, Krzysztof Kozlowski wrote:
> > > On Wed, Jan 28, 2026 at 09:57:08AM +0100, Janne Grunau wrote:
> > > > AIC version 3 as found on the Apple M3 (t8122) is very similar to AICv2
> > > > in its base functionality. It can use the same device tree bindings as
> > > > AICv2 so add it to the AICv2 bindings.
> > > > This interrupt controller is used on all Apple SoCs starting with M3 up
> > > > to at least M5.
> > > > The only apparent difference is the increased IRQ config offset. Apple's
> > > > device tree codes this new offset as property of the "aic" node but the
> > > > value stayed constant for all SoCs with "aic,3". Since the SoC specific
> > > > compatible "apple,t8122-aic3" will be used in the driver this offset can
> > > > remain a driver implementation detail.
> > > >
> > > > Signed-off-by: Janne Grunau <j@xxxxxxxxxx>
> > > > ---
> > > >    .../bindings/interrupt-controller/apple,aic2.yaml  | 45 +++++++++++++++++-----
> > > >    1 file changed, 36 insertions(+), 9 deletions(-)
> > > >
> > > > diff --git a/Documentation/devicetree/bindings/interrupt-controller/apple,aic2.yaml b/Documentation/devicetree/bindings/interrupt-controller/apple,aic2.yaml
> > > > index ee5a0dfff437816056bda0de5523bf38be4f49ba..a6e2251fcc111340c0a27ab6912452f6b1255be2 100644
> > > > --- a/Documentation/devicetree/bindings/interrupt-controller/apple,aic2.yaml
> > > > +++ b/Documentation/devicetree/bindings/interrupt-controller/apple,aic2.yaml
> > > > @@ -4,10 +4,10 @@
> > > >    $id: http://devicetree.org/schemas/interrupt-controller/apple,aic2.yaml#
> > > >    $schema: http://devicetree.org/meta-schemas/core.yaml#
> > > >    
> > > > -title: Apple Interrupt Controller 2
> > > > +title: Apple Interrupt Controller 2 and 3
> > > >    
> > > >    maintainers:
> > > > -  - Hector Martin <marcan@xxxxxxxxx>
> > > > +  - Janne Grunau <j@xxxxxxxxxx>
> > > >    
> > > >    description: |
> > > >      The Apple Interrupt Controller 2 is a simple interrupt controller present on
> > > > @@ -28,14 +28,24 @@ description: |
> > > >      which do not go through a discrete interrupt controller. It also handles
> > > >      FIQ-based Fast IPIs.
> > > >    
> > > > +  The Apple Interrupt Controller 3 is in its base functionality very similar to
> > > > +  the Apple Interrupt Controller 2 and uses the same device tree bindings. It is
> > > > +  found on Apple ARM SoCs platforms starting with t8122 (M3).
> > > > +
> > > >    properties:
> > > >      compatible:
> > > > -    items:
> > > > -      - enum:
> > > > -          - apple,t8112-aic
> > > > -          - apple,t6000-aic
> > > > -          - apple,t6020-aic
> > > > -      - const: apple,aic2
> > > > +    oneOf:
> > > > +      - items:
> > > > +          - enum:
> > > > +              - apple,t8112-aic
> > > > +              - apple,t6000-aic
> > > > +              - apple,t6020-aic
> > >
> > > If you are re-shuffling these, you can as well sort alphanumerically.
> >
> > I don't have a strong opinion here but the last time this came up we
> > agreed that we'd go with release order IIRC for consistency since that
> > order is used in every other binding for this platform.
>
> The intended scheme as documented in some commit messages is first base
> Mx SoCs in release order and then Mx Pro/Max/Ultra SoCs in release
> order. This scheme isn't used consistently though. I think I fixed that
> in one or two bindings already but there are other using a different
> order.
> I think these unecessarily complex rules are guaranteed to lead to
> inconsistencies. I'll resort the entries alphabetically as that's the
> only rule with a chance of consistent use.

Sure, I'm totally fine with just using alphanumeric order everywhere as 
well.


Sven