2023-11-16 17:39:26

by Rob Herring (Arm)

[permalink] [raw]
Subject: Re: [PATCH v4 1/5] dt-bindings: interrupt-controller: loongson,liointc: Standardize the naming of 'loongson,parent-int-map'

On Mon, Nov 13, 2023 at 10:36:21AM +0800, Binbin Zhou wrote:
> Since the 'loongson,parent_int_map' attribute naming is non-standard, we
> should use 'loongson,parent-int-map' instead.
> Also, 'loongson,parent_int_map' will be marked as deprecated.

Again, as I said last time, it is not worth adding
'loongson,parent-int-map' and having to support both forms in the kernel
forever. If you can define a time that we keep 'loongson,parent_int_map'
that's less than "forever", then may be it would be worth changing.

To put this another way, using '-' instead of '_' is for *new* bindings,
not existing users.

Rob


2023-11-17 09:07:14

by Binbin Zhou

[permalink] [raw]
Subject: Re: [PATCH v4 1/5] dt-bindings: interrupt-controller: loongson,liointc: Standardize the naming of 'loongson,parent-int-map'

On Thu, Nov 16, 2023 at 11:39 PM Rob Herring <[email protected]> wrote:
>
> On Mon, Nov 13, 2023 at 10:36:21AM +0800, Binbin Zhou wrote:
> > Since the 'loongson,parent_int_map' attribute naming is non-standard, we
> > should use 'loongson,parent-int-map' instead.
> > Also, 'loongson,parent_int_map' will be marked as deprecated.
>
> Again, as I said last time, it is not worth adding
> 'loongson,parent-int-map' and having to support both forms in the kernel
> forever. If you can define a time that we keep 'loongson,parent_int_map'
> that's less than "forever", then may be it would be worth changing.
>
> To put this another way, using '-' instead of '_' is for *new* bindings,
> not existing users.
>
OK, I see, and I will fix it in the next version.

Thanks.
Binbin

> Rob