Re: [PATCH RESEND] arm64: dts: qcom: sm8550: Add support for camss

From: Vladimir Zapolskiy
Date: Tue Jun 10 2025 - 18:52:49 EST




On 6/11/25 01:17, Bryan O'Donoghue wrote:
On 10/06/2025 22:13, Vladimir Zapolskiy wrote:
Hi Konrad.

On 6/11/25 00:04, Konrad Dybcio wrote:
On 6/10/25 11:02 PM, Vladimir Zapolskiy wrote:
On 6/10/25 22:02, Bryan O'Donoghue wrote:
On 10/06/2025 13:49, Vladimir Zapolskiy wrote:

List feedback from DT people is isp@ is the correct prefix.


My bad, but I don't understand this comment, it seems irrelevant...

The expressed concern is about the novel label name.

I mean to say the feedback from Krzysztof was that we should use isp@
not camss@ and I agree.


Let me repeat it thrice, it's okay...

I don't object against the properly selected device tree node name
"isp",
here I object against a never used and very questionable label name
"isp".

Please feel free to ask more questions, if you still find it confusing.

Again, I may missed a discussion about the need to get and use a novel
label name, then please share a link to it, it'll be very much
appreciated.

To hopefully help out:

label: node-name@unit-address {
    property = value;
};


Thank you, here is a link to the wanted section of the dt specification
for Bryan's comprehension:

* https://github.com/devicetree-org/devicetree-specification/blob/main/
source/chapter6-source-language.rst.

If for whatever reason a proposed "isp" label is preferred, then
since a label rename is not an ABI change, it would make sense to
do a massive change of renaming all camss labels. Otherwise there will
be an outstanding incorrespondence/confusion of the label names in
board .dts files, and that's bad.

--
Best wishes,
Vladimir

Ah the label, I thought you meant node.

I'm trying to do my best in expressing myself by means of the second
signaling system. As an example when I write "a label" repeatedly, I mean
to transmit "a label" symbol, hence I hope it should not be overly
complicated to understand me.

It's great that the understanding is reached now, it would be better,
if we can save some time in future.

There is no bug introduced in this particular change, however it shall be
fixed and be resubmitted.

--
Best wishes,
Vladimir