On Wed, Feb 01, 2023 at 08:23:48AM +0100, Krzysztof Kozlowski wrote:
> On 01/02/2023 02:57, Saurabh Singh Sengar wrote:
> > On Tue, Jan 31, 2023 at 07:54:56PM +0100, Krzysztof Kozlowski wrote:
> >> On 31/01/2023 19:10, Saurabh Sengar wrote:
> >>>
> >>> Signed-off-by: Saurabh Sengar <ssengar@xxxxxxxxxxxxxxxxxxx>
> >>
> >>> + - |
> >>> + / {
> >>> + compatible = "foo";
> >>> + model = "foo";
> >>> + #address-cells = <0x02>;
> >>> + #size-cells = <0x02>;
> >>
> >> Except previous comments (all of them were ignored),
> >
> > Thank you for your comments, I have tried to address them all in this version
> > but I may have missed few. I will go and look again all the emails, but if
> > there is any thing which you can point again and we can start a new dicussion
> > from here will be very helpful.
> >
> > I think one concern was related to use of 'reg' or 'ranges', and I
> > thought this patch will give clarity that I intend to use 'ranges'
> > without any child node. Is this acceptable ?
>
> There were several comments.
Ok, let me reply to all of your comments from previous thread to avoid
any confusion. Hope this is fine.
>
> >
> >
> >> also:
> >> Drop entire part. Not related, not correct, not helping and you cannot
> >> have top level nodes in example.
> >
> > If I dont use the top level node, the parent address cells are assumed to be 1,
> > and I get below warning. If there is better way to address this warning, please
> > suggest I will work on it.
> >
> > Warning (ranges_format): /example-0/vmbus@ff0000000:ranges: "ranges" property has invalid length (20 bytes) (parent #address-cells == 1, child #address-cells == 2, #size-cells == 1)
>
> Then use soc just like every other example.
Thanks for suggestion, I will fix this in V3
Regards,
Saurabh
>
>
> Best regards,
> Krzysztof
