[PATCH 1/2] arm64: dts: juno/vexpress: fix node name unit-address presence warnings

[Rob Herring]

On Fri, Mar 18, 2016 at 5:32 AM, Jon Medhurst (Tixy) <tixy at linaro.org> wrote:
> On Thu, 2016-03-17 at 16:44 -0500, Rob Herring wrote:
>> On Mon, Mar 7, 2016 at 6:12 AM, Sudeep Holla <sudeep.holla at arm.com> wrote:
>> > Commit fa38a82096a1 ("scripts/dtc: Update to upstream version
>> > 53bf130b1cdd") added warnings on node name unit-address presence/absence
>> > mismatch in device trees.
>> >
>> > This patch fixes those warning on all the juno/vexpress platforms where
>> > unit-address is present in node name while the reg/ranges property is
>> > not present.
>> >
>> > Signed-off-by: Sudeep Holla <sudeep.holla at arm.com>
>> > ---
>> >  arch/arm64/boot/dts/arm/foundation-v8.dtsi         |  2 +-
>> >  arch/arm64/boot/dts/arm/juno-base.dtsi             | 14 +++++-----
>> >  arch/arm64/boot/dts/arm/juno-motherboard.dtsi      | 30 +++++++++++-----------
>> >  arch/arm64/boot/dts/arm/rtsm_ve-aemv8a.dts         |  2 +-
>> >  arch/arm64/boot/dts/arm/rtsm_ve-motherboard.dtsi   | 14 +++++-----
>> >  .../boot/dts/arm/vexpress-v2f-1xv7-ca53x2.dts      | 10 ++++----
>> >  6 files changed, 36 insertions(+), 36 deletions(-)
>> >
>> > Hi Rob,
>> >
>> > I am not sure if the intention of commit fa38a82096a1 is to fix even
>> > existing warnings or not. I saw warning with today's -next branch.
>>
>> Yes. I'm dropping it for 4.6 though.
>>
>> > Let me know.
>> >
>> > Regards,
>> > Sudeep
>> >
>> > diff --git a/arch/arm64/boot/dts/arm/foundation-v8.dtsi b/arch/arm64/boot/dts/arm/foundation-v8.dtsi
>> > index 9314f3943269..ba03c934a861 100644
>> > --- a/arch/arm64/boot/dts/arm/foundation-v8.dtsi
>> > +++ b/arch/arm64/boot/dts/arm/foundation-v8.dtsi
>> > @@ -89,7 +89,7 @@
>> >                              <0 63 4>;
>> >         };
>> >
>> > -       smb {
>> > +       smb at 0,08000000 {
>>
>> Drop the comma and leading zeros. Comma is for distinct fields like
>> bus,dev,func on PCI.
>
> So what is the correct way of representing 2 cell addresses? Not just
> those that are logically 64-bit addresses, (which I guess could be just
> a single number), but things like vexpress motherboard where we have
> Chip-select and Offset values?

64-bit address is just a single number.
2 cells with distinct fields should generally be ...@<cell1>,<cell2>

If you have your own bus type, you can devise whatever you like as
long as it is reviewed.

> I looked at ePAPR but it just says 'The unit-address must
> match the first address specified in the reg property of the node.' and
> lists valid characters for unit address, which includes ,._+-

Yes, we need to clarify this some.

Rob