Hello!
On 10/08/2018 07:40 PM, Geert Uytterhoeven wrote:
>>>> Describe THS/CIVM in the R8A77970 device trees.
>>>> Based on the original (and large) patches by Vladimir Barinov.
>>>>
>>>> Signed-off-by: Vladimir Barinov <vladimir.barinov@xxxxxxxxxxxxxxxxxx>
>>>> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@xxxxxxxxxxxxxxxxxx>
>>>>
>>>> ---
>>>> This patch is against the 'renesas-devel-20181004-v4.19-rc6' tag of Simon
>>>> Horman's 'renesas.git' repo.
>>>>
>>>> The thermal driver/bindings patches have been posted yesterday...
>>>>
>>>> arch/arm64/boot/dts/renesas/r8a77970.dtsi | 32 ++++++++++++++++++++++++++++++
>>>> 1 file changed, 32 insertions(+)
>>>>
>>>> Index: renesas/arch/arm64/boot/dts/renesas/r8a77970.dtsi
>>>> ===================================================================
>>>> --- renesas.orig/arch/arm64/boot/dts/renesas/r8a77970.dtsi
>>>> +++ renesas/arch/arm64/boot/dts/renesas/r8a77970.dtsi
>>>> @@ -300,6 +300,19 @@
>>>> #power-domain-cells = <1>;
>>>> };
>>>>
>>>> + thermal: thermal@e6190000 {
>>>> + compatible = "renesas,thermal-r8a77970";
>>>> + reg = <0 0xe6190000 0 0x14
>>>
>>> 0x14 was appropriate for R-Mobile APE6...
>>
>> Copy&paste is to blame here, I guess... I'll fix to 0x10.
>
> OK.
>
>>>> + 0 0xe6190100 0 0x38>;
>>>
>>> What about the CIVM status register? DT describes hardware, not driver
>>> limitations.
>>
>> I wasn't sure whether to put it into a separate "reg" tuple (which would confuse
>> the driver) or not. After looking into the manual again, I'm going to extend the
>> 2nd "reg" tuple...
>
> The CIVM Status Register indicates the chip internal voltage.
> As such it's not a per-channel property, and IMHO doesn't belong in the second
> tuple
I was looking a the block diagrams (both in the chapters 10A and 10B) and I got a totally different impression...
> (e.g. R-Mobile APE6 has 3 channels).
The driver handles that alright. It's the adding the CIVM as a separate tuple
that would break it.
> Perhaps extending the bindings to handle more reg tuples, possibly using
> reg-names?
You mean teaching the driver about one more kind of "reg" tuples? I would like
to avoid that of course -- due to the need to still handle the old DTs as well...
> Gr{oetje,eeting}s,
>
> Geert
>
MBR, Sergei
