On Thu, Oct 17, 2019 at 12:10:15PM -0400, Thara Gopinath wrote: > On 10/17/2019 11:43 AM, Ulf Hansson wrote: > > On Thu, 17 Oct 2019 at 17:28, Thara Gopinath <thara.gopinath@xxxxxxxxxx> wrote: > >> > >> Hello Ulf, > >> Thanks for the review! > >> > >> On 10/17/2019 05:04 AM, Ulf Hansson wrote: > >>> On Wed, 16 Oct 2019 at 21:37, Thara Gopinath <thara.gopinath@xxxxxxxxxx> wrote: > >>>> > >>>> RPMh power controller hosts mx domain that can be used as thermal > >>>> warming device. Add a sub-node to specify this. > >>>> > >>>> Signed-off-by: Thara Gopinath <thara.gopinath@xxxxxxxxxx> > >>>> --- > >>>> Documentation/devicetree/bindings/power/qcom,rpmpd.txt | 10 ++++++++++ > >>>> 1 file changed, 10 insertions(+) > >>>> > >>>> diff --git a/Documentation/devicetree/bindings/power/qcom,rpmpd.txt b/Documentation/devicetree/bindings/power/qcom,rpmpd.txt > >>>> index eb35b22..fff695d 100644 > >>>> --- a/Documentation/devicetree/bindings/power/qcom,rpmpd.txt > >>>> +++ b/Documentation/devicetree/bindings/power/qcom,rpmpd.txt > >>>> @@ -18,6 +18,16 @@ Required Properties: > >>>> Refer to <dt-bindings/power/qcom-rpmpd.h> for the level values for > >>>> various OPPs for different platforms as well as Power domain indexes > >>>> > >>>> += SUBNODES > >>>> +RPMh alsp hosts power domains that can behave as thermal warming device. > >>>> +These are expressed as subnodes of the RPMh. The name of the node is used > >>>> +to identify the power domain and must therefor be "mx". > >>>> + > >>>> +- #cooling-cells: > >>>> + Usage: optional > >>>> + Value type: <u32> > >>>> + Definition: must be 2 > >>>> + > >>> > >>> Just wanted to express a minor thought about this. In general we use > >>> subnodes of PM domain providers to represent the topology of PM > >>> domains (subdomains), this is something different, which I guess is > >>> fine. > >>> > >>> I assume the #cooling-cells is here tells us this is not a PM domain > >>> provider, but a "cooling device provider"? > >> Yep. > >>> > >>> Also, I wonder if it would be fine to specify "power-domains" here, > >>> rather than using "name" as I think that is kind of awkward!? > >> Do you mean "power-domain-names" ? I am using this to match against the > >> genpd names defined in the provider driver. > > > > No. If you are using "power-domains" it means that you allow to > > describe the specifier for the provider. > Yep. But won't this look funny in DT ? The provider node will have a sub > node with a power domain referencing to itself Like below: Is this ok ? > > rpmhpd: power-controller { > compatible = "qcom,sdm845-rpmhpd"; > #power-domain-cells = <1>; > > ... > ... > mx_cdev: mx { > #cooling-cells = <2>; > power-domains = <&rpmhpd SDM845_MX>; > }; > The whole concept here seems all wrong to me. Isn't it what's in the power domain that's the cooling device. A CPU power domain is not a cooling device, the CPU is. Or we wouldn't make a clock a cooling device, but what the clock drives. Rob