Re: [PATCH v2 3/8] dt-bindings: clock: Add YAML schemas for the QCOM GPUCC clock bindings

From: Stephen Boyd
Date: Thu Dec 19 2019 - 00:32:47 EST


Quoting Taniya Das (2019-11-26 20:06:49)
>
>
> On 11/26/2019 11:41 PM, Stephen Boyd wrote:
> > Quoting Jeffrey Hugo (2019-11-15 07:11:01)
> >> On Fri, Nov 15, 2019 at 3:07 AM Taniya Das <tdas@xxxxxxxxxxxxxx> wrote:
> >>> diff --git a/Documentation/devicetree/bindings/clock/qcom,gpucc.yaml b/Documentation/devicetree/bindings/clock/qcom,gpucc.yaml
> >>> new file mode 100644
> >>> index 0000000..c2d6243
> >>> --- /dev/null
> >>> +++ b/Documentation/devicetree/bindings/clock/qcom,gpucc.yaml
> >>> + - description: GPLL0 source from GCC
> >>
> >> This is not an accurate conversion. GPLL0 was not valid for 845, and
> >> is required for 8998.
> >
> > Thanks for checking Jeff.
> >
> > It looks like on 845 there are two gpll0 clocks going to gpucc. From
> > gpu_cc_parent_map_0:
> >
> > "gcc_gpu_gpll0_clk_src",
> > "gcc_gpu_gpll0_div_clk_src",
> >
>
> There are branches of GPLL0 which would be connected to most external
> CCs. It is upto to the external CCs to either use them to source a
> frequency or not.

Yes, they can decide to use them or not, but they really do go to the
CCs so the DT should describe that.