[PATCH v6 1/4] dt-bindings: clock: sophgo: Add clock controller of SG2000 series SoC

Inochi Amaoto inochiama at outlook.com
Mon Jan 15 16:27:59 PST 2024


>On Sun, Jan 14, 2024 at 12:16:58PM +0800, Inochi Amaoto wrote:
>> SG2000 series SoC has the same clock as CV1810 series, but the clock
>> related to A53 is functional in SG2000 series. So a new compatible
>> string is needed for the new SoC.
>>
>> Add definition for the clock controller of the SG2000 series SoC.
>>
>> Signed-off-by: Inochi Amaoto <inochiama at outlook.com>
>> Link: https://github.com/sophgo/sophgo-doc/releases/tag/sg2000-datasheet-v1.0-alpha
>> ---
>>  Documentation/devicetree/bindings/clock/sophgo,cv1800-clk.yaml | 3 ++-
>>  1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/Documentation/devicetree/bindings/clock/sophgo,cv1800-clk.yaml b/Documentation/devicetree/bindings/clock/sophgo,cv1800-clk.yaml
>> index c1dc24673c0d..59ef41adb539 100644
>> --- a/Documentation/devicetree/bindings/clock/sophgo,cv1800-clk.yaml
>> +++ b/Documentation/devicetree/bindings/clock/sophgo,cv1800-clk.yaml
>> @@ -4,7 +4,7 @@
>>  $id: http://devicetree.org/schemas/clock/sophgo,cv1800-clk.yaml#
>>  $schema: http://devicetree.org/meta-schemas/core.yaml#
>>
>> -title: Sophgo CV1800 Series Clock Controller
>> +title: Sophgo CV1800/SG2000 Series Clock Controller
>>
>>  maintainers:
>>    - Inochi Amaoto <inochiama at outlook.com>
>> @@ -14,6 +14,7 @@ properties:
>>      enum:
>>        - sophgo,cv1800-clk
>>        - sophgo,cv1810-clk
>> +      - sophgo,sg2000-clk
>
>I recall before you mentioned that the Sophgo folks were considering
>renaming one of their devices. Is the sg2000 the renamed one, or a
>different chip?
>
>Thanks,
>Conor.

Hi Conor,

The sg2000/sg2002 SoCs have one A53 core which cv1812/cv1813 SoCs
don't have. I prefer sg2000/sg2002 are different chips, or at least
an enhanced version of existed cv1812/cv1813. It is not a simple
rename.

For this patch, the sg2000 doesn't need to disable A53 related clocks
like cv18xx series. So this compatible is needed to bind to this new
logic.

Thanks,
Inochi



More information about the linux-riscv mailing list