On 02/01/2025 17:37, Rohit Visavalia wrote:
> Updated VCU binding for reset GPIO pin as optional property.
Subject and here: everything is an update. Be specific and drop all
redundant things making this unnecessary long: add reset GPIO
> It is marked as optional as some of the ZynqMP designs are having vcu_reset
> (reset pin of VCU IP) is driven by proc_sys_reset, proc_sys_reset is another
"are having is" looks like two verbs.
I don't get here mainly why SoC has something driven by its own GPIO.
That's unusual pattern.
> PL IP driven by the PS pl_reset. So, here the VCU reset is not driven by
> axi_gpio or PS GPIO so there will be no GPIO entry.
Anyway, this has to be constrained per SoC.
>
> Signed-off-by: Rohit Visavalia <rohit.visavalia@amd.com>
> ---
> Documentation/devicetree/bindings/clock/xlnx,vcu.yaml | 6 ++++++
> 1 file changed, 6 insertions(+)
>
> diff --git a/Documentation/devicetree/bindings/clock/xlnx,vcu.yaml b/Documentation/devicetree/bindings/clock/xlnx,vcu.yaml
> index bdb14594c40b..b3061309f8dd 100644
> --- a/Documentation/devicetree/bindings/clock/xlnx,vcu.yaml
> +++ b/Documentation/devicetree/bindings/clock/xlnx,vcu.yaml
> @@ -36,6 +36,11 @@ properties:
> - const: pll_ref
> - const: aclk
>
> + reset-gpios:
> + description: Optional GPIO used to reset the VCU, if available. Need use this
Drop redundant parts. Not being part of required defines "optional"
already. Don't repeat the schema but say something which we cannot
deduce from this.
> + reset gpio when in design 'vcu_resetn' is driven by gpio.
> + maxItems: 1
> +
> required:
> - reg
> - clocks
> @@ -52,6 +57,7 @@ examples:
> xlnx_vcu: vcu@a0040000 {
> compatible = "xlnx,vcu-logicoreip-1.0";
> reg = <0x0 0xa0040000 0x0 0x1000>;
> + reset-gpios = <&gpio 0x4e GPIO_ACTIVE_HIGH>;
GPIO numbers are not hex... unless this is not GPIO :/
> clocks = <&si570_1>, <&clkc 71>;
> clock-names = "pll_ref", "aclk";
> };
Best regards,
Krzysztof