Re: [PATCH 2/8] arm64: dts: qcom: lemans: Add t-power-on-us property for PCIe root ports

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [High] Logic Error / Dead Code: The newly added `t-power-on-us` property is completely ignored
- [Low] Missing DT Binding Documentation
--

commit 821321f53e293af9b1a675f69406aacca48b199d
Author: Krishna Chaitanya Chundru <krishna.chundru@oss.qualcomm.com>

arm64: dts: qcom: lemans: Add t-power-on-us property for PCIe root ports

This commit attempts to fix improper L1.2 exit behavior by programming the
T_POWER_ON value. It adds the t-power-on-us property to the PCIe root port
nodes in the lemans device tree.

> diff --git a/arch/arm64/boot/dts/qcom/lemans.dtsi b/arch/arm64/boot/dts/qcom/lemans.dtsi
> index 695eae1b72569..615a9bfda2377 100644
> --- a/arch/arm64/boot/dts/qcom/lemans.dtsi
> +++ b/arch/arm64/boot/dts/qcom/lemans.dtsi
> @@ -2770,6 +2770,8 @@ pcieport0: pcie@0 {
>  				#address-cells = <3>;
>  				#size-cells = <2>;
>  				ranges;
> +
> +				t-power-on-us = <80>;

[Severity: Low]
Is this missing a corresponding update to the device tree binding
documentation? This introduces the t-power-on-us property without
documenting it in any Documentation/devicetree/bindings/ YAML file, which
will cause dtbs_check to fail.

[Severity: High]
Will this property actually be parsed by the driver as structured?

Since this child node lacks a phys property, devm_of_phy_get() will return
-ENODEV when qcom_pcie_parse_port() iterates over child nodes to parse the
port properties.

The fallback path, qcom_pcie_parse_legacy_binding(), successfully reads
the PHY from the parent node, but it appears to never parse the new
t-power-on-us property.

Does this cause the value to be completely ignored during device probe,
preventing the T_POWER_ON value from being programmed into hardware?

>  			};
>  		};
>  
> @@ -2943,6 +2945,8 @@ pcie@0 {
>  				#address-cells = <3>;
>  				#size-cells = <2>;
>  				ranges;
> +
> +				t-power-on-us = <80>;

[Severity: High]
Does this second child node also suffer from the same parsing fallback
issue? Without a phys property on this node, the driver will likely
ignore the new t-power-on-us property here as well.

>  			};
>  		};
>

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260911-t_pwr_on-v1-0-c9eee197d082@oss.qualcomm.com?part=2