Re: [PATCH 3/3] arm64: dts: qcom: Add Lenovo IdeaPad Slim 5x 14Q8X9

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 4 potential issue(s) to consider:
- [Medium] Missing deletion of the VADC die temperature channel for the removed pmc8380_6 PMIC.
- [Medium] Invalid attempt to delete a node using `/delete-property/ port@1;`.
- [Medium] Misnamed `perst-gpios` property in `&pcie3_port0` prevents PCIe3 reset.
- [Medium] Peripheral device pin controls (touchpad and keyboard interrupts) are incorrectly attached to the I2C bus controller.
--

commit 3876cbc8da2512403990a017c13a7d2eb531d04c
Author: Jens Glathe <jens.glathe@oldschoolsolutions.biz>

arm64: dts: qcom: Add Lenovo IdeaPad Slim 5x 14Q8X9

Add a common DTSI plus LCD and OLED board files for the Lenovo
IdeaPad Slim 5 14Q8X9 (machine type 83HL) on X1P42100.

> diff --git a/arch/arm64/boot/dts/qcom/x1p42100-lenovo-ideapad-slim5x.dtsi b/arch/arm64/boot/dts/qcom/x1p42100-lenovo-ideapad-slim5x.dtsi
> new file mode 100644
> index 0000000000000..1f9e048119221
> --- /dev/null
> +++ b/arch/arm64/boot/dts/qcom/x1p42100-lenovo-ideapad-slim5x.dtsi
> @@ -0,0 +1,1695 @@
> +// SPDX-License-Identifier: BSD-3-Clause
> +/*
> + * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved.
> + * Copyright (c) 2024, Linaro Limited
> + * Copyright (c) 2026 Jens Glathe
> + */
> +
> +#include <dt-bindings/gpio/gpio.h>
> +#include <dt-bindings/input/gpio-keys.h>
> +#include <dt-bindings/input/input.h>
> +#include <dt-bindings/pinctrl/qcom,pmic-gpio.h>
> +#include <dt-bindings/regulator/qcom,rpmh-regulator.h>
> +
> +#include "purwa.dtsi"
> +#include "hamoa-pmics.dtsi"
> +
> +/delete-node/ &pmc8380_6;
> +/delete-node/ &pmc8380_6_thermal;

[Severity: Medium]
Should this block also delete the VADC die temperature channel
&pmc8380_6_die_temp? Since the PMIC is unpopulated, leaving
&pmc8380_6_die_temp active can cause the qcom-spmi-adc5 driver to attempt to
read the temperature of a physically absent PMIC.

[ ... ]

> +&i2c0 {
> +	clock-frequency = <400000>;
> +
> +	pinctrl-0 = <&qup_i2c0_data_clk>, <&tpad_default>, <&kybd_default>;
> +	pinctrl-names = "default";
> +	status = "okay";

[Severity: Medium]
Are the peripheral pinctrl states improperly assigned to the parent I2C bus
controller here? Attaching &tpad_default and &kybd_default to the bus
controller prevents the peripheral devices from independently managing their
interrupt and wake pins during suspend/resume.

[ ... ]

> +&pcie3_port0 {
> +	perst-gpios = <&tlmm 143 GPIO_ACTIVE_LOW>;
> +	wake-gpios = <&tlmm 145 GPIO_ACTIVE_LOW>;
> +};

[Severity: Medium]
Should perst-gpios be named reset-gpios since it is placed on the port0 node?
The generic PCI port handling expects reset-gpios, while perst-gpios is only
parsed when placed on the host controller node (&pcie3). This leaves the
device in reset.

[ ... ]

> +&usb_1_ss2 {
> +	dr_mode = "host";
> +	maximum-speed = "high-speed";
> +	phys = <&usb_1_ss2_hsphy>;
> +	phy-names = "usb2-phy";
> +	qcom,select-utmi-as-pipe-clk;
> +	status = "okay";
> +	/delete-property/ port@1;
> +};

[Severity: Medium]
Should this use /delete-node/ port@1 instead of /delete-property/ port@1?
In device tree bindings, port@1 is a node representing a Type-C connection,
not a property, so /delete-property/ will leave the node intact in the final
DTB.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260906-x1-ideapad-slim5x-v1-0-7ccffbe6bc4a@oldschoolsolutions.biz?part=3