Re: [PATCH v3 3/3] arm64: dts: qcom: qcs6490-rb3gen2: Add LVDS display overlay

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:
- [High] The DT overlay relies on hardware compatible strings that do not exist in the upstream kernel, preventing probe.
- [High] The 'lcd_disp_bias' fixed regulator claims a 5.5V output while being powered by a 1.8V PMIC LDO input rail.
- [Medium] The `lvds_bridge` reset GPIO lacks a corresponding `pinctrl` configuration.
--

commit 80f1496cde2f06d27ebc2ce6e50c133cce0ed74c
Author: Vishnu Saini <vishnu.saini@oss.qualcomm.com>

arm64: dts: qcom: qcs6490-rb3gen2: Add LVDS display overlay

This commit adds a DTBO overlay for the QCS6490 RB3 Gen2 industrial mezzanine 
to support LVDS display output. It disables HDMI and wires up a Lontium 
LT9211C DSI-to-LVDS bridge alongside a BOE DV215FHM-R01 panel.

[ ... ]

> diff --git a/arch/arm64/boot/dts/qcom/qcs6490-rb3gen2-industrial-mezzanine-lvds-boe,dv215fhm-r01.dtso b/arch/arm64/boot/dts/qcom/qcs6490-rb3gen2-industrial-mezzanine-lvds-boe,dv215fhm-r01.dtso
> new file mode 100644
> index 0000000000000..5501ea09f43ba
> --- /dev/null
> +++ b/arch/arm64/boot/dts/qcom/qcs6490-rb3gen2-industrial-mezzanine-lvds-boe,dv215fhm-r01.dtso
> @@ -0,0 +1,106 @@
> +// SPDX-License-Identifier: BSD-3-Clause
> +/*
> + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries.
> + */
> +
> +/dts-v1/;
> +/plugin/;
> +
> +#include <dt-bindings/gpio/gpio.h>
> +
> +&{/} {
> +	model = "Qualcomm Technologies, Inc. Industrial RB3gen2 LVDS";
> +
> +	hdmi-connector {
> +		status = "disabled";
> +	};
> +
> +	panel_lvds: panel-lvds {
> +		compatible = "boe,dv215fhm-r01";

[Severity: High]
Are these compatible strings present in the upstream kernel? It appears
that "boe,dv215fhm-r01" (and "lontium,lt9211c" below in the lvds_bridge
node) do not exist in mainline. Without these drivers and bindings, the
devices will silently fail to probe and the overlay will fail dtbs_check
validation when booted.

> +		power-supply = <&lcd_disp_bias>;
> +
> +		ports {

[ ... ]

> +	lcd_disp_bias: regulator-lcd-disp-bias {
> +		compatible = "regulator-fixed";
> +		regulator-name = "lcd_disp_bias";
> +		regulator-min-microvolt = <5500000>;
> +		regulator-max-microvolt = <5500000>;
> +		vin-supply = <&vreg_l18b_1p8>;

[Severity: High]
Is this physical regulator configuration correct? The lcd_disp_bias fixed
regulator claims a 5.5V output while being powered by a 1.8V PMIC LDO input
rail (vreg_l18b_1p8). A fixed regulator cannot step up 1.8V to 5.5V, and
the PMIC LDO cannot supply the wattage required for a 21.5" panel logic
rail. This will likely result in panel power failure and a potential PMIC
overcurrent when the panel driver enables the power-supply during display
initialization.

> +        };
> +};
> +
> +&i2c1 {
> +	#address-cells = <1>;
> +	#size-cells = <0>;
> +
> +	status = "okay";
> +
> +	lvds_bridge: lvds-bridge@29 {
> +		compatible = "lontium,lt9211c";
> +		reg = <0x29>;
> +		reset-gpios = <&tlmm 117 GPIO_ACTIVE_HIGH>;

[Severity: Medium]
Does this reset GPIO need a corresponding pinctrl configuration to define
its hardware state? The GPIO is requested for the bridge reset but lacks
guaranteed pin muxing. Without pinctrl, the pin might be left in a
high-impedance, undefined, or bootloader-specific state, potentially
preventing the reset signal from successfully reaching the bridge chip
during probe.

> +		vccio-supply = <&vreg_l18b_1p8>;
> +
> +		ports {

[ ... ]

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260905-lvds-support-v3-0-b40a1b03f63c@oss.qualcomm.com?part=3