Re: [PATCH] arm64: dts: qcom: talos-lyra-evk: Add CYPD6129 Type-C support

From: Konrad Dybcio

Date: Thu Oct 15 2026 - 10:01:35 EST


On 10/12/26 3:11 PM, Akash Kumar wrote:
> Add support for the Cypress CYPD6129 dual Type-C PD controller on the
> Talos Lyra EVK.
>
> Enable the controller on I2C5 at address 0x40, configure its GPIO86
> interrupt and wakeup support, and describe both USB Type-C connectors.
> Connect the connectors to USB1 and USB2 through USB role-switch
> remote-endpoint links.
>
> Configure both USB controllers in OTG mode with usb-role-switch support
> so that the CYPD6129 controller can manage their USB data roles.
>
> USB1 is routed to the daughter-card board-to-board connector. Add a
> fixed USB1 VBUS regulator controlled by PM8150 GPIO10 to provide and
> control VBUS for the USB port on the daughter card. Configure the
> corresponding PM8150 GPIO pinctrl state and keep the regulator enabled
> for the board configuration.
>
> Add the TLMM GPIO64 VBUS boost enable as a GPIO hog. This enables the
> 5 V boost converter required to generate the VBUS supply for the USB
> daughter-card connector. The boost converter output is set to 5 V by
> the feedback network consisting of R2403 and R2401:
>
> VOUT = 0.6 * (1 + R2403 / R2401)
> = 0.6 * (1 + 75K / 10K)
> = 5 V
>
> The changes also add the I2C5 alias, CYPD6129 pinctrl configuration,
> wakeup-source properties, USB endpoint connections, and the required
> VBUS power-control GPIO descriptions.
>
> Signed-off-by: Akash Kumar <akakum@xxxxxxxxxxxxxxxx>
> ---
> arch/arm64/boot/dts/qcom/talos-lyra-evk.dts | 119 +++++++++++++++++++-
> 1 file changed, 117 insertions(+), 2 deletions(-)
> ---
> base-commit: b7e62b4d28bc0d2f16f03904616da78e4478aa88
> change-id: 20261012-talos-lyra-evk-v2-v2-0-dcc9316ebb5
> base commit link: https://lore.kernel.org/all/20261012-talos-lyra-evk-v2-v2-0-dcc9316ebb59@xxxxxxxxxxxxxxxx/
>
> Link to Driver and DT-Binding change: https://lore.kernel.org/all/20261012091658.635441-1-akash.kumar@xxxxxxxxxxxxxxxx/

This depends on both the driver+dt change *and* the missing base
DT, making this effectively unmergeable. Please either send the
changes together or refrain from sending them until that's possible

Konrad