Re: [PATCH v4 1/2] dt-bindings: display: panel: add Novatek NT36525
From: Krzysztof Kozlowski
Date: Sat Nov 21 2026 - 06:09:38 EST
On Fri, Nov 20, 2026 at 02:40:59PM +0300, Adil Hüsrev Tepe wrote:
> Add a DT binding for MIPI-DSI video-mode panels based on the Novatek
> NT36525 controller, as fitted on the Samsung Galaxy XCover5 (SM-G525F,
> Exynos850). The panel reset (RESX) is active-low, and the panel needs
> three separate supplies (vdd/vpos/vneg) plus an optional backlight.
>
> Signed-off-by: Adil Hüsrev Tepe <tadilhusrev@xxxxxxxxx>
Still incorrectly threaded.
I asked to use b4 for a reason. If you insist on not to, then at least
follow standard process and standard guides of upstreaming work.
> ---
> v3 -> v4: no changes to this binding.
>
> v2 -> v3: no content change. Resent as a fresh, standalone patch
> (previously v2 1/2 was a diff on top of v1's own patch instead of a
> complete patch against mainline, which Krzysztof Kozlowski correctly
> flagged as a git-management mistake on my part; see the cover letter).
>
> v1 -> v2 (addressing Sashiko AI's review of v1,
> https://sashiko.dev/#/patchset/20261116070212.314228-1-tadilhusrev@xxxxxxxxx?part=1):
> - reset-gpios is the physical RESX pin, which is active-low (asserted
> low, released high); describe it as such (GPIO_ACTIVE_LOW in the
> example) instead of GPIO_ACTIVE_HIGH. The matching driver fix is in
> "drm/panel: add Novatek NT36525 panel driver".
> - Add vpos-supply and vneg-supply: the driver requests three separate
> regulators ("vdd", "vpos", "vneg"), but the binding only declared
> "vdd-supply".
> - Add backlight: true so dtbs_check accepts the backlight property the
> driver already parses via drm_panel_of_backlight(), instead of
> switching additionalProperties to unevaluatedProperties (which would
> also pull in every other panel-common.yaml property, not just this
> one).
> - Fold the compatible string into a single, driver-matching value
> (novatek,nt36525-xcover5) instead of samsung,xcover5-nt36525, which
> never matched the driver's of_match_table at all.
> ---
> .../display/panel/novatek,nt36525.yaml | 83 +++++++++++++++++++
> 1 file changed, 83 insertions(+)
> create mode 100644 Documentation/devicetree/bindings/display/panel/novatek,nt36525.yaml
>
> diff --git a/Documentation/devicetree/bindings/display/panel/novatek,nt36525.yaml b/Documentation/devicetree/bindings/display/panel/novatek,nt36525.yaml
> new file mode 100644
> index 000000000..9158125c9
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/display/panel/novatek,nt36525.yaml
> @@ -0,0 +1,83 @@
> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
> +%YAML 1.2
> +---
> +$id: http://devicetree.org/schemas/display/panel/novatek,nt36525.yaml#
> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> +
> +title: Novatek NT36525 MIPI-DSI panel
> +
> +maintainers:
> + - Adil Hüsrev Tepe <tadilhusrev@xxxxxxxxx>
> +
> +description:
> + A MIPI-DSI video-mode panel driven by a Novatek NT36525 controller,
> + 720x1480. The first compatible string names the specific panel as fitted
> + in a device; the second is the NT36525 controller it is based on. The
> + panel reset (RESX) is active low.
> +
> +allOf:
> + - $ref: panel-common.yaml#
> +
> +properties:
> + compatible:
> + items:
> + - enum:
> + - novatek,nt36525-xcover5
I don't get it, what is xcover5? This claims it is Novatek device. Why
isn't this actual panel manufacturer?
> + - const: novatek,nt36525
> +
> + reg:
> + maxItems: 1
> + description: DSI virtual channel
> +
> + reset-gpios:
> + maxItems: 1
> +
> + vdd-supply:
> + description: panel logic power supply (vdd_ldo28, 1.8V)
> +
> + vpos-supply:
> + description: LCD positive bias supply
> +
> + vneg-supply:
> + description: LCD negative bias supply
> +
> + backlight: true
> +
> + port: true
I don't understand why this cannot be part of other bindings... looks
exactly like nt36536 or nt36672a
> +
> +required:
> + - compatible
> + - reg
> + - reset-gpios
> + - vdd-supply
> + - vpos-supply
> + - vneg-supply
> + - port
Drop port
> +
> +additionalProperties: false
Best regards,
Krzysztof