Re: [PATCH v2 01/14] dt-bindings: phy: Add MIPI CSI PHY for i.MX8Q
From: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Date: 2025-02-06 21:18:15
Also in:
imx, linux-devicetree, linux-media, linux-phy, lkml
Hi Frank, Thank you for the patch. On Wed, Feb 05, 2025 at 12:18:10PM -0500, Frank Li wrote:
Add MIPI CSI phy binding doc for i.MX8QXP, i.MX8QM and i.MX8ULP.
s/CSI/CSI-2/ in the subject line, here and below. s/phy/PHY/
quoted hunk ↗ jump to hunk
Signed-off-by: Frank Li <Frank.Li@nxp.com> --- change from v1 to v2 - Add missed fsl,imx8qm-mipi-cphy, which failback to fsl,imx8qxp-mipi-cphy - Move reg to required. Previous 8ulp use fsl,offset in downstream version. which should be reg. So move it to required --- .../bindings/phy/fsl,imx8qxp-mipi-cphy.yaml | 57 ++++++++++++++++++++++ 1 file changed, 57 insertions(+)diff --git a/Documentation/devicetree/bindings/phy/fsl,imx8qxp-mipi-cphy.yaml b/Documentation/devicetree/bindings/phy/fsl,imx8qxp-mipi-cphy.yaml new file mode 100644 index 0000000000000..7335b9262d0e7 --- /dev/null +++ b/Documentation/devicetree/bindings/phy/fsl,imx8qxp-mipi-cphy.yaml@@ -0,0 +1,57 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/phy/fsl,imx8qxp-mipi-cphy.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Freescale i.MX8 SoC MIPI CSI PHY + +maintainers: + - Frank Li <Frank.Li@nxp.com> + +properties: + "#phy-cells": + const: 0 + + compatible: + oneOf: + - enum: + - fsl,imx8qxp-mipi-cphy + - fsl,imx8ulp-mipi-cphy + - items: + - const: fsl,imx8qm-mipi-cphy + - const: fsl,imx8qxp-mipi-cphy
Why are those called cphy when, as far as I can tell from the documentation, they are D-PHYs ? Does that stand for *C*SI PHY ? I find it slightly confusing, but not so much that I'd ask for a change. It's just a name at the end of the day. Apart from that the binding looks fairly OK. Except maybe from the fact that this device is not a PHY :-( It has two PHY control registers, but the rest seems related to the glue logic at the output of the CSI-2 receiver. I wonder if we should go the syscon route.
+
+ reg:
+ maxItems: 1
+
+ power-domains:
+ maxItems: 1
+
+required:
+ - "#phy-cells"
+ - compatible
+ - reg
+
+allOf:
+ - if:
+ properties:
+ compatible:
+ contains:
+ enum:
+ - fsl,imx8qxp-mipi-cphy
+ then:
+ required:
+ - power-domains
+
+additionalProperties: false
+
+examples:
+ - |
+ phy@58221000 {
+ compatible = "fsl,imx8qxp-mipi-cphy";
+ reg = <0x58221000 0x10000>;
+ #phy-cells = <0>;
+ power-domains = <&pd 0>;
+ };
+-- Regards, Laurent Pinchart