Re: [PATCH 3/4] media: dt-bindings: add bindings for Toshiba TC358746

[Laurent Pinchart]

Hi Marco,

On Thu, Sep 15, 2022 at 03:19:33PM +0200, Marco Felsch wrote:
> On 22-09-05, Laurent Pinchart wrote:
> > On Thu, Aug 18, 2022 at 04:33:06PM +0200, Marco Felsch wrote:
> > > Add the bindings for the Toshiba TC358746 Parallel <-> MIPI-CSI bridge
> > > driver.
> > > 
> > > Signed-off-by: Marco Felsch <m.felsch@xxxxxxxxxxxxxx>
> > > ---
> > >  .../bindings/media/i2c/toshiba,tc358746.yaml  | 157 ++++++++++++++++++
> > >  1 file changed, 157 insertions(+)
> > >  create mode 100644 Documentation/devicetree/bindings/media/i2c/toshiba,tc358746.yaml
> > > 
> > > diff --git a/Documentation/devicetree/bindings/media/i2c/toshiba,tc358746.yaml b/Documentation/devicetree/bindings/media/i2c/toshiba,tc358746.yaml
> > > new file mode 100644
> > > index 000000000000..9783cca363c6
> > > --- /dev/null
> > > +++ b/Documentation/devicetree/bindings/media/i2c/toshiba,tc358746.yaml
> > > @@ -0,0 +1,157 @@
> > > +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
> > > +%YAML 1.2
> > > +---
> > > +$id: http://devicetree.org/schemas/media/i2c/toshiba,tc358746.yaml#
> > > +$schema: http://devicetree.org/meta-schemas/core.yaml#
> > > +
> > > +title: Toshiba TC358746 Parallel to MIPI CSI2 Bridge
> > > +
> > > +maintainers:
> > > +  - Marco Felsch <kernel@xxxxxxxxxxxxxx>
> > > +
> > > +description: |-
> > > +  The Toshiba TC358746 converts a parallel video stream into a MIPI CSI-2
> > > +  stream. The direction can be either parallel-in -> csi-out or csi-in ->
> > > +  parallel-out The chip is programmable trough I2C and SPI but the SPI
> > > +  interface is only supported in parallel-in -> csi-out mode.
> > > +
> > > +  Note that the current device tree bindings only support the
> > > +  parallel-in -> csi-out path.
> > > +
> > > +properties:
> > > +  compatible:
> > > +    const: toshiba,tc358746
> > > +
> > > +  reg:
> > > +    maxItems: 1
> > > +
> > > +  clocks:
> > > +    description:
> > > +      The phandle to the reference clock source. This corresponds to the
> > > +      hardware pin REFCLK.
> > > +    maxItems: 1
> > > +
> > > +  clock-names:
> > > +    const: refclk
> > 
> > As there's a single clock, should we omit the clock name ?
> 
> Don't know, I would rather keep it to make it explicite. Many other go
> that way too.

I don't mind either way.

> > > +# The bridge can act as clock provider for the sensor. To enable this support
> > > +# #clock-cells must be specified. Attention if this feature is used than the
> > 
> > s/than/then/
> 
> Sure, thanks.
> 
> > > +# mclk rate must be at least: (2 * link-frequency) / 8
> > > +#                             `------------------´   ^
> > > +#                             internal PLL rate   smallest possible mclk-div
> > 
> > Could this be captured in the description of the property instead of a
> > comment ?
> 
> Hm.. a few (1,2) bindings using comments like this but I'm not against
> it. If it belongs to the description, which property should I use? Since
> there is no: clock-controller property like gpio-controller or
> interrupt-controller. The clock provider will be activated based on the
> #clock-cells property. But this property has nothing to with the
> description since this property specifies the cell size.

I would have put it in the description of the #clock-cells property:

  "#clock-cells":
    description: |
      The bridge can act as clock provider for the sensor. To enable
      this support #clock-cells must be specified. Attention ...
    const: 0

but if it becomes an issue, I'm OK with the comment too.

> > > +  "#clock-cells":
> > > +    const: 0
> > > +
> > > +  clock-output-names:
> > > +    description:
> > > +      The clock name of the MCLK output, the default name is tc358746-mclk.
> > > +    maxItems: 1
> > > +
> > > +  vddc-supply:
> > > +    description: Digital core voltage supply, 1.2 volts
> > > +
> > > +  vddio-supply:
> > > +    description: Digital I/O voltage supply, 1.8 volts
> > > +
> > > +  vddmipi-supply:
> > > +    description: MIPI CSI phy voltage supply, 1.2 volts
> > > +
> > > +  reset-gpios:
> > > +    description:
> > > +      The phandle and specifier for the GPIO that controls the chip reset.
> > > +      This corresponds to the hardware pin RESX which is physically active low.
> > > +    maxItems: 1
> > > +
> > > +  ports:
> > > +    $ref: /schemas/graph.yaml#/properties/ports
> > > +    properties:
> > > +      port@0:
> > > +        $ref: /schemas/graph.yaml#/properties/port
> > > +        description: Input port
> > 
> > Are the H/V sync signal polarities fixed, or should they be specified
> > here ?
> 
> At the moment they are fixed.

If the hardware supports different polarities, let's add them to the
device tree binding, even if the driver only supports a single option at
the moment.

> I can describe it if you want.

Thanks.

> > Does the chip support external sync only, or also BT.656 ? In the
> > latter case this needs a bus-type.
> 
> Yes, the chip also supports BT.656 but the driver doesn't support it
> yet. You're right, that we should make it explicite albeit it would a
> bit overhead yet since the only mode is: parallel-in with externa syncs.

Let's prepare for BT.656 support in the DT bindings, or we may regret it
later :-)

> > > +
> > > +      port@1:
> > > +        $ref: /schemas/graph.yaml#/$defs/port-base
> > > +        description: Output port
> > > +
> > > +        properties:
> > > +          endpoint:
> > > +            $ref: /schemas/media/video-interfaces.yaml#
> > > +            unevaluatedProperties: false
> > > +
> > > +            properties:
> > > +              data-lanes:
> > > +                minItems: 1
> > > +                maxItems: 4
> > > +
> > > +              clock-noncontinuous: true
> > > +              link-frequencies: true
> > > +
> > > +            required:
> > > +              - data-lanes
> > > +              - link-frequencies
> > > +
> > > +    required:
> > > +      - port@0
> > > +      - port@1
> > > +
> > > +required:
> > > +  - compatible
> > > +  - reg
> > > +  - clocks
> > > +  - clock-names
> > > +  - vddc-supply
> > > +  - vddio-supply
> > > +  - vddmipi-supply
> > > +  - ports
> > > +
> > > +additionalProperties: false
> > > +
> > > +examples:
> > > +  - |
> > > +    #include <dt-bindings/gpio/gpio.h>
> > > +
> > > +    i2c {
> > > +      #address-cells = <1>;
> > > +      #size-cells = <0>;
> > > +
> > > +      csi-bridge@e {
> > > +        compatible = "toshiba,tc358746";
> > > +        reg = <0xe>;
> > > +
> > > +        clocks = <&refclk>;
> > > +        clock-names = "refclk";
> > > +
> > > +        reset-gpios = <&gpio 2 GPIO_ACTIVE_LOW>;
> > > +
> > > +        vddc-supply = <&v1_2d>;
> > > +        vddio-supply = <&v1_8d>;
> > > +        vddmipi-supply = <&v1_2d>;
> > > +
> > > +        /* sensor mclk provider */
> > > +        #clock-cells = <0>;
> > > +
> > > +        ports {
> > > +          #address-cells = <1>;
> > > +          #size-cells = <0>;
> > > +
> > > +          /* Input */
> > > +          port@0 {
> > > +            reg = <0>;
> > > +            tc358746_in: endpoint {
> > > +              remote-endpoint = <&sensor_out>;
> > > +              };
> > 
> > Wrong indentation here.
> 
> Yes, thanks.
> 
> > > +          };
> > > +
> > > +          /* Output */
> > > +          port@1 {
> > > +            reg = <1>;
> > > +            tc358746_out: endpoint {
> > > +              remote-endpoint = <&mipi_csi2_in>;
> > > +              data-lanes = <1 2>;
> > > +              clock-noncontinuous;
> > > +              link-frequencies = /bits/ 64 <216000000>;
> > > +            };
> > > +          };
> > > +        };
> > > +      };
> > > +    };

-- 
Regards,

Laurent Pinchart