[PATCH v6 1/2] dt-bindings: hwmon: Add Sophgo SG2042 external hardware monitor support

Inochi Amaoto inochiama at outlook.com
Sat Jul 6 16:34:15 PDT 2024


On Sat, Jul 06, 2024 at 09:48:58AM GMT, Guenter Roeck wrote:
> On Wed, Jul 03, 2024 at 10:30:43AM +0800, Inochi Amaoto wrote:
> > Due to the design, Sophgo SG2042 use an external MCU to provide
> > hardware information, thermal information and reset control.
> > 
> > Add bindings for this monitor device.
> > 
> > Signed-off-by: Inochi Amaoto <inochiama at outlook.com>
> > Reviewed-by: Conor Dooley <conor.dooley at microchip.com>
> > ---
> >  .../hwmon/sophgo,sg2042-hwmon-mcu.yaml        | 43 +++++++++++++++++++
> >  1 file changed, 43 insertions(+)
> >  create mode 100644 Documentation/devicetree/bindings/hwmon/sophgo,sg2042-hwmon-mcu.yaml
> > 
> > --
> > 2.45.2
> > 
> > diff --git a/Documentation/devicetree/bindings/hwmon/sophgo,sg2042-hwmon-mcu.yaml b/Documentation/devicetree/bindings/hwmon/sophgo,sg2042-hwmon-mcu.yaml
> > new file mode 100644
> > index 000000000000..f0667ac41d75
> > --- /dev/null
> > +++ b/Documentation/devicetree/bindings/hwmon/sophgo,sg2042-hwmon-mcu.yaml
> > @@ -0,0 +1,43 @@
> > +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
> > +%YAML 1.2
> > +---
> > +$id: http://devicetree.org/schemas/hwmon/sophgo,sg2042-hwmon-mcu.yaml#
> > +$schema: http://devicetree.org/meta-schemas/core.yaml#
> > +
> > +title: Sophgo SG2042 onboard MCU support
> > +
> > +maintainers:
> > +  - Inochi Amaoto <inochiama at outlook.com>
> > +
> > +properties:
> > +  compatible:
> > +    const: sophgo,sg2042-hwmon-mcu
> 
> According to the other patch, this actually covers four
> distinct models/devices.
> 
> static const struct sg2042_mcu_board_data sg2042_boards_data[] = {
> > +	{
> > +		.id = 0x80,
> > +		.name = "SG2042 evb x8",
> > +	},
> > +	{
> > +		.id = 0x81,
> > +		.name = "SG2042R evb",
> > +	},
> > +	{
> > +		.id = 0x83,
> > +		.name = "SG2042 evb x4",
> > +	},
> > +	{
> > +		.id = 0x90,
> > +		.name = "Milk-V Pioneer",
> > +	},
> > +};
> > +
> 
> Is it really appropriate to use a single compatible property for all of those ?
> 
> Guenter

These board can only be detected at running time (even this should on
a specific board). On real world, it can only sees a MCU onboard.
I don't think it is a good idea to add some bindings to cover these
model. It seems better to remove this array and let userspace to parse
these ids.

Regards,
Inochi



More information about the linux-riscv mailing list