On 27/11/2022 18:42, Jonathan Cameron wrote: > On Sun, 27 Nov 2022 13:51:19 +0100 > Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx> wrote: > >> On 25/11/2022 23:09, Samuel Holland wrote: >>> Linux has a driver for these ADCs at drivers/iio/adc/ti-adc081c.c, but >>> the compatible strings were undocumented. Add a binding for them. The >>> hardware has an alert interrupt output, but existing ti,adc081c users >>> do not provide the 'interrupts' property, so leave it as optional. >>> >>> Signed-off-by: Samuel Holland <samuel@xxxxxxxxxxxx> >>> --- >>> >>> .../bindings/iio/adc/ti,adc081c.yaml | 55 +++++++++++++++++++ >>> 1 file changed, 55 insertions(+) >>> create mode 100644 Documentation/devicetree/bindings/iio/adc/ti,adc081c.yaml >>> >>> diff --git a/Documentation/devicetree/bindings/iio/adc/ti,adc081c.yaml b/Documentation/devicetree/bindings/iio/adc/ti,adc081c.yaml >>> new file mode 100644 >>> index 000000000000..caaad777580c >>> --- /dev/null >>> +++ b/Documentation/devicetree/bindings/iio/adc/ti,adc081c.yaml >>> @@ -0,0 +1,55 @@ >>> +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause) >>> +%YAML 1.2 >>> +--- >>> +$id: http://devicetree.org/schemas/iio/adc/ti,adc081c.yaml# >>> +$schema: http://devicetree.org/meta-schemas/core.yaml# >>> + >>> +title: TI Single-channel I2C ADCs >>> + >>> +maintainers: >>> + - Jonathan Cameron <jic23@xxxxxxxxxx> >>> + - Lars-Peter Clausen <lars@xxxxxxxxxx> >>> + >>> +description: | >>> + Single-channel ADC supporting 8, 10, or 12-bit samples and high/low alerts. >>> + >>> +properties: >>> + compatible: >>> + enum: >>> + - ti,adc081c >>> + - ti,adc101c >>> + - ti,adc121c >>> + >>> + reg: >>> + maxItems: 1 >>> + >>> + interrupts: >>> + maxItems: 1 >>> + >>> + vref-supply: >>> + description: >>> + Regulator for the combined power supply and voltage reference >>> + >>> + "#io-channel-cells": >>> + const: 1 >>> + >>> +required: >>> + - compatible >>> + - reg >> >> Why not requiring io-channel-cells? If it is an IIO ADC provider, you >> need the cells, right? > > Only if anyone is using it as a provider. If it's purely being used via > IIO then there are no consumers registered. > > So historically I've left it up to those defining the binding to decide if > they think #io-channel-cells should be required or optional. > > It gets a bit non obvious with some of the more complex special ADCs on whether > they will ever be consumed. This one is generic, so quite likely it will be. I remember I asked some time ago and got the same answer... need to write it down into my notes :) Best regards, Krzysztof
