1. Patchew
  2. linux
  3. [PATCH 0/2] Support ROHM BU79100G ADC
  4. View patch

[PATCH 1/2] dt-bindings: iio: adc: Add ROHM BD79100G

Matti Vaittinen posted 2 patches 9 months ago
[PATCH 1/2] dt-bindings: iio: adc: Add ROHM BD79100G
Posted by Matti Vaittinen 9 months ago 
The ROHM BD79100G is a 12-bit ADC which can be read over SPI. Device has
no MOSI pin. ADC results can be read from MISO by clocking in 16 bits.
The 4 leading bits will be zero, last 12 containig the data.

Device has only VCC supply pin, which acts also as a VFS, determining the
voltage for full 12-bits. Specifying it is mandatory.

Signed-off-by: Matti Vaittinen <mazziesaccount@gmail.com>
---
 Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml b/Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml
index 44c671eeda73..10fb6e14c2d0 100644
--- a/Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml
+++ b/Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml
@@ -46,6 +46,7 @@ properties:
       - ti,ads7867
       - ti,ads7868
       - lltc,ltc2314-14
+      - rohm,bu79100g
 
   reg:
     maxItems: 1
@@ -96,6 +97,7 @@ allOf:
               - ti,adc121s
               - ti,ads7866
               - ti,ads7868
+              - rohm,bu79100g
     then:
       required:
         - vcc-supply
-- 
2.49.0
Re: [PATCH 1/2] dt-bindings: iio: adc: Add ROHM BD79100G
Posted by Conor Dooley 9 months ago 
On Tue, May 13, 2025 at 11:26:27AM +0300, Matti Vaittinen wrote:
> The ROHM BD79100G is a 12-bit ADC which can be read over SPI. Device has
> no MOSI pin. ADC results can be read from MISO by clocking in 16 bits.
> The 4 leading bits will be zero, last 12 containig the data.

I think it is probably worth mentioning why a rohm device is going into
this binding (clone?) and that the 12-bit thing is a differentiator that
is why you're not using a fallback.

> 
> Device has only VCC supply pin, which acts also as a VFS, determining the
> voltage for full 12-bits. Specifying it is mandatory.
> 
> Signed-off-by: Matti Vaittinen <mazziesaccount@gmail.com>
> ---
>  Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml b/Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml
> index 44c671eeda73..10fb6e14c2d0 100644
> --- a/Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml
> +++ b/Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml
> @@ -46,6 +46,7 @@ properties:
>        - ti,ads7867
>        - ti,ads7868
>        - lltc,ltc2314-14
> +      - rohm,bu79100g
>  
>    reg:
>      maxItems: 1
> @@ -96,6 +97,7 @@ allOf:
>                - ti,adc121s
>                - ti,ads7866
>                - ti,ads7868
> +              - rohm,bu79100g
>      then:
>        required:
>          - vcc-supply
> -- 
> 2.49.0
>
Re: [PATCH 1/2] dt-bindings: iio: adc: Add ROHM BD79100G
Posted by Matti Vaittinen 9 months ago 
On 13/05/2025 17:39, Conor Dooley wrote:
> On Tue, May 13, 2025 at 11:26:27AM +0300, Matti Vaittinen wrote:
>> The ROHM BD79100G is a 12-bit ADC which can be read over SPI. Device has
>> no MOSI pin. ADC results can be read from MISO by clocking in 16 bits.
>> The 4 leading bits will be zero, last 12 containig the data.
> 
> I think it is probably worth mentioning why a rohm device is going into
> this binding (clone?) and that the 12-bit thing is a differentiator that
> is why you're not using a fallback.

Thanks for mentioning the fallback option Conor! You're a hero :)

Now that you mentioned using a fallback, I believe I can ditch the 
driver changes and make BU79100G to use adc101s as a fallback!

I didn't even consider if some of the existing devices were (from SW 
perspective) identical. I was just happy when I found there was a driver 
supporting these simple SPI ADCs. Then I picked the right macro for 
doing register data conversion and correct shift, dumped in the bit 
width and extended the longish list of devices. I never checked if 
another device in the driver had similar set of "IC specific values".

Yours,
	-- Matti
Re: [PATCH 1/2] dt-bindings: iio: adc: Add ROHM BD79100G
Posted by Matti Vaittinen 9 months ago 
On 14/05/2025 08:36, Matti Vaittinen wrote:
> On 13/05/2025 17:39, Conor Dooley wrote:
>> On Tue, May 13, 2025 at 11:26:27AM +0300, Matti Vaittinen wrote:
>>> The ROHM BD79100G is a 12-bit ADC which can be read over SPI. Device has
>>> no MOSI pin. ADC results can be read from MISO by clocking in 16 bits.
>>> The 4 leading bits will be zero, last 12 containig the data.
>>
>> I think it is probably worth mentioning why a rohm device is going into
>> this binding (clone?) and that the 12-bit thing is a differentiator that
>> is why you're not using a fallback.
> 
> Thanks for mentioning the fallback option Conor! You're a hero :)
> 
> Now that you mentioned using a fallback, I believe I can ditch the 
> driver changes and make BU79100G to use adc101s as a fallback!

I mean, ads7866 as a fallback. Sorry.

> 
> I didn't even consider if some of the existing devices were (from SW 
> perspective) identical. I was just happy when I found there was a driver 
> supporting these simple SPI ADCs. Then I picked the right macro for 
> doing register data conversion and correct shift, dumped in the bit 
> width and extended the longish list of devices. I never checked if 
> another device in the driver had similar set of "IC specific values".
> 
> Yours,
>      -- Matti
Re: [PATCH 1/2] dt-bindings: iio: adc: Add ROHM BD79100G
Posted by Conor Dooley 9 months ago 
On Tue, May 13, 2025 at 03:39:11PM +0100, Conor Dooley wrote:
> On Tue, May 13, 2025 at 11:26:27AM +0300, Matti Vaittinen wrote:
> > The ROHM BD79100G is a 12-bit ADC which can be read over SPI. Device has
> > no MOSI pin. ADC results can be read from MISO by clocking in 16 bits.
> > The 4 leading bits will be zero, last 12 containig the data.
> 
> I think it is probably worth mentioning why a rohm device is going into
> this binding (clone?) and that the 12-bit thing is a differentiator that
> is why you're not using a fallback.

Sent too soon, meant to say: with that
Acked-by: Conor Dooley <conor.dooley@microchip.com>

> 
> > 
> > Device has only VCC supply pin, which acts also as a VFS, determining the
> > voltage for full 12-bits. Specifying it is mandatory.
> > 
> > Signed-off-by: Matti Vaittinen <mazziesaccount@gmail.com>
> > ---
> >  Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml | 2 ++
> >  1 file changed, 2 insertions(+)
> > 
> > diff --git a/Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml b/Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml
> > index 44c671eeda73..10fb6e14c2d0 100644
> > --- a/Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml
> > +++ b/Documentation/devicetree/bindings/iio/adc/adi,ad7476.yaml
> > @@ -46,6 +46,7 @@ properties:
> >        - ti,ads7867
> >        - ti,ads7868
> >        - lltc,ltc2314-14
> > +      - rohm,bu79100g
> >  
> >    reg:
> >      maxItems: 1
> > @@ -96,6 +97,7 @@ allOf:
> >                - ti,adc121s
> >                - ti,ads7866
> >                - ti,ads7868
> > +              - rohm,bu79100g
> >      then:
> >        required:
> >          - vcc-supply
> > -- 
> > 2.49.0
> > 
> 
>

Patchew 2.1-devel-b67e4c2

© 2016 - 2026 Red Hat, Inc.