On Sat 18 Mar 06:33 PDT 2017, Linus Walleij wrote:
> In order to accommodate in a logical manner for the premuxed channels
> in PM8921 and the similarly addressed channels in later PMICs, we
> need a twocell arrangement with premux and analog mux setting as
> a tuple to uniquely identify a hardware channel.
>
> These bindings are not yet in use, so it should be find to augment
s/find/fine?
> them before we actually start using it in drivers and device trees.
>
> This scheme came out of lengthy discussions and reverse-engineering
> and reading of the few information sources we have.
>
As there's no documentation of how this works I find it quite important
to make something that is intuitive - so more than you and I can use
this binding. So I like this.
> Suggested-by: Björn Andersson <bjorn.andersson@xxxxxxxxxx>
> Signed-off-by: Linus Walleij <linus.walleij@xxxxxxxxxx>
> ---
> .../bindings/iio/adc/qcom,pm8xxx-xoadc.txt | 117 ++++++++++++---------
> 1 file changed, 69 insertions(+), 48 deletions(-)
>
> diff --git a/Documentation/devicetree/bindings/iio/adc/qcom,pm8xxx-xoadc.txt b/Documentation/devicetree/bindings/iio/adc/qcom,pm8xxx-xoadc.txt
> index 53cd146d8096..680bb7a29dd5 100644
> --- a/Documentation/devicetree/bindings/iio/adc/qcom,pm8xxx-xoadc.txt
> +++ b/Documentation/devicetree/bindings/iio/adc/qcom,pm8xxx-xoadc.txt
> @@ -19,32 +19,42 @@ Required properties:
> with PMIC variant but is typically something like 2.2 or 1.8V.
>
> The following required properties are standard for IO channels, see
> -iio-bindings.txt for more details:
> +iio-bindings.txt for more details, but noitice that this particular
> +ADC has a special adressing scheme that require two cells for
> +identifying each ADC channel:
>
> -- #address-cells: should be set to <1>
> +- #address-cells: should be set to <2>, the first cell is the
> + prescaler (on PM8058) or premux (on PM8921) with two valid bits
> + so legal values are 0x00, 0x01 or 0x02. The second cell
> + is the main analog mux setting (0x00..0x0f). The combination
> + of prescaler/premux and analog mux uniquely addresses a hardware
> + channel on all systems.
>
> - #size-cells: should be set to <0>
>
> -- #io-channel-cells: should be set to <1>
> +- #io-channel-cells: should be set to <2>, again the cells are
> + precaler or premux followed by the analog muxing line.
>
> - interrupts: should refer to the parent PMIC interrupt controller
> and reference the proper ADC interrupt.
>
> Required subnodes:
>
> -The ADC channels are configured as subnodes of the ADC. Since some of
> -them are used for calibrating the ADC, these nodes are compulsory:
> +The ADC channels are configured as subnodes of the ADC.
>
> -adc-channel@c {
> - reg = <0x0c>;
> +Since some of them are used for calibrating the ADC, these nodes are
> +compulsory:
> +
> +adc-channel@0c {
Shouldn't the @ be followed by the value of the first cell in "reg"?
(Which will collide if we name them all "adc-channel").
> + reg = <0x00 0x0c>;
> };
>
> -adc-channel@d {
> - reg = <0x0d>;
> +adc-channel@0d {
> + reg = <0x00 0x0d>;
> };
>
> -adc-channel@f {
> - reg = <0x0f>;
> +adc-channel@0f {
> + reg = <0x00 0x0f>;
> };
>
> These three nodes are used for absolute and ratiometric calibration
> @@ -52,13 +62,26 @@ and only need to have these reg values: they are by hardware definition
> 1:1 ratio converters that sample 625, 1250 and 0 milliV and create
> an interpolation calibration for all other ADCs.
>
> -Optional subnodes: any channels other than channel 0x0c, 0x0d and
> -0x0f are optional.
> +Optional subnodes: any channels other than channels [0x00 0x0c],
> +[0x00 0x0d] and [0x00 0x0f] are optional.
>
> Required channel node properties:
>
> - reg: should contain the hardware channel number in the range
> - 0 .. 0x0f (4 bits). The hardware only supports 16 channels.
> + 0 .. 0xff (8 bits).
> +
> + On PM8058 the hardware only supports 16 channels, but we get the same
> + channels repeating with its input divided down by 1 or 3. Channels 00,
> + 10, 20, ... f0 are the raw values, 04, 14, 24 .. f4 are "unity" channels
> + divided by 1, and 08, 18, 28 .. f8 are channels divided by 3. Bits 0
> + and 1 of the channel index should always be 0.
> +
> + On PM8921 the hardware supports more than 16 channels through a complex
> + routing matrix using a premux, so 00, 10, 20 .. f0 are the basic raw
> + channels while another set of channels appear for 04, 14, 24 .. f4,
> + and again some of the same channels appear again divided down by 3
> + in 08, 18, 28 .. f8. Again bits 0 and 1 of the channel index should
> + always be 0.
While I believe documenting this is a good thing I do not think it adds
value to this binding document (other than showing Rob the absurdness of
the addressing scheme). Please consider moving it to the driver (haven't
checked how you commented it yet) and drop it from here.
Regards,
Bjorn
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
