Re: [linux-sunxi] Re: [PATCH v4 1/6] dt-bindings: update the Allwinner GPADC device tree binding for H3
From: <hidden>
Date: 2017-09-18 15:47:30
Also in:
linux-arm-kernel, linux-iio, lkml
在 2017-09-18 16:30,Maxime Ripard 写道:
On Mon, Sep 18, 2017 at 03:36:43PM +0800, Icenowy Zheng wrote:quoted
于 2017年9月18日 GMT+08:00 下午3:33:36, Maxime Ripard [off-list ref] 写到:quoted
On Thu, Sep 14, 2017 at 10:52:46PM +0800, Icenowy Zheng wrote:quoted
Allwinner H3 features a thermal sensor like the one in A33, but hasitsquoted
register re-arranged, the clock divider moved to CCU (originally the clock divider is in ADC) and added a pair of bus clock and reset. Update the binding document to cover H3. Signed-off-by: Icenowy Zheng <redacted> Reviewed-by: Chen-Yu Tsai <redacted> --- Changes in v4: - Add nvmem calibration data (not yet used by the driver) Changes in v3: - Clock name changes. - Example node name changes. - Add interupts (not yet used by the driver). .../devicetree/bindings/mfd/sun4i-gpadc.txt | 30++++++++++++++++++++--quoted
1 file changed, 28 insertions(+), 2 deletions(-)diff --git a/Documentation/devicetree/bindings/mfd/sun4i-gpadc.txtb/Documentation/devicetree/bindings/mfd/sun4i-gpadc.txtquoted
index badff3611a98..6c470d584bf9 100644--- a/Documentation/devicetree/bindings/mfd/sun4i-gpadc.txt +++ b/Documentation/devicetree/bindings/mfd/sun4i-gpadc.txt@@ -4,12 +4,26 @@ The Allwinner SoCs all have an ADC that can alsoact as a thermal sensorquoted
and sometimes as a touchscreen controller. Required properties: - - compatible: "allwinner,sun8i-a33-ths", + - compatible: must contain one of the following compatibles: + - "allwinner,sun8i-a33-ths" + - "allwinner,sun8i-h3-ths" - reg: mmio address range of the chip, - #thermal-sensor-cells: shall be 0, - #io-channel-cells: shall be 0, -Example: +Optional properties: + - nvmem-cells: A phandle to the calibration data provided by anvmem device.quoted
+ If unspecified default values shall be used. + - nvmem-cell-names: Should be "calibration-data"I'd prefer to have which sensor it applies to here. It wouldn't change anything for the H3, but it definitely does for example for the A83t that has two sensors, one for each cluster, and one for the GPU, each with calibration data. What about cluster0-calibration?
I prefer sensor0-calibration to sensor3-calibration now. (Theortically the new generation THS can support up to 4 sensors)
quoted
The calibration data is in fact a 2 word (8 bytes) zone, which is reserved for 4 sensors on all SoCs, even on H3. It's half word per sensor. I prefer to just assume a 2 word cell for every SoC.You have three different data sources, it should be reprensented as such. Otherwise, the client has to get some knowledge of how the data are stored in the provider, which is an abstraction violation. Maxime -- Maxime Ripard, Free Electrons Embedded Linux and Kernel engineering http://free-electrons.com