On 10/8/21 1:00 AM, Krzysztof Kozlowski wrote:
On 24/09/2021 13:51, Guenter Roeck wrote:
On Fri, Sep 24, 2021 at 08:57:44AM +0200, Krzysztof Kozlowski wrote:
On 23/09/2021 23:16, Rob Herring wrote:
On Mon, Sep 20, 2021 at 08:21:14PM +0200, Krzysztof Kozlowski wrote:
Document bindings for NXP SE97B, a DDR memory module temperature sensor
with integrated SPD and EEPROM via Atmel's AT24 interface.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxxxxx>
---
Documentation/devicetree/bindings/hwmon/jedec,jc42.yaml | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/Documentation/devicetree/bindings/hwmon/jedec,jc42.yaml b/Documentation/devicetree/bindings/hwmon/jedec,jc42.yaml
index a7bb4e3a1c46..0e49b3901161 100644
--- a/Documentation/devicetree/bindings/hwmon/jedec,jc42.yaml
+++ b/Documentation/devicetree/bindings/hwmon/jedec,jc42.yaml
@@ -10,6 +10,14 @@ maintainers:
- Jean Delvare <jdelvare@xxxxxxxx>
- Guenter Roeck <linux@xxxxxxxxxxxx>
+select:
+ properties:
+ compatible:
+ const: jedec,jc-42.4-temp
+
+ required:
+ - compatible
+
Is this supposed to be in the last patch? And why is it needed?
Yes, this is here on purpose because of nxp,se97b which is sensor with
at24-compatible EEPROM.
arch/arm/boot/dts/at91-nattis-2-natte-2.dts:
169 temp@18 {
170 compatible = "nxp,se97b", "jedec,jc-42.4-temp";
171 reg = <0x18>;
172 smbus-timeout-disable;
173 };
174
175 eeprom@50 {
176 compatible = "nxp,se97b", "atmel,24c02";
How would that be handled anyway ? Yes, the chip includes both a temperature
sensor and an eeprom, but this node should most definitely not instantiate as
temperature sensor.
I am not sure if I understand the problem you are mentioning. You have
two nods in DT, two different compatible sets and two difference
devices. One eeprom and other one a temperature sensor.
I didn't realize that the driver is supposed to bind to "jedec,jc-42.4-temp"
and that "nxp,se97b" is really informational.
Sorry for the confusion.
Guenter
