Hello Krzysztof, thanks for the discussion and the review. On Mon, Jun 20, 2022 at 08:05:48PM +0200, Krzysztof Kozlowski wrote: > On 20/06/2022 19:43, Francesco Dolcini wrote: > > On Mon, Jun 20, 2022 at 06:44:23PM +0200, Krzysztof Kozlowski wrote: > >> On 20/06/2022 17:48, Francesco Dolcini wrote: > >>> Ideally one could just implement the imx6/7 thermal sensor reading and > >>> just make use of the thermal_of driver, however that would break > >>> compatibility with a lot of existing system ... to me this is just a > >>> no-go. > >> > >> This I did not understand... What is not implemented in thermal sensor > >> which would solve the issue? And why it cannot be implemented in > >> backwards compatible way? > > > > Currently the imx_thermal driver defines its own trip points. How would > > you change the code to work with old device tree binaries using the > > generic thermal_of driver? imx_thermal would need to be changed to be a > > thermal sensor device, all the thermal trip point code removed. > > The driver is using thermal_zone_device_register(). > > > > Maybe I am missing an obvious solution, just correct me if I am wrong. > > Probably you would need to support both solutions in the same driver, > based on presence of thermal-sensor-cells property. It won't be > particularly easy code, but maybe it is worth anyway... I am quite > surprised to see that IMX thermal driver does not use generic framework > and does not support generic bindings. #thermal-sensor-cells is already present in all dtsi using the imx_thermal driver without reason (see my previous comment on the topic [0]). >From my point of view the situation is somehow a mess, not sure how to proceed to be honest. Francesco [0] https://lore.kernel.org/all/acbf8ed3-0b8c-a0b2-88ef-7b13ad0908d5@xxxxxxxxxx/
