On 9/5/2024 9:23 AM, Andrew Lunn wrote: >> If the QUPs yaml changes are not included in the same series with >> i2c,serial yaml changes, you see these errors: >> >> /builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/soc/qcom/qcom,geni-se.example.dtb: geniqup@9c0000: serial@990000:compatible:0: 'qcom,sa8255p-geni-uart' is not one of ['qcom,geni-uart', 'qcom,geni-debug-uart'] >> /builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/soc/qcom/qcom,geni-se.example.dtb: geniqup@9c0000: i2c@984000:compatible:0: 'qcom,sa8255p-geni-i2c' is not one of ['qcom,geni-i2c', 'qcom,geni-i2c-master-hub'] > So you have a couple of options: > > 1) It sounds like you should get the QUP changes merged first. Then > submit the i2c,serial changes. Is there a reason you cannot do > this? Is there a mutual dependency between these two series, or > just a one way dependency? The ask in this thread is to create new yaml files since existing one is using generic compatibles. With new yaml, we would need to provide example and can't avoid it. If we have to provide example of QUP node, IMO, we should provide a few subnodes as well since just QUP node without subnodes(i2c/serial/spi) will not be very useful. We can possibly skip all 3 subnode and only keep one subsystem(e.g. serial) so QUP and UART yaml can go together(still need two subsystems) while SPI and I2C can go independently after QUP series is accepted. Not sure if that is acceptable to maintainers though. QUP node in actual DT will have all 3 types of subnodes(i2c,spi, serial) so example in this case won't be complete. > > 2) Explain in the commit message that following errors are expected > because ... And explain in detail why the dependency cannot be > broken to avoid the errors. > > Andrew
