[as this mentions nvmem layouts it would have been nice to include me] > NVMEM subsystem looks for fixed NVMEM cells (specified in DT) by > default. This behaviour made sense in early days before adding support > for dynamic cells. Why is that? It still makes sense, doesn't it? > With every new supported NVMEM device with dynamic cells current > behaviour becomes non-optimal. It results in unneeded iterating over DT > nodes and may result in false discovery of cells (depending on used DT > properties). What false discoveries? > This behaviour has actually caused a problem already with the MTD > subsystem. MTD subpartitions were incorrectly treated as NVMEM cells. But this is solved, correct? > Also with upcoming support for NVMEM layouts no new binding or driver > should support fixed cells defined in device node. How would you support older device trees with newer kernels or the other way around? I'm not sure I get your motivation to drop handling the fixed cells. > Solve this by modifying drivers for bindings that support specifying > fixed NVMEM cells in DT. Make them explicitly tell NVMEM subsystem to > read cells from DT. How can a driver know when there are fixed cells and when not? IOW. only new ones can be affected. If you want to get rid of the schema for *new* drivers then what about having a new child node, something like "nvmem-fixed-cells", similar to "nvmem-layout". And then you'd tell the new drivers to use the new-style dt binding. But there are no new drivers yet. So I'm still not sure I get your motivation. -michael
