Am Mittwoch, 14. Oktober 2026, 09:20:59 Mitteleuropäische Sommerzeit schrieb 
Navid Ghahremani:

> +      - zte,zx29-qspi
> +      - zte,zx279128s-spifc

Unless there is a hardware difference that matters to the driver I think one 
compatible should be enough. Probably zte,zx-qspi. Although since this 
controller is very specific for flash chips zte,zx-spifc or zte,spifc are worthy 
considerations too. As far as I can see this controller only has flash memory 
ops and can't even send a generic byte stream.

Note that at least my board, and probably yours too, has a "normal" SPI 
controller too. In my case it can talk to SPI displays or a SLIC analog phone 
controller. It has a different hardware interface and needs a separate driver.

> +
> +  clocks:
> +    items:
> +      - description: work clock, the source of the SPI clock
> +      - description: register (APB) clock
> +
> +  clock-names:
> +    items:
> +      - const: wclk
> +      - const: pclk
> +
> +required:
> +  - compatible
> +  - reg
> +  - clocks
> +  - clock-names

I think the interrupt should be required by the binding too. The driver is 
still free to ignore it, but the binding is supposed to describe the hardware, 
not the driver.

My hardware has two DMA lines too. I don't know if they are always there. For 
pragmatic reasons (the DMA controller is its own beast and for usability less 
important than the flash controller) I'd make the DMA properties optional.