On Fri, Nov 22, 2019 at 1:47 PM Bartosz Golaszewski <bgolaszewski@xxxxxxxxxxxx> wrote: > what about the existing bindings for at24 that don't mandate the > active-low flag? I'm afraid this would break the support for this > specific chip or lead to code duplication if we had this in both nvmem > and at24 with different logic. Hm yeah I realized this when I read patches 3 & 4. I would to like this: 1. Add a new generic property writeprotect-gpios that mandates to use GPIO_ACTIVE_LOW and use this in the new example 2. Deprecate wp-gpios in the binding, keep it around but deprecated. 3. Add a quirk to gpiolib-of in the manner of the other quirks there (like for SPI) so that if we are dealing with some EEPROM node like at24 and the flag is zero, tag on GPIO_ACTIVE_LOW on the descriptor. The driver will now handle the semantic of both cases with gpiolib-of providing a quirk for the old binding. This is how we solved this type of problem before. Yours, Linus Walleij
