On Sun, 23 Oct 2022 19:23:09 -0400 Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx> wrote: > On 24/10/2022 13:50, Rajat Khandelwal wrote: > > The datasheet states that the slave address for the device is 0x20 > > when the pins A0 and A1 are ground. The DT binding has been using > > 0x10 as the value and I think it should be 0x20 as per datasheet. > > > > Signed-off-by: Rajat Khandelwal <rajat.khandelwal@xxxxxxxxxxxxxxx> > > --- > > > Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx> > hmm. This is curious. So the datasheet indeed provides a table saying grounding both pins sets the address to 0x20, however take a look at Figure 2 which says the address is A1 | A0 | 1 | 0 | 0 | 0 | 0 or 0x10 as per the example. My guess is someone forgot that i2c addresses are 7 bits and the lowest bit of the first byte is used for R/W control. So unless we have this verified on hardware (implying that the address table is correct in this sense) I'm not keen to take this. I doubt that is the case given it has 8 bit addresses (0xe0) and i2c addresses are 7 bits. Jonathan > Best regards, > Krzysztof >
