On Tue, Aug 10, 2021 at 10:39 PM Matt Johnston <matt@xxxxxxxxxxxxxxxxxxxx> wrote: > > On Mon, 2021-08-02 at 10:45 -0600, Rob Herring wrote: > > On Sun, Aug 1, 2021 at 10:12 PM Matt Johnston <matt@xxxxxxxxxxxxxxxxxxxx> > > wrote: > > > > > > + slave functionality. The reg address must include > > > I2C_OWN_SLAVE_ADDRESS. > > > > This constraint can be a schema. > > The flag is already described in i2c.txt, is it OK to just make reference > to that? I know it is, but you are saying the only addresses valid must have that. Sounds like a constraint to me, so it should be a schema. Why make a user have to debug that they forgot to set that bit? Rob
