Hi, On Mon, Mar 24, 2025 at 9:40 AM Laurent Pinchart <laurent.pinch...@ideasonboard.com> wrote: > > On Tue, Mar 18, 2025 at 10:00:31PM +0100, Wolfram Sang wrote: > > Hi Laurent, > > > > > > Read out and check the ID registers, so we can bail out if I2C > > > > communication does not work or if the device is unknown. > > > > > > What's the advantage of that, what are you trying to guard against ? > > > > That a random chip at address 0x2c will be used. > > Is that really a problem ? That would only occur with a broken DT, is it > worth guarding against a development-time issue with a runtime check > that will increase boot time for every user ?
FWIW, this can also happen simply due to broken / damaged hardware. If a board gets stressed and causes a pin to become disconnected or if a regulator ages and stops providing power then we can also end up in this state. Getting a nice obvious error at probe when the device isn't responding at all can make problems like this much easier to debug. It's not uncommon for i2c devices to probe to make sure the device is really there at bootup. Checking for the full 8-byte ID is probably a bit overkill, but at the same time if we're going to probe something the ID is not a terrible thing to probe and reading 8 bytes won't really take much longer than reading 1. -Doug
