On Mon, Sep 08, 2008 at 01:41:25PM -0700, Paul Goyette wrote: > BTW, would it be reasonable to add a chip-verification routine to the > adt7467 driver? Currently the driver's match routine blindly returns > "1". Reasonable, sure. Worth the time, maybe not. > The adt7463 driver has a "chip verify" routine to check the chip's > company_id and device_id registers. The 7467 has similar registers > (documented in the data sheet) and hard-coded values. Since both chips > (and possibly other dbCool chips) share the same i2c bus address, it > seems to me to be a worthwhile thing to check. The thing with indirect configuration is that a match function that fails might have already made damage to a device at the probed address. There's nothing you can do about that fact, and the way I2C works, the user pretty much always have to know what is where; therefore, the drivers tend to trust the kernel configuration (which is what returning 1 in that context means). > I've actually written the code to do it, so if someone actually has a > machine with an adt7467 I love to have them verify that my verify code > actually works. If it gets tested, it's ok to go in. -- Quentin Garnier - cube%cubidou.net@localhost - cube%NetBSD.org@localhost "See the look on my face from staying too long in one place [...] every time the morning breaks I know I'm closer to falling" KT Tunstall, Saving My Face, Drastic Fantastic, 2007.
Attachment:
pgpw6dOvsBFeW.pgp
Description: PGP signature