On Wed, Sep 08, 2010 at 11:29:02AM -0400, Jean Delvare wrote: > On Wed, 8 Sep 2010 06:56:54 -0700, Guenter Roeck wrote: > > Too bad - registers 0x16 and 0x17 exist on both 6658 and 6659. So the only way to detect 6659 > > would be the address (0x4d or 0x4e), and we would mis-detect it on 0x4c. Is that worth it ? > > I'd say adding support for the MAX6659 is worth it. Just don't add > detection. That is, all of MAX6657, 6658 and 6658 should be detected as > max6657, which has the minimum set of features. But if someone declares > a "max6659" device either as part of the platform data or from > user-space, then the driver should expose all the chip features. > > Deal? > I'd say yes, but then we would deliberately mis-detect the 6659 on address 0x4d and 0x4e, which kind of hurts my consciousness. How about a middle ground - mis-detect it on address 0x4c, but detect it correctly on 0x4d and 0x4e ? Should be ok if we add a note into the file and into the documentation, and we would do as good as we can. Guenter _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
