On Wed, 2012-05-09 at 12:53 -0400, R, Durgadoss wrote: > Hi Guenter, > > [snip.] > > > > > > > can discuss/decide and I can work on it to submit it again, with > > > changes needed, > > > > > > if any. Kindly help. > > > > > Hi Durga, > > > > problem is that neither Jean nor me are happy with it. I think it is > > something that should, by its nature, reside in the thermal subsystem. > > I agree with you. If 'configuring the thresholds' were to reside in the Thermal > subsystem, then, one driver in hwmon will report CPU temperature and another > driver in thermal will configure the thresholds for that temperature. Is this > a good approach ? > > Or, we can include the threshold configuration in coretemp under #ifdef > CONFIG_THERMAL. This way even though the code resides in hwmon, > the functionality is enabled only by the Thermal Subsystem. > > Or, we can register coretemp with thermal_sys.c, and expose 2 trip points, > (for Threshold1 and Threshold2) like other thermal drivers do. > Actually, the thermal subsystem registers itself with the hwmon subsystem. I don't see a hwmon driver registering itself as thermal device. I was thinking more along the mine of drivers/platform/x86/intel_mid_thermal.c, though I don't really know if that model applies. Thanks, Guenter _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
