On Fri, Nov 02, 2012 at 06:31:34PM +0100, Jean Delvare wrote: > On Fri, 2 Nov 2012 10:19:25 -0700, Guenter Roeck wrote: > > On Fri, Nov 02, 2012 at 04:24:04PM +0100, Jean Delvare wrote: > > > temp3_offset value of 116000 makes them match. > > > > Good, so at least we know that we have the correct register. The only remaining > > question is if we should make tempX_offset unsigned if a sensor is configured > > for AMDTSI. Kind of odd, though, since it is signed for everything else and > > might thus change sign when/if the sensor type is changed. Not sure what the > > best approach might be. > > Treating it as unsigned when the type is set to AMD-SI is probably the > way to go for now, although I would love to get more samples from users Agreed. > of AMD CPUs to make sure. I suppose we'll get more feedback after the > patch series you send earlier this week hits mainline. > and agreed. I'll come up with one more patch to add support for AMD-SI to it8728f. Hopefully sometime this weekend. Thanks, Guenter _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
