On Wed, Aug 14, 2013 at 04:45:43PM -0400, Dev, Kapil wrote: [ ... ] > Kapil: After making above changes, I think it seems to work now. Now I am > getting following output from sensors: > f71869-isa-0e80 > Adapter: ISA adapter > in0: +1.69 V > in1: +0.90 V > in2: +1.21 V > in3: +1.22 V > in4: +1.49 V > in5: +1.22 V > in6: +1.19 V > in7: +1.73 V > in8: +1.66 V > fan1: 2177 RPM > fan2: 0 RPM ALARM > fan3: 0 RPM ALARM > temp1: +24.0 C (high = +85.0 C, hyst = +81.0 C) > (crit = +100.0 C, hyst = +96.0 C) sensor = thermistor > temp2: +32.0 C (high = +85.0 C, hyst = +81.0 C) > (crit = +100.0 C, hyst = +96.0 C) sensor = thermistor > temp3: +48.0 C (high = +70.0 C, hyst = +68.0 C) > (crit = +85.0 C, hyst = +83.0 C) sensor = thermistor > > k10temp-pci-00c3 > Adapter: PCI adapter > temp1: +4.5 C (high = +70.0 C, crit = +70.0 C) > > I believe temp1 from PCI adapter denotes the core-temperature? But, it is Correct. > definitely off by 20-30C; its hard to believe that the temperature would be > 4.5C under fan based cooling system. I ran a SPEC hmmer benchmark and > plotted the temperature over time. temp1 seems to change now. I am > attaching the plots herewith. > Yes, that is pretty common with AMD chips. > Please let me know if there is a way to calibrate the core-temperature. > Yes, you'll have to do it in /etc/sensors3.conf. Basically you'll have to figure out what you think the real temperature is and add it to the raw temperature provided by the driver. Guenter _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
