On Mon, 2012-06-04 at 15:23 -0400, Ger vermeulen wrote: > Added requested cpuinfo at the bottom > Hi, Please don't top-post. > I also noticed a logging in the syslog "Unable to read TjMax from CPU", > Maybe gives a hint to reason. > > root@atom:~# rmmod coretemp > root@atom:~# modprobe coretemp > > Jun 4 21:06:48 atom kernel: [88459.638772] coretemp coretemp.0: Unable to > read TjMax from CPU. > Jun 4 21:06:48 atom kernel: [88459.640633] coretemp coretemp.1: Unable to > read TjMax from CPU. > This is "normal" for Atom CPUs. > root@atom:~# sensors > coretemp-isa-0000 > Adapter: ISA adapter > Core0: -1.0°C (crit = +90.0°C) > > coretemp-isa-0001 > Adapter: ISA adapter > Core1: +10.0°C (crit = +90.0°C) > > root@atom:~# cat /sys/devices/platform/coretemp.0/temp1_input > -1000 > root@atom:~# cat /sys/devices/platform/coretemp.1/temp1_input > 10000 > Problem is that TjMax is not correctly detected. I submitted an RFC patch to see if we can improve on it. That won't be made available in your version of the kernel, though, so you would either have to import the coretemp module from a later version of the kernel, or use the workaround I suggested below. TjMax for the Atom 330 CPU is 125 degrees C, so the real temperatures are 34 degrees C and 45 degrees C. I can not explain the differences in the reported temperatures vs. the work load. Maybe Core0 and Core1 are getting swapped somewhere. Overall, though, keep in mind that the reported temperatures are very unreliable for temperatures as far below TjMax as you experience in your system. Guenter > root@atom:~# cat /proc/cpuinfo > processor : 0 > vendor_id : GenuineIntel > cpu family : 6 > model : 28 > model name : Intel(R) Atom(TM) CPU 330 @ 1.60GHz > stepping : 2 > cpu MHz : 1599.991 > cache size : 512 KB > physical id : 0 > siblings : 2 > core id : 0 > cpu cores : 2 > apicid : 0 > initial apicid : 0 > fdiv_bug : no > hlt_bug : no > f00f_bug : no > coma_bug : no > fpu : yes > fpu_exception : yes > cpuid level : 10 > wp : yes > flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca > cmov pat clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm constant_tsc > arch_perfmon pebs bts aperfmperf pni dtes64 monitor ds_cpl tm2 ssse3 cx16 > xtpr pdcm movbe lahf_lm dts > bogomips : 3199.98 > clflush size : 64 > cache_alignment : 64 > address sizes : 32 bits physical, 48 bits virtual > power management: > > processor : 1 > vendor_id : GenuineIntel > cpu family : 6 > model : 28 > model name : Intel(R) Atom(TM) CPU 330 @ 1.60GHz > stepping : 2 > cpu MHz : 1599.991 > cache size : 512 KB > physical id : 0 > siblings : 2 > core id : 1 > cpu cores : 2 > apicid : 2 > initial apicid : 2 > fdiv_bug : no > hlt_bug : no > f00f_bug : no > coma_bug : no > fpu : yes > fpu_exception : yes > cpuid level : 10 > wp : yes > flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca > cmov pat clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm constant_tsc > arch_perfmon pebs bts aperfmperf pni dtes64 monitor ds_cpl tm2 ssse3 cx16 > xtpr pdcm movbe lahf_lm dts > bogomips : 3199.56 > clflush size : 64 > cache_alignment : 64 > address sizes : 32 bits physical, 48 bits virtual > power management: > > > > -----Original Message----- > > From: Guenter Roeck [mailto:guenter.roeck@xxxxxxxxxxxx] > > Sent: Monday, 04 June, 2012 01:17 > > To: geve002@xxxxxxxxx > > Cc: lm-sensors@xxxxxxxxxxxxxx > > Subject: Re: System: "Point of View ION-MB330-1" produces > > wrong values with lm-sensors 3.3.2 > > > > On Sun, Jun 03, 2012 at 01:50:14PM -0400, geve001@xxxxxxxxx wrote: > > > On slackware, sensors produce a wrong value for CPU temperature for the > > MB > > > "Point of View ION-MB330-1" > > > > > > > > > > > > It is off at least 20 degrees. > > > > > > CPU-0 and CPU-1 differ 10 degrees which is odd. CPU-0 has more load and > > lower > > > temp (see mpstat) > > > > > > Critical temperature should be 125 according to datasheets > > > > > > > > > > > > Some Google topics suggest to add 20 degrees, but in those cases the > > cpu’s have > > > the same low temp. > > > > > > > > > > > > ==> How to tackle this problem?? > > > > > > > > > > > > ------------- Linux kernel 2.6.37.6 > > > > > > root@atom:~# uname -a > > > > > > Linux atom 2.6.37.6-smp #2 SMP Sat Apr 9 23:39:07 CDT 2011 i686 Intel(R) > > Atom > > > (TM) CPU 330 @ 1.60GHz GenuineIntel GNU/Linux > > > > > > > > > > > Looks like the CPU does not provide a value for TjMax itself, and we don't > > detect > > the CPU type correctly to adjust it. > > > > Can you provide the output of /proc/cpuinfo ? > > > > As workaround, you can set the module parameter force_tjmax to 125. > > > > Thanks, > > Guenter > _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
