[ltp] CPU running – but why?…
Arno Trautmann
linux-thinkpad@linux-thinkpad.org
Sat, 07 Jun 2008 19:40:11 +0200
Hi,
some times now it happened that my CPUs were running nearly 50% in C0,
the rest in C3 – but no prgramm was active, and there was absolutely no
reason for high CPU activity. The CPUs heated up over 70°C (I didn’t
wait to see how high it would get, but shut down the system then), while
they normally are at 40–50°C. Also, sometimes when this happens, the
frequency is at 1200MHz instead of the expected 800MHz.
I’m running Arch Linux on a X61s, using speedstep; normally, the CPUs
are in C3 for about 95%, 800MHz:
---
Cn Avg residency P-states (frequencies)
C0 (cpu running) ( 4.8%) 1.61 Ghz 0.0%
C1 0.0ms ( 0.0%) 1.60 Ghz 0.0%
C2 0.0ms ( 0.0%) 1200 Mhz 0.0%
C3 3.9ms (95.2%) 800 Mhz 100.0%
---
When it is heating up, powertop gives me something like:
---
Cn Avg residency P-states (frequencies)
C0 (cpu running) (42.1%) 1.61 Ghz 0.0%
C1 0.0ms ( 0.0%) 1.60 Ghz 1.4%
C2 0.0ms ( 8.3%) 1200 Mhz 0.0%
C3 0.0ms (49.6%) 800 Mhz 98.6%
Wakeups-from-idle per second : 16359.3 interval: 5.0s
no ACPI power usage estimate available
Top causes for wakeups:
35.8% ( 87.8) <kernel IPI> : Rescheduling interrupts
26.2% ( 64.2) firefox-bin : schedule_timeout (process_timeout)
19.2% ( 47.0) <interrupt> : HDA Intel, ohci1394
7.4% ( 18.2) <interrupt> : extra timer interrupt
7.3% ( 18.0) <interrupt> : acpi
---
and the output of top:
---
top - 21:06:04 up 1:18, 1 user, load average: 0.40, 0.27, 0.21
Tasks: 86 total, 2 running, 84 sleeping, 0 stopped, 0 zombie
Cpu(s): 6.0%us, 0.6%sy, 0.0%ni, 93.4%id, 0.0%wa, 0.0%hi, 0.0%si,
0.0%st
Mem: 1009680k total, 558744k used, 450936k free, 27964k buffers
Swap: 1461872k total, 0k used, 1461872k free, 199776k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
2840 arno 20 0 532m 152m 27m S 6 15.5 10:15.16 firefox-bin
2707 root 20 0 361m 47m 7924 S 3 4.8 2:33.98 Xorg
2796 arno 20 0 41852 10m 7416 S 0 1.1 0:01.62 xfce4-mixer
2798 arno 20 0 44388 15m 9456 R 0 1.6 0:01.29 terminal
3102 root 20 0 2132 1040 816 R 0 0.1 0:00.04 top
---
Most time everything is absolutely fine, and this behaviour is not
reproduceable. Can anyone give me a hint what could cause this? [Happens
both when on AC and on battery]
Thanks,
Arno