FS#18901 - CPU stuck at 800Mhz

Attached to Project: Arch Linux
Opened by Keerthi (keerthi) - Thursday, 01 April 2010, 01:08 GMT
Last edited by Ionut Biru (wonder) - Monday, 10 May 2010, 21:04 GMT
Task Type Bug Report
Category Packages: Core
Status Closed
Assigned To No-one
Architecture i686
Severity High
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 2
Private No

Details

Description:
After recent kernel updates, my CPU stopped scaling. It appears to permanently stuck at 800MHz.

Additional info:
* package version(s)
kernel26 2.6.32.10-1

* config and/or log files etc.
keerthi ~ $ cpufreq-info
cpufrequtils 007: cpufreq-info (C) Dominik Brodowski 2004-2009
Report errors and bugs to cpufreq@vger.kernel.org, please.
analyzing CPU 0:
driver: acpi-cpufreq
CPUs which run at the same hardware frequency: 0
CPUs which need to have their frequency coordinated by software: 0
maximum transition latency: 10.0 us.
hardware limits: 800 MHz - 2.13 GHz
available frequency steps: 2.13 GHz, 1.87 GHz, 1.60 GHz, 1.33 GHz, 1.07 GHz, 800 MHz
available cpufreq governors: ondemand, performance
current policy: frequency should be within 800 MHz and 800 MHz.
The governor "ondemand" may decide which speed to use
within this range.
current CPU frequency is 800 MHz.
--------------
keerthi ~ $ cat /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 13
model name : Intel(R) Pentium(R) M processor 2.13GHz
stepping : 8
cpu MHz : 800.000
cache size : 2048 KB
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 2
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov clflush dts acpi mmx fxsr sse sse2 ss tm pbe nx up bts est tm2
bogomips : 1596.71
clflush size : 64
cache_alignment : 64
address sizes : 32 bits physical, 32 bits virtual
power management:
--------------

keerthi ~ $ uname -a
Linux DELL-XPS-G2 2.6.32-ARCH #1 SMP PREEMPT Mon Mar 15 20:08:25 UTC 2010 i686 Intel(R) Pentium(R) M processor 2.13GHz GenuineIntel GNU/Linux
--------------

keerthi ~ $ cat /etc/rc.conf

# /etc/rc.conf - Main Configuration for Arch Linux
LOCALE="en_US.UTF-8"
HARDWARECLOCK="localtime"
TIMEZONE="Asia/Kolkata"
KEYMAP="us"
CONSOLEFONT=
CONSOLEMAP=
USECOLOR="yes"

# HARDWARE
MOD_AUTOLOAD="yes"
MODULES=(fuse acpi_cpufreq cpufreq_ondemand vboxdrv vboxnetflt !snd_intel8x0 !snd_intel8x0m !pcspkr ipw2200 !tg3)

USELVM="no"

# NETWORKING
HOSTNAME="DELL-XPS-G2"
INTERFACES=(!eth0 !eth1)
eth0="dhcp"
gateway="default gw 192.168.1.1"
ROUTES=(!gateway)

# DAEMONS
DAEMONS=(syslog-ng hal @fam @crond @alsa @dkms_autoinstaller @networkmanager @postfix !mysqld kdm)

Steps to reproduce:
1. Boot and Login to DE
2. Observer CPU's frequency
This task depends upon

Closed by  Ionut Biru (wonder)
Monday, 10 May 2010, 21:04 GMT
Reason for closing:  Duplicate
Additional comments about closing:   FS#17107 
Comment by Svein Tonning (sveinemann) - Saturday, 03 April 2010, 05:02 GMT
Have you tried setting the frequency manually?
cpufreq-set -c 0 -f 2.13GHz
Comment by marc (Bomb) - Monday, 05 April 2010, 16:02 GMT
I have the same Problem with a Intel Q6600,

With the same Kernel and with the TestingKernel.

After load the modul acpi-cpufreq,only the modul, with no gov. my Mhz goes down from 2400mhz to 900mhz.

With Debian SID (2.6.32-trunk64, tested with a seperate HDD) it works fine

Comment by Keerthi (keerthi) - Thursday, 08 April 2010, 14:52 GMT
Svein, Didn't help. Thanks anyways.

The issue appears to be sporadic across reboots. Sometimes it scales, sometimes stuck at 800Mhz.

Comment by Matthew Gyurgyik (pyther) - Saturday, 10 April 2010, 02:45 GMT
Seems like you might want to file a bug on the kernel bug tracker
Comment by Mark (voidzero) - Sunday, 11 April 2010, 22:09 GMT
I had this problem too. In my case it happened when I loaded p4-clockmod when I actually needed acpi-cpufreq. As a hopeful suggestion, maybe there is another cpu scaling module you could try to see if a similar problem exists for you. Also, did you load cpufreq / cpufreqd or laptop-mode on boot? Because this is also something that can force the cpu into a paticular speed. Just some suggestions, sorry if I'm stepping on anybody's toes.
Comment by Andrej Gelenberg (nikel) - Monday, 12 April 2010, 08:12 GMT
please post output of:
for i in /sys/devices/system/cpu/cpu0/cpufreq/*; do echo $i; cat $i ; done
Comment by Keerthi (keerthi) - Tuesday, 13 April 2010, 14:28 GMT
for i in /sys/devices/system/cpu/cpu0/cpufreq/*; do echo $i; cat $i ; done
--------------------------------------------------------------------------\

/sys/devices/system/cpu/cpu0/cpufreq/affected_cpus
0
/sys/devices/system/cpu/cpu0/cpufreq/bios_limit
800000
/sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_cur_freq
800000
/sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_max_freq
2133000
/sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_min_freq
800000
/sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_transition_latency
10000
/sys/devices/system/cpu/cpu0/cpufreq/ondemand
cat: /sys/devices/system/cpu/cpu0/cpufreq/ondemand: Is a directory
/sys/devices/system/cpu/cpu0/cpufreq/related_cpus
0
/sys/devices/system/cpu/cpu0/cpufreq/scaling_available_frequencies
2133000 1867000 1600000 1333000 1067000 800000
/sys/devices/system/cpu/cpu0/cpufreq/scaling_available_governors
ondemand performance
/sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq
800000
/sys/devices/system/cpu/cpu0/cpufreq/scaling_driver
acpi-cpufreq
/sys/devices/system/cpu/cpu0/cpufreq/scaling_governor
ondemand
/sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq
800000
/sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq
800000
/sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed
<unsupported>
Comment by Andrej Gelenberg (nikel) - Wednesday, 14 April 2010, 13:18 GMT
Try as root:
echo performance > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor
and see, if freqency go up:
cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq

Sometimes is nice load ignored:
cat /sys/devices/system/cpu/cpufreq/ondemand/ignore_nice_load

Install lm_sensors and load coretemp kernel module. Watch, if your cpu is too hot:
watch sensors
Comment by Keerthi (keerthi) - Wednesday, 14 April 2010, 13:42 GMT
After installing kernel 2.6.33, CPU is now permanently stuck at 800Mhz (it was sporadic earlier as stated in my previous comments)

Ambient temperature in 39 degree Centigrade. CPU temp is only 40 degree Centigrade.

[quote]
echo performance > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor
[/quote]
-- Nothing changes

[quote]
cat /sys/devices/system/cpu/cpufreq/ondemand/ignore_nice_load
[/quote]
-- No such file or directory
Comment by Keerthi (keerthi) - Wednesday, 14 April 2010, 14:06 GMT
Strange. I removed '@' from every daemon that I start in my rc.conf and the CPU appears to scale properly now.

Before:
---------
DAEMONS=(syslog-ng hal @fam @crond @alsa @dkms_autoinstaller @networkmanager @postfix !mysqld kdm)

Now:
------
DAEMONS=(syslog-ng hal fam crond alsa dkms_autoinstaller networkmanager postfix kdm)

I'm gonna have to observe this for a couple of days and see if the issue has been permanently resolved.
Comment by Keerthi (keerthi) - Sunday, 18 April 2010, 18:52 GMT
No. The issue exists still. Sometimes scales (upon reboot). Most of the time stuck at 800Mhz.
Comment by Keerthi (keerthi) - Tuesday, 20 April 2010, 01:47 GMT
I find this in dmesg:

CPUFREQ: Per core ondemand sysfs interface is deprecated - up_threshold

Does this have anything to do with the scaling issue?
Comment by Keerthi (keerthi) - Tuesday, 20 April 2010, 17:38 GMT
Found Pentium-m specific kernel with speedstep patch in the AUR. http://aur.archlinux.org/packages.php?ID=33104
This seems to resolve the issue. Scales nicely, laptop runs a lot cooler. The stock kernel ought to be problematic.
Comment by Arnau Sanchez (tokland) - Friday, 23 April 2010, 09:56 GMT
My two cents: Not sure if this has something with this issue, but "bios_limit" was stuck to 800MHz in my laptop (probably due to a suspension).

$ cat /sys/devices/system/cpu/cpu0/cpufreq/bios_limit
800000
$ sudo cpufreq-set -c0 --max 2.4G
$ cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq
800000

"bios_limit" is obviously read-only so I all I tried was unplugging and re-plugging the laptop. Then:

$ cat /sys/devices/system/cpu/cpu0/cpufreq/bios_limit
24001000
$ sudo cpufreq-set -c0 --max 2.4G
$ cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq
2400000

Now if I unplug again, "bios_limit" is not stuck anymore (it fluctuates 0.8-2.4GHz as expected)
Comment by Keerthi (keerthi) - Friday, 30 April 2010, 22:50 GMT
This seems to be an issue in modern kernels. Installed Lucid Lynx (ubuntu 10.04). Still stuck at 800 Mhz

---------------------------
Linux ubuntu 2.6.32-21-generic #32-Ubuntu SMP Fri Apr 16 08:10:02 UTC 2010 i686 GNU/Lin
---------------------------
Comment by marc (Bomb) - Saturday, 01 May 2010, 11:31 GMT
After a Biosupdate, i have no Probs anymore.

<Sry for my bad English>
Comment by Nuno (quarkup) - Saturday, 01 May 2010, 11:42 GMT
check this bug.
As far as I know of, it may be related: http://bugs.archlinux.org/task/17107

Loading...