bug? acpi p-state + ondemand keeps dropping max freq
I've consistently experienced the following bizarre problem since 2.6.20, all the way up to 2.6.25.3 (regressed yesterday and each of these kernels exposes this behaviour):
/sys/devices/system/cpu/cpu0/cpufreq # grep . * affected_cpus:0
cpuinfo_cur_freq:800000
cpuinfo_max_freq:1866000
cpuinfo_min_freq:800000 scaling_available_frequencies:1866000 1600000 1333000 1066000 800000 scaling_available_governors:ondemand performance scaling_cur_freq:800000
scaling_driver:acpi-cpufreq
scaling_governor:ondemand
scaling_max_freq:800000
scaling_min_freq:800000
Notice that scaling_mx_freq dropped down to the lowest possible value and as such my CPU is only working at 800MHz. At boot time this field properly displays 1866MHz and everything works OK. After a certain period (?) this value drops down and I cannot manually elevate it back to the normal level:
# echo 1866000 > scaling_max_freq ; cat scaling_max_freq 800000 # echo 1866000 > scaling_max_freq ; cat scaling_max_freq 800000
This renders my Dothan to utterly poor speeds. (standard T43)
performance cpufreq governor makes no difference - I still can't change the frequency upper/lower values.
Auke
Quoted 1 line · Auke
This renders my Dothan to utterly poor speeds. (standard T43)
Which ThinkPad T43 model and BIOS revision?
I have a T43 2687DDU with the latest BIOS, and I have never seen anything like that happening. Is it overheating or something like that?
Signature · 4 lines
"One disk to rule them all, One disk to find them. One disk to bring them all and in the darkness grind them. In the Land of Redmond where the shadows lie." -- The Silicon Valley Tarot Henrique Holschuh
Quoted 8 lines · Henrique de Moraes Holschuh
Quoted 1 line · Auke
This renders my Dothan to utterly poor speeds. (standard T43)
Which ThinkPad T43 model and BIOS revision?
I have a T43 2687DDU with the latest BIOS, and I have never seen anything like that happening. Is it overheating or something like that?
T43 2668NU3 Version: 1YET59WW (1.24 )
hardly overheating:
Thermal zone 1 : ok, 49 C
it goes maybe up to 56C during work, but that's it.
Auke
Quoted 10 lines · Auke
Quoted 7 lines · Henrique de Moraes Holschuh
Quoted 1 line · Auke
This renders my Dothan to utterly poor speeds. (standard T43)
Which ThinkPad T43 model and BIOS revision?
I have a T43 2687DDU with the latest BIOS, and I have never seen anything like that happening. Is it overheating or something like that?
T43 2668NU3 Version: 1YET59WW (1.24 )
Older version of the planar card, but should be close enough to the 2687 I have...
Your BIOS is horribly old, and your EC firmware is very old too. You really should upgrade to BIOS 1.29 (1YET65WW) and EC 1.06. Lenovo has CDs you can use to upgrade even without Windows. thinkwiki.org has the links to the support pages with the downloads.
After you upgrade, go into the BIOS configuration screen, and set everything related to Speedstep and performance management to highest performance. You can let the BIOS do BUS power management, and you should let it do screen brightness power management, but don't let it mess with the processor speed
or the disks. It can have bad interactions with Linux power management.
Quoted 3 lines
hardly overheating:
Thermal zone 1 : ok, 49 C
It is probably a bad interaction from SMBIOS power management with the native Linux power management.
BTW: install thinkpad-acpi and lm-sensors 3.0, and you will be able to see the 11 thermal zones the EC controls in a T43.
Signature · 4 lines
"One disk to rule them all, One disk to find them. One disk to bring them all and in the darkness grind them. In the Land of Redmond where the shadows lie." -- The Silicon Valley Tarot Henrique Holschuh
Len Brown
reviewerQuoted 31 lines · Auke
I've consistently experienced the following bizarre problem since 2.6.20, all the way up to 2.6.25.3 (regressed yesterday and each of these kernels exposes this behaviour):
/sys/devices/system/cpu/cpu0/cpufreq# grep . * affected_cpus:0
cpuinfo_cur_freq:800000
cpuinfo_max_freq:1866000
cpuinfo_min_freq:800000 scaling_available_frequencies:1866000 1600000 1333000 1066000 800000 scaling_available_governors:ondemand performance scaling_cur_freq:800000
scaling_driver:acpi-cpufreq
scaling_governor:ondemand
scaling_max_freq:800000
scaling_min_freq:800000Notice that
scaling_mx_freqdropped down to the lowest possible value and as such my CPU is only working at 800MHz. At boot time this field properly displays 1866MHz and everything works OK. After a certain period (?) this value drops down and I cannot manually elevate it back to the normal level:# echo 1866000 > scaling_max_freq ; cat scaling_max_freq 800000 # echo 1866000 > scaling_max_freq ; cat scaling_max_freq 800000
Quoted 4 lines
This renders my Dothan to utterly poor speeds. (standard T43)
performance cpufreq governor makes no difference - I still can't change the frequency upper/lower values.
could be two causes:
1. thermal
2. user space
for #1....
watch grep . /proc/acpi/thermal_zone/THM*/* and see if anythign is climbing
you can build w/o CONFIG_ACPI_THERMAL and see if they symptom goes away -- for that is the module which would force cpufreq to Pn.
for #2...
see if it happens in single user mode without hal or distro daemons running.
cheers,
-Len
Quoted 5 lines · Auke
I've consistently experienced the following bizarre problem since 2.6.20, all the way up to 2.6.25.3 (regressed yesterday and each of these kernels exposes this behaviour):
Quoted 13 lines
/sys/devices/system/cpu/cpu0/cpufreq# grep . * affected_cpus:0
cpuinfo_cur_freq:800000
cpuinfo_max_freq:1866000
cpuinfo_min_freq:800000 scaling_available_frequencies:1866000 1600000 1333000 1066000 800000 scaling_available_governors:ondemand performance scaling_cur_freq:800000
scaling_driver:acpi-cpufreq
scaling_governor:ondemand
scaling_max_freq:800000
scaling_min_freq:800000
I also saw that on my T43, didn't have time to investigate any further and forgot about it shortly after :-(
Quoted 11 lines
Notice that
scaling_mx_freqdropped down to the lowest possible value and as such my CPU is only working at 800MHz. At boot time this field properly displays 1866MHz and everything works OK. After a certain period (?) this value drops down and I cannot manually elevate it back to the normal level:# echo 1866000 > scaling_max_freq ; cat scaling_max_freq 800000 # echo 1866000 > scaling_max_freq ; cat scaling_max_freq 800000
Try with "echo -n", seems that sysfs (or at least that file) doesn't like newlines.
Björn
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html