Skip to contentSkip to conversation

kernel: acpi-cpufreq: Transition failed

DISCUSSION Clemens Orthacker 3 participants 7 messages Feb 2, 2005 – Feb 6, 2005

lore
Share & tools
lore
#0 lore raw

Hi!

Since I'm using the 2.6.10 kernel I constantly get the following error:
kernel: acpi-cpufreq: Transition failed

Pentium4 M
acpi-cpufreq driver (userspace governor, powernowd)

Is this a configuration problem? (i read about this problem in a previous posting)

Thanks,
Clemens

(btw. bogomips fails, and I have no idea why..)

#
# Power management options (ACPI, APM)
#
CONFIG_PM=y
# CONFIG_PM_DEBUG is not set
CONFIG_SOFTWARE_SUSPEND=y
CONFIG_PM_STD_PARTITION="/dev/hda5"

#
# ACPI (Advanced Configuration and Power Interface) Support #
CONFIG_ACPI=y
CONFIG_ACPI_BOOT=y
CONFIG_ACPI_INTERPRETER=y
CONFIG_ACPI_SLEEP=y
CONFIG_ACPI_SLEEP_PROC_FS=y
CONFIG_ACPI_AC=m
CONFIG_ACPI_BATTERY=m
CONFIG_ACPI_BUTTON=m
CONFIG_ACPI_VIDEO=m
CONFIG_ACPI_FAN=m
CONFIG_ACPI_PROCESSOR=m
CONFIG_ACPI_THERMAL=m
CONFIG_ACPI_ASUS=m
# CONFIG_ACPI_IBM is not set
CONFIG_ACPI_TOSHIBA=m
CONFIG_ACPI_BLACKLIST_YEAR=0
# CONFIG_ACPI_DEBUG is not set
CONFIG_ACPI_BUS=y
CONFIG_ACPI_EC=y
CONFIG_ACPI_POWER=y
CONFIG_ACPI_PCI=y
CONFIG_ACPI_SYSTEM=y
CONFIG_X86_PM_TIMER=y

#
# APM (Advanced Power Management) BIOS Support
#
# CONFIG_APM is not set

#
# CPU Frequency scaling
#
CONFIG_CPU_FREQ=y
# CONFIG_CPU_FREQ_DEBUG is not set CONFIG_CPU_FREQ_PROC_INTF=m
# CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE is not set CONFIG_CPU_FREQ_DEFAULT_GOV_USERSPACE=y CONFIG_CPU_FREQ_GOV_PERFORMANCE=y CONFIG_CPU_FREQ_GOV_POWERSAVE=m CONFIG_CPU_FREQ_GOV_USERSPACE=y
# CONFIG_CPU_FREQ_24_API is not set CONFIG_CPU_FREQ_GOV_ONDEMAND=m CONFIG_CPU_FREQ_TABLE=y

#
# CPUFreq processor drivers
#
CONFIG_X86_ACPI_CPUFREQ=m
# CONFIG_X86_POWERNOW_K6 is not set
# CONFIG_X86_POWERNOW_K7 is not set
# CONFIG_X86_POWERNOW_K8 is not set
# CONFIG_X86_GX_SUSPMOD is not set
# CONFIG_X86_SPEEDSTEP_CENTRINO is not set
CONFIG_X86_SPEEDSTEP_ICH=m
# CONFIG_X86_SPEEDSTEP_SMI is not set
# CONFIG_X86_P4_CLOCKMOD is not set
# CONFIG_X86_CPUFREQ_NFORCE2 is not set
# CONFIG_X86_LONGRUN is not set
# CONFIG_X86_LONGHAUL is not set

#

# shared options
#
# CONFIG_X86_ACPI_CPUFREQ_PROC_INTF is not set
CONFIG_X86_SPEEDSTEP_LIB=m
CONFIG_X86_SPEEDSTEP_RELAXED_CAP_CHECK=y

cat /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 15
model : 2
model name : Mobile Intel(R) Pentium(R) 4 - M CPU 1.90GHz
stepping : 7
cpu MHz : 1196.242
cache size : 512 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 sep mtrr pge mca cmov pat
pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe cid xtpr
bogomips : 2367.48

Signature · 13 lines
| Clemens Orthacker
| e-mail: [email protected]
| phone: +43/650/6693393

-- 
| Clemens Orthacker
| e-mail: [email protected]
| phone: +43/650/6693393

-- 
| Clemens Orthacker
| e-mail: [email protected]
| phone: +43/650/6693393

_______________________________________________
Cpufreq mailing list
[email protected]
http://lists.linux.org.uk/mailman/listinfo/cpufreq

#1 lore raw

Am Mittwoch, 2. Februar 2005 11:58 schrieb Clemens Orthacker:

Quoted 3 lines

Pentium4 M
acpi-cpufreq driver (userspace governor, powernowd)

if I change the governor to performance, the frequency changes (from 1200000) to 1900000 without problems..

Signature · 3 lines
| Clemens Orthacker
| e-mail: [email protected]
| phone: +43/650/6693393
#2 lore raw
In reply to Clemens Orthacker

Hi,

Quoted 1 line · Clemens Orthacker

Since I'm using the 2.6.10 kernel

Did it work correctly when you used earlier kernels than 2.6.10? Even if using powernowd?

Quoted 7 lines

I constantly get the following error:
kernel: acpi-cpufreq: Transition failed

Pentium4 M
acpi-cpufreq driver (userspace governor, powernowd)

Is this a configuration problem?

If speedstep-ich works, you should use that module instead -- other than that, this doesn't look like a configuration problem. The BIOS tells us how the kernel should ask it for CPU frequency changes; the kernel does it, but then the BIOS simply reports "error". It might be that the BIOS does provide the kernel with wrong information on what the return value means -- i.e. the kernel thinks it is an error, but in reality it isn't.

What chipset does your notebook use (plain "lspci" output, please)?

Quoted 1 line

(btw. bogomips fails, and I have no idea why..)

I'll investigate that.

Dominik

#3 lore raw

Am Freitag, 4. Februar 2005 21:27 schrieb Dominik Brodowski:

Quoted 2 lines

Did it work correctly when you used earlier kernels than 2.6.10? Even if using powernowd?

Yes. At least there were no error messages. (2.6.9)

Quoted 1 line

If speedstep-ich works, you should use that module instead --

I don't have a Intel chipset, so I thought I should use acpi-cpufreq.

Quoted 9 lines

other than
that, this doesn't look like a configuration problem. The BIOS tells us how the kernel should ask it for CPU frequency changes; the kernel does it, but then the BIOS simply reports "error". It might be that the BIOS does provide the kernel with wrong information on what the return value means -- i.e. the kernel thinks it is an error, but in reality it isn't.

What chipset does your notebook use (plain "lspci" output, please)?

Sorry, I forgot to add this in the first mail.

0000:00:00.0 Host bridge: ALi Corporation M1671 Super P4 Northbridge [AGP4X,PCI and SDR/DDR] (rev 02)
0000:00:01.0 PCI bridge: ALi Corporation PCI to AGP Controller 0000:00:06.0 Multimedia audio controller: ALi Corporation M5451 PCI AC-Link Controller Audio Device (rev 02)
0000:00:07.0 ISA bridge: ALi Corporation M1533 PCI to ISA Bridge [Aladdin IV] 0000:00:08.0 Modem: ALi Corporation M5457 AC'97 Modem Controller 0000:00:0a.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10)
0000:00:0b.0 FireWire (IEEE 1394): VIA Technologies, Inc. IEEE 1394 Host Controller (rev 46)
0000:00:0c.0 Network controller: Intersil Corporation Prism 2.5 Wavelan chipset (rev 01)
0000:00:0d.0 USB Controller: NEC Corporation USB (rev 41) 0000:00:0d.1 USB Controller: NEC Corporation USB (rev 41) 0000:00:0d.2 USB Controller: NEC Corporation USB 2.0 (rev 02) 0000:00:10.0 IDE interface: ALi Corporation M5229 IDE (rev c4) 0000:00:11.0 Bridge: ALi Corporation M7101 Power Management Controller [PMU] 0000:00:13.0 CardBus bridge: O2 Micro, Inc. OZ6912 Cardbus Controller 0000:01:00.0 VGA compatible controller: nVidia Corporation NV11 [GeForce2 Go] (rev b2)

Signature · 3 lines
| Clemens Orthacker
| e-mail: [email protected]
| phone: +43/650/6693393
#4 lore raw
Quoted 6 lines · Clemens Orthacker

Am Freitag, 4. Februar 2005 21:27 schrieb Dominik Brodowski:

Quoted 2 lines

Did it work correctly when you used earlier kernels than 2.6.10? Even if using powernowd?

Yes. At least there were no error messages. (2.6.9)

This is most interesting. There have been the following changes between 2.6.9 and 2.6.10 relating to the acpi-cpufreq module:

- [core,drivers] add cpumask_t cpus for better SMT or CPU groups handling => this is non-HT, so unrelated

- [acpi-io] rename acpi.c to acpi-cpufreq.c
=> this _can't_ do any harm

- [acpi-cpufreq] debug modifications and advancements => I doubt this has any effect

- [acpi-cpufreq] fix x86_64 symlink breakage [Dave Jones] => wrong architecture

- [acpi-perflib] disable SMM access to SpeedStep using ACPI-FADT pstate_cnt => so this is what's left over.

Could you remove the line in arch/i386/kernel/cpu/cpufreq/acpi-cpufreq.c which states:

        acpi_processor_notify_smm(THIS_MODULE);

recompile, re-install, and re-test, please?

Quoted 3 lines
Quoted 1 line

If speedstep-ich works, you should use that module instead --

I don't have a Intel chipset, so I thought I should use acpi-cpufreq.

That's correct.

Dominik

#5 lore raw

Am Samstag, 5. Februar 2005 11:53 schrieben Sie:

Quoted 5 lines

Could you remove the line in arch/i386/kernel/cpu/cpufreq/acpi-cpufreq.c which states:

        acpi_processor_notify_smm(THIS_MODULE);

recompile, re-install, and re-test, please?

Gives the same result as before. (3 error messages, 22 min uptime)

ps. While recompiling the kernel the frequency was changing to max without problems. Probably there is some problem with powernowd. Can you recommand any other userspace tool?
Will there be a cpufrequtils debian package?

Thanks

Signature · 3 lines
| Clemens Orthacker
| e-mail: [email protected]
| phone: +43/650/6693393
#6 lore raw

On Sun, Feb 06, 2005 at 08:34:16PM +0100, Clemens Orthacker wrote:
[...]

Quoted 1 line

Will there be a cpufrequtils debian package?

There is! :) It's in the NEW queue waiting for approval. you can get a preview from http://oioio.altervista.org/debian/binary/

bye

Signature · 2 lines
mattia
:wq!