/proc/cpuinfo reports wrong frequeny for AMD Athlon XP 3000+

Same here with a Barton core Athlon XP Mobile 2400+ that should have 1800 Mhz more or less and /proc/cpuinfo gives 1473 Mhz.

kernel: 2.6.17-11-generic

Tried the same Athlon XP-M 2400+ with Feisty - 3 March snapshot and also with this is detected as 1472-3 Mhz, see syslog attached, please fix this if possible.

My CPU:
http://www.cpu-world.com/CPUs/K7/AMD-Mobile%20Athlon%20XP-M%202400%2B%20-%20AXMA2400FUT4C.html

From SYSLOG:
Mar 4 13:15:07 ubuntu kernel: [ 132.272544] powernow: PowerNOW! Technology present. Can scale: frequency and voltage.
Mar 4 13:15:07 ubuntu kernel: [ 132.282694] powernow: Trying ACPI perflib
Mar 4 13:15:07 ubuntu kernel: [ 132.291331] powernow: Minimum speed 1472 MHz. Maximum speed 1472 MHz.

also it say I haven't PowerNOW but accordingly to the link I have it! PowerNOW or kernel bug?

syslog file

Sorry I have wrong, it recognize powernow technology but it doesn't recognize correctly min and max frequencies.

Tim if you set to "needs info" tell me what info do you need first!

thanx :=)

Sorry. Please attach dmesg and lspci output from Feisty Herd5. It contains some ACPI updates that might affect your frequency settings.

dmesg > dmesg.txt
lspc -vv > lspci.txt

I know you asked me from Herd 5 but for now I have only Edgy 6.10, with Feisty I will test next week.

Looking for a cpufreq problem, namely in /sys/devices/system/cpu/cpu0/cpufreq I have:

 - scaling_max_freq < cpuinfo_max_freq
 - scaling_max_freq is not in the list of available frequencies

$> cat scaling_available_frequencies
1391439 1192662 1060144 927626 662590
$> cat scaling_governor cpuinfo_max_freq scaling_max_freq cpuinfo_cur_freq
performance
1391439
1050000
927626

The bug
https://launchpad.net/ubuntu/+source/linux-source-2.6.15/+bug/36014 points here, and I found that after a

$> cat /proc/cpuinfo
processor : 0
vendor_id : AuthenticAMD
cpu family : 6
model : 8
model name : mobile AMD Athlon(tm) XP 1600+
stepping : 0
cpu MHz : 662.590
cache size : 256 KB
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 1
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov pat pse36 mmx fxsr sse syscall mp mmxext 3dnowext 3dnow up ts fid vid
bogomips : 1325.90
clflush size : 32

I got the information on the current speed not in the maximum speed.

In previous Linux Distros, namely FC5 running 2.6.19, there was no problem at all with cpufreq or whatsoever, and I could get the most of my Athlon XP 1600+, ie 1391439 Hz.

I think cpuinfo reflects wrong detection by kernel/powernow

Sorry for the last "lspci" that was without -vv if you need I can provide it again with feisty beta.

I don't know if it can be an hint but with the same identical notebook but another Athlon XP Mobile the bug is not present.

With Athlon XP-M 2000+ core Thorobred then CORRECT speed -> 1600 Mhz
With Athlon XP-M 2400+ core Barton then WRONG speed -> 1473 Mhz instead of 1800 Mhz

updated lspci log
kubuntu feisty beta lspci -vv

_seems_ that for me is not a bug, but then a chipset issue that don't recognize well barton core for athlon xp.
I will do more test, but in this case I think this isn't a linux bug.

This is taken from 2.6.21.2 can be the fix for our problems?

commit ca5cf888be6e9261cdcf0b0de30c7237aa87f366
Author: Daniel Drake <email address hidden>
Date: Tue May 15 19:59:03 2007 +0000

    [PATCH] CPUFREQ: powernow-k7: fix MHz rounding issue with perflib

    When the PST tables are broken, powernow-k7 uses ACPI's processor_perflib to
    deduce the available frequency multipliers from the _PSS tables.

    Upon frequency change, processor_perflib performs some verification on the
    frequency (checks that it's within allowable bounds).

    powernow-k7 deals with absolute frequencies in KHz, whereas perflib only
    deals with MHz values. When performing the above verification, perflib
    multiplies the MHz values by 1000 to obtain the KHz value.

    We then end up with situations like the following:
     - powernow-k7 multiplies the multiplier by the FSB, and obtains a value
       such as 1266768 KHz
     - perflib belives the same state has frequency of 1266 MHz
     - acpi_processor_ppc_notifier calls cpufreq_verify_within_limits to verify
       that 1266768 is in the allowable range of 0 to 1266000 (i.e. 1266 * 1000)
     - it's not, so that frequency is rejected
     - the maximum CPU frequency is not reachable

    This patch solves the problem by rounding up the MHz values stored in perflib's
    tables. Additionally it corrects a broken URL.

    It also fixes http://bugzilla.kernel.org/show_bug.cgi?id=8255 although this
    case is a bit different: the frequencies in the _PSS tables are wildly wrong,
    but we get better results if we force ACPI to respect the fsb * multiplier
    calculations (even though it seems that the multiplier values aren't entirely
    correct either).

    Signed-off-by: Daniel Drake <email address hidden>
    Signed-off-by: Dave Jones <email address hidden>
    Signed-off-by: Chris Wright <email address hidden>

My problems where completely solved using linux-image-2.6.22-9-generic from Gusty Gibbon.

I have an Athlon 1600+, and now frecuency scaling works smoothly.

Can anyone else confirm this being fixed with newer versions of the kernel?

The 18 month support period for Edgy Eft 6.10 has reached it's end of life. As a result, we are closing the linux-source-2.6.17 Edgy Eft kernel task. However, please note that this report will remain open against the actively developed kernel. Thank you for your continued support and help as we debug this issue.

Hardy Heron 8.04 was recently released. It would be helpful if you could test the new release and verify if this is still an issue - http://www.ubuntu.com/getubuntu/download . You should be able to test your bug using the LiveCD. Please let us know your results. Thanks.

The Ubuntu Kernel Team is planning to move to the 2.6.27 kernel for the upcoming Intrepid Ibex 8.10 release. As a result, the kernel team would appreciate it if you could please test this newer 2.6.27 Ubuntu kernel. There are one of two ways you should be able to test:

1) If you are comfortable installing packages on your own, the linux-image-2.6.27-* package is currently available for you to install and test.

--or--

2) The upcoming Alpha5 for Intrepid Ibex 8.10 will contain this newer 2.6.27 Ubuntu kernel. Alpha5 is set to be released Thursday Sept 4. Please watch http://www.ubuntu.com/testing for Alpha5 to be announced. You should then be able to test via a LiveCD.

Please let us know immediately if this newer 2.6.27 kernel resolves the bug reported here or if the issue remains. More importantly, please open a new bug report for each new bug/regression introduced by the 2.6.27 kernel and tag the bug report with 'linux-2.6.27'. Also, please specifically note if the issue does or does not appear in the 2.6.26 kernel. Thanks again, we really appreicate your help and feedback.

Unfortunately this bug report is being closed because we received no response to the last inquiry for information. However, the Intrepid Ibex 8.10 Beta release was most recently announced - http://www.ubuntu.com/testing/intrepid/beta . If you are able to confirm this is still an issue with this most recent release please feel free to reopen this report. To reopen the bug report you can click on the current status, under the Status column, and change the Status back to "New".

Thanks,
Markus

Additionally, I'm closing the linux-source-2.6.20 task as this does not qualify for a Stable Release Update. Please refer to http://wiki.ubuntu.com/StableReleaseUpdates for more information. As Markus noted above, if you can confirm this is still an issue with the actively developed kernel please feel free to reopen the "linux" task. Thanks.

Yes, on my Athlon 5600, linux-2.6.27-7 it occurs too. /proc/cpuinfo shows frequency 1000 Mhz for each core.
/sys/devices/system/cpu/cpu0/cpufreq/cpu_min_freq: 1000000
/sys/devices/system/cpu/cpu0/cpufreq/cpu_cur_freq: 1000000
/sys/devices/system/cpu/cpu0/cpufreq/cpu_max_freq: 2800000

I am using Ubuntu Intrepid beta.

> Yes, on my Athlon 5600, linux-2.6.27-7 it occurs too. /proc/cpuinfo shows frequency 1000 Mhz for each core.
Sorry, it's not a bug. Powernowd was slowing down the CPU, scaling works correctly.

Per a decision made by the Ubuntu Kernel Team, bugs will longer be assigned to the ubuntu-kernel-team in Launchpad as part of the bug triage process. The ubuntu-kernel-team is being unassigned from this bug report. Refer to https://wiki.ubuntu.com/KernelTeamBugPolicies for more information. Thanks.