nForce2 cpufreq driver does not load by default

[Paul F]

== Quantal SRU Justification ==
cpufreq-nforce2 is not autoloadable. We built this in for past releases (eg Precise 12.04). Reinstate this as built in rather than enabled as a module.

== Test Case ==
Boot a kernel with cpufreq-nforce2 driver enabled as a module and confirm the cpufreq-nforce2 doesn't load by default. Because the cpufreq driver is not loaded, there is no /sys/devices/system/cpu/cpu0/cpufreq/ directory. This is used by CPU monitoring and/or scaling software to report or change CPU speed, governor etc. All software using this interface will fail unless the cpufreq driver is explicitly loaded post boot.

Then boot a kernel with the cpufreq-nforce2 driver built in and confirm the described issues above are resolved.

----------

nForce2 cpufreq driver does not load by default

On upgrading to Quantal 12.10 from Precise 12.04.1 I noticed that the cpufreq nforce2 driver (cpufreq-nforce2) no longer loads at boot.

The driver loads fine manually with modprobe or by adding an explicit /etc/default/loadcpufreq file. Once loaded all works as expected.

Incidentally, trying to load it using kernel command line parameter cpufreq_driver=cpufreq-nforce2 fails. Is this a bug? Am I misusing it? Is it related to this problem?

I also tried a clean install of Quantal to rule out upgrade jip. Same result.

The last kernel published to updates on Precise before my upgrade was:
% uname -a
Linux NF7-S1 3.2.0-32-generic-pae #51-Ubuntu SMP Wed Sep 26 21:54:23 UTC 2012 i686 athlon i686 GNU/Linux

which works as expected:

% cpufreq-info
cpufrequtils 008: cpufreq-info (C) Dominik Brodowski 2004-2009
Report errors and bugs to <email address hidden>, please.
analyzing CPU 0:
  driver: nforce2
  CPUs which run at the same hardware frequency: 0
  CPUs which need to have their frequency coordinated by software: 0
  maximum transition latency: 0.00 ms.
  hardware limits: 1.46 GHz - 2.09 GHz
  available cpufreq governors: conservative, ondemand, userspace, powersave, performance
  current policy: frequency should be within 1.46 GHz and 2.09 GHz.
                  The governor "performance" may decide which speed to use
                  within this range.
  current CPU frequency is 2.09 GHz.

The Quantal kernel is:
% uname -a
Linux NF7-S1 3.5.0-18-generic #29-Ubuntu SMP Fri Oct 19 10:27:31 UTC 2012 i686 athlon i686 GNU/Linux

which doesn't load the nforce2 driver on boot.

% cpufreq-info
cpufrequtils 008: cpufreq-info (C) Dominik Brodowski 2004-2009
Report errors and bugs to <email address hidden>, please.
analyzing CPU 0:
  no or unknown cpufreq driver is active on this CPU
  maximum transition latency: 0.00 ms.

There has been some builtin / module shuffling since the 3.2.0 kernel wrt cpufreq. For 3.2.0:
% ls /lib/modules/3.2.0-32-generic-pae/kernel/drivers/cpufreq

p4-clockmod.ko
pcc-cpufreq.ko

For 3.5.0:
% ls /lib/modules/3.5.0-18-generic/kernel/drivers/cpufreq/

cpufreq-nforce2.ko
gx-suspmod.ko
longhaul.ko
longrun.ko
p4-clockmod.ko
powernow-k6.ko
powernow-k7.ko

So, cpufreq-nforce2 was a builtin for the 3.2.0 kernel and is now a module. Coincidence? Looks suspicious.

Consequences
Because the cpufreq driver is not loaded, there is no /sys/devices/system/cpu/cpu0/cpufreq/ directory. This is used by CPU monitoring and/or scaling software to report or change CPU speed, governor et cetera. All s/w using this interface will fail unless the cpufreq driver is explicitly loaded post boot. I noticed the problem because my conky stopped working -- it references /sys/devices/system/cpu/cpu0/cpufreq/*. Once the driver is manually loaded, the conky works as expected.

To reproduce
This will require access to old metal. This bug report comes from an Abit NF7-S V2 with an AMD Athlon XP-M which has been running Ubuntu since Lucid without problem. I would expect any nForce2 chipset motherboard to reproduce this bug.

There is a similar bug (fix released) for the PCC cpufreq driver:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1061126

Thank you.

ProblemType: Bug
DistroRelease: Ubuntu 12.10
Package: linux-image-3.5.0-18-generic 3.5.0-18.29
ProcVersionSignature: Ubuntu 3.5.0-18.29-generic 3.5.7
Uname: Linux 3.5.0-18-generic i686
ApportVersion: 2.6.1-0ubuntu6
Architecture: i386
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: p 1911 F.... pulseaudio
Date: Fri Nov 16 21:47:57 2012
HibernationDevice: RESUME=UUID=b3f266c2-a7bf-4448-87f7-ecbc65e1f22b
InstallationDate: Installed on 2012-05-19 (180 days ago)
InstallationMedia: Ubuntu 12.04 LTS "Precise Pangolin" - Release i386 (20120423)
MarkForUpload: True
ProcFB: 0 radeondrmfb
ProcKernelCmdLine: BOOT_IMAGE=/vmlinuz-3.5.0-18-generic root=UUID=1814a781-3e1c-468c-abff-8b99f9b8f6d9 ro rootdelay=50 vt.handoff=7
RelatedPackageVersions:
 linux-restricted-modules-3.5.0-18-generic N/A
 linux-backports-modules-3.5.0-18-generic N/A
 linux-firmware 1.95
RfKill:
 0: phy0: Wireless LAN
  Soft blocked: no
  Hard blocked: no
SourcePackage: linux
UpgradeStatus: Upgraded to quantal on 2012-11-11 (5 days ago)
dmi.bios.date: 11/22/2004
dmi.bios.vendor: Phoenix Technologies, LTD
dmi.bios.version: 6.00 PG
dmi.board.name: NF7-S/NF7,NF7-V (nVidia-nForce2)
dmi.board.vendor: http://www.abit.com.tw/
dmi.board.version: 2.X,1.0
dmi.chassis.type: 3
dmi.modalias: dmi:bvnPhoenixTechnologies,LTD:bvr6.00PG:bd11/22/2004:svn:pn:pvr:rvnhttp//www.abit.com.tw/:rnNF7-S/NF7,NF7-V(nVidia-nForce2):rvr2.X,1.0:cvn:ct3:cvr:

[Brad Figg (brad-figg)]

This change was made by a bot.

[Joseph Salisbury (jsalisbury)]

Would it be possible for you to test the latest upstream kernel? Refer to https://wiki.ubuntu.com/KernelMainlineBuilds . Please test the latest v3.7 kernel[0] (Not a kernel in the daily directory) and install both the linux-image and linux-image-extra .deb packages.

If this bug is fixed in the mainline kernel, please add the following tag 'kernel-fixed-upstream'.

If the mainline kernel does not fix this bug, please add the tag: 'kernel-bug-exists-upstream'.

If you are unable to test the mainline kernel, for example it will not boot, please add the tag: 'kernel-unable-to-test-upstream'.
Once testing of the upstream kernel is complete, please mark this bug as "Confirmed".

Thanks in advance.

[0] http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.7-rc5-raring/

[Paul F (boxjunk)]

Tested against latest upstream kernel per #3

% uname -a
Linux NF7-S1 3.7.0-030700rc6-generic #201211162135 SMP Sat Nov 17 02:44:28 UTC 2012 i686 athlon i686 GNU/Linux

The bug is not fixed in the mainline kernel.

[Joseph Salisbury (jsalisbury)]

Does the module load automatically if you boot back into the Precise(12.04) kernel?

[Paul F (boxjunk)]

Yes, booting kernel 3.2.0-32-generic-pae #51under Quantal loads the module automagically as expected.

[Paul F (boxjunk)]

CORRECTION to #6

The nforce2 cpufreq driver is a builtin under 3.2.0-32-generic-pae (see original bug desc), so the answer to your question is no, the \em module \ is not loaded, but the driver is present and works as expected. Sorry for confusion re #5.

[Launchpad Janitor (janitor)]

This bug was fixed in the package linux - 3.7.0-3.9

---------------
linux (3.7.0-3.9) raring; urgency=low

  [ Leann Ogasawara ]

  * [Config] Enable CONFIG_X86_CPUFREQ_NFORCE2=y
    - LP: #1079900
  * Add nfsv3 to nfs-modules udeb

  [ Paolo Pisati ]

  * [Config] SND_OMAP_SOC*=y
    - LP: #1019321

  [ Stefan Bader ]

  * SAUCE: (no-up) xen/netfront: handle compound page fragments on transmit
    - LP: #1078926

  [ Tim Gardner ]

  * Revert "SAUCE: SECCOMP: audit: always report seccomp violations"
    - LP: #1079469
  * Revert "SAUCE: omap3 clocks .dev_id = NULL"
  * rebase to v3.7-rc6
  * SAUCE: script to detect obsolete firmware
  * SAUCE: Remove yam files duplicated in linux-firmware
  * SAUCE: Remove tehuti files duplicated in linux-firmware
  * SAUCE: Remove matrox files duplicated in linux-firmware
  * SAUCE: Remove cxgb3 files duplicated in linux-firmware
  * SAUCE: Remove r128 files duplicated in linux-firmware
  * SAUCE: Remove acenic files duplicated in linux-firmware
  * SAUCE: Remove keyspan files duplicated in linux-firmware
  * SAUCE: Remove sun files duplicated in linux-firmware
  * SAUCE: Remove radeon files duplicated in linux-firmware
  * SAUCE: Update bnx2x firmware to 7.8.2.0
  * [Config] generic.inclusion-list: econet has disappeared

  [ Upstream Kernel Changes ]

  * seccomp: forcing auditing of kill condition
    - LP: #1079469
  * rebase to v3.7-rc6
 -- Leann Ogasawara <email address hidden> Tue, 20 Nov 2012 12:28:55 -0800

[Paul F (boxjunk)]

I have discovered a (loosely) related issue. The connection is that the new bug LP: #1084262, has surfaced due to cpufreq driver kernel module file name changes.

[Paul F (boxjunk)]

Fixing LP: #1084262 would resolve this bug.

Proposal: Fix LP: #1084262 and then unwind the fixes here.

This is a better solution than the current config fix committed for the 3.7.0-3.9 kernel build:

[Config] Enable CONFIG_X86_CPUFREQ_NFORCE2=y

The main advantage is a reduction in kernel footprint since the built-in will revert to a module which was probably the rationale for making this change in the first place. Multiple built-ins can potentially be modularised, eg the PCC cpufreq driver (see LP: #1061126) which appears to have had the same problem.

The cpufrequtils package maintainer would ideally fix this upstream (see LP: #1084262 for patch).

[Adam Conrad (adconrad)]

The verification of this Stable Release Update has completed successfully and the package has now been released to -updates. Subsequently, the Ubuntu Stable Release Updates Team is being unsubscribed and will not receive messages about this bug report. In the event that you encounter a regression using the package from -updates please report a new bug using ubuntu-bug and tag the bug report regression-update so we can easily find any regresssions.

[Joseph Salisbury (jsalisbury)]

This bug was nominated against a series that is no longer supported, ie quantal. The bug task representing the quantal nomination is being closed as Won't Fix.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.