Fehler bei cpufreq-info

Fehler bei cpufreq-info

[Rainer Maier]

cpufrequtils 007: cpufreq-info (C) Dominik Brodowski 2004-2009
Bitte melden Sie Fehler an cpufreq@vger.kernel.org.
analysiere CPU 0:
   Treiber: p4-clockmod
   Folgende CPUs laufen mit der gleichen Hardware-Taktfrequenz: 0
   Die Taktfrequenz folgender CPUs werden per Software koordiniert: 0
   Maximale Dauer eines Taktfrequenzwechsels: 10.00 ms.
   Hardwarebedingte Grenzen der Taktfrequenz: 317 MHz - 2.53 GHz
   mögliche Taktfrequenzen: 317 MHz, 633 MHz, 950 MHz, 1.27 GHz, 1.58 GHz, 1.90 GHz, 2.22 GHz, 2.53 GHz
   mögliche Regler: userspace, powersave, conservative, ondemand, performance
   momentane Taktik: die Frequenz soll innerhalb 317 MHz und 2.53 GHz.
                     liegen. Der Regler "performance" kann frei entscheiden,
                     welche Taktfrequenz innerhalb dieser Grenze verwendet wird.
   momentane Taktfrequenz ist 2.53 GHz  (verifiziert durch Nachfrage bei der Hardware).
   Statistik:317 MHz:0,00%, 633 MHz:0,00%, 950 MHz:0,00%, 1.27 GHz:0,00%, 1.58 GHz:0,00%, 1.90 GHz:0,00%, 2.22 GHz:0,00%, 2.53 GHz:100,00%

-- 
Rainer Maier
Fabriciusstr. 46
65933 Frankfurt
T: +49 69 3904 7935
M: +49 176 3472 4262

Re: Fehler bei cpufreq-info

[Viresh Kumar]

2013/3/29 Rainer Maier <RaMaier@gmx.de>:
> cpufrequtils 007: cpufreq-info (C) Dominik Brodowski 2004-2009
> Bitte melden Sie Fehler an cpufreq@vger.kernel.org.
> analysiere CPU 0:
>   Treiber: p4-clockmod
>   Folgende CPUs laufen mit der gleichen Hardware-Taktfrequenz: 0
>   Die Taktfrequenz folgender CPUs werden per Software koordiniert: 0
>   Maximale Dauer eines Taktfrequenzwechsels: 10.00 ms.
>   Hardwarebedingte Grenzen der Taktfrequenz: 317 MHz - 2.53 GHz
>   mögliche Taktfrequenzen: 317 MHz, 633 MHz, 950 MHz, 1.27 GHz, 1.58 GHz,
> 1.90 GHz, 2.22 GHz, 2.53 GHz
>   mögliche Regler: userspace, powersave, conservative, ondemand, performance
>   momentane Taktik: die Frequenz soll innerhalb 317 MHz und 2.53 GHz.
>                     liegen. Der Regler "performance" kann frei entscheiden,
>                     welche Taktfrequenz innerhalb dieser Grenze verwendet
> wird.
>   momentane Taktfrequenz ist 2.53 GHz  (verifiziert durch Nachfrage bei der
> Hardware).
>   Statistik:317 MHz:0,00%, 633 MHz:0,00%, 950 MHz:0,00%, 1.27 GHz:0,00%,
> 1.58 GHz:0,00%, 1.90 GHz:0,00%, 2.22 GHz:0,00%, 2.53 GHz:100,00%

What's the context of this mail? And why is it written in German?

Re: Fehler bei cpufreq-info

[Thomas Renninger]

Hi,

On Friday, March 29, 2013 01:53:48 PM Rainer Maier wrote:
> cpufrequtils 007: cpufreq-info (C) Dominik Brodowski 2004-2009
cpufrequtils is outdated.
Please use cpupower which is the successor.

> Bitte melden Sie Fehler an cpufreq@vger.kernel.org.
> analysiere CPU 0:
>    Treiber: p4-clockmod
P4-clockmod is evil and does not save much/any power.
If you have trouble with it (lagging applications/performance), best
blacklist p4-clockmod driver.

       Thomas

Re: Fehler bei cpufreq-info

[Rainer Maier]

Hi Thomas,
thanks for your info.
I removed cpufrequtils, but when I tried to install cpupower by aptitude it didn't know the package.
I searched for it, but couldn't find one.
Would you perhaps know which packet to install ?

Thanks
Rainer

Am 02.04.2013 10:42, schrieb Thomas Renninger:
> Hi,
>
> On Friday, March 29, 2013 01:53:48 PM Rainer Maier wrote:
>> cpufrequtils 007: cpufreq-info (C) Dominik Brodowski 2004-2009
> cpufrequtils is outdated.
> Please use cpupower which is the successor.
>
>> Bitte melden Sie Fehler an cpufreq@vger.kernel.org.
>> analysiere CPU 0:
>>     Treiber: p4-clockmod
> P4-clockmod is evil and does not save much/any power.
> If you have trouble with it (lagging applications/performance), best
> blacklist p4-clockmod driver.
>
>         Thomas
>

-- 
Rainer Maier
Fabriciusstr. 46
65933 Frankfurt
T: +49 69 3904 7935
M: +49 176 3472 4262

Re: Fehler bei cpufreq-info

[Petr Šabata]

[-- Attachment #1: Type: text/plain, Size: 1411 bytes --]

On Tue, Apr 02, 2013 at 11:29:23PM +0200, Rainer Maier wrote:
> Hi Thomas,
> thanks for your info.
> I removed cpufrequtils, but when I tried to install cpupower by aptitude it didn't know the package.
> I searched for it, but couldn't find one.
> Would you perhaps know which packet to install ?
> 
> Thanks
> Rainer

The cpupower utilities are part of the kernel tree now and
included in the kernel-tools package on Fedora.  I suppose it
will be something similar in your distribution.

Petr

> 
> Am 02.04.2013 10:42, schrieb Thomas Renninger:
> >Hi,
> >
> >On Friday, March 29, 2013 01:53:48 PM Rainer Maier wrote:
> >>cpufrequtils 007: cpufreq-info (C) Dominik Brodowski 2004-2009
> >cpufrequtils is outdated.
> >Please use cpupower which is the successor.
> >
> >>Bitte melden Sie Fehler an cpufreq@vger.kernel.org.
> >>analysiere CPU 0:
> >>    Treiber: p4-clockmod
> >P4-clockmod is evil and does not save much/any power.
> >If you have trouble with it (lagging applications/performance), best
> >blacklist p4-clockmod driver.
> >
> >        Thomas
> >
> 
> -- 
> Rainer Maier
> Fabriciusstr. 46
> 65933 Frankfurt
> T: +49 69 3904 7935
> M: +49 176 3472 4262
> --
> To unsubscribe from this list: send the line "unsubscribe cpufreq" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

[-- Attachment #2: Type: application/pgp-signature, Size: 230 bytes --]

Re: Fehler bei cpufreq-info

[Thomas Renninger]

On Wednesday, April 03, 2013 09:53:04 AM Petr Å abata wrote:
> On Tue, Apr 02, 2013 at 11:29:23PM +0200, Rainer Maier wrote:
> > Hi Thomas,
> > thanks for your info.
> > I removed cpufrequtils, but when I tried to install cpupower by aptitude
> > it didn't know the package. I searched for it, but couldn't find one.
> > Would you perhaps know which packet to install ?
> > 
> > Thanks
> > Rainer
> 
> The cpupower utilities are part of the kernel tree now and
> included in the kernel-tools package on Fedora.  I suppose it
> will be something similar in your distribution.

Sigh.
I guess you have a separate perf package at least?
It would be great if cpupower gets packaged separately for two
reasons:
1) cpuidle and cpufreq are used by a lot archs nowadays:
    ppc, arm, afaik S390 at least they tried,...   
    I expect the kernel-tools package will only compile on X86?
2) To avoid confusion like above, so that this tool can easily
    be found by people searching for it.

Petr: Would you mind forward this to the maintainer.
I once made sure cpupower compiles on ppc iirc, if someone
runs into arch specific compile (or runtime) issues, please let
me know, this should get fixed then.

Thanks,

   Thomas

Rainer: Again, if you have any cpufreq related problems it's probably
to p4-clockmod which is not a real cpufreq driver. The technique used
should only be used to avoid critical cpu temperature and does not
save you power, more the other way around (things should be processed
quickly so that CPU sleep states can be entered which are most efficient).
It even can be that p4-clockmode interferes with another interface doing
the same (CPU throttling, T-states) then your system becomes even more
laggy up to unusable.

Re: Fehler bei cpufreq-info

[Petr Šabata]

[-- Attachment #1: Type: text/plain, Size: 1888 bytes --]

On Fri, Apr 05, 2013 at 04:01:20PM +0200, Thomas Renninger wrote:
> On Wednesday, April 03, 2013 09:53:04 AM Petr Šabata wrote:
> > On Tue, Apr 02, 2013 at 11:29:23PM +0200, Rainer Maier wrote:
> > > Hi Thomas,
> > > thanks for your info.
> > > I removed cpufrequtils, but when I tried to install cpupower by aptitude
> > > it didn't know the package. I searched for it, but couldn't find one.
> > > Would you perhaps know which packet to install ?
> > > 
> > > Thanks
> > > Rainer
> > 
> > The cpupower utilities are part of the kernel tree now and
> > included in the kernel-tools package on Fedora.  I suppose it
> > will be something similar in your distribution.
> 
> Sigh.
> I guess you have a separate perf package at least?

Yes, but technically it's a kernel subpackage.

> It would be great if cpupower gets packaged separately for two
> reasons:
> 1) cpuidle and cpufreq are used by a lot archs nowadays:
>     ppc, arm, afaik S390 at least they tried,...   
>     I expect the kernel-tools package will only compile on X86?

They used to be a separate package before the code moved
to kernel.  I see the package is build for ppc[64] but not
on s390[x].  I have no information on arm[64] at the moment
but if there are issues, I'll let you know :)

> 2) To avoid confusion like above, so that this tool can easily
>     be found by people searching for it.

This is handled by package manager.  Besides the list of of
binaries the tools package should also act as a replacement
for former cpufrequtils and cpupowerutils packages.

> Petr: Would you mind forward this to the maintainer.
> I once made sure cpupower compiles on ppc iirc, if someone
> runs into arch specific compile (or runtime) issues, please let
> me know, this should get fixed then.

See above.

I'll forward this to kernel owners in Fedora.

Regards,
Petr

[-- Attachment #2: Type: application/pgp-signature, Size: 230 bytes --]

Re: Fehler bei cpufreq-info

[Thomas Renninger]

[-- Attachment #1: Type: text/plain, Size: 2187 bytes --]

On Friday, April 05, 2013 04:31:09 PM Petr Šabata wrote:
> On Fri, Apr 05, 2013 at 04:01:20PM +0200, Thomas Renninger wrote:
> > On Wednesday, April 03, 2013 09:53:04 AM Petr Šabata wrote:
> > > On Tue, Apr 02, 2013 at 11:29:23PM +0200, Rainer Maier wrote:
> > > > Hi Thomas,
> > > > thanks for your info.
> > > > I removed cpufrequtils, but when I tried to install cpupower by
> > > > aptitude
> > > > it didn't know the package. I searched for it, but couldn't find one.
> > > > Would you perhaps know which packet to install ?
> > > > 
> > > > Thanks
> > > > Rainer
> > > 
> > > The cpupower utilities are part of the kernel tree now and
> > > included in the kernel-tools package on Fedora.  I suppose it
> > > will be something similar in your distribution.
> > 
> > Sigh.
> > I guess you have a separate perf package at least?
> 
> Yes, but technically it's a kernel subpackage.
Sure, but I doubt it makes sense to put every tool in there.
For bigger ones like perf, and IMO also cpupower it makes
sense to package them separately.
 
> > It would be great if cpupower gets packaged separately for two
> > reasons:
> > 
> > 1) cpuidle and cpufreq are used by a lot archs nowadays:
> >     ppc, arm, afaik S390 at least they tried,...
> >     I expect the kernel-tools package will only compile on X86?
> 
> They used to be a separate package before the code moved
> to kernel.  I see the package is build for ppc[64] but not
> on s390[x].
Yep, I double checked: cpupower builds on ppc here too.

> I have no information on arm[64] at the moment
> but if there are issues, I'll let you know :)
> 
> > 2) To avoid confusion like above, so that this tool can easily
> > 
> >     be found by people searching for it.
> 
> This is handled by package manager.  Besides the list of of
> binaries the tools package should also act as a replacement
> for former cpufrequtils and cpupowerutils packages.

It would really be great if this gets packaged separately.
Find attached my .spec file.
If someone looks at this, he can freely take over whatever he wants.
...
> I'll forward this to kernel owners in Fedora.

Thanks,

   Thomas

[-- Attachment #2: cpupower.spec --]
[-- Type: text/x-rpm-spec, Size: 4212 bytes --]

#
# spec file for package cpupower
#
# Copyright (c) 2013 SUSE LINUX Products GmbH, Nuernberg, Germany.
# Author: Thomas Renninger <trenn@suse.de>
#
# All modifications and additions to the file contributed by third parties
# remain the property of their copyright owners, unless otherwise agreed
# upon. The license for this file, and modifications and additions to the
# file, is the same license as for the pristine package itself (unless the
# license for the pristine package is not an Open Source License, in which
# case the license is the MIT License). An "Open Source License" is a
# license that conforms to the Open Source Definition (Version 1.9)
# published by the Open Source Initiative.

# Please submit bugfixes or comments via http://bugs.opensuse.org/
#



Name:           cpupower
Url:            http://www.kernel.org/pub/linux/utils/kernel/cpufreq/cpufrequtils.html
# Use this as version when things are in mainline kernel
%define version %(rpm -q --qf '%{VERSION}' kernel-source)
Version:        3.8
Release:        0
Summary:        Tools to determine and set CPU Power related Settings
License:        GPL-2.0
Group:          System/Base
Source:         %{name}-%{version}.tar.bz2
Source1:        cpupower_export_tarball_from_git.sh
Patch1:         cpupower-haswell_support.patch
BuildRoot:      %{_tmppath}/%{name}-%{version}-build
BuildRequires:  gettext-tools
BuildRequires:  pciutils
BuildRequires:  pciutils-devel
Obsoletes:      cpufrequtils < %version
Provides:       cpufrequtils = %version

%description
This tool is to make access to the Linux kernel's processor power subsystems
like CPU frequency switching (cpufreq) or CPU sleep states (cpuidle) for users
and userspace tools easier.

%package -n libcpupower0
Summary:        Obsolete processor frequency related C-library
Group:          System/Base

%description -n libcpupower0
Contains libcpupower and soon possibly other CPU power related
C libraries.

%package devel
Summary:        Include files and C-libraries
Group:          Development/Languages/C and C++
Requires:       libcpupower0 = %{version}

%description devel
Include files and C-libraries for C/C++ development

%package bench
Summary:        CPU frequency micro benchmark
Group:          System/Benchmark
Obsoletes:      cpufrequtils-bench < %version
Provides:       cpufrequtils-bench = %version

%description bench
Helps to test the condition of a given kernel cpufreq
governor (e.g. ondemand, userspace, conservative) and the cpufreq HW driver
(e.g. powernow-k8, acpi-cpufreq, ...).
For that purpose, it compares the performance governor to a configured
powersave module.

%prep
%setup -q
%patch1 -p1

%build
# This package failed when testing with -Wl,-as-needed being default.
# So we disable it here, if you want to retest, just delete this comment and the line below.
export SUSE_ASNEEDED=0
CONF="PACKAGE_BUGREPORT=http://bugs.opensuse.org mandir=%_mandir libdir=%_libdir CPUFRQ_BENCH=true STRIP=true VERSION=%{version}"
export CFLAGS="$RPM_OPT_FLAGS"
make $CONF %{?_smp_mflags}

%install
export SUSE_ASNEEDED=0
CONF="PACKAGE_BUGREPORT=http://bugs.opensuse.org mandir=%_mandir libdir=%_libdir CPUFRQ_BENCH=true DESTDIR=$RPM_BUILD_ROOT sbindir=/usr/sbin docdir=%{_docdir}/%{name} confdir=/etc VERSION=%{version}"
make install DESTDIR=$RPM_BUILD_ROOT $CONF
# copy to examples doc dir to avoid complains from the build
# system about an executable in the doc dir.
mkdir -p $RPM_BUILD_ROOT/%{_docdir}/%{name}/examples
mv $RPM_BUILD_ROOT/%{_docdir}/%{name}/cpufreq-bench_script.sh $RPM_BUILD_ROOT/%{_docdir}/%{name}/examples
%find_lang %{name}

%post -n libcpupower0 -p /sbin/ldconfig

%postun -n libcpupower0 -p /sbin/ldconfig

%files -f %{name}.lang
%defattr (-,root,root)
%{_mandir}/man1/cpupower*
/usr/bin/cpupower

%files bench
%defattr (-,root,root)
%config /etc/cpufreq-bench.conf
/usr/sbin/cpufreq-bench
/usr/bin/cpufreq-bench_plot.sh
%dir %{_docdir}/%{name}
%dir %{_docdir}/%{name}/examples
%{_docdir}/%{name}/examples/cpufreq-bench_script.sh
%{_docdir}/%{name}/README-BENCH

%files -n libcpupower0
%defattr(-,root,root)
%{_libdir}/libcpupower*.so.*

%files devel
%defattr(-,root,root)
/usr/include/cpufreq.h
%{_libdir}/libcpu*.so

%changelog