[Bug 54761] New: Variable cpu frequency causes to loose performance

[Bug 54761] New: Variable cpu frequency causes to loose performance

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=54761

           Summary: Variable cpu frequency causes to loose performance
           Product: Power Management
           Version: 2.5
    Kernel Version: Ubuntu 3.8.0-9.18-generic 3.8.1
          Platform: All
        OS/Version: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: cpufreq
        AssignedTo: cpufreq@vger.kernel.org
        ReportedBy: sworddragon2@aol.com
        Regression: No


Created an attachment (id=94381)
 --> (https://bugzilla.kernel.org/attachment.cgi?id=94381)
Example application

In my sytsme is an AMD Phenom II X6 1045T with 6 cores and a scaling frequency
of "2700000 2000000 1400000 800000". I have noticed if a process is using ~the
half of cpu time of a core I'm getting heavy performance impacts if I'm using
the ondemand governor. Switching to the performance governor solves this
problem.

I have written a testcase in C which demonstrates this problem (compiled with
"gcc -o governor governor.c -lpthread"). The application creates 20 threads
(change NUM_THREADS in the code if needed) - every of them uses only a little
of cpu time. All threads will be executed on the same core. Here are the
differences between both governors:


ondemand:
The frequency of the related core goes up to 1400 MHz and the used cpu time of
this core is ~50%. The application needs ~7.17 seconds to be finished.

performance:
The frequency of the related core is constantly 2700 MHz and the used cpu time
of this core is ~25%. The application needs ~5.99 seconds to be finished.


The difference between the used cpu time is logical since the frequency is as
double so high. But both runs are far away from a 100% usage so theoretically
they should be as same fast.

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

[Bug 54761] Variable cpu frequency causes to loose performance

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=54761


Aaron Lu <aaron.lu@intel.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |aaron.lu@intel.com
         AssignedTo|cpufreq@vger.kernel.org     |aaron.lu@intel.com




--- Comment #1 from Aaron Lu <aaron.lu@intel.com>  2013-03-26 07:43:18 ---
Hi,

Does the problem occur on previous kernels? If possible, can you please
identify from which kernel this problem starts to occur, thanks.

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

[Bug 54761] Variable cpu frequency causes to loose performance

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=54761


Aaron Lu <aaron.lu@intel.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|aaron.lu@intel.com          |cpufreq@vger.kernel.org




-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

[Bug 54761] Variable cpu frequency causes to loose performance

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=54761





--- Comment #3 from sworddragon2@aol.com  2013-03-26 09:15:00 ---
Does this mean that 1400 MHz vs 2700 MHz could be slower with ondemand even if
both variants are below 100% of cpu usage? If yes this means it is intended
that ondemand has sometimes performance impacts and this ticket can be marked
as invalid.

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

[Bug 54761] Variable cpu frequency causes to loose performance

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=54761





--- Comment #4 from Viresh Kumar <viresh.kumar@linaro.org>  2013-03-26 09:41:40 ---
1400 MHz would always be slower than 2700 MHz. And ondemand can't simply switch
to max without much load. So, you can say it that way "Ondemand sometimes have
performance impact" :)

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

[Bug 54761] Variable cpu frequency causes to loose performance

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=54761





--- Comment #5 from sworddragon2@aol.com  2013-03-26 13:24:50 ---
I thought 1400 MHz @ ~50% cpu load vs 2700 MHz @ ~25% cpu load would have
mathematically ~the same performance but it doesn't seem so. What is the reason
for this? Maybe because of short 100% peaks that aren't enough for an
upscaling?

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

[Bug 54761] Variable cpu frequency causes to loose performance

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=54761


Lan Tianyu <tianyu.lan@intel.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |tianyu.lan@intel.com
         AssignedTo|cpufreq@vger.kernel.org     |tianyu.lan@intel.com




--- Comment #6 from Lan Tianyu <tianyu.lan@intel.com>  2013-04-18 09:12:32 ---
This one is as same as bug 55261. Not an actual bug. So mark as invalid.
You can continue to discuss in the bug 55261.

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.