Bug#421879: cpufreq_ondemand: cpufreq_ondemand responds too slowly

[Greg Kochanski <gpk@kochanski.plus.com>]

I got the following response from Intel.   It's a little confused,
perhaps.   Or, maybe if you have ACPI turned on properly, perhaps
the kernel can read the latency from the processor itself.

Anyhow, he says that the latency is 10 - 100 microseconds for Intel
processors.    Since the values of policy->cpuinfo.transition_latency
in the kernel seem to be much longer than that
(based on http://lists.debian.org/debian-devel/2006/12/msg00184.html ,
e.g. the Pentium 4 is used as if it has a 1000 microsecond latency),
it looks like they ought to be reduced to the correct values.


> Hi Greg,
>
> The author responded:
>
> The latency to switch P-state depends on the processor. You can look at the latency value on your particular processor by
> #cat /proc/acpi/processor/CPU*/performance
> ## Note this may need some particular kernel configuration
>  
> Typically this value is 10uS or 100uS. Ondemand, by default, does the monitoring based on this latency. Default sampling interval of ondemand is 1000 times this latency. So, if you have a CPU with latency 100uS, ondemand is already polling at 100 mS (10 times per sec). You may see freq changing lesser number of times than that. But, that will be because thats when CPU load actually changed and not because ondemand is polling once per second. On all Intel systems I have seen ondemand samples atleast 9 times per second.
>  
> You can check ondemand sampling value and increase/decrease it by looking at the tunables in
> /sys/devices/system/cpu/cpu*/cpufreq/ondemand/ directory.
>
> Regards,
>
> Lexi S.
> Intel(R) Software Network Support
> http://www.intel.com/software
> email:  ISN.support@intel.com
>
> Intel is a registered trademark of Intel Corporation or its subsidiaries in the United States and other countries.