[procps] Re: C-States handling - new switch?
- From: Craig Small <csmall-procps@xxxxxxxxxx>
- To: procps@xxxxxxxxxxxxx
- Date: Fri, 10 Feb 2012 10:01:02 +1100
On Thu, Feb 09, 2012 at 07:02:18AM -0500, Jaromir Capik wrote:
> > > 1.) with frequency set to 2.66GHz the load was approximately 34%
> > > 2.) with frequency set to 1.73GHz the load was approximately 52%
> > > Whilst the load should be 34% in both cases ...
Just so its clear in my head, you have a program that consumes
0.9044 GHZ of CPU (2.66 x 34%) in the first instance and 0.8996 GHz
of CPU (1.73 x 52%).
I assume they are supposed to be, given how top works today, the same
and its a rounding error for argument's sake.
So, for a constant load of 0.5 GHz on a program on a 2GHz CPU top says
the load is 25%. If the CPU drops to 1GHz the load is now 50% even
though the "real load" hasn't changed, it is still 0.5 GHz?
The problem is that top sees 500 user ticks out of 2000 user ticks in
the first instance and 500 user ticks out of 1000 user ticks in the
second.
Is that the problem here?
It sounds like Jaromir needs the kernel to have a "CPU bludging because
its running slower" tick.
If so, is there something in the kernel to see this?
- Craig
--
Craig Small VK2XLZ http://enc.com.au/ csmall at : enc.com.au
Debian GNU/Linux http://www.debian.org/ csmall at : debian.org
GPG fingerprint: 5D2F B320 B825 D939 04D2 0519 3938 F96B DF50 FEA5
Other related posts:
