Re: [PATCH 2/2] CFS CGroup: Report usage

[Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>]

On Tue, Oct 23, 2007 at 09:41:49AM -0700, Paul Menage wrote:
> > > Adds a cpu.usage file to the CFS cgroup that reports CPU usage in
> > > milliseconds for that cgroup's tasks
> >
> > It would be nice to split this into user and sys time at some point.
> 
> Sounds reasonable - but does CFS track this?

No, not for a group. We could extend account_user_time() and
account_systime_time() in this regard.

> > We have also received request to provide idle time for a
> > container/cgroup.
> 
> The semantics of "idle time" for a cgroup on a shared system seem a
> bit fuzzy. How would you define it?

I think the percentage of time when it didn't have any runnable task in
its runqueues.

> Suppose you have two cgroups that would each want to use, say, 55% of
> a CPU - technically they should each be regarded as having 45% idle
> time, but if they run on a the same CPU the chances are that they will
> both always have some processes on their runqueue due to contention
> with the other group. So how would you measure the difference between
> this and a cgroup that really is trying to use 100%?

Good point. I think we need to subtract out the time it was waiting on runqueue
when calculating idle time.

	|------- . . . . . . ---------zzzzzzzzzzzz.......-----------|
	t0     t1	     t2	      t3 	 t4     t5          t6


	----	-> Running time
	....	-> Waiting time (to get on the cpu)
	zzzz	-> Sleeping time (when it didnt want to run because of
		   lack of tasks)

So, in this case, 

	idle time = (t4 - t3) / [ (t6 - t1) - (t2-t1) - (t5-t4) 

?

This idle time will be a per-cpu stat for every cgroup and needs to be
consolidated across cpus into a single idle-stat number, just like how
top does it.

-- 
Regards,
vatsa