Re: [Xen-devel] A question on the credit scheduler

[George Dunlap <George.Dunlap@xxxxxxxxxxxxx>]

2011/12/17 gavin <gbtux@xxxxxxx>:
>
> At 2011-12-16 23:58:26,"George Dunlap" <George.Dunlap@xxxxxxxxxxxxx> wrote:
>>2011/12/16 gavin <gbtux@xxxxxxx>:
>>> At 2011-12-16 19:04:19,"George Dunlap" <George.Dunlap@xxxxxxxxxxxxx> wrote:
>>>
>>>>2011/12/16 zhikai <gbtux@xxxxxxx>:
>>>>> Hi All,
>>>>>
>>>>> In the credit scheduler, the scheduling decision function 
>>>>>csched_schedule()
>>>>> is called in the schedule function in scheduler.c, such as the following.
>>>>> next_slice = sched->do_schedule(sched, now, tasklet_work_scheduled);
>>>>>
>>>>> But, how often the csched_schedule() is called and to run? Does this
>>>>> frequency have something to do with the slice of credit scheduler that is
>>>>> 30ms?
>>>>
>>>>The scheduler runs whenever the SCHEDULE_SOFTIRQ is raised.  If you
>>>>grep through the source code fro that string, you can find all the
>>>>places where it's raised.
>>>>
>>>>Some examples include:
>>>>* When the 30ms timeslice is finished
>>>>* When a sleeping vcpu of higher priority than what's currently running 
>>>>wakes up
>>>>* When a vcpu blocks
>>>>* When a vcpu is migrated from one cpu to another
>>>>
>>>>30ms is actually a pretty long time; in typical workloads, vcpus block
>>>>or are preempted by other waking vcpus without using up their full
>>>>timeslice.
>>>
>>> Thank you very much for your reply.
>>>
>>> So, the vcpu is very likely to be preempted whenever the SCHEDULE_SOFTIRQ is
>>> raised.
>>
>>It depends; if you have a cpu-burning vcpu running on a cpu all by
>>itself, then after its 30ms timeslice, Xen will have no one else to
>>run, and so will let it run again.
>>
>>But yes, if there are other vcpus on the runqueue, or the host is
>>moderately busy, it's likely that SCHEDULE_SOFTIRQ will cause a
>>context-switch.
>>
>>> And we cannot find a small timeslice, such as a(ms), which makes the
>>> time any vcpu spending on running phase is k*a(ms), k is integer here. There
>>> is no such a small timeslice. Is it right?
>>
>>I'm sorry, I don't really understand your question.  Perhaps if you
>>told me what you're trying to accomplish?
>
> I try to describe my idea as the following clearly. But I really don't know
> if it will work. Please give me some advice if possible.
>
> According to the credit scheduler in Xen, a vCPU can run a 30ms timeslice
> when it is scheduled on the physical CPU. And, a vCPU with the BOOST
> priority will preempt the running one and run additional 10ms. So, what I
> think is if we monitor the physical CPU every 10ms and we can get the
> mapping information of a physical CPU and a vCPU. And also, we can get the
> un-mapping information that a physical CPU isn’t mapped to any vCPU. Thus,
> we can get the CPU usage by calculating the proportion of the mapping
> information to the total time when we monitored.
>
> For example, if we monitor the physical CPUs every 10ms and we can get 100
> pairs of pCPU and vCPU in a second, such as (pCPU_id, vCPU_id). If there is
> 60 mapping pairs that the pCPU is mapped to a valid vPCU, and 40 un-mapping
> pairs that we cannot find the pCPU to be mapped a valid vCPU. So, we can get
> the usage of the physical CPUs that is 60%.
>
> Here, we monitor the physical CPUs every 10ms. We also can monitor them once
> less than the 10ms interval, such as 1ms interval. Whatever interval we
> choose, we must make sure no CPU content switch in the interval or the
> context switch always occur at the edge of interval. Only in this condition,
> can this idea work.
>
> So, I am not sure whether we can find such a time interval that can meet
> this condition. In other words, whether we can find such a time interval
> that ensures all the CPU content switch occur at the edge of interval.

You still haven't described exactly what it is you're trying to
accomplish: what is your end goal?  It seems to be related somehow to
measuring how busy the system is (i.e., the number of active pcpus and
idle pcpus); but as I don't know what you want to do with that
information, I can't tell you the best way to get it.

Regarding a map of pcpus to vcpus, that already exists.  The
scheduling code will keep track of the currently running vcpu here:
  per_cpu(schedule_data, pcpu_id).curr

You can see examples of the above structure used in
xen/common/sched_credit2.c.  If "is_idle(per_cpu(schedule_data,
pcpu).curr)" is false, then the cpu is running a vcpu; if it is true,
then the pcpu is idle (although it may be running a tasklet).

Additionally, if all you want is the number of non-idle cpus, the
credit1 scheduler keeps track of the idle and non-idle cpus in
prv->idlers.  You could easily use "cpumask_weight(&prv->idlers)" to
find out how many idle cpus there are at any given time.  If you know
how many online cpus there are, that will give you the busy-ness of
the system.

So now that you have this instantaneous percentage, what do you want
to do with it?

 -George

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel