lkml.org 
[lkml]   [2012]   [Feb]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH v5] sched: Avoid unnecessary work in reweight_entity
On 02/27/2012 01:10 PM, Srivatsa Vaddagiri wrote:

> * Michael Wang <wangyun@linux.vnet.ibm.com> [2012-02-27 13:07:28]:
>
>>> Is that benchmark run in root (cpu) cgroup? If so, reweight_entity() should not
>>> kick in at all.
>>
>>
>> That's right, if no children group, 'reweight_entity' won't be called, so I have
>> created a cpuset group under root group named 'rg1', and created a memory group
>> under 'rg1' named 'sub', I attached the current shell to the 'sub' cgroup.
>
> 'cpu' and 'cpuset' cgroup resource controllers are separate. By above
> steps, you would still be running kern-bench in root cgroup as far as cpu
> cgroup controller is concerned ..


I think I really need some study on cgroup first...I've done some totally useless
test :( , but still confused that why sys time reduced?

And I got a server with Redhat now, so I will do:

1. cgcreate -g cpu:/subcg1
2. echo $$ > /cgroup/cpu/subcg1/tasks
3. open another shell
4. do 1~3 multi times, plan to get 7 cpu cgroup with 7 shell attached to each one.
5. run kernbench in each shell

Please tell me if I missed some thing :)

Regards,
Michael Wang

>
> - vatsa




\
 
 \ /
  Last update: 2012-02-27 07:25    [W:0.047 / U:0.016 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site