Kohei san ..
The CPU is one scheduler that needs to patch the kernel and can
not be loaded as module. It is quite pervasive at the lowest level
of the CPU scheduling routine in order to track class progress.
-- Hubertus
Kaigai Kohei wrote:
> Hello, Shailabh and Hubertus.
> Thanks for your comments.
>
> I have tried to apply CKRM-Core and CPU-Resource-Controller patchs to
> original 2.6.8.1 kernel.
> Then, this patched kernel was booted up on my desktop.
> --------
>
>>>I tried to run the three infinite-CPU-consuming tasks(one belongs 'gold'
>>>class,
>>>and others belong 'silver' class) on SuSE Linux ES9 kernel, but each tasks
>>>consume
>>>33% of CPU-time. I expected the gold task consumes 50% of CPU-time, and the
>>>silver
>>>tasks consume 25% each.
>
> --------
> Of course, the above fair-scheduling is available and fine.
>
> But my problem was whether the vanilla SuSE9ES kernel make it possible to
> control
> the CPU-resource,etc... or impossible, Since it's hopeful for me not to apply
> additional patches and not to rebuild.
>
> Thanks.
> --[ top on my desktop 2.6.8.1+CKRM+CPU-Ctlr]-------------------------------
> top - 19:07:11 up 5 min, 2 users, load average: 6.24, 2.68, 1.08
> Tasks: 61 total, 4 running, 57 sleeping, 0 stopped, 0 zombie
> Cpu(s): 100.0% us, 0.0% sy, 0.0% ni, 0.0% id, 0.0% wa, 0.0% hi, 0.0% si
> Mem: 320672k total, 82504k used, 238168k free, 9828k buffers
> Swap: 522104k total, 0k used, 522104k free, 44396k cached
>
> PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
> 3178 kaigai 25 0 1284 304 1260 R 52.2 0.1 0:40.51 gold
> 3179 kaigai 25 0 1284 304 1260 R 27.3 0.1 1:07.31 silver
> 3180 kaigai 25 0 1284 304 1260 R 23.6 0.1 0:46.50 silver
> 3234 kaigai 16 0 1860 896 1652 R 0.3 0.3 0:00.25 top
> 1 root 16 0 1584 484 1396 S 0.0 0.2 0:01.01 init
> 2 root 34 19 0 0 0 S 0.0 0.0 0:00.00 ksoftirqd/0
> 3 root 5 -10 0 0 0 S 0.0 0.0 0:00.00 events/0
> 4 root 8 -10 0 0 0 S 0.0 0.0 0:00.03 khelper
> 5 root 5 -10 0 0 0 S 0.0 0.0 0:00.03 kacpid
> 23 root 5 -10 0 0 0 S 0.0 0.0 0:00.02 kblockd/0
> 24 root 15 0 0 0 0 S 0.0 0.0 0:00.05 khubd
> 36 root 20 0 0 0 0 S 0.0 0.0 0:00.00 pdflush
> ---------------------------------------------------------------------------
>
> Shailabh Nagar wrote:
>
>>Kaigai Kohei wrote:
>>
>>
>>>Hi,
>>>
>>>I'm trying to use CKRM on SuSE Linux ES9, but I failed to use
>>>fair-scheduling of
>>>CPU resources.
>>>
>>>I tried to run the three infinite-CPU-consuming tasks(one belongs 'gold'
>>>class,
>>>and others belong 'silver' class) on SuSE Linux ES9 kernel, but each tasks
>>>consume
>>>33% of CPU-time. I expected the gold task consumes 50% of CPU-time, and the
>>>silver
>>>tasks consume 25% each.
>>>
>>>Then I researched the kernel sources of SuSE Linux ES9:
>>>CKRM patches were applied, but the resource-controlleres like CPU were not
>>>included.
>>>(the CKRM-related patches don't modify kernel/sched.c in spite of the demand
>>>for
>>>implementation of CPU-controller.)
>>>IMO, process-grouping is possible, but resource-control is impossible.
>>>
>>>Is there someone who has suceeded the same thing on SuSE Linux ES9?
>>>
>>>Thanks.
>>>
>>
>>
>>Kaigai-san,
>>
>>The latest CPU scheduler patch has been uploaded as
>> cpu.ckrm-e16.v9rc1.patch
>>
>>at
>> http://sourceforge.net/project/showfiles.php?group_id=85838
>>
>>Please apply this patch to the SLES9 kernel and try.
>>
>>Thanks,
>>Shailabh
>
>
-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://productguide.itmanagersjournal.com/
_______________________________________________
ckrm-tech mailing list
https://lists.sourceforge.net/lists/listinfo/ckrm-tech
