Process & threads - Karmic vs past kernels

[Tim Gardner]

On 03/12/2010 07:17 AM, Peter Matulis wrote:
> Chase Douglas wrote:
>> On Wed, Mar 10, 2010 at 3:45 PM, Peter Matulis
>> <peter.matulis at canonical.com>  wrote:
>>> Can anyone tell me whether the Karmic kernel has implemented a different
>>> way of what it considers a process (as opposed to threads)?
>>>
>>> I have a situation where CPU load is zero on Karmic but considerably
>>> higher in Jaunty and earlier.
>>>
>>> The scenario is a single java process with many threads.  The system has
>>> 4 cores and only one java process should logically produce a negligible
>>> CPU load but why was this not the case with earlier kernels?  Has
>>> something changed in Karmic that would explain what I'm seeing?
>>
>> I doubt that to be the case. Have you been able to get more data for
>> your bug [1]?
>>
>> Thanks,
>> Chase
>>
>> [1] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/513848
>
> Yes, but I couldn't get as much CPU usage out of the Karmic test.  My
> 'top' results show the following:
>
> Cpu1  :  0.3%us,  0.6%sy
> Cpu2  :  2.4%us,  0.6%sy
> Cpu3  :  9.1%us,  1.8%sy
> Cpu4  : 16.0%us,  7.4%sy
>
> With a load of 0.00 across the board.
>
> Now since CPU1 has such a low usage it makes sense that load is
> negligible since that CPU is always available.  It seems that the
> question is now:
>
> Why the CPU usage is so much different between Karmic and Jaunty in this
>   (single process/multiple thread) scenario.
>

There was a pretty major change in the process and I/O schedulers 
between Jaunty and Karmic. Lucid is different yet again.

-- 
Tim Gardner tim.gardner at canonical.com