Increase the degree of interactivity ULE scheduler
jroberson at jroberson.net
Thu Nov 3 21:22:20 UTC 2011
On Thu, 3 Nov 2011, Ivan Klymenko wrote:
> Thank you for taking the time to answer me.
> ? Thu, 3 Nov 2011 10:21:48 -1000 (HST)
> Jeff Roberson <jroberson at jroberson.net> ?????:
>> On Sat, 22 Oct 2011, Ivan Klymenko wrote:
>>> Hello people!
>>> I have:
>>> CPU: Intel(R) Core(TM)2 Duo CPU T7250 @ 2.00GHz (1994.48-MHz
>>> K8-class CPU) FreeBSD 10.0-CURRENT r226607 amd64
>>> For example during the building of the port lang/gcc46 in four
>>> streams (-j 4) with a heavy load on the processor - use the system
>>> was nearly impossible - responsiveness was terrible - the mouse
>>> cursor sometimes froze on the spot a few seconds...
>> Am I right in understanding that you have only two cores?
>> What else is running that achieves poor interactivity?
> This is mainly a compilation with make option -j >= ncpu*2
> And as an example - launching a large number of programs
> This patch allows me to make do with ULE nearly as well as with FBFS
> Without the patch, somewhere in the middle of the time with ULE has
> been difficult to control the mouse cursor.
>> What is the cpu utilization of your x server at this time?
> ~2.00% - 20.00% WCPU time... But sometimes there are up to 79%...
> Upon unloading the CPU returns to normal...
When the x server is down at 20% is it laggy? Can you tell me the
priorities of the x server and the compile tasks? You can use the 'pri'
keyword with ps and write a short script to log all priorities once per
second during your test. That would be most helpful. Let me know if you
need assistance with that.
>>> I managed to achieve a significant increase in the degree of
>>> interactivity ULE scheduler due to the following changes:
>> This patch probably breaks nice, adaptive idling, and slows the
>> interactivity computation. That being said I'm not sure why it helps
>> It seems that there are increasing reports of bad interactivity
>> creeping in to ULE over the last year. If people can help provide me
>> with data I can look into this more.
> I'll be glad to provide data
>> Thanks for your report.
> How to repeat your tests on my system?
> Sorry for my english.
More information about the freebsd-current