> > Date: Mon, 14 Nov 2016 13:11:58 +0100 > > From: Martin Pieuchot <[email protected]> > > > > On 12/11/16(Sat) 15:52, patrick keshishian wrote: > > > Ahh... seems the culprit is softclock_thread added 2016/09/22 > > > (kern/kern_timeout.c mpi@). > > > > I'd suggest we simply skip kernel thread when calculating the load. > > Hmm, I'd say that counting runnable kernel threads towards the load is > the right thing to do. > > > Since we're slowly moving code executed in software interrupt > > context to kernel threads this will keep the original behavior. > > A different way of viewing is that this move now enables us to > properly take into account the work done by the network stack.
Ah, it is Patrick again, wanting low load averages rather than accurate load averages. Not understanding the loadavg comes without any promise as looks. The load average is an ABSTRACT measurement of work that is queued to be done. As more parts of the kernel become desyncronized from the big lock, that number will change. If you want the old load average number, run old code with the old behaviours. I agree with Mark. The kernel threads should be counted.
