fair clock use in CFS
Quoted 9 lines · Ingo Molnar
the current task is recalculated at scheduler tick time and put into the tree at its new position. At a million tasks the fair-clock will advance little (or not at all - which at these load levels is our smallest problem anyway) so during a scheduling tick in
kernel/sched_fair.cupdate_curr()we will have a 'delta_mine' and 'delta_fair' of near zero and a 'delta_exec' of ~1 million, socurr->wait_runtimewill be decreased at 'full speed': delta_exec-delta_mine, by almost a full tick. So preemption will occur everysched_granularity(rounded up to the next tick) points in time, in wall-clock time.
The only problem I have with this fairness is the server workload that services requests by fork/thread creation. In such a case, this fairness is completely counter-productive, as running tasks unfairly inhibit the creation of peers.
Giving fork/thread creation special priority may alleviate this problem.
Thanks!
--
Al