Messages in this thread | | | Date | Fri, 4 Sep 2009 14:40:54 +0200 | From | Ingo Molnar <> | Subject | Re: CPU scheduler weirdness? |
| |
* Peter Zijlstra <peterz@infradead.org> wrote:
> On Thu, 2009-09-03 at 23:57 +0200, Marton Balint wrote: > > > sched: fine-tune SD_MC_INIT: > > > 14800984706bf6936bbec5187f736e928be5c218 > > > > > > If I add again the removed SD_BALANCE_NEWIDLE to flags, then everything works > > > as expected. So what would be the correct fix for this bug? Revert the patch? > > > Or just add SD_BALANCE_NEWIDLE to flags? > > > > > > Ingo, Peter, could any of you guys have a look at the commit > > that caused this bug? Is it OK to revert it? Or a fix somewhere > > else is necessary? I'm pushing this because I hope that this bug > > will get fixed in the upcoming stable kernel... > > I'm fine with re-adding SD_BALANCE_IDLE and SD_WAKE_IDLE on > SMT/MC/CPU levels. > > Ingo?
Ok, agreed. I have re-benchmarked a couple of key workload and it seems like a good change, on top of your load-balancer fixes.
Marton, could you please double check the latest -tip tree:
http://people.redhat.com/mingo/tip.git/README
Does it resolve the problem?
Thanks,
Ingo
| |