Re: [PATCH v2 4/8] sched/fair: rework load_balance

From: Valentin Schneider
Date: Wed Aug 07 2019 - 07:16:52 EST


On 06/08/2019 18:17, Valentin Schneider wrote:
>> @@ -8765,7 +8942,7 @@ static int load_balance(int this_cpu, struct rq *this_rq,
>> env.src_rq = busiest;
>>
>> ld_moved = 0;
>> - if (busiest->cfs.h_nr_running > 1) {
>> + if (busiest->nr_running > 1) {
>
> Shouldn't that stay h_nr_running ? We can't do much if those aren't CFS
> tasks.
>

Wait, so that seems to be a correction of an over-zealous rename in patch
2/8, but I think we actually *do* want it to be a cfs.h_nr_running check
here.

And actually this made me have a think about our active balance checks,
I'm cooking something up in that regards.