Messages in this thread | | | Subject | Re: [PATCH v2 5/6] sched/deadline: Make DL capacity-aware | From | Dietmar Eggemann <> | Date | Tue, 5 May 2020 20:02:00 +0200 |
| |
On 04/05/2020 05:58, Pavan Kondeti wrote: > On Fri, May 01, 2020 at 06:12:07PM +0200, Dietmar Eggemann wrote: >> On 30/04/2020 15:10, Pavan Kondeti wrote: >>> On Mon, Apr 27, 2020 at 10:37:08AM +0200, Dietmar Eggemann wrote: >>>> From: Luca Abeni <luca.abeni@santannapisa.it> >> >> [...] >> >>>> @@ -1653,10 +1654,19 @@ select_task_rq_dl(struct task_struct *p, int cpu, int sd_flag, int flags) >>>> * other hand, if it has a shorter deadline, we >>>> * try to make it stay here, it might be important. >>>> */ >>>> - if (unlikely(dl_task(curr)) && >>>> - (curr->nr_cpus_allowed < 2 || >>>> - !dl_entity_preempt(&p->dl, &curr->dl)) && >>>> - (p->nr_cpus_allowed > 1)) { >>>> + select_rq = unlikely(dl_task(curr)) && >>>> + (curr->nr_cpus_allowed < 2 || >>>> + !dl_entity_preempt(&p->dl, &curr->dl)) && >>>> + p->nr_cpus_allowed > 1; >>>> + >>>> + /* >>>> + * Take the capacity of the CPU into account to >>>> + * ensure it fits the requirement of the task. >>>> + */ >>>> + if (static_branch_unlikely(&sched_asym_cpucapacity)) >>>> + select_rq |= !dl_task_fits_capacity(p, cpu); >>>> + >>>> + if (select_rq) { >>>> int target = find_later_rq(p); >>> >>> I see that find_later_rq() checks if the previous CPU is part of >>> later_mask and returns it immediately. So we don't migrate the >>> task in the case where there previous CPU can't fit the task and >>> there are no idle CPUs on which the task can fit. LGTM. >> >> Hope I understand you here. I don't think that [patch 6/6] provides this >> already. >> >> In case 'later_mask' has no fitting CPUs, 'max_cpu' is set in the >> otherwise empty 'later_mask'. But 'max_cpu' is not necessary task_cpu(p). >> >> Example on Juno [L b b L L L] with thread0-0 (big task) >> >> cpudl_find [thread0-0 2117] orig later_mask=0,3-4 later_mask=0 >> find_later_rq [thread0-0 2117] task_cpu=2 later_mask=0 >> >> A tweak could be added favor task_cpu(p) in case it is amongst the CPUs >> with the maximum capacity in cpudl_find() for the !fit case. >> > > You are right. max_cpu can be other than task_cpu(p) in which case we > migrate the task though it won't fit on the new CPU. While introducing > capacity awareness in RT, Quais made the below change to avoid the > migration. We can do something similar here also. > > commit b28bc1e002c2 (sched/rt: Re-instate old behavior in select_task_rq_rt())
I'm not sure something like this is necessary here.
With DL capacity awareness we reduce the later_mask returned by cpudl_find() in find_later_rq() to those idle CPUs which can handle p or in case there is none to (the/a) 'non-fitting CPU w/ max capacity'.
We just have to favor task_cpu(p) in [patch 6/6] in case it is part of the initial later_mask and among these 'non-fitting CPUs w/ max capacity'.
This will make sure that it gets chosen so find_later_rq() returns it before the 'for_each_domain()' loop.
And I guess we still want to migrate if there is a non-fitting CPU w/ a higher CPU capacity than task_cpu() (tri-gear).
@@ -137,7 +137,8 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p,
cap = capacity_orig_of(cpu);
- if (cap > max_cap) { + if (cap > max_cap || + (cpu == task_cpu(p) && cap == max_cap)) { max_cap = cap; max_cpu = cpu;
In case task_cpu() is not part of later_cpu, the 'max CPU capacity CPU' is returned as 'best_cpu'.
[...]
| |