Re: [PATCH v2 1/7] sched/fair: Generalize asym_packing logic for SMT local sched group

[Dietmar Eggemann]

On 22/12/2022 05:32, Ricardo Neri wrote:
> On Wed, Dec 21, 2022 at 02:03:15PM +0100, Dietmar Eggemann wrote:
>> On 12/12/2022 18:53, Ricardo Neri wrote:
>>> On Tue, Dec 06, 2022 at 06:22:41PM +0100, Dietmar Eggemann wrote:
>>>> On 22/11/2022 21:35, Ricardo Neri wrote:
>>
>> [...]
>>
>>>> I'm not sure why you change asym_smt_can_pull_tasks() together with
>>>> removing SD_ASYM_PACKING from SMT level (patch 5/7)?
>>>
>>> In x86 we have SD_ASYM_PACKING at the MC, CLS* and, before my patches, SMT
>>> sched domains.
>>>
>>>>
>>>> update_sg_lb_stats()
>>>>
>>>>   ... && env->sd->flags & SD_ASYM_PACKING && .. && sched_asym()
>>>>                                                    ^^^^^^^^^^^^
>>>>     sched_asym()
>>>>
>>>>       if ((sds->local->flags & SD_SHARE_CPUCAPACITY) ||
>>>>           (group->flags & SD_SHARE_CPUCAPACITY))
>>>>         return asym_smt_can_pull_tasks()
>>>>                ^^^^^^^^^^^^^^^^^^^^^^^^^
>>>>
>>>> So x86 won't have a sched domain with SD_SHARE_CPUCAPACITY and
>>>> SD_ASYM_PACKING anymore. So sched_asym() would call sched_asym_prefer()
>>>> directly on MC. What do I miss here?
>>>
>>> asym_smt_can_pull_tasks() is used above the SMT level *and* when either the
>>> local or sg sched groups are composed of CPUs that are SMT siblings.
>>
>> OK.
>>
>>> In fact, asym_smt_can_pull_tasks() can only be called above the SMT level.
>>> This is because the flags of a sched_group in a sched_domain are equal to
>>> the flags of the child sched_domain. Since SMT is the lowest sched_domain,
>>> its groups' flags are 0.
>>
>> I see. I forgot about `[PATCH v5 0/6] sched/fair: Fix load balancing of
>> SMT siblings with ASYM_PACKING` from Sept 21 (specifically [PATCH v5
>> 2/6] sched/topology: Introduce sched_group::flags).
>>
>>> sched_asym() calls sched_asym_prefer() directly if balancing at the
>>> SMT level and, at higher domains, if the child domain is not SMT.
>>
>> OK.
>>
>>> This meets the requirement of Power7, where SMT siblings have different
>>> priorities; and of x86, where physical cores have different priorities.
>>>
>>> Thanks and BR,
>>> Ricardo
>>>
>>> * The target of these patches is Intel hybrid processors, on which cluster
>>>   scheduling is disabled - cabdc3a8475b ("sched,x86: Don't use cluster
>>>   topology for x86 hybrid CPUs"). Also, I have not observed topologies in
>>>   which CPUs of the same cluster have different priorities.
>>
>> OK.
>>
>> IMHO, the function header of asym_smt_can_pull_tasks() (3rd and 4th
>> paragraph ...  `If both @dst_cpu and @sg have SMT siblings` and
> 
> Agreed. I changed the behavior of the function. I will update the
> description.
> 
>> `If @sg does not have SMT siblings` still reflect the old code layout.
> 
> But this behavior did not change. The check covers both SMT and non-SMT
> cases:

The condition to call sched_asym_prefer() seems to have changed slightly
though (including the explanation that busy_cpus_delta >= 2 handling
should be done by fbg().:

sds->local_stat.sum_nr_running (A)
busy_cpus_delta = sg_busy_cpus - local_busy_cpus (B)
sg_busy_cpus = sgs->group_weight - sgs->idle_cpus (C)

>From ((smt && B == 1) || (!smt && !A)) to (C == 1 && !A)

> 
> 	 /*
> 	  * non-SMT @sg can only have 1 busy CPU. We only care SMT @sg
> 	  * has exactly one busy sibling
> 	  */
> 	if (sg_busy_cpus == 1 && 
> 	    /* local group is fully idle, SMT and non-SMT. */
> 	    !sds->local_stat.sum_nr_running)
> 
> Perhaps I can collapse the two paragraphs into one.

Sounds good to me.

[...]