Re: [RFC PATCH 2/2] NUMA balancing: avoid to migrate task to CPU-less node

[Huang, Ying]

Srikar Dronamraju <srikar@xxxxxxxxxxxxxxxxxx> writes:

> * Huang, Ying <ying.huang@xxxxxxxxx> [2022-01-28 15:51:36]:
>
>> Srikar Dronamraju <srikar@xxxxxxxxxxxxxxxxxx> writes:
>> 
>> > * Huang Ying <ying.huang@xxxxxxxxx> [2022-01-28 10:38:42]:
>> >
>> This sounds reasonable.  How about the following solution?  If a
>> CPU-less node is selected as migration target, we select a nearest node
>> with CPU instead?  That is, something like the below patch.
>> 
>> Best Regards,
>> Huang, Ying
>> 
>> ------------------------------8<---------------------------------
>> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
>> index 5146163bfabb..52d926d8cbdb 100644
>> --- a/kernel/sched/fair.c
>> +++ b/kernel/sched/fair.c
>> @@ -2401,6 +2401,23 @@ static void task_numa_placement(struct task_struct *p)
>>  		}
>>  	}
>> 
>> +	/* Cannot migrate task to CPU-less node */
>> +	if (!node_state(max_nid, N_CPU)) {
>> +		int near_nid = max_nid;
>> +		int distance, near_distance = INT_MAX;
>> +
>> +		for_each_online_node(nid) {
>> +			if (!node_state(nid, N_CPU))
>> +				continue;
>> +			distance = node_distance(max_nid, nid);
>> +			if (distance < near_distance) {
>> +				near_nid = nid;
>> +				near_distance = distance;
>> +			}
>> +		}
>> +		max_nid = near_nid;
>> +	}
>> +
>
>
> This looks good. but should we move this into preferred_group_nid()?

Yes.  We need to take care of preferred_group_nid() too.  Will do that
in the next version.

> i.e should we care for !ng case, since those would mean only private faults.

IMO we need to care for !ng case.  If the fault number of the CPU-less
node is the max, we need to migrate to the nearest node with CPU
instead.

Best Regards,
Huang, Ying

>>  	if (ng) {
>>  		numa_group_count_active_nodes(ng);
>>  		spin_unlock_irq(group_lock);