Re: [PATCH 00/13] Reconcile NUMA balancing decisions with the load balancer v6

[Mel Gorman]

On Wed, May 13, 2020 at 04:57:15PM +0200, Jirka Hladky wrote:
> Hi Mel,
> 
> we have tried the kernel with adjust_numa_imbalance() crippled to just
> return the imbalance it's given.
> 
> It has solved all the performance problems I have reported.
> Performance is the same as with 5.6 kernel (before the patch was
> applied).
> 
> * solved the performance drop upto 20%  with single instance
> SPECjbb2005 benchmark on 8 NUMA node servers (particularly on AMD EPYC
> Rome systems) => this performance drop was INCREASING with higher
> threads counts (10% for 16 threads and 20 % for 32 threads)
> * solved the performance drop for low load scenarios (SPECjvm2008 and NAS)
> 
> Any suggestions on how to proceed? One approach is to turn
> "imbalance_min" into the kernel tunable. Any other ideas?
> 
> https://github.com/torvalds/linux/blob/4f8a3cc1183c442daee6cc65360e3385021131e4/kernel/sched/fair.c#L8914
> 

Complete shot in the dark but restore adjust_numa_imbalance() and try
this

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 1a9983da4408..0b31f4468d5b 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2393,7 +2393,7 @@ static void ttwu_queue(struct task_struct *p, int cpu, int wake_flags)
 	struct rq_flags rf;
 
 #if defined(CONFIG_SMP)
-	if (sched_feat(TTWU_QUEUE) && !cpus_share_cache(smp_processor_id(), cpu)) {
+	if (sched_feat(TTWU_QUEUE)) {
 		sched_clock_cpu(cpu); /* Sync clocks across CPUs */
 		ttwu_queue_remote(p, cpu, wake_flags);
 		return;

-- 
Mel Gorman
SUSE Labs