Commit 70ce3ea9 authored by Mel Gorman's avatar Mel Gorman Committed by Peter Zijlstra
Browse files

sched/numa: Initialise numa_migrate_retry 



On clone, numa_migrate_retry is inherited from the parent which means
that the first NUMA placement of a task is non-deterministic. This
affects when load balancing recognises numa tasks and whether to
migrate "regular", "remote" or "all" tasks between NUMA scheduler
domains.

Signed-off-by: default avatarMel Gorman <mgorman@techsingularity.net>
Signed-off-by: default avatarPeter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: default avatarK Prateek Nayak <kprateek.nayak@amd.com>
Link: https://lore.kernel.org/r/20220520103519.1863-2-mgorman@techsingularity.net
parent b13baccc

kernel/sched/fair.c

+1 −0
Original line number Diff line number Diff line
@@ -2885,6 +2885,7 @@ void init_numa_balancing(unsigned long clone_flags, struct task_struct *p) 
	p->node_stamp			= 0;

	p->numa_scan_seq		= mm ? mm->numa_scan_seq : 0;

	p->numa_scan_period		= sysctl_numa_balancing_scan_delay;

	p->numa_migrate_retry		= 0;

	/* Protect against double add, see task_tick_numa and task_numa_work */

	p->numa_work.next		= &p->numa_work;

	p->numa_faults			= NULL;