sched/numa: Use group's max nid as task's preferred nid
authorRik van Riel <riel@redhat.com>
Mon, 23 Jun 2014 15:41:29 +0000 (11:41 -0400)
committerIngo Molnar <mingo@kernel.org>
Sat, 5 Jul 2014 09:17:33 +0000 (11:17 +0200)
commitf0b8a4afd6a8c500161e45065a91738b490bf5ae
treeba510d83c047b38a726ee8f0bb4a65ba143a9216
parent4486edd12b5ac8a9af7a5e16e4b9eeb3b8339c10
sched/numa: Use group's max nid as task's preferred nid

From task_numa_placement, always try to consolidate the tasks
in a group on the group's top nid.

In case this task is part of a group that is interleaved over
multiple nodes, task_numa_migrate will set the task's preferred
nid to the best node it could find for the task, so this patch
will cause at most one run through task_numa_migrate.

Signed-off-by: Rik van Riel <riel@redhat.com>
Cc: mgorman@suse.de
Cc: chegu_vinod@hp.com
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1403538095-31256-2-git-send-email-riel@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
kernel/sched/fair.c