sched/numa: Cure update_numa_stats() vs. hotplug
authorPeter Zijlstra <peterz@infradead.org>
Wed, 6 Nov 2013 17:47:57 +0000 (18:47 +0100)
committerIngo Molnar <mingo@kernel.org>
Wed, 13 Nov 2013 12:33:52 +0000 (13:33 +0100)
commit5eca82a9ac2c961cfbd26a4b6f43e6e3747a71dd
treec63a3495a562c32b447bd4dec093eca98e9a584b
parent46a73e8a1c1720f7713b5e2df68e9dd272015b5d
sched/numa: Cure update_numa_stats() vs. hotplug

Because we're completely unserialized against hotplug its well
possible to try and generate numa stats for an offlined node.

Bail out early (and avoid a /0) in this case. The resulting stats are
all 0 which should result in an undesirable balance target -- not to
mention that actually trying to migrate to an offline CPU will fail.

Reported-by: Prarit Bhargava <prarit@redhat.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Link: http://lkml.kernel.org/n/tip-orja0qylcvyhxfsuebcyL5sI@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
kernel/sched/fair.c