lkml.org 
[lkml]   [2013]   [Nov]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
Subject[tip:sched/urgent] sched/numa: Cure update_numa_stats() vs. hotplug
Commit-ID:  5eca82a9ac2c961cfbd26a4b6f43e6e3747a71dd
Gitweb: http://git.kernel.org/tip/5eca82a9ac2c961cfbd26a4b6f43e6e3747a71dd
Author: Peter Zijlstra <peterz@infradead.org>
AuthorDate: Wed, 6 Nov 2013 18:47:57 +0100
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Wed, 13 Nov 2013 13:33:52 +0100

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 | 15 ++++++++++++++-
1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index c11e36f..201be78 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -1000,7 +1000,7 @@ struct numa_stats {
*/
static void update_numa_stats(struct numa_stats *ns, int nid)
{
- int cpu;
+ int cpu, cpus = 0;

memset(ns, 0, sizeof(*ns));
for_each_cpu(cpu, cpumask_of_node(nid)) {
@@ -1009,8 +1009,21 @@ static void update_numa_stats(struct numa_stats *ns, int nid)
ns->nr_running += rq->nr_running;
ns->load += weighted_cpuload(cpu);
ns->power += power_of(cpu);
+
+ cpus++;
}

+ /*
+ * If we raced with hotplug and there are no CPUs left in our mask
+ * the @ns structure is NULL'ed and task_numa_compare() will
+ * not find this node attractive.
+ *
+ * We'll either bail at !has_capacity, or we'll detect a huge imbalance
+ * and bail there.
+ */
+ if (!cpus)
+ return;
+
ns->load = (ns->load * SCHED_POWER_SCALE) / ns->power;
ns->capacity = DIV_ROUND_CLOSEST(ns->power, SCHED_POWER_SCALE);
ns->has_capacity = (ns->nr_running < ns->capacity);

\
 
 \ /
  Last update: 2013-11-13 19:01    [W:0.036 / U:0.044 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site