[lkml]   [2017]   [Sep]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
Patch in this message
Subject[tip:sched/urgent] sched/fair: Avoid newidle balance for !active CPUs
Commit-ID:  2800486ee34825d954f64c6f98037daea328f121
Author: Peter Zijlstra <>
AuthorDate: Thu, 7 Sep 2017 17:03:50 +0200
Committer: Ingo Molnar <>
CommitDate: Tue, 12 Sep 2017 17:41:03 +0200

sched/fair: Avoid newidle balance for !active CPUs

On CPU hot unplug, when parking the last kthread we'll try and
schedule into idle to kill the CPU. This last schedule can (and does)
trigger newidle balance because at this point the sched domains are
still up because of commit:

77d1dfda0e79 ("sched/topology, cpuset: Avoid spurious/wrong domain rebuilds")

Obviously pulling tasks to an already offline CPU is a bad idea, and
all balancing operations _should_ be subject to cpu_active_mask, make
it so.

Reported-by: Thomas Gleixner <>
Signed-off-by: Peter Zijlstra (Intel) <>
Cc: Linus Torvalds <>
Cc: Peter Zijlstra <>
Fixes: 77d1dfda0e79 ("sched/topology, cpuset: Avoid spurious/wrong domain rebuilds")
Signed-off-by: Ingo Molnar <>
kernel/sched/fair.c | 6 ++++++
1 file changed, 6 insertions(+)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 8415d1e..3bcea40 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -8448,6 +8448,12 @@ static int idle_balance(struct rq *this_rq, struct rq_flags *rf)
this_rq->idle_stamp = rq_clock(this_rq);

+ * Do not pull tasks towards !active CPUs...
+ */
+ if (!cpu_active(this_cpu))
+ return 0;
+ /*
* This is OK, because current is on_cpu, which avoids it being picked
* for load-balance and preemption/IRQs are still disabled avoiding
* further scheduler activity on it and we're being very careful to
 \ /
  Last update: 2017-09-12 20:09    [W:0.093 / U:3.908 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site