[PATCH tip] [PATCH]sched/fair: Fix endless loop in idle_balance()

[PATCH tip] [PATCH]sched/fair: Fix endless loop in idle_balance()

[Kirill Tkhai]

This patch must go on top of the series:

[tip 1/2] sched/fair: Push down check for high priority class task into idle_balance()
[tip 2/2] sched/core: Fix endless loop in pick_next_task()

Check for fair tasks number to decide, that we've pulled a task.
rq's nr_running may contain throttled RT tasks.

Signed-off-by: Kirill Tkhai <ktkhai@parallels.com>
CC: Peter Zijlstra <peterz@infradead.org>
CC: Ingo Molnar <mingo@redhat.com>

Peter, this finaly fixes problem with RT throttling.
Should I send all three patches as one series?
---
 kernel/sched/fair.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 10db4a8..f1eedae 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6710,7 +6710,7 @@ static int idle_balance(struct rq *this_rq)
 	 * While browsing the domains, we released the rq lock.
 	 * A task could have be enqueued in the meantime
 	 */
-	if (this_rq->nr_running && !pulled_task) {
+	if (this_rq->cfs.h_nr_running && !pulled_task) {
 		pulled_task = 1;
 		goto out;
 	}

Re: [PATCH tip] [PATCH]sched/fair: Fix endless loop in idle_balance()

[Peter Zijlstra]

On Thu, Mar 06, 2014 at 07:16:15PM +0400, Kirill Tkhai wrote:
> Peter, this finaly fixes problem with RT throttling.
> Should I send all three patches as one series?

No need; got them. Thanks!

[tip:sched/core] sched/fair: Fix endless loop in idle_balance()

[tip-bot for Kirill Tkhai]

Commit-ID:  35805ff8f4fc535ac85330170d3c56829c87c677
Gitweb:     http://git.kernel.org/tip/35805ff8f4fc535ac85330170d3c56829c87c677
Author:     Kirill Tkhai <ktkhai@parallels.com>
AuthorDate: Thu, 6 Mar 2014 19:16:15 +0400
Committer:  Ingo Molnar <mingo@kernel.org>
CommitDate: Tue, 11 Mar 2014 12:05:41 +0100

sched/fair: Fix endless loop in idle_balance()

Check for fair tasks number to decide, that we've pulled a task.
rq's nr_running may contain throttled RT tasks.

Signed-off-by: Kirill Tkhai <ktkhai@parallels.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1394118975.19290.104.camel@tkhai
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
 kernel/sched/fair.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 10db4a8..f1eedae 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6710,7 +6710,7 @@ static int idle_balance(struct rq *this_rq)
 	 * While browsing the domains, we released the rq lock.
 	 * A task could have be enqueued in the meantime
 	 */
-	if (this_rq->nr_running && !pulled_task) {
+	if (this_rq->cfs.h_nr_running && !pulled_task) {
 		pulled_task = 1;
 		goto out;
 	}