[tip: sched/urgent] sched: Prepare to use balance_push in ttwu()

[tip-bot2 for Peter Zijlstra]

The following commit has been merged into the sched/urgent branch of tip:

Commit-ID:     975707f227b07a8212060f94447171d15d7a681b
Gitweb:        https://git.kernel.org/tip/975707f227b07a8212060f94447171d15d7a681b
Author:        Peter Zijlstra <peterz@xxxxxxxxxxxxx>
AuthorDate:    Wed, 20 Jan 2021 15:05:41 +01:00
Committer:     Peter Zijlstra <peterz@xxxxxxxxxxxxx>
CommitterDate: Fri, 22 Jan 2021 15:09:43 +01:00

sched: Prepare to use balance_push in ttwu()

In preparation of using the balance_push state in ttwu() we need it to
provide a reliable and consistent state.

The immediate problem is that rq->balance_callback gets cleared every
schedule() and then re-set in the balance_push_callback() itself. This
is not a reliable signal, so add a variable that stays set during the
entire time.

Also move setting it before the synchronize_rcu() in
sched_cpu_deactivate(), such that we get guaranteed visibility to
ttwu(), which is a preempt-disable region.

Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
Reviewed-by: Valentin Schneider <valentin.schneider@xxxxxxx>
Tested-by: Valentin Schneider <valentin.schneider@xxxxxxx>
Link: https://lkml.kernel.org/r/20210121103506.966069627@xxxxxxxxxxxxx
---
 kernel/sched/core.c  | 11 ++++++-----
 kernel/sched/sched.h |  1 +
 2 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 8da0fd7..16946b5 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -7320,6 +7320,7 @@ static void balance_push_set(int cpu, bool on)
 	struct rq_flags rf;
 
 	rq_lock_irqsave(rq, &rf);
+	rq->balance_push = on;
 	if (on) {
 		WARN_ON_ONCE(rq->balance_callback);
 		rq->balance_callback = &balance_push_callback;
@@ -7489,17 +7490,17 @@ int sched_cpu_deactivate(unsigned int cpu)
 	int ret;
 
 	set_cpu_active(cpu, false);
+	balance_push_set(cpu, true);
+
 	/*
-	 * We've cleared cpu_active_mask, wait for all preempt-disabled and RCU
-	 * users of this state to go away such that all new such users will
-	 * observe it.
+	 * We've cleared cpu_active_mask / set balance_push, wait for all
+	 * preempt-disabled and RCU users of this state to go away such that
+	 * all new such users will observe it.
 	 *
 	 * Do sync before park smpboot threads to take care the rcu boost case.
 	 */
 	synchronize_rcu();
 
-	balance_push_set(cpu, true);
-
 	rq_lock_irqsave(rq, &rf);
 	if (rq->rd) {
 		update_rq_clock(rq);
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 12ada79..bb09988 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -975,6 +975,7 @@ struct rq {
 	unsigned long		cpu_capacity_orig;
 
 	struct callback_head	*balance_callback;
+	unsigned char		balance_push;
 
 	unsigned char		nohz_idle_balance;
 	unsigned char		idle_balance;