[PATCH AUTOSEL 5.15 18/41] sched: Avoid double preemption in __cond_resched_*lock*()

From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Peter Zijlstra <peterz@infradead.org>,
	xuhaifeng <xuhaifeng@oppo.com>, Sasha Levin <sashal@kernel.org>,
	mingo@redhat.com, juri.lelli@redhat.com,
	vincent.guittot@linaro.org
Subject: [PATCH AUTOSEL 5.15 18/41] sched: Avoid double preemption in __cond_resched_*lock*()
Date: Thu,  3 Feb 2022 15:32:22 -0500	[thread overview]
Message-ID: <20220203203245.3007-18-sashal@kernel.org> (raw)
In-Reply-To: <20220203203245.3007-1-sashal@kernel.org>

From: Peter Zijlstra <peterz@infradead.org>

[ Upstream commit 7e406d1ff39b8ee574036418a5043c86723170cf ]

For PREEMPT/DYNAMIC_PREEMPT the *_unlock() will already trigger a
preemption, no point in then calling preempt_schedule_common()
*again*.

Use _cond_resched() instead, since this is a NOP for the preemptible
configs while it provide a preemption point for the others.

Reported-by: xuhaifeng <xuhaifeng@oppo.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/YcGnvDEYBwOiV0cR@hirez.programming.kicks-ass.net
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 kernel/sched/core.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 0d12ec7be3017..c2dec6ce98091 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -8199,9 +8199,7 @@ int __cond_resched_lock(spinlock_t *lock)
 
 	if (spin_needbreak(lock) || resched) {
 		spin_unlock(lock);
-		if (resched)
-			preempt_schedule_common();
-		else
+		if (!_cond_resched())
 			cpu_relax();
 		ret = 1;
 		spin_lock(lock);
@@ -8219,9 +8217,7 @@ int __cond_resched_rwlock_read(rwlock_t *lock)
 
 	if (rwlock_needbreak(lock) || resched) {
 		read_unlock(lock);
-		if (resched)
-			preempt_schedule_common();
-		else
+		if (!_cond_resched())
 			cpu_relax();
 		ret = 1;
 		read_lock(lock);
@@ -8239,9 +8235,7 @@ int __cond_resched_rwlock_write(rwlock_t *lock)
 
 	if (rwlock_needbreak(lock) || resched) {
 		write_unlock(lock);
-		if (resched)
-			preempt_schedule_common();
-		else
+		if (!_cond_resched())
 			cpu_relax();
 		ret = 1;
 		write_lock(lock);
-- 
2.34.1

