[PATCH v2 08/13] locking/qspinlock: Use smp_cond_load_relaxed to wait for next node

[Will Deacon]

When a locker reaches the head of the queue and takes the lock, a
concurrent locker may enqueue and force the lock holder to spin
whilst its node->next field is initialised. Rather than open-code
a READ_ONCE/cpu_relax() loop, this can be implemented using
smp_cond_load_relaxed instead.

Cc: Peter Zijlstra <peterz at infradead.org>
Cc: Ingo Molnar <mingo at kernel.org>
Signed-off-by: Will Deacon <will.deacon at arm.com>
---
 kernel/locking/qspinlock.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/kernel/locking/qspinlock.c b/kernel/locking/qspinlock.c
index 648a16a2cd23..c781ddbe59a6 100644
--- a/kernel/locking/qspinlock.c
+++ b/kernel/locking/qspinlock.c
@@ -523,10 +523,8 @@ void queued_spin_lock_slowpath(struct qspinlock *lock, u32 val)
 	/*
 	 * contended path; wait for next if not observed yet, release.
 	 */
-	if (!next) {
-		while (!(next = READ_ONCE(node->next)))
-			cpu_relax();
-	}
+	if (!next)
+		next = smp_cond_load_relaxed(&node->next, (VAL));
 
 	arch_mcs_spin_unlock_contended(&next->locked);
 	pv_kick_node(lock, next);
-- 
2.1.4