[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1524738868-31318-10-git-send-email-will.deacon@arm.com>
Date: Thu, 26 Apr 2018 11:34:23 +0100
From: Will Deacon <will.deacon@....com>
To: linux-kernel@...r.kernel.org
Cc: linux-arm-kernel@...ts.infradead.org, peterz@...radead.org,
mingo@...nel.org, boqun.feng@...il.com, paulmck@...ux.vnet.ibm.com,
longman@...hat.com, will.deacon@....com
Subject: [PATCH v3 09/14] locking/qspinlock: Use smp_cond_load_relaxed to wait for next node
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@...radead.org>
Cc: Ingo Molnar <mingo@...nel.org>
Signed-off-by: Will Deacon <will.deacon@....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 5e383cfe4cce..7b8c81ebb15e 100644
--- a/kernel/locking/qspinlock.c
+++ b/kernel/locking/qspinlock.c
@@ -483,10 +483,8 @@ 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
Powered by blists - more mailing lists