[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <tip-87ff19cb2f1aa55a5d8b691e6690cc059a59d2ec@git.kernel.org>
Date: Mon, 21 Jan 2019 03:28:53 -0800
From: tip-bot for Davidlohr Bueso <tipbot@...or.com>
To: linux-tip-commits@...r.kernel.org
Cc: mingo@...nel.org, linux-kernel@...r.kernel.org, longman@...hat.com,
akpm@...ux-foundation.org, dave@...olabs.net, peterz@...radead.org,
tglx@...utronix.de, will.deacon@....com, elohimes@...il.com,
dbueso@...e.de, hpa@...or.com, paulmck@...ux.vnet.ibm.com,
torvalds@...ux-foundation.org
Subject: [tip:locking/core] sched/wake_q: Add branch prediction hint to
wake_q_add() cmpxchg
Commit-ID: 87ff19cb2f1aa55a5d8b691e6690cc059a59d2ec
Gitweb: https://git.kernel.org/tip/87ff19cb2f1aa55a5d8b691e6690cc059a59d2ec
Author: Davidlohr Bueso <dave@...olabs.net>
AuthorDate: Sun, 2 Dec 2018 21:31:30 -0800
Committer: Ingo Molnar <mingo@...nel.org>
CommitDate: Mon, 21 Jan 2019 11:18:50 +0100
sched/wake_q: Add branch prediction hint to wake_q_add() cmpxchg
The cmpxchg() will fail when the task is already in the process
of waking up, and as such is an extremely rare occurrence.
Micro-optimize the call and put an unlikely() around it.
To no surprise, when using CONFIG_PROFILE_ANNOTATED_BRANCHES
under a number of workloads the incorrect rate was a mere 1-2%.
Signed-off-by: Davidlohr Bueso <dbueso@...e.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
Acked-by: Waiman Long <longman@...hat.com>
Cc: Andrew Morton <akpm@...ux-foundation.org>
Cc: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: Paul E. McKenney <paulmck@...ux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@...radead.org>
Cc: Thomas Gleixner <tglx@...utronix.de>
Cc: Will Deacon <will.deacon@....com>
Cc: Yongji Xie <elohimes@...il.com>
Cc: andrea.parri@...rulasolutions.com
Cc: lilin24@...du.com
Cc: liuqi16@...du.com
Cc: nixun@...du.com
Cc: xieyongji@...du.com
Cc: yuanlinsi01@...du.com
Cc: zhangyu31@...du.com
Link: https://lkml.kernel.org/r/20181203053130.gwkw6kg72azt2npb@linux-r8p5
Signed-off-by: Ingo Molnar <mingo@...nel.org>
---
kernel/sched/core.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index d8d76a65cfdd..b05eef7d7a1f 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -421,7 +421,7 @@ void wake_q_add(struct wake_q_head *head, struct task_struct *task)
* state, even in the failed case, an explicit smp_mb() must be used.
*/
smp_mb__before_atomic();
- if (cmpxchg_relaxed(&node->next, NULL, WAKE_Q_TAIL))
+ if (unlikely(cmpxchg_relaxed(&node->next, NULL, WAKE_Q_TAIL)))
return;
get_task_struct(task);
Powered by blists - more mailing lists