[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20210901230320.3pogmlyzfgwpvw77@offworld>
Date: Wed, 1 Sep 2021 16:03:20 -0700
From: Davidlohr Bueso <dave@...olabs.net>
To: tglx@...utronix.de
Cc: peterz@...radead.org, mingo@...nel.org, rostedt@...dmis.org,
longman@...hat.com, bigeasy@...utronix.de, boqun.feng@...il.com,
linux-kernel@...r.kernel.org, Davidlohr Bueso <dbueso@...e.de>
Subject: Re: [PATCH 2/2] locking/rwbase_rt: Lockless reader waking up a writer
On Wed, 01 Sep 2021, Bueso wrote:
>As with the rest of our sleeping locks, use a wake_q to
>allow waking up the writer without having to hold the
>wait_lock across the operation. While this is ideally
>for batching wakeups, single wakeup usage as still shown
>to be beneficial vs the cost of try_to_wakeup when the
>lock is contended.
>
>Signed-off-by: Davidlohr Bueso <dbueso@...e.de>
>---
> kernel/locking/rwbase_rt.c | 4 +++-
> kernel/sched/core.c | 16 +++++++++++++---
Bleh, this of course is missing:
diff --git a/include/linux/sched/wake_q.h b/include/linux/sched/wake_q.h
index 06cd8fb2f409..695efd987b56 100644
--- a/include/linux/sched/wake_q.h
+++ b/include/linux/sched/wake_q.h
@@ -62,5 +62,6 @@ static inline bool wake_q_empty(struct wake_q_head *head)
extern void wake_q_add(struct wake_q_head *head, struct task_struct *task);
extern void wake_q_add_safe(struct wake_q_head *head, struct task_struct *task);
extern void wake_up_q(struct wake_q_head *head);
+extern void wake_up_q_state(struct wake_q_head *head, unsigned int state);
#endif /* _LINUX_SCHED_WAKE_Q_H */
Powered by blists - more mailing lists