[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87ilz4fyzi.ffs@tglx>
Date: Mon, 13 Sep 2021 14:20:49 +0200
From: Thomas Gleixner <tglx@...utronix.de>
To: Davidlohr Bueso <dave@...olabs.net>
Cc: peterz@...radead.org, mingo@...nel.org, rostedt@...dmis.org,
longman@...hat.com, bigeasy@...utronix.de, boqun.feng@...il.com,
dave@...olabs.net, 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, Sep 01 2021 at 15:28, Davidlohr Bueso wrote:
> diff --git a/kernel/locking/rwbase_rt.c b/kernel/locking/rwbase_rt.c
> index 4ba15088e640..3444bc709973 100644
> --- a/kernel/locking/rwbase_rt.c
> +++ b/kernel/locking/rwbase_rt.c
> @@ -141,6 +141,7 @@ static void __sched __rwbase_read_unlock(struct rwbase_rt *rwb,
> {
> struct rt_mutex_base *rtm = &rwb->rtmutex;
> struct task_struct *owner;
> + DEFINE_WAKE_Q(wake_q);
>
> raw_spin_lock_irq(&rtm->wait_lock);
> /*
> @@ -151,9 +152,10 @@ static void __sched __rwbase_read_unlock(struct rwbase_rt *rwb,
> */
> owner = rt_mutex_owner(rtm);
> if (owner)
> - wake_up_state(owner, state);
> + wake_q_add(&wake_q, owner);
That's broken for rw_locks. See commit 456cfbc65cd072f4f53936ee5a37eb1447a7d3ba.
Thanks,
tglx
Powered by blists - more mailing lists