Re: [PATCH tip/sched/core v2] sched/rt: Simplify the IPI rt balancing logic

From: Steven Rostedt <rostedt@goodmis.org>
To: Peter Zijlstra <peterz@infradead.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@kernel.org>,
	Clark Williams <williams@redhat.com>,
	Daniel Bristot de Oliveira <bristot@redhat.com>,
	John Kacur <jkacur@redhat.com>, Scott Wood <swood@redhat.com>
Subject: Re: [PATCH tip/sched/core v2] sched/rt: Simplify the IPI rt balancing logic
Date: Thu, 4 May 2017 11:01:55 -0400	[thread overview]
Message-ID: <20170504110155.2ed7b158@gandalf.local.home> (raw)
In-Reply-To: <20170504144128.k2a5vxebtyly7qlv@hirez.programming.kicks-ass.net>

On Thu, 4 May 2017 16:41:28 +0200
Peter Zijlstra <peterz@infradead.org> wrote:

> On Mon, Apr 24, 2017 at 11:47:32AM -0400, Steven Rostedt wrote:
> > +	/* Keep the loop going if the IPI is currently active */
> > +	atomic_inc_return(&rq->rd->rto_loop_next);  
> 
> What's the return for?

-ENOCONTEXT (too much cut from email)

/me goes and finds his email that he sent.

	/* Keep the loop going if the IPI is currently active */
	atomic_inc_return(&rq->rd->rto_loop_next);

	/* Only one CPU can initiate a loop at a time */
	if (!rto_start_trylock(&rq->rd->rto_loop_start))
		return;

Ah, it's not needed. I think I had that to supply a full memory
barrier for a previous version, but now that rto_start_trylock() is:

static inline bool rto_start_trylock(atomic_t *v)
{
	return !atomic_cmpxchg(v, 0, 1);
}

Which supplies its own memory barrier, it can now be a simple

   atomic_inc().

Thanks, will update.

-- Steve