Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Tue, 16 Oct 2018 14:35:17 +0200
From:   Peter Zijlstra <peterz@infradead.org>
To:     Steven Rostedt <rostedt@goodmis.org>
Cc:     Peng Hao <peng.hao2@zte.com.cn>, mingo@redhat.com,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3] sched/rt : return accurate release rq lock info
Message-ID: <20181016123517.GB2537@hirez.programming.kicks-ass.net>
References: <1538778131-44406-1-git-send-email-peng.hao2@zte.com.cn>
 <20181015092032.GO9867@hirez.programming.kicks-ass.net>
 <20181015114220.70c3598d@gandalf.local.home>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20181015114220.70c3598d@gandalf.local.home>
User-Agent: Mutt/1.10.1 (2018-07-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Oct 15, 2018 at 11:42:20AM -0400, Steven Rostedt wrote:
> On Mon, 15 Oct 2018 11:20:32 +0200
> Peter Zijlstra <peterz@infradead.org> wrote:
> 
> > > index 2e2955a..be0fc43 100644
> > > --- a/kernel/sched/rt.c
> > > +++ b/kernel/sched/rt.c
> > > @@ -1754,7 +1754,7 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq)
> > >  				     !task_on_rq_queued(task))) {
> > >  
> > >  				double_unlock_balance(rq, lowest_rq);
> > > -				lowest_rq = NULL;
> > > +				lowest_rq = RETRY_TASK;
> > >  				break;
> > >  			}
> > >  		}  
> > 
> > I'm confused.. should not:
> > 
> > 		/* try again */
> > 		double_unlock_balance(rq, lowest_rq);
> > 		lowest_rq = NULL;
> > 
> > also return RETRY_TASK? That also is in the double_lock_balance() path
> > and will this have had rq->lock() released.
> 
> I thought the same thing at first, but this is in the loop path, where
> it does everything again. But now looking closer, I think there's a bug
> in the original code.

So I find that whole thing utterly confusing; what about we start with
something like so?

---
 kernel/sched/rt.c | 51 +++++++++++++++++++++++----------------------------
 1 file changed, 23 insertions(+), 28 deletions(-)

diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index 2e2955a8cf8f..237c84c2b042 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -1714,6 +1714,26 @@ static int find_lowest_rq(struct task_struct *task)
 	return -1;
 }
 
+static struct task_struct *first_pushable_task(struct rq *rq)
+{
+	struct task_struct *p;
+
+	if (!has_pushable_tasks(rq))
+		return NULL;
+
+	p = plist_first_entry(&rq->rt.pushable_tasks,
+			      struct task_struct, pushable_tasks);
+
+	BUG_ON(rq->cpu != task_cpu(p));
+	BUG_ON(task_current(rq, p));
+	BUG_ON(p->nr_cpus_allowed <= 1);
+
+	BUG_ON(!task_on_rq_queued(p));
+	BUG_ON(!rt_task(p));
+
+	return p;
+}
+
 /* Will lock the rq it finds */
 static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq)
 {
@@ -1747,12 +1767,7 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq)
 			 * migrated already or had its affinity changed.
 			 * Also make sure that it wasn't scheduled on its rq.
 			 */
-			if (unlikely(task_rq(task) != rq ||
-				     !cpumask_test_cpu(lowest_rq->cpu, &task->cpus_allowed) ||
-				     task_running(rq, task) ||
-				     !rt_task(task) ||
-				     !task_on_rq_queued(task))) {
-
+			if (first_pushable_task(rq) != task)
 				double_unlock_balance(rq, lowest_rq);
 				lowest_rq = NULL;
 				break;
@@ -1771,26 +1786,6 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq)
 	return lowest_rq;
 }
 
-static struct task_struct *pick_next_pushable_task(struct rq *rq)
-{
-	struct task_struct *p;
-
-	if (!has_pushable_tasks(rq))
-		return NULL;
-
-	p = plist_first_entry(&rq->rt.pushable_tasks,
-			      struct task_struct, pushable_tasks);
-
-	BUG_ON(rq->cpu != task_cpu(p));
-	BUG_ON(task_current(rq, p));
-	BUG_ON(p->nr_cpus_allowed <= 1);
-
-	BUG_ON(!task_on_rq_queued(p));
-	BUG_ON(!rt_task(p));
-
-	return p;
-}
-
 /*
  * If the current CPU has more than one RT task, see if the non
  * running task can migrate over to a CPU that is running a task
@@ -1805,7 +1800,7 @@ static int push_rt_task(struct rq *rq)
 	if (!rq->rt.overloaded)
 		return 0;
 
-	next_task = pick_next_pushable_task(rq);
+	next_task = first_pushable_task(rq);
 	if (!next_task)
 		return 0;
 
@@ -1840,7 +1835,7 @@ static int push_rt_task(struct rq *rq)
 		 * run-queue and is also still the next task eligible for
 		 * pushing.
 		 */
-		task = pick_next_pushable_task(rq);
+		task = first_pushable_task(rq);
 		if (task == next_task) {
 			/*
 			 * The task hasn't migrated, and is still the next