FROMLIST: sched/fair: Use wake_q length as a hint for wake_wide

(from https://patchwork.kernel.org/patch/9895261/)

This patch adds a parameter to select_task_rq, sibling_count_hint
allowing the caller, where it has this information, to inform the
sched_class the number of tasks that are being woken up as part of
the same event.

The wake_q mechanism is one case where this information is available.

select_task_rq_fair can then use the information to detect that it
needs to widen the search space for task placement in order to avoid
overloading the last-level cache domain's CPUs.

                               * * *

The reason I am investigating this change is the following use case
on ARM big.LITTLE (asymmetrical CPU capacity): 1 task per CPU, which
all repeatedly do X amount of work then
pthread_barrier_wait (i.e. sleep until the last task finishes its X
and hits the barrier). On big.LITTLE, the tasks which get a "big" CPU
finish faster, and then those CPUs pull over the tasks that are still
running:

     v CPU v           ->time->

                    -------------
   0  (big)         11111  /333
                    -------------
   1  (big)         22222   /444|
                    -------------
   2  (LITTLE)      333333/
                    -------------
   3  (LITTLE)      444444/
                    -------------

Now when task 4 hits the barrier (at |) and wakes the others up,
there are 4 tasks with prev_cpu=<big> and 0 tasks with
prev_cpu=<little>. want_affine therefore means that we'll only look
in CPUs 0 and 1 (sd_llc), so tasks will be unnecessarily coscheduled
on the bigs until the next load balance, something like this:

     v CPU v           ->time->

                    ------------------------
   0  (big)         11111  /333  31313\33333
                    ------------------------
   1  (big)         22222   /444|424\4444444
                    ------------------------
   2  (LITTLE)      333333/          \222222
                    ------------------------
   3  (LITTLE)      444444/            \1111
                    ------------------------
                                 ^^^
                           underutilization

So, I'm trying to get want_affine = 0 for these tasks.

I don't _think_ any incarnation of the wakee_flips mechanism can help
us here because which task is waker and which tasks are wakees
generally changes with each iteration.

However pthread_barrier_wait (or more accurately FUTEX_WAKE) has the
nice property that we know exactly how many tasks are being woken, so
we can cheat.

It might be a disadvantage that we "widen" _every_ task that's woken in
an event, while select_idle_sibling would work fine for the first
sd_llc_size - 1 tasks.

IIUC, if wake_affine() behaves correctly this trick wouldn't be
necessary on SMP systems, so it might be best guarded by the presence
of SD_ASYM_CPUCAPACITY?

                               * * *

Final note..

In order to observe "perfect" behaviour for this use case, I also had
to disable the TTWU_QUEUE sched feature. Suppose during the wakeup
above we are working through the work queue and have placed tasks 3
and 2, and are about to place task 1:

     v CPU v           ->time->

                    --------------
   0  (big)         11111  /333  3
                    --------------
   1  (big)         22222   /444|4
                    --------------
   2  (LITTLE)      333333/      2
                    --------------
   3  (LITTLE)      444444/          <- Task 1 should go here
                    --------------

If TTWU_QUEUE is enabled, we will not yet have enqueued task
2 (having instead sent a reschedule IPI) or attached its load to CPU
2. So we are likely to also place task 1 on cpu 2. Disabling
TTWU_QUEUE means that we enqueue task 2 before placing task 1,
solving this issue. TTWU_QUEUE is there to minimise rq lock
contention, and I guess that this contention is less of an issue on
big.LITTLE systems since they have relatively few CPUs, which
suggests the trade-off makes sense here.

Change-Id: I2080302839a263e0841a89efea8589ea53bbda9c
Signed-off-by: Brendan Jackman <brendan.jackman@arm.com>
Signed-off-by: Chris Redpath <chris.redpath@arm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Josef Bacik <josef@toxicpanda.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Matt Fleming <matt@codeblueprint.co.uk>

Author: Brendan Jackman

include/linux/sched.h

@@ -993,12 +993,13 @@ struct wake_q_node {
 struct wake_q_head {
 	struct wake_q_node *first;
 	struct wake_q_node **lastp;
+	int count;
 };
 
 #define WAKE_Q_TAIL ((struct wake_q_node *) 0x01)
 
 #define WAKE_Q(name)					\
-	struct wake_q_head name = { WAKE_Q_TAIL, &name.first }
+	struct wake_q_head name = { WAKE_Q_TAIL, &name.first, 0 }
 
 extern void wake_q_add(struct wake_q_head *head,
 		       struct task_struct *task);

kernel/sched/core.c

@@ -546,6 +546,8 @@ void wake_q_add(struct wake_q_head *head, struct task_struct *task)
 	if (cmpxchg(&node->next, NULL, WAKE_Q_TAIL))
 		return;
 
+	head->count++;
+
 	get_task_struct(task);
 
 	/*
@@ -555,6 +557,10 @@ void wake_q_add(struct wake_q_head *head, struct task_struct *task)
 	head->lastp = &node->next;
 }
 
+static int
+try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags,
+	       int sibling_count_hint);
+
 void wake_up_q(struct wake_q_head *head)
 {
 	struct wake_q_node *node = head->first;
@@ -569,10 +575,10 @@ void wake_up_q(struct wake_q_head *head)
 		task->wake_q.next = NULL;
 
 		/*
-		 * wake_up_process() implies a wmb() to pair with the queueing
+		 * try_to_wake_up() implies a wmb() to pair with the queueing
 		 * in wake_q_add() so as not to miss wakeups.
 		 */
-		wake_up_process(task);
+		try_to_wake_up(task, TASK_NORMAL, 0, head->count);
 		put_task_struct(task);
 	}
 }
@@ -1642,12 +1648,14 @@ static int select_fallback_rq(int cpu, struct task_struct *p)
  * The caller (fork, wakeup) owns p->pi_lock, ->cpus_allowed is stable.
  */
 static inline
-int select_task_rq(struct task_struct *p, int cpu, int sd_flags, int wake_flags)
+int select_task_rq(struct task_struct *p, int cpu, int sd_flags, int wake_flags,
+		   int sibling_count_hint)
 {
 	lockdep_assert_held(&p->pi_lock);
 
 	if (p->nr_cpus_allowed > 1)
-		cpu = p->sched_class->select_task_rq(p, cpu, sd_flags, wake_flags);
+		cpu = p->sched_class->select_task_rq(p, cpu, sd_flags, wake_flags,
+						     sibling_count_hint);
 
 	/*
 	 * In order not to call set_task_cpu() on a blocking task we need
@@ -1932,6 +1940,8 @@ static void ttwu_queue(struct task_struct *p, int cpu)
  * @p: the thread to be awakened
  * @state: the mask of task states that can be woken
  * @wake_flags: wake modifier flags (WF_*)
+ * @sibling_count_hint: A hint at the number of threads that are being woken up
+ *                      in this event.
  *
  * Put it on the run-queue if it's not already there. The "current"
  * thread is always on the run-queue (except when the actual
@@ -1943,7 +1953,8 @@ static void ttwu_queue(struct task_struct *p, int cpu)
  * or @state didn't match @p's state.
  */
 static int
-try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
+try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags,
+	       int sibling_count_hint)
 {
 	unsigned long flags;
 	int cpu, success = 0;
@@ -2044,8 +2055,8 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
 	if (p->sched_class->task_waking)
 		p->sched_class->task_waking(p);
 
-	cpu = select_task_rq(p, p->wake_cpu, SD_BALANCE_WAKE, wake_flags);
-
+	cpu = select_task_rq(p, p->wake_cpu, SD_BALANCE_WAKE, wake_flags,
+			     sibling_count_hint);
 	if (task_cpu(p) != cpu) {
 		wake_flags |= WF_MIGRATED;
 		set_task_cpu(p, cpu);
@@ -2127,13 +2138,13 @@ static void try_to_wake_up_local(struct task_struct *p)
  */
 int wake_up_process(struct task_struct *p)
 {
-	return try_to_wake_up(p, TASK_NORMAL, 0);
+	return try_to_wake_up(p, TASK_NORMAL, 0, 1);
 }
 EXPORT_SYMBOL(wake_up_process);
 
 int wake_up_state(struct task_struct *p, unsigned int state)
 {
-	return try_to_wake_up(p, state, 0);
+	return try_to_wake_up(p, state, 0, 1);
 }
 
 /*
@@ -2467,7 +2478,7 @@ void wake_up_new_task(struct task_struct *p)
 	 * Use __set_task_cpu() to avoid calling sched_class::migrate_task_rq,
 	 * as we're not fully set-up yet.
 	 */
-	__set_task_cpu(p, select_task_rq(p, task_cpu(p), SD_BALANCE_FORK, 0));
+	__set_task_cpu(p, select_task_rq(p, task_cpu(p), SD_BALANCE_FORK, 0, 1));
 #endif
 	rq = __task_rq_lock(p);
 	update_rq_clock(rq);
@@ -2905,7 +2916,7 @@ void sched_exec(void)
 	int dest_cpu;
 
 	raw_spin_lock_irqsave(&p->pi_lock, flags);
-	dest_cpu = p->sched_class->select_task_rq(p, task_cpu(p), SD_BALANCE_EXEC, 0);
+	dest_cpu = p->sched_class->select_task_rq(p, task_cpu(p), SD_BALANCE_EXEC, 0, 1);
 	if (dest_cpu == smp_processor_id())
 		goto unlock;
 
@@ -3560,7 +3571,7 @@ asmlinkage __visible void __sched preempt_schedule_irq(void)
 int default_wake_function(wait_queue_t *curr, unsigned mode, int wake_flags,
 			  void *key)
 {
-	return try_to_wake_up(curr->private, mode, wake_flags);
+	return try_to_wake_up(curr->private, mode, wake_flags, 1);
 }
 EXPORT_SYMBOL(default_wake_function);
 

kernel/sched/deadline.c

@@ -1070,7 +1070,8 @@ static void yield_task_dl(struct rq *rq)
 static int find_later_rq(struct task_struct *task);
 
 static int
-select_task_rq_dl(struct task_struct *p, int cpu, int sd_flag, int flags)
+select_task_rq_dl(struct task_struct *p, int cpu, int sd_flag, int flags,
+		  int sibling_count_hint)
 {
 	struct task_struct *curr;
 	struct rq *rq;

kernel/sched/fair.c

@@ -5773,15 +5773,18 @@ energy_diff(struct energy_env *eenv)
  * being client/server, worker/dispatcher, interrupt source or whatever is
  * irrelevant, spread criteria is apparent partner count exceeds socket size.
  */
-static int wake_wide(struct task_struct *p)
+static int wake_wide(struct task_struct *p, int sibling_count_hint)
 {
 	unsigned int master = current->wakee_flips;
 	unsigned int slave = p->wakee_flips;
-	int factor = this_cpu_read(sd_llc_size);
+	int llc_size = this_cpu_read(sd_llc_size);
+
+	if (sibling_count_hint >= llc_size)
+		return 1;
 
 	if (master < slave)
 		swap(master, slave);
-	if (slave < factor || master < slave * factor)
+	if (slave < llc_size || master < slave * llc_size)
 		return 0;
 	return 1;
 }
@@ -6754,17 +6757,21 @@ static int select_energy_cpu_brute(struct task_struct *p, int prev_cpu, int sync
  * preempt must be disabled.
  */
 static int
-select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_flags)
+select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_flags,
+		    int sibling_count_hint)
 {
 	struct sched_domain *tmp, *affine_sd = NULL, *sd = NULL;
 	int cpu = smp_processor_id();
 	int new_cpu = prev_cpu;
 	int want_affine = 0;
 	int sync = wake_flags & WF_SYNC;
 
-	if (sd_flag & SD_BALANCE_WAKE)
-		want_affine = !wake_wide(p) && !wake_cap(p, cpu, prev_cpu)
-			      && cpumask_test_cpu(cpu, tsk_cpus_allowed(p));
+	if (sd_flag & SD_BALANCE_WAKE) {
+		record_wakee(p);
+		want_affine = !wake_wide(p, sibling_count_hint) &&
+			      !wake_cap(p, cpu, prev_cpu) &&
+			      cpumask_test_cpu(cpu, &p->cpus_allowed);
+	}
 
 	if (energy_aware() && !(cpu_rq(prev_cpu)->rd->overutilized))
 		return select_energy_cpu_brute(p, prev_cpu, sync);

kernel/sched/idle_task.c

@@ -9,7 +9,8 @@
 
 #ifdef CONFIG_SMP
 static int
-select_task_rq_idle(struct task_struct *p, int cpu, int sd_flag, int flags)
+select_task_rq_idle(struct task_struct *p, int cpu, int sd_flag, int flags,
+		    int sibling_count_hint)
 {
 	return task_cpu(p); /* IDLE tasks as never migrated */
 }

kernel/sched/rt.c

@@ -1372,7 +1372,8 @@ static void yield_task_rt(struct rq *rq)
 static int find_lowest_rq(struct task_struct *task);
 
 static int
-select_task_rq_rt(struct task_struct *p, int cpu, int sd_flag, int flags)
+select_task_rq_rt(struct task_struct *p, int cpu, int sd_flag, int flags,
+		  int sibling_count_hint)
 {
 	struct task_struct *curr;
 	struct rq *rq;

kernel/sched/sched.h

@@ -1261,7 +1261,8 @@ struct sched_class {
 	void (*put_prev_task) (struct rq *rq, struct task_struct *p);
 
 #ifdef CONFIG_SMP
-	int  (*select_task_rq)(struct task_struct *p, int task_cpu, int sd_flag, int flags);
+	int  (*select_task_rq)(struct task_struct *p, int task_cpu, int sd_flag, int flags,
+			       int subling_count_hint);
 	void (*migrate_task_rq)(struct task_struct *p);
 
 	void (*task_waking) (struct task_struct *task);

kernel/sched/stop_task.c

@@ -12,7 +12,8 @@
 
 #ifdef CONFIG_SMP
 static int
-select_task_rq_stop(struct task_struct *p, int cpu, int sd_flag, int flags)
+select_task_rq_stop(struct task_struct *p, int cpu, int sd_flag, int flags,
+		    int sibling_count_hint)
 {
 	return task_cpu(p); /* stop tasks as never migrate */
 }