Re: [PATCH 7/7] Propagate negative bias

From: Hongyan Xia
Date: Tue Jun 25 2024 - 06:30:37 EST


Hi,

Thanks for taking a look!

On 25/06/2024 05:48, K Prateek Nayak wrote:
Hello Hongyan,

On 6/24/2024 3:53 PM, Hongyan Xia wrote:
Negative bias is interesting, because dequeuing such a task will
actually increase utilization.

Solve by applying PELT decay to negative biases as well. This in fact
can be implemented easily with some math tricks.

Signed-off-by: Hongyan Xia <hongyan.xia2@xxxxxxx>
---
  kernel/sched/fair.c  | 40 ++++++++++++++++++++++++++++++++++++++++
  kernel/sched/sched.h |  4 ++++
  2 files changed, 44 insertions(+)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 3bb077df52ae..d09af6abf464 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -4878,6 +4878,45 @@ static inline unsigned long root_cfs_util_uclamp(struct rq *rq)
      return max(ret, 0L);
  }
+
+/*
+ * Negative biases are tricky. If we remove them right away then dequeuing a
+ * uclamp_max task has the interesting effect that dequeuing results in a higher
+ * rq utilization. Solve this by applying PELT decay to the bias itself.
+ *
+ * Keeping track of a PELT-decayed negative bias is extra overhead. However, we
+ * observe this interesting math property, where y is the decay factor and p is
+ * the number of periods elapsed:
+ *
+ *    util_new = util_old * y^p - neg_bias * y^p
+ *         = (util_old - neg_bias) * y^p
+ *
+ * Therefore, we simply subtract the negative bias from util_avg the moment we
+ * dequeue, then the PELT signal itself is the total of util_avg and the decayed
+ * negative bias, and we no longer need to track the decayed bias separately.
+ */
+static void propagate_negative_bias(struct task_struct *p)
+{
+    if (task_util_bias(p) < 0 && !task_on_rq_migrating(p)) {
+        unsigned long neg_bias = -task_util_bias(p);
+        struct sched_entity *se = &p->se;
+        struct cfs_rq *cfs_rq;
+
+        p->se.avg.util_avg_bias = 0;
+
+        for_each_sched_entity(se) {
+            u32 divider, neg_sum;
+
+            cfs_rq = cfs_rq_of(se);
+            divider = get_pelt_divider(&cfs_rq->avg);
+            neg_sum = neg_bias * divider;
+            sub_positive(&se->avg.util_avg, neg_bias);
+            sub_positive(&se->avg.util_sum, neg_sum);

Most cases where I've seen "get_pelt_divider()" followed by
"add_positive()" or "sub_positive()" on "util_avg" and "util_sum" I've
seen a correction step that does:

    util_sum = max_t(u32, util_sum, util_avg * PELT_MIN_DIVIDER)

There is a comment on its significance in "update_cfs_rq_load_avg()".
Would it also apply in this case?


That's a good point. The problem in update_cfs_rq_load_avg() should also be possible here. I can add the guard logic in the next rev.

But if we change the code in a way suggested below, then this problem is solved anyway.

+            sub_positive(&cfs_rq->avg.util_avg, neg_bias);
+            sub_positive(&cfs_rq->avg.util_sum, neg_sum);
+        }
+    }
+}
  #else
  static inline long task_util_bias(struct task_struct *p)
  {
@@ -6869,6 +6908,7 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
      /* At this point se is NULL and we are at root level*/
      sub_nr_running(rq, 1);
      util_bias_dequeue(rq, p);
+    propagate_negative_bias(p);

Perhaps I'm pointing to a premature optimization but since the hierarchy
is traversed above in "dequeue_task_fair()", could the "neg_bias" and
"neg_sum" removal be done along the way above instead of
"propagate_negative_bias()" traversing the hierarchy again? I don't see
a dependency on "util_bias_dequeue()" (which modifies
"rq->cfs.avg.util_avg_bias") for "propagate_negative_bias()" (which
works purely with task_util_bias() or "p->se.avg.util_avg_bias") but if
I'm missing something please do let me know.

Since you mentioned this patch isn't strictly necessary in the cover
letter, I would wait for other folks to chime in before changing this :)

I've been thinking about similar things for both enqueue() and dequeue(). Currently this series makes util_avg_bias completely separate from util_avg to ease review, acting more like util_est, but like you said we do things twice in a couple of places.

enqueue_task_fair():
for_each_sched_entity()
enqueue_entity()
if root_cfs()
cpufreq_update_util()
util_bias_enqueue(p)
cpufreq_update_util() // duplicate cpufreq update

dequeue_task_fair():
for_each_sched_entity()
dequeue_entity()
if root_cfs()
cpufreq_update_util()
util_bias_dequeue(p)
propagate_negative_bias() // duplicate tree traversal
cpufreq_update_util() // duplicate cpufreq update

But we can integrate the bias closer into the hierarchy, like this:

enqueue_task_fair():
for_each_sched_entity()
enqueue_entity()
if (entity_is_task())
util_bias_enqueue(p)
if root_cfs()
// No duplicate cpufreq updates
cpufreq_update_util()

dequeue_task_fair():
for_each_sched_entity()
dequeue_entity()
if (entity_is_task())
util_bias_dequeue(p)
// No need to traverse twice.
propagate_negative_bias(p)
if root_cfs()
// No duplicate cpufreq updates
cpufreq_update_util()

This new structure will address both of your concerns.

      /* balance early to pull high priority tasks */
      if (unlikely(!was_sched_idle && sched_idle_rq(rq)))
[..snip..]