[PATCH v2 1/3] sched/fair: Add asymmetric CPU capacity wakeup scan

[Valentin Schneider]

From: Morten Rasmussen <morten.rasmussen@xxxxxxx>

On asymmetric CPU capacity topologies, we currently rely on wake_cap() to
drive select_task_rq_fair() towards either
- its slow-path (find_idlest_cpu()) if either the previous or
  current (waking) CPU has too little capacity for the waking task
- its fast-path (select_idle_sibling()) otherwise

Commit 3273163c6775 ("sched/fair: Let asymmetric CPU configurations balance
at wake-up") points out that this relies on the assumption that "[...]the
CPU capacities within an SD_SHARE_PKG_RESOURCES domain (sd_llc) are
homogeneous".

This assumption no longer holds on newer generations of big.LITTLE
systems (DynamIQ), which can accommodate CPUs of different compute capacity
within a single LLC domain. To hopefully paint a better picture, a regular
big.LITTLE topology would look like this:

  +---------+ +---------+
  |   L2    | |   L2    |
  +----+----+ +----+----+
  |CPU0|CPU1| |CPU2|CPU3|
  +----+----+ +----+----+
      ^^^         ^^^
    LITTLEs      bigs

which would result in the following scheduler topology:

  DIE [         ] <- sd_asym_cpucapacity
  MC  [   ] [   ] <- sd_llc
       0 1   2 3

Conversely, a DynamIQ topology could look like:

  +-------------------+
  |        L3         |
  +----+----+----+----+
  | L2 | L2 | L2 | L2 |
  +----+----+----+----+
  |CPU0|CPU1|CPU2|CPU3|
  +----+----+----+----+
     ^^^^^     ^^^^^
    LITTLEs    bigs

which would result in the following scheduler topology:

  MC [       ] <- sd_llc, sd_asym_cpucapacity
      0 1 2 3

What this means is that, on DynamIQ systems, we could pass the wake_cap()
test (IOW presume the waking task fits on the CPU capacities of some LLC
domain), thus go through select_idle_sibling().
This function operates on an LLC domain, which here spans both bigs and
LITTLEs, so it could very well pick a CPU of too small capacity for the
task, despite there being fitting idle CPUs - it very much depends on the
CPU iteration order, on which we have absolutely no guarantees
capacity-wise.

Introduce yet another select_idle_sibling() helper function that takes CPU
capacity into account. The policy is basically to pick the first idle CPU
which is big enough for the task (task_util * margin < cpu_capacity).

Unlike other select_idle_sibling() helpers, this one operates on the
sd_asym_cpucapacity sched_domain pointer, which is guaranteed to span all
known CPU capacities in the system. As such, this will work for both
"legacy" big.LITTLE (LITTLEs & bigs split at MC, joined at DIE) and for
newer DynamIQ systems (e.g. LITTLEs and bigs in the same MC domain).

Co-authored-by: Valentin Schneider <valentin.schneider@xxxxxxx>
Signed-off-by: Morten Rasmussen <morten.rasmussen@xxxxxxx>
Signed-off-by: Valentin Schneider <valentin.schneider@xxxxxxx>
---
 kernel/sched/fair.c | 37 +++++++++++++++++++++++++++++++++++++
 1 file changed, 37 insertions(+)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index fe4e0d7753756..c44b135f61ad0 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5894,6 +5894,38 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int t
 	return cpu;
 }
 
+/*
+ * Scan the asym_capacity domain for idle CPUs; pick the first idle one on which
+ * the task fits.
+ */
+static int select_idle_capacity(struct task_struct *p, int target)
+{
+	struct sched_domain *sd;
+	struct cpumask *cpus;
+	int cpu;
+
+	if (!static_branch_unlikely(&sched_asym_cpucapacity))
+		return -1;
+
+	sd = rcu_dereference(per_cpu(sd_asym_cpucapacity, target));
+	if (!sd)
+		return -1;
+
+	cpus = this_cpu_cpumask_var_ptr(select_idle_mask);
+	cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr);
+
+	for_each_cpu_wrap(cpu, cpus, target) {
+		if (!available_idle_cpu(cpu))
+			continue;
+		if (!task_fits_capacity(p, capacity_of(cpu)))
+			continue;
+
+		return cpu;
+	}
+
+	return -1;
+}
+
 /*
  * Try and locate an idle core/thread in the LLC cache domain.
  */
@@ -5902,6 +5934,11 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target)
 	struct sched_domain *sd;
 	int i, recent_used_cpu;
 
+	/* For asymmetric capacities, try to be smart about the placement */
+	i = select_idle_capacity(p, target);
+	if ((unsigned)i < nr_cpumask_bits)
+		return i;
+
 	if (available_idle_cpu(target) || sched_idle_cpu(target))
 		return target;
 
-- 
2.24.0