[PATCH sched_ext/for-6.13] sched_ext: Do not enable LLC/NUMA optimizations when domains overlap

From: Andrea Righi
Date: Thu Oct 31 2024 - 03:34:20 EST


When the LLC and NUMA domains fully overlap, enabling both optimizations
in the built-in idle CPU selection policy is redundant, as it leads to
searching for an idle CPU within the same domain twice.

Likewise, if all online CPUs are within a single LLC domain, LLC
optimization is unnecessary.

Therefore, detect overlapping domains and enable topology optimizations
only when necessary.

Fixes: 860a45219bce ("sched_ext: Introduce NUMA awareness to the default idle selection policy")
Signed-off-by: Andrea Righi <arighi@xxxxxxxxxx>
---
kernel/sched/ext.c | 23 ++++++++++++++++++-----
1 file changed, 18 insertions(+), 5 deletions(-)

diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c
index fc7f15eefe54..82acbaffd5a7 100644
--- a/kernel/sched/ext.c
+++ b/kernel/sched/ext.c
@@ -3140,7 +3140,7 @@ static void update_selcpu_topology(void)
{
bool enable_llc = false, enable_numa = false;
struct sched_domain *sd;
- const struct cpumask *cpus;
+ const struct cpumask *llc_cpus = NULL, *numa_cpus = NULL;
s32 cpu = cpumask_first(cpu_online_mask);

/*
@@ -3154,16 +3154,29 @@ static void update_selcpu_topology(void)
rcu_read_lock();
sd = rcu_dereference(per_cpu(sd_llc, cpu));
if (sd) {
- cpus = sched_domain_span(sd);
- if (cpumask_weight(cpus) < num_possible_cpus())
+ llc_cpus = sched_domain_span(sd);
+ if (cpumask_weight(llc_cpus) < num_possible_cpus())
enable_llc = true;
}
sd = highest_flag_domain(cpu, SD_NUMA);
if (sd) {
- cpus = sched_group_span(sd->groups);
- if (cpumask_weight(cpus) < num_possible_cpus())
+ numa_cpus = sched_group_span(sd->groups);
+ if (cpumask_weight(numa_cpus) < num_possible_cpus())
enable_numa = true;
}
+ /*
+ * If the NUMA domain perfectly overlaps with the LLC domain, enable
+ * LLC optimization only, as checking for an idle CPU in the same
+ * domain twice is redundant.
+ */
+ if (enable_numa && enable_llc && cpumask_equal(numa_cpus, llc_cpus))
+ enable_numa = false;
+ /*
+ * If all the online CPUs are in the same LLC domain, there is no
+ * reason to enable LLC optimization.
+ */
+ if (enable_llc && cpumask_equal(llc_cpus, cpu_online_mask))
+ enable_llc = false;
rcu_read_unlock();

pr_debug("sched_ext: LLC idle selection %s\n",
--
2.47.0