diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index b231ed3e6c74..15a8e35f0198 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -5885,6 +5885,8 @@ static int init_rootdomain(struct root_domain *rd)
 
 	if (cpupri_init(&rd->cpupri) != 0)
 		goto free_rto_mask;
+
+	init_max_cpu_capacity(&rd->max_cpu_capacity);
 	return 0;
 
 free_rto_mask:
@@ -7131,20 +7133,10 @@ static int build_sched_domains(const struct cpumask *cpu_map,
 	for_each_cpu(i, cpu_map) {
 		rq = cpu_rq(i);
 		sd = *per_cpu_ptr(d.sd, i);
-
-		/* Use READ_ONCE()/WRITE_ONCE() to avoid load/store tearing: */
-		if (rq->cpu_capacity_orig > READ_ONCE(d.rd->max_cpu_capacity))
-			WRITE_ONCE(d.rd->max_cpu_capacity, rq->cpu_capacity_orig);
-
 		cpu_attach_domain(sd, d.rd, i);
 	}
 	rcu_read_unlock();
 
-	if (rq && sched_debug_enabled) {
-		pr_info("span: %*pbl (max cpu_capacity = %lu)\n",
-			cpumask_pr_args(cpu_map), rq->rd->max_cpu_capacity);
-	}
-
 	ret = 0;
 error:
 	__free_domain_allocs(&d, alloc_state, cpu_map);
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 2672e7e6934a..bfd407869ef0 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5534,7 +5534,7 @@ static inline bool __task_fits(struct task_struct *p, int cpu, int util)
 static inline bool task_fits_max(struct task_struct *p, int cpu)
 {
 	unsigned long capacity = capacity_of(cpu);
-	unsigned long max_capacity = cpu_rq(cpu)->rd->max_cpu_capacity;
+	unsigned long max_capacity = cpu_rq(cpu)->rd->max_cpu_capacity.val;
 
 	if (capacity == max_capacity)
 		return true;
@@ -7246,13 +7246,43 @@ static unsigned long scale_rt_capacity(int cpu)
 	return 1;
 }
 
+void init_max_cpu_capacity(struct max_cpu_capacity *mcc)
+{
+	raw_spin_lock_init(&mcc->lock);
+	mcc->val = 0;
+	mcc->cpu = -1;
+}
+
 static void update_cpu_capacity(struct sched_domain *sd, int cpu)
 {
 	unsigned long capacity = arch_scale_cpu_capacity(sd, cpu);
 	struct sched_group *sdg = sd->groups;
+	struct max_cpu_capacity *mcc;
+	unsigned long max_capacity;
+	int max_cap_cpu;
+	unsigned long flags;
 
 	cpu_rq(cpu)->cpu_capacity_orig = capacity;
 
+	mcc = &cpu_rq(cpu)->rd->max_cpu_capacity;
+
+	raw_spin_lock_irqsave(&mcc->lock, flags);
+	max_capacity = mcc->val;
+	max_cap_cpu = mcc->cpu;
+
+	if ((max_capacity > capacity && max_cap_cpu == cpu) ||
+	    (max_capacity < capacity)) {
+		mcc->val = capacity;
+		mcc->cpu = cpu;
+#ifdef CONFIG_SCHED_DEBUG
+		raw_spin_unlock_irqrestore(&mcc->lock, flags);
+		pr_info("CPU%d: update max cpu_capacity %lu\n", cpu, capacity);
+		goto skip_unlock;
+#endif
+	}
+	raw_spin_unlock_irqrestore(&mcc->lock, flags);
+
+skip_unlock: __attribute__ ((unused));
 	capacity *= scale_rt_capacity(cpu);
 	capacity >>= SCHED_CAPACITY_SHIFT;
 
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 86512afe0711..d08f3b054194 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -539,6 +539,12 @@ struct dl_rq {
 
 #ifdef CONFIG_SMP
 
+struct max_cpu_capacity {
+	raw_spinlock_t lock;
+	unsigned long val;
+	int cpu;
+};
+
 /*
  * We add the notion of a root-domain which will be used to define per-domain
  * variables. Each exclusive cpuset essentially defines an island domain by
@@ -576,7 +582,8 @@ struct root_domain {
 	cpumask_var_t rto_mask;
 	struct cpupri cpupri;
 
-	unsigned long max_cpu_capacity;
+	/* Maximum cpu capacity in the system. */
+	struct max_cpu_capacity max_cpu_capacity;
 };
 
 extern struct root_domain def_root_domain;
@@ -1414,6 +1421,8 @@ static inline void sched_update_tick_dependency(struct rq *rq)
 static inline void sched_update_tick_dependency(struct rq *rq) { }
 #endif
 
+extern void init_max_cpu_capacity(struct max_cpu_capacity *mcc);
+
 static inline void add_nr_running(struct rq *rq, unsigned count)
 {
 	unsigned prev_nr = rq->nr_running;