diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 19da937f4ae6..ce7c9b3f2760 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -11121,6 +11121,27 @@ static inline bool nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle
 static inline void nohz_newidle_balance(struct rq *this_rq) { }
 #endif /* CONFIG_NO_HZ_COMMON */
 
+#ifdef CONFIG_SCHED_WALT
+static bool silver_has_big_tasks(void)
+{
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		if (!is_min_capacity_cpu(cpu))
+			break;
+		if (cpu_rq(cpu)->walt_stats.nr_big_tasks)
+			return true;
+	}
+
+	return false;
+}
+#else
+static inline bool silver_has_big_tasks(void)
+{
+	return false;
+}
+#endif
+
 /*
  * idle_balance is called by schedule() if this_cpu is about to become
  * idle. Attempts to pull tasks from other CPUs.
@@ -11132,6 +11153,7 @@ int newidle_balance(struct rq *this_rq, struct rq_flags *rf)
 	struct sched_domain *sd;
 	int pulled_task = 0;
 	u64 curr_cost = 0;
+	u64 avg_idle = this_rq->avg_idle;
 
 	if (cpu_isolated(this_cpu))
 		return 0;
@@ -11148,7 +11170,8 @@ int newidle_balance(struct rq *this_rq, struct rq_flags *rf)
 	 */
 	if (!cpu_active(this_cpu))
 		return 0;
-
+	if (!is_min_capacity_cpu(this_cpu) && silver_has_big_tasks())
+		avg_idle = ULLONG_MAX;
 	/*
 	 * This is OK, because current is on_cpu, which avoids it being picked
 	 * for load-balance and preemption/IRQs are still disabled avoiding
@@ -11157,7 +11180,7 @@ int newidle_balance(struct rq *this_rq, struct rq_flags *rf)
 	 */
 	rq_unpin_lock(this_rq, rf);
 
-	if (this_rq->avg_idle < sysctl_sched_migration_cost ||
+	if (avg_idle < sysctl_sched_migration_cost ||
 	    !READ_ONCE(this_rq->rd->overload)) {
 
 		rcu_read_lock();
@@ -11182,7 +11205,7 @@ int newidle_balance(struct rq *this_rq, struct rq_flags *rf)
 		if (!(sd->flags & SD_LOAD_BALANCE))
 			continue;
 
-		if (this_rq->avg_idle < curr_cost + sd->max_newidle_lb_cost) {
+		if (avg_idle < curr_cost + sd->max_newidle_lb_cost) {
 			update_next_balance(sd, &next_balance);
 			break;
 		}