diff --git a/[refs] b/[refs]
index 0a4bc7a9be05..f98444abdca5 100644
--- a/[refs]
+++ b/[refs]
@@ -1,2 +1,2 @@
 ---
-refs/heads/master: 5926c50b83b626991c8c38efbca2020ee96b215f
+refs/heads/master: f9153ee6c71cb9ab38de3b8ed66b1c3fa27c3f7d
diff --git a/trunk/kernel/sched.c b/trunk/kernel/sched.c
index 3332bbb5d5cf..a7284bc79cdf 100644
--- a/trunk/kernel/sched.c
+++ b/trunk/kernel/sched.c
@@ -736,7 +736,9 @@ static void update_curr_load(struct rq *rq, u64 now)
  *
  * The "10% effect" is relative and cumulative: from _any_ nice level,
  * if you go up 1 level, it's -10% CPU usage, if you go down 1 level
- * it's +10% CPU usage.
+ * it's +10% CPU usage. (to achieve that we use a multiplier of 1.25.
+ * If a task goes up by ~10% and another task goes down by ~10% then
+ * the relative distance between them is ~25%.)
  */
 static const int prio_to_weight[40] = {
 /* -20 */ 88818, 71054, 56843, 45475, 36380, 29104, 23283, 18626, 14901, 11921,