sched/eevdf: Fix avg_vruntime()

The expectation is that placing a task at avg_vruntime() makes it
eligible. Turns out there is a corner case where this is not the case.

Specifically, avg_vruntime() relies on the fact that integer division
is a flooring function (eg. it discards the remainder). By this
property the value returned is slightly left of the true average.

However! when the average is a negative (relative to min_vruntime) the
effect is flipped and it becomes a ceil, with the result that the
returned value is just right of the average and thus not eligible.

Fixes: af4cf40470c2 ("sched/fair: Add cfs_rq::avg_vruntime")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 7d73652acbb23512a7050c9515ba2b60ea70f586..ef7490c4b8b40ded1c02a709bdc684a24ae7a69b 100644 (file)
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -664,6 +664,10 @@ void avg_vruntime_update(struct cfs_rq *cfs_rq, s64 delta)
        cfs_rq->avg_vruntime -= cfs_rq->avg_load * delta;
 }
 
+/*
+ * Specifically: avg_runtime() + 0 must result in entity_eligible() := true
+ * For this to be so, the result of this function must have a left bias.
+ */
 u64 avg_vruntime(struct cfs_rq *cfs_rq)
 {
        struct sched_entity *curr = cfs_rq->curr;
@@ -677,8 +681,12 @@ u64 avg_vruntime(struct cfs_rq *cfs_rq)
                load += weight;
        }
 
-       if (load)
+       if (load) {
+               /* sign flips effective floor / ceil */
+               if (avg < 0)
+                       avg -= (load - 1);
                avg = div_s64(avg, load);
+       }
 
        return cfs_rq->min_vruntime + avg;
 }