[tip: sched/core] sched,fair: Alternative sched_slice()

[tip-bot2 for Peter Zijlstra]

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     0c2de3f054a59f15e01804b75a04355c48de628c
Gitweb:        https://git.kernel.org/tip/0c2de3f054a59f15e01804b75a04355c48de628c
Author:        Peter Zijlstra <peterz@xxxxxxxxxxxxx>
AuthorDate:    Thu, 25 Mar 2021 13:44:46 +01:00
Committer:     Peter Zijlstra <peterz@xxxxxxxxxxxxx>
CommitterDate: Fri, 16 Apr 2021 17:06:35 +02:00

sched,fair: Alternative sched_slice()

The current sched_slice() seems to have issues; there's two possible
things that could be improved:

 - the 'nr_running' used for __sched_period() is daft when cgroups are
   considered. Using the RQ wide h_nr_running seems like a much more
   consistent number.

 - (esp) cgroups can slice it real fine, which makes for easy
   over-scheduling, ensure min_gran is what the name says.

Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
Tested-by: Valentin Schneider <valentin.schneider@xxxxxxx>
Link: https://lkml.kernel.org/r/20210412102001.611897312@xxxxxxxxxxxxx
---
 kernel/sched/fair.c     | 12 +++++++++++-
 kernel/sched/features.h |  3 +++
 2 files changed, 14 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index b3ea14c..49636a4 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -687,7 +687,13 @@ static u64 __sched_period(unsigned long nr_running)
  */
 static u64 sched_slice(struct cfs_rq *cfs_rq, struct sched_entity *se)
 {
-	u64 slice = __sched_period(cfs_rq->nr_running + !se->on_rq);
+	unsigned int nr_running = cfs_rq->nr_running;
+	u64 slice;
+
+	if (sched_feat(ALT_PERIOD))
+		nr_running = rq_of(cfs_rq)->cfs.h_nr_running;
+
+	slice = __sched_period(nr_running + !se->on_rq);
 
 	for_each_sched_entity(se) {
 		struct load_weight *load;
@@ -704,6 +710,10 @@ static u64 sched_slice(struct cfs_rq *cfs_rq, struct sched_entity *se)
 		}
 		slice = __calc_delta(slice, se->load.weight, load);
 	}
+
+	if (sched_feat(BASE_SLICE))
+		slice = max(slice, (u64)sysctl_sched_min_granularity);
+
 	return slice;
 }
 
diff --git a/kernel/sched/features.h b/kernel/sched/features.h
index 422fa68..011c5ec 100644
--- a/kernel/sched/features.h
+++ b/kernel/sched/features.h
@@ -90,3 +90,6 @@ SCHED_FEAT(WA_BIAS, true)
  */
 SCHED_FEAT(UTIL_EST, true)
 SCHED_FEAT(UTIL_EST_FASTUP, true)
+
+SCHED_FEAT(ALT_PERIOD, true)
+SCHED_FEAT(BASE_SLICE, true)