2020-04-23 21:46:37

by Wei Yang

Subject: [PATCH] sched/core: not necessary to init root_task_group's common part for each cpu

Currently root_task_group.shares and cfs_bandwidth are initialized for
each online cpu, which not necessary.

Let's take it out to do it only once.

Signed-off-by: Wei Yang <[email protected]>
---
kernel/sched/core.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 5ca567adfcb9..613d6e892d80 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -6642,6 +6642,8 @@ void __init sched_init(void)
root_task_group.cfs_rq = (struct cfs_rq **)ptr;
ptr += nr_cpu_ids * sizeof(void **);

+ root_task_group.shares = ROOT_TASK_GROUP_LOAD;
+ init_cfs_bandwidth(&root_task_group.cfs_bandwidth);
#endif /* CONFIG_FAIR_GROUP_SCHED */
#ifdef CONFIG_RT_GROUP_SCHED
root_task_group.rt_se = (struct sched_rt_entity **)ptr;
@@ -6694,7 +6696,6 @@ void __init sched_init(void)
init_rt_rq(&rq->rt);
init_dl_rq(&rq->dl);
#ifdef CONFIG_FAIR_GROUP_SCHED
- root_task_group.shares = ROOT_TASK_GROUP_LOAD;
INIT_LIST_HEAD(&rq->leaf_cfs_rq_list);
rq->tmp_alone_branch = &rq->leaf_cfs_rq_list;
/*
@@ -6716,7 +6717,6 @@ void __init sched_init(void)
* We achieve this by letting root_task_group's tasks sit
* directly in rq->cfs (i.e root_task_group->se[] = NULL).
*/
- init_cfs_bandwidth(&root_task_group.cfs_bandwidth);
init_tg_cfs_entry(&root_task_group, &rq->cfs, NULL, i, NULL);
#endif /* CONFIG_FAIR_GROUP_SCHED */

--
2.23.0

2020-05-01 18:26:59

by tip-bot2 for Eric Biggers

Subject: [tip: sched/core] sched/core: Simplify sched_init()

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

Commit-ID: b1d1779e5ef7a60b192b61fd97201f322e1e9303
Gitweb: https://git.kernel.org/tip/b1d1779e5ef7a60b192b61fd97201f322e1e9303
Author: Wei Yang <[email protected]>
AuthorDate: Thu, 23 Apr 2020 21:44:43
Committer: Peter Zijlstra <[email protected]>
CommitterDate: Thu, 30 Apr 2020 20:14:42 +02:00

sched/core: Simplify sched_init()

Currently root_task_group.shares and cfs_bandwidth are initialized for
each online cpu, which not necessary.

Let's take it out to do it only once.

Signed-off-by: Wei Yang <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
---
kernel/sched/core.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index f6ae262..b58efb1 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -6597,6 +6597,8 @@ void __init sched_init(void)
root_task_group.cfs_rq = (struct cfs_rq **)ptr;
ptr += nr_cpu_ids * sizeof(void **);

+ root_task_group.shares = ROOT_TASK_GROUP_LOAD;
+ init_cfs_bandwidth(&root_task_group.cfs_bandwidth);
#endif /* CONFIG_FAIR_GROUP_SCHED */
#ifdef CONFIG_RT_GROUP_SCHED
root_task_group.rt_se = (struct sched_rt_entity **)ptr;
@@ -6649,7 +6651,6 @@ void __init sched_init(void)
init_rt_rq(&rq->rt);
init_dl_rq(&rq->dl);
#ifdef CONFIG_FAIR_GROUP_SCHED
- root_task_group.shares = ROOT_TASK_GROUP_LOAD;
INIT_LIST_HEAD(&rq->leaf_cfs_rq_list);
rq->tmp_alone_branch = &rq->leaf_cfs_rq_list;
/*
@@ -6671,7 +6672,6 @@ void __init sched_init(void)
* We achieve this by letting root_task_group's tasks sit
* directly in rq->cfs (i.e root_task_group->se[] = NULL).
*/
- init_cfs_bandwidth(&root_task_group.cfs_bandwidth);
init_tg_cfs_entry(&root_task_group, &rq->cfs, NULL, i, NULL);
#endif /* CONFIG_FAIR_GROUP_SCHED */