Skip to contentSkip to conversation

sched/fair: Fix sched_entity::avg::decay_count initialization

PATCH In review Kirill Tkhai 3 participants 3 messages Dec 15, 2014 – Jan 14, 2015 Scheduler

lore
Share & tools
lore

Kirill Tkhai

author
#0 lore raw

PATCH 1/1 sched/fair: Fix sched_entity::avg::decay_count initialization

Child has the same decay_count as parent. If it's not zero, we add it to parent's cfs_rq->removed_load:

wake_up_new_task()->set_task_cpu()->migrate_task_rq_fair().

Child's load is a just garbade after copying of parent, it hasn't been on cfs_rq yet, and it must not be added to cfs_rq::removed_load in migrate_task_rq_fair().

The patch moves sched_entity::avg::decay_count intialization in sched_fork(). So, migrate_task_rq_fair() does not change removed_load.

Signed-off-by Kirill Tkhai <[email protected]>
---
 kernel/sched/core.c |    1 +
 kernel/sched/fair.c |    1 -
 2 files changed, 1 insertion(+), 1 deletion(-)
kernel/sched/core.c +1 -0
index bb398c0..2894b69 100644
@@ -1832,6 +1832,7 @@ static void __sched_fork(unsigned long clone_flags, struct task_struct *p)
 	p->se.prev_sum_exec_runtime	= 0;
 	p->se.nr_migrations		= 0;
 	p->se.vruntime			= 0;
+	p->se.avg.decay_count		= 0;
 	INIT_LIST_HEAD(&p->se.group_node);
 
 #ifdef CONFIG_SCHEDSTATS
kernel/sched/fair.c +0 -1
index df2cdf7..5f3b5a7 100644
@@ -676,7 +676,6 @@ void init_task_runnable_average(struct task_struct *p)
 {
 	u32 slice;
 
-	p->se.avg.decay_count = 0;
 	slice = sched_slice(task_cfs_rq(p), &p->se) >> 10;
 	p->se.avg.runnable_avg_sum = slice;
 	p->se.avg.runnable_avg_period = slice;
reviewer Reviewed-by on v1 1/1
#1 lore raw
Quoted 9 lines · Kirill Tkhai

Child has the same decay_count as parent. If it's not zero, we add it to parent's cfs_rq->removed_load:

wake_up_new_task()->set_task_cpu()->migrate_task_rq_fair().

Child's load is a just garbade after copying of parent, it hasn't been on cfs_rq yet, and it must not be added to cfs_rq::removed_load in migrate_task_rq_fair().

Quoted 5 lines

The patch moves sched_entity::avg::decay_count intialization in sched_fork(). So, migrate_task_rq_fair() does not change removed_load.

Signed-off-by Kirill Tkhai <[email protected]>
Reviewed-by Ben Segall <[email protected]>
Quoted 37 lines
---
 kernel/sched/core.c |    1 +
 kernel/sched/fair.c |    1 -
 2 files changed, 1 insertion(+), 1 deletion(-)
kernel/sched/core.c +1 -0
index bb398c0..2894b69 100644
@@ -1832,6 +1832,7 @@ static void __sched_fork(unsigned long clone_flags, struct task_struct *p)
 	p->se.prev_sum_exec_runtime	= 0;
 	p->se.nr_migrations		= 0;
 	p->se.vruntime			= 0;
+	p->se.avg.decay_count		= 0;
 	INIT_LIST_HEAD(&p->se.group_node);
 
 #ifdef CONFIG_SCHEDSTATS
kernel/sched/fair.c +0 -2
index df2cdf7..5f3b5a7 100644
@@ -676,7 +676,6 @@ void init_task_runnable_average(struct task_struct *p)
 {
 	u32 slice;
 
-	p->se.avg.decay_count = 0;
 	slice = sched_slice(task_cfs_rq(p), &p->se) >> 10;
 	p->se.avg.runnable_avg_sum = slice;
 	p->se.avg.runnable_avg_period = slice;
 
 
 
--

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

merge bot
#2 lore raw

[tip:sched/core] sched/fair: Fix sched_entity::avg:: decay_count initialization

In reply to Kirill Tkhai
sched/fair: Fix sched_entity::avg::decay_count initialization
Commit-ID:  bb04159df99fa353d0fb524574aca03ce2c6515b
Gitweb:     http://git.kernel.org/tip/bb04159df99fa353d0fb524574aca03ce2c6515b
Author:     Kirill Tkhai <[email protected]>
AuthorDate: Mon, 15 Dec 2014 14:56:58 +0300
Committer:  Ingo Molnar <[email protected]>
CommitDate: Wed, 14 Jan 2015 13:34:16 +0100

sched/fair: Fix sched_entity::avg::decay_count initialization

Child has the same decay_count as parent. If it's not zero, we add it to parent's cfs_rq->removed_load:

wake_up_new_task()->set_task_cpu()->migrate_task_rq_fair().

Child's load is a just garbade after copying of parent, it hasn't been on cfs_rq yet, and it must not be added to cfs_rq::removed_load in migrate_task_rq_fair().

The patch moves sched_entity::avg::decay_count intialization in sched_fork(). So, migrate_task_rq_fair() does not change removed_load.

Signed-off-by Kirill Tkhai <[email protected]>
Signed-off-by Peter Zijlstra (Intel) <[email protected]>
Reviewed-by Ben Segall <[email protected]>
Cc Linus Torvalds <[email protected]>
Link http://lkml.kernel.org/r/1418644618.6074.13.camel@tkhai open here
Signed-off-by Ingo Molnar <[email protected]>
---
 kernel/sched/core.c | 3 +++
 kernel/sched/fair.c | 1 -
 2 files changed, 3 insertions(+), 1 deletion(-)
kernel/sched/core.c +3 -0
index 816c172..95ac795 100644
@@ -1832,6 +1832,9 @@ static void __sched_fork(unsigned long clone_flags, struct task_struct *p)
 	p->se.prev_sum_exec_runtime	= 0;
 	p->se.nr_migrations		= 0;
 	p->se.vruntime			= 0;
+#ifdef CONFIG_SMP
+	p->se.avg.decay_count		= 0;
+#endif
 	INIT_LIST_HEAD(&p->se.group_node);
 
 #ifdef CONFIG_SCHEDSTATS
kernel/sched/fair.c +0 -1
index 97000a9..2a0b302 100644
@@ -676,7 +676,6 @@ void init_task_runnable_average(struct task_struct *p)
 {
 	u32 slice;
 
-	p->se.avg.decay_count = 0;
 	slice = sched_slice(task_cfs_rq(p), &p->se) >> 10;
 	p->se.avg.runnable_avg_sum = slice;
 	p->se.avg.runnable_avg_period = slice;