sched/rt: Sum number of all children tasks in hierarhy at rt_nr_running
PATCH 1/4 sched/rt: Sum number of all children tasks in hierarhy at rt_nr_running
{inc,dec}_rt_tasks used to count entities which are directly queued on rt_rq. If an entity was not a task (i.e., it is some queue), its children were not counted.
There is no problem here, but now we want to count number of all tasks which are actually queued under the rt_rq in all the hierarhy (except throttled rt queues).
Empty queues are not able to be queued and all of the places, which use rt_nr_running, just compare it with zero, so we do not break anything here.
--- kernel/sched/rt.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-)
kernel/sched/rt.c +13 -2
@@ -1045,12 +1045,23 @@ void dec_rt_group(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) {}#endif /* CONFIG_RT_GROUP_SCHED */ static inline +unsigned int rt_se_nr_running (struct sched_rt_entity *rt_se) +{ +struct rt_rq *group_rq =group_rt_rq (rt_se); + +if (group_rq) +return group_rq->rt_nr_running; +else +return 1 ; +} + +static inline void inc_rt_tasks (struct sched_rt_entity *rt_se,struct rt_rq *rt_rq) {int prio =rt_se_prio (rt_se);WARN_ON (!rt_prio (prio)); - rt_rq->rt_nr_running++; + rt_rq->rt_nr_running +=rt_se_nr_running (rt_se);inc_rt_prio (rt_rq, prio);inc_rt_migration (rt_se, rt_rq);
@@ -1062,7 +1073,7 @@ void dec_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) {WARN_ON (!rt_prio (rt_se_prio (rt_se)));WARN_ON (!rt_rq->rt_nr_running); - rt_rq->rt_nr_running--; + rt_rq->rt_nr_running -=rt_se_nr_running (rt_se);dec_rt_prio (rt_rq,rt_se_prio (rt_se));dec_rt_migration (rt_se, rt_rq);
Quoted 3 lines · Kirill Tkhai
{inc,dec}_rt_tasks used to count entities which are directly queued on
rt_rq. If an entity was not a task (i.e., it is some queue), its children were not counted.
Its always the case that a task is queued right, never a sched entity? When a task is queued, the nr_running of every rt_rq in the hierarchy of sched entities which are parents of this task is incremented by 1. Similar is with dequeue of a sched_entity. If the sched_entity has just 1 task on it, then its dequeued from its parent queue and its number decremented for every rq in the hierarchy. But you would never dequeue a sched_entity if it has more than 1 task in it. The granularity of enqueue and dequeue of sched_entities is one task at a time. You can extend this to enqueue and dequeue of a sched_entity only if it has just one task in its queue.
Regards
Preeti U Murthy
Quoted 63 lines
There is no problem here, but now we want to count number of all tasks which are actually queued under the
rt_rqin all the hierarhy (except throttled rt queues).Empty queues are not able to be queued and all of the places, which use
rt_nr_running, just compare it with zero, so we do not break anything here.Signed-off-by Kirill Tkhai <[email protected]>CC Peter Zijlstra <[email protected]>CC Ingo Molnar <[email protected]>--- kernel/sched/rt.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-)kernel/sched/rt.c +13
-3@@ -1045,12 +1045,23 @@ void dec_rt_group(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) {}#endif /* CONFIG_RT_GROUP_SCHED */ static inline +unsigned int rt_se_nr_running (struct sched_rt_entity *rt_se) +{ +struct rt_rq *group_rq =group_rt_rq (rt_se); + +if (group_rq) +return group_rq->rt_nr_running; +else +return 1 ; +} + +static inline void inc_rt_tasks (struct sched_rt_entity *rt_se,struct rt_rq *rt_rq) {int prio =rt_se_prio (rt_se);WARN_ON (!rt_prio (prio)); - rt_rq->rt_nr_running++; + rt_rq->rt_nr_running +=rt_se_nr_running (rt_se);inc_rt_prio (rt_rq, prio);inc_rt_migration (rt_se, rt_rq);@@ -1062,7 +1073,7 @@ void dec_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) {WARN_ON (!rt_prio (rt_se_prio (rt_se)));WARN_ON (!rt_rq->rt_nr_running); - rt_rq->rt_nr_running--; + rt_rq->rt_nr_running -=rt_se_nr_running (rt_se);dec_rt_prio (rt_rq,rt_se_prio (rt_se));dec_rt_migration (rt_se, rt_rq); --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/
18.03.2014, 15:08, "Preeti Murthy" <[email protected]>:
Quoted 7 lines · Kirill Tkhai
Quoted 4 lines
{inc,dec}_rt_tasks used to count entities which are directly queued on
rt_rq. If an entity was not a task (i.e., it is some queue), its children were not counted.Its always the case that a task is queued right, never a sched entity?
With patch applied, when sched entity is group queue, we add number of its child tasks instead of "1".
Quoted 2 lines
When a task is queued, the
nr_runningof everyrt_rqin the hierarchy of sched entities which are parents of this task is incremented by 1.
Only if they had not had a queued task before.
Quoted 7 lines
Similar is with dequeue of a
sched_entity. If thesched_entityhas just 1 task on it, then its dequeued from its parent queue and its number decremented for every rq in the hierarchy. But you would never dequeue asched_entityif it has more than 1 task in it. The granularity of enqueue and dequeue ofsched_entitiesis one task at a time. You can extend this to enqueue and dequeue of asched_entityonly if it has just one task in its queue.
We do not queue entites, which are empty. In __enqueue_rt_entity():
if (group_rq && (rt_rq_throttled(group_rq) || !group_rq->rt_nr_running))
return;
Where do you see a collision? Please explain, if so.
Quoted 62 lines
Regards
Preeti U MurthyQuoted 59 lines
There is no problem here, but now we want to count number of all tasks which are actually queued under the
rt_rqin all the hierarhy (except throttled rt queues).Empty queues are not able to be queued and all of the places, which use
rt_nr_running, just compare it with zero, so we do not break anything here.Signed-off-by Kirill Tkhai <[email protected]>CC Peter Zijlstra <[email protected]>CC Ingo Molnar <[email protected]>---
kernel/sched/rt.c | 15 +++++++++++++--1 file changed, 13 insertions(+), 2 deletions(-)diff --git
a/kernel/sched/rt.cb/kernel/sched/rt.c
index d8cdf16..e4def13 100644
---a/kernel/sched/rt.c
+++b/kernel/sched/rt.c
@@ -1045,12 +1045,23 @@ voiddec_rt_group(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq){} #endif /*CONFIG_RT_GROUP_SCHED*/static inline
+unsigned intrt_se_nr_running(struct sched_rt_entity *rt_se)+{
+ structrt_rq*group_rq=group_rt_rq(rt_se);
+
+ if (group_rq)
+ returngroup_rq->rt_nr_running;
+ else
+ return 1;
+}
+
+static inline
voidinc_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq){
int prio =rt_se_prio(rt_se);WARN_ON(!rt_prio(prio));
- rt_rq->rt_nr_running++;
+ rt_rq->rt_nr_running +=rt_se_nr_running(rt_se);inc_rt_prio(rt_rq, prio);
inc_rt_migration(rt_se,rt_rq);
@@ -1062,7 +1073,7 @@ voiddec_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq){
WARN_ON(!rt_prio(rt_se_prio(rt_se)));WARN_ON(!rt_rq->rt_nr_running);
- rt_rq->rt_nr_running--;
+ rt_rq->rt_nr_running -=rt_se_nr_running(rt_se);dec_rt_prio(rt_rq,
rt_se_prio(rt_se));
dec_rt_migration(rt_se,rt_rq);--
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/
Preeti U Murthy
Reviewed-byQuoted 27 lines · Kirill Tkhai
18.03.2014, 15:08, "Preeti Murthy" <[email protected]>:
Quoted 7 lines · Kirill Tkhai
Quoted 4 lines
{inc,dec}_rt_tasks used to count entities which are directly queued on
rt_rq. If an entity was not a task (i.e., it is some queue), its children were not counted.Its always the case that a task is queued right, never a sched entity?
With patch applied, when sched entity is group queue, we add number of its child tasks instead of "1".
Quoted 2 lines
When a task is queued, the
nr_runningof everyrt_rqin the hierarchy of sched entities which are parents of this task is incremented by 1.Only if they had not had a queued task before.
Quoted 7 lines
Similar is with dequeue of a
sched_entity. If thesched_entityhas just 1 task on it, then its dequeued from its parent queue and its number decremented for every rq in the hierarchy. But you would never dequeue asched_entityif it has more than 1 task in it. The granularity of enqueue and dequeue ofsched_entitiesis one task at a time. You can extend this to enqueue and dequeue of asched_entityonly if it has just one task in its queue.
Quoted 6 lines
We do not queue entites, which are empty. In
__enqueue_rt_entity():if (
group_rq&& (rt_rq_throttled(group_rq)|| !group_rq->rt_nr_running))return;Where do you see a collision? Please explain, if so.
Ok I went through the patchset more closely and I understand it better now. So this patchset targets the throttled rt runqueues specifically. I am sorry I missed this focal point. So this patch looks good to me.
Quoted 64 lines
Quoted 63 lines
Regards
Preeti U MurthyQuoted 59 lines
There is no problem here, but now we want to count number of all tasks which are actually queued under the
rt_rqin all the hierarhy (except throttled rt queues).Empty queues are not able to be queued and all of the places, which use
rt_nr_running, just compare it with zero, so we do not break anything here.Signed-off-by Kirill Tkhai <[email protected]>CC Peter Zijlstra <[email protected]>CC Ingo Molnar <[email protected]>---
kernel/sched/rt.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-)diff --git
a/kernel/sched/rt.cb/kernel/sched/rt.c
index d8cdf16..e4def13 100644
---a/kernel/sched/rt.c
+++b/kernel/sched/rt.c
@@ -1045,12 +1045,23 @@ voiddec_rt_group(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq){}
#endif /*CONFIG_RT_GROUP_SCHED*/static inline
+unsigned intrt_se_nr_running(struct sched_rt_entity *rt_se)+{
+ structrt_rq*group_rq=group_rt_rq(rt_se);
+
+ if (group_rq)
+ returngroup_rq->rt_nr_running;
+ else
+ return 1;
+}
+
+static inline
voidinc_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq){ int prio = rt_se_prio(rt_se); WARN_ON(!rt_prio(prio));- rt_rq->rt_nr_running++;
+rt_rq->rt_nr_running+=rt_se_nr_running(rt_se);inc_rt_prio(rt_rq, prio); inc_rt_migration(rt_se, rt_rq); @@ -1062,7 +1073,7 @@ void dec_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) { WARN_ON(!rt_prio(rt_se_prio(rt_se))); WARN_ON(!rt_rq->rt_nr_running);- rt_rq->rt_nr_running--;
+rt_rq->rt_nr_running-=rt_se_nr_running(rt_se);dec_rt_prio(rt_rq, rt_se_prio(rt_se)); dec_rt_migration(rt_se, rt_rq);--
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/
tip-bot for Kirill Tkhai
merge bot[tip:sched/core] sched/rt: Sum number of all children tasks in hierarhy at ->rt_nr_running
In reply to Kirill Tkhaisched/rt: Sum number of all children tasks in hierarhy at ->rt_nr_running
Commit-ID: 22abdef37cebcdd4933c72339401a174b7d87768 Gitweb: http://git.kernel.org/tip/22abdef37cebcdd4933c72339401a174b7d87768 Author: Kirill Tkhai <[email protected]> AuthorDate: Sat, 15 Mar 2014 02:14:49 +0400 Committer: Ingo Molnar <[email protected]> CommitDate: Fri, 18 Apr 2014 12:07:25 +0200
sched/rt: Sum number of all children tasks in hierarhy at ->rt_nr_running
{inc,dec}_rt_tasks() used to count entities which are directly queued on the rt_rq. If an entity was not a task (i.e., it is some queue), its children were not counted.
There is no problem here, but now we want to count number of all tasks which are actually queued under the rt_rq in all the hierarchy (except throttled rt queues).
Empty queues are not able to be queued and all of the places, which use ->rt_nr_running, just compare it with zero, so we do not break anything here.
--- kernel/sched/rt.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-)
kernel/sched/rt.c +13 -2
@@ -1045,12 +1045,23 @@ void dec_rt_group(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) {}#endif /* CONFIG_RT_GROUP_SCHED */ static inline +unsigned int rt_se_nr_running (struct sched_rt_entity *rt_se) +{ +struct rt_rq *group_rq =group_rt_rq (rt_se); + +if (group_rq) +return group_rq->rt_nr_running; +else +return 1 ; +} + +static inline void inc_rt_tasks (struct sched_rt_entity *rt_se,struct rt_rq *rt_rq) {int prio =rt_se_prio (rt_se);WARN_ON (!rt_prio (prio)); - rt_rq->rt_nr_running++; + rt_rq->rt_nr_running +=rt_se_nr_running (rt_se);inc_rt_prio (rt_rq, prio);inc_rt_migration (rt_se, rt_rq);
@@ -1062,7 +1073,7 @@ void dec_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) {WARN_ON (!rt_prio (rt_se_prio (rt_se)));WARN_ON (!rt_rq->rt_nr_running); - rt_rq->rt_nr_running--; + rt_rq->rt_nr_running -=rt_se_nr_running (rt_se);dec_rt_prio (rt_rq,rt_se_prio (rt_se));dec_rt_migration (rt_se, rt_rq);