Re: [PATCH] sched/rt: Unthrottle the highest RT task of the rq if there are no another available tasks to be picked

2013-02-12 Thread Steven Rostedt
On Tue, 2013-02-12 at 08:06 +0100, Mike Galbraith wrote: > On Tue, 2013-02-12 at 01:31 +0400, Kirill Tkhai wrote: > > It's possible a situation when rq->rt is throttled or > > it has no child entities and there are RT tasks ready > > for execution in the rq which are the only tasks > > of

Re: [PATCH] sched/rt: Unthrottle the highest RT task of the rq if there are no another available tasks to be picked

2013-02-12 Thread Mike Galbraith
On Tue, 2013-02-12 at 09:12 +0100, Stanislav Meduna wrote: > On 12.02.2013 08:06, Mike Galbraith wrote: > > >> In this case pick_next_task takes idle tasks and idle wastes cpu > >> time. > > > That's not a waste of CPU time, that's utilization enforcement the thing > > it is designed to do. >

Re: [PATCH] sched/rt: Unthrottle the highest RT task of the rq if there are no another available tasks to be picked

2013-02-12 Thread Stanislav Meduna
On 12.02.2013 08:06, Mike Galbraith wrote: >> In this case pick_next_task takes idle tasks and idle wastes cpu >> time. > That's not a waste of CPU time, that's utilization enforcement the thing > it is designed to do. Well this is a philosophical question and the opinions will IMHO vary

Re: [PATCH] sched/rt: Unthrottle the highest RT task of the rq if there are no another available tasks to be picked

2013-02-12 Thread Stanislav Meduna
On 12.02.2013 08:06, Mike Galbraith wrote: In this case pick_next_task takes idle tasks and idle wastes cpu time. That's not a waste of CPU time, that's utilization enforcement the thing it is designed to do. Well this is a philosophical question and the opinions will IMHO vary strongly. If

Re: [PATCH] sched/rt: Unthrottle the highest RT task of the rq if there are no another available tasks to be picked

2013-02-12 Thread Mike Galbraith
On Tue, 2013-02-12 at 09:12 +0100, Stanislav Meduna wrote: On 12.02.2013 08:06, Mike Galbraith wrote: In this case pick_next_task takes idle tasks and idle wastes cpu time. That's not a waste of CPU time, that's utilization enforcement the thing it is designed to do. Well this is

Re: [PATCH] sched/rt: Unthrottle the highest RT task of the rq if there are no another available tasks to be picked

2013-02-12 Thread Steven Rostedt
On Tue, 2013-02-12 at 08:06 +0100, Mike Galbraith wrote: On Tue, 2013-02-12 at 01:31 +0400, Kirill Tkhai wrote: It's possible a situation when rq-rt is throttled or it has no child entities and there are RT tasks ready for execution in the rq which are the only tasks of TASK_RUNNING

Re: [PATCH] sched/rt: Unthrottle the highest RT task of the rq if there are no another available tasks to be picked

2013-02-11 Thread Mike Galbraith
On Tue, 2013-02-12 at 01:31 +0400, Kirill Tkhai wrote: > It's possible a situation when rq->rt is throttled or > it has no child entities and there are RT tasks ready > for execution in the rq which are the only tasks > of TASK_RUNNING state. In this case pick_next_task > takes idle tasks and

[PATCH] sched/rt: Unthrottle the highest RT task of the rq if there are no another available tasks to be picked

2013-02-11 Thread Kirill Tkhai
It's possible a situation when rq->rt is throttled or it has no child entities and there are RT tasks ready for execution in the rq which are the only tasks of TASK_RUNNING state. In this case pick_next_task takes idle tasks and idle wastes cpu time. The patch change logic of pre_schedule a

[PATCH] sched/rt: Unthrottle the highest RT task of the rq if there are no another available tasks to be picked

2013-02-11 Thread Kirill Tkhai
It's possible a situation when rq-rt is throttled or it has no child entities and there are RT tasks ready for execution in the rq which are the only tasks of TASK_RUNNING state. In this case pick_next_task takes idle tasks and idle wastes cpu time. The patch change logic of pre_schedule a little

Re: [PATCH] sched/rt: Unthrottle the highest RT task of the rq if there are no another available tasks to be picked

2013-02-11 Thread Mike Galbraith
On Tue, 2013-02-12 at 01:31 +0400, Kirill Tkhai wrote: It's possible a situation when rq-rt is throttled or it has no child entities and there are RT tasks ready for execution in the rq which are the only tasks of TASK_RUNNING state. In this case pick_next_task takes idle tasks and idle