[PATCH]sched/stop_task: Do not account zero delta_exec in put_prev_task_stop

[PATCH]sched/stop_task: Do not account zero delta_exec in put_prev_task_stop

[Kirill Tkhai]

Simple return

Signed-off-by: Kirill V Tkhai <tkhai@yandex.ru>
CC: Steven Rostedt <rostedt@goodmis.org>
CC: Ingo Molnar <mingo@kernel.org>
CC: Peter Zijlstra <peterz@infradead.org>
CC: linux-rt-users <linux-rt-users@vger.kernel.org>
---
 kernel/sched/stop_task.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/kernel/sched/stop_task.c b/kernel/sched/stop_task.c
index da5eb5b..78d0458 100644
--- a/kernel/sched/stop_task.c
+++ b/kernel/sched/stop_task.c
@@ -58,8 +58,8 @@ static void put_prev_task_stop(struct rq *rq, struct task_struct *prev)
 	u64 delta_exec;
 
 	delta_exec = rq->clock_task - curr->se.exec_start;
-	if (unlikely((s64)delta_exec < 0))
-		delta_exec = 0;
+	if (unlikely((s64)delta_exec <= 0))
+		return;
 
 	schedstat_set(curr->se.statistics.exec_max,
 			max(curr->se.statistics.exec_max, delta_exec));

Re: [PATCH]sched/stop_task: Do not account zero delta_exec in put_prev_task_stop

[Namhyung Kim]

Hi Kirill,

On Tue, 12 Feb 2013 01:40:04 +0400, Kirill Tkhai wrote:
> Simple return
>
> Signed-off-by: Kirill V Tkhai <tkhai@yandex.ru>
> CC: Steven Rostedt <rostedt@goodmis.org>
> CC: Ingo Molnar <mingo@kernel.org>
> CC: Peter Zijlstra <peterz@infradead.org>
> CC: linux-rt-users <linux-rt-users@vger.kernel.org>
> ---
>  kernel/sched/stop_task.c |    4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> diff --git a/kernel/sched/stop_task.c b/kernel/sched/stop_task.c
> index da5eb5b..78d0458 100644
> --- a/kernel/sched/stop_task.c
> +++ b/kernel/sched/stop_task.c
> @@ -58,8 +58,8 @@ static void put_prev_task_stop(struct rq *rq, struct task_struct *prev)
>  	u64 delta_exec;
>  
>  	delta_exec = rq->clock_task - curr->se.exec_start;
> -	if (unlikely((s64)delta_exec < 0))
> -		delta_exec = 0;
> +	if (unlikely((s64)delta_exec <= 0))
> +		return;

It seems it needs to update curr->se.exec_start then.

Thanks,
Namhyung

>  
>  	schedstat_set(curr->se.statistics.exec_max,
>  			max(curr->se.statistics.exec_max, delta_exec));
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH]sched/stop_task: Do not account zero delta_exec in put_prev_task_stop

[Steven Rostedt]

On Tue, 2013-02-12 at 01:40 +0400, Kirill Tkhai wrote:
> Simple return
> 

I'll ack the patch, but I nack the change log. Please write something up
that's a little more descriptive.

Thanks,

-- Steve

> Signed-off-by: Kirill V Tkhai <tkhai@yandex.ru>
> CC: Steven Rostedt <rostedt@goodmis.org>
> CC: Ingo Molnar <mingo@kernel.org>
> CC: Peter Zijlstra <peterz@infradead.org>
> CC: linux-rt-users <linux-rt-users@vger.kernel.org>
> ---
>  kernel/sched/stop_task.c |    4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> diff --git a/kernel/sched/stop_task.c b/kernel/sched/stop_task.c
> index da5eb5b..78d0458 100644
> --- a/kernel/sched/stop_task.c
> +++ b/kernel/sched/stop_task.c
> @@ -58,8 +58,8 @@ static void put_prev_task_stop(struct rq *rq, struct task_struct *prev)
>  	u64 delta_exec;
>  
>  	delta_exec = rq->clock_task - curr->se.exec_start;
> -	if (unlikely((s64)delta_exec < 0))
> -		delta_exec = 0;
> +	if (unlikely((s64)delta_exec <= 0))
> +		return;
>  
>  	schedstat_set(curr->se.statistics.exec_max,
>  			max(curr->se.statistics.exec_max, delta_exec));

Re: [PATCH]sched/stop_task: Do not account zero delta_exec in put_prev_task_stop

[Steven Rostedt]

On Wed, 2013-02-13 at 15:15 +0900, Namhyung Kim wrote:
> ---
> >  kernel/sched/stop_task.c |    4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > diff --git a/kernel/sched/stop_task.c b/kernel/sched/stop_task.c
> > index da5eb5b..78d0458 100644
> > --- a/kernel/sched/stop_task.c
> > +++ b/kernel/sched/stop_task.c
> > @@ -58,8 +58,8 @@ static void put_prev_task_stop(struct rq *rq, struct task_struct *prev)
> >  	u64 delta_exec;
> >  
> >  	delta_exec = rq->clock_task - curr->se.exec_start;
> > -	if (unlikely((s64)delta_exec < 0))
> > -		delta_exec = 0;
> > +	if (unlikely((s64)delta_exec <= 0))
> > +		return;
> 
> It seems it needs to update curr->se.exec_start then.
> 

Ah, if it's less than zero, it does.

A better patch would be:

	if (unlikely((s64)delta_exec <= 0)) {
		curr->se.exec_start = rq->clock_task;
		return;
	}

Thanks!

-- Steve